Techno-economic analysis and optimization of ... › smash › get › diva2:1247170 ›...
Transcript of Techno-economic analysis and optimization of ... › smash › get › diva2:1247170 ›...
Techno-economic analysis
and optimization of
electrochemical energy
storage solutions
GUIDO LORENZI
KTH Royal Institute of Technology
Industrial Engineering and Management
Department of Energy Technology
Heat and Power Division
Brinellvägen 68, 100 44, Stockholm,
Sweden
IST Instituto Superior Técnico
Department of Mechanical Engineering
IN+ Center for Innovation, Technology
and Policy Research
Avenida Rovisco Pais 1, 1049-001,
Lisbon, Portugal
Printed by Universitetsservice US-AB Drottning Kristinas väg 53B SE-114 28 Stockholm
Sweden.
TRITA-ITM-AVL 2018:43
ISBN: 978-91-7729-940-0
© Guido Lorenzi, 2018
To my grandparents Maria, Giuseppe and Mara
“On ne reçoit pas la sagesse, il faut la découvrir soi-même après un
trajet que personne ne peut faire pour nous, ne peut nous épargner, car
elle est un point de vue sur les choses.”
“We do not receive wisdom, we must discover it for ourselves, after a
journey through the wilderness which no one else can make for us, which
no one else can spare us, for our wisdom is the point of view from which
we come at last to regard the world.”
“La saggezza non la si riceve, bisogna scoprirla da soli, dopo un
tragitto che nessuno può fare per noi, né può risparmiarci, perché è un
punto di vista sulle cose.”
Marcel Proust
i
Abstract
The need to integrate the rapidly growing share of variable renewable
energy sources in the power sector requires solutions that are capable of
mitigating the intermittent nature of these sources. They are expected to
constitute the backbone of the electricity generation system in the coming
years in order to reach ambitious goals in terms of energy security and
reduction of environmental impact. Energy storage appears a promising
solution to improve the capability of variable renewable plants to meet the
energy demand at all times, mainly through the re-allocation of the
generation surplus.
Among the several options that can favor the integration of variable
renewables, electrochemical storage technologies - batteries and
electrolysis cells - constitute the focus of this dissertation. These
approaches can be used to defer substantial quantities of energy for
medium to long time intervals, are largely location independent, and are
considered a strategic part of the decarbonization pathway in the European
Union, which represents the geo-political framework under investigation.
Battery storage systems are analyzed in both small- and large-scale
settings to quantify the energetic and economic benefits deriving from a
more efficient use of renewable energy. A small-scale battery system
connected to a residential PV plant is analyzed and the results are
compared to a demand response strategy for load shifting. The integration
of a large-scale battery facility in the energy system of an island is also
simulated and the results show a much lower level of renewable energy
curtailment. In both the situations the projected costs of the battery
technologies are used to assess their techno-economic performance.
Solid oxide electrolysis cells (SOECs) as employed in power-to-gas
upgrading of biogas constitute the second electrochemical energy storage
pathway that was studied. The upgrading process sought to increase the
methane content in the biogas by directly converting the embedded carbon
dioxide through high-temperature electrolysis and methanation. This
process showed energy conversion efficiencies higher than 80%. However,
its economic viability depends on the cost of electricity, the cost of the core
components, and the price of natural gas.
ii
Keywords:
Electrical Energy Storage, Battery Energy Storage, Lithium-ion Batteries,
Vanadium Flow Batteries, Power-to-Gas, Electrolysis, Biogas Upgrading
iii
Sammanfattning
Behovet för integrering av förnybar energis ökande andel inom
elkraftssektorn innebär lösningar som kan hantera dessa energikällors
oregelbundenhet. Förnybara energikällor förväntas att bidra till elkrafts
ryggrad inom de kommande åren för att kunna nå de ambitiösa målen
gällande energisäkerhet och förbättrad miljöprestanda. Energilagring är
en lovande teknik för att underlätta förnybara energikällors möjligheter att
tillgodose energibehoven dygnet runt, primärt genom omallokeringen av
överskottsel.
Bland de olika alternativen som kan öka integreringsgraden av
förnybara energikällor är elektrokemiska lagringstekniker – batterier och
elektrolysörer – i fokus i denna avhandling. Dessa enheter kan användas
för att flytta en del av energin inom mellan- och långtidsintervaller, är i
huvud sak oberoende av placeringen och är en strategisk del av EU:s plan
för minskning av koldioxidutsläpp. Det senare omfattar avhandlingens
geopolitiska ramverk.
Batterilagringssystem är analyserade i både små- och storskaliga faller
för att kvantifiera de energimässiga och ekonomiska fördelarna med en
mer effektiv användning av förnybar energi. Ett småskaligt batterisystem
kopplat till villabaserade solceller analyserades och resultaten jämfördes
mot en efterfrågans-baserad strategi. Integrering av en storskalig
batterianläggning för ett energisystem p en ö simulerades och resultaten
visar en lägre nivå av slösad elkraft. I både fallen används projicerade
kostnader för att utvärdera den tekno-ekonomiska presentandan.
Fastoxid elektrolysörer (SOFC) inom ramen av elkraft-till-gas
uppgradering av biogas omfattar det andra energilagringssätt som
studerades. Uppgraderingsprocessens mål var att öka metanhalten hos
biogasen med direktomvandling av koldioxid genom högtemperatur-
elektrolys och -metanisering Denna process visade verkningsgrader för
energiomvandling högre än 80%. Däremot är den ekonomiska
konkurrenskraften en funktion av elkostnaden, kostnaderna hos de olika
komponenterna och gaskostnaden.
Nyckelord
Elenergilagring, Batterilagring, Litiumjonbatterier,
Vanadinflödesbatterier, Elkraft-till-gas, Elektrolys, Biogasuppgradering
iv
v
Resumo
A necessidade de integrar o rápido crescimento dos recursos renováveis
no setor da geração da energia elétrica exige soluções capazes de mitigar a
intermitência destes recursos. As energias renováveis têm o potencial para
se tornar a espinha dorsal da produção elétrica nos próximos anos para
alcançar os objetivos ambiciosos que dizem respeito à segurança energética
e à reducão do impacte ambiental. O armazenamento da energia parece
uma solução promissora para melhorar a capacidade das renováveis
variáveis (eólica e fotovoltaico) de satisfazerem continuamente a procura
de energia elétrica, principalmente através da disponibilização do excesso
de energia gerada.
Entre as diferentes opções que podem favorecer a integração dos
recursos renováveis intermitentes, as tecnologias para o armazenamento
eletroquímico – baterias e células de electrólise – constituem o foco desta
tese. Estas abordagens permitem adiar o uso de grandes quantidades de
energia a médio-longo prazo, são independentes da localização e são, além
do mais, consideradas uma componente estratégica do plano para a
descarbonização na União Europeia – a entidade geo-política sobre a qual
o trabalho se concentra.
Os sistemas de armazenamento com baterias são analisados para casos
de estudo de pequena e grande escala para quantificar os beneficios
energético e económico que derivam de um uso mais eficiente dos recursos
renovaveis. Para a pequena escala apresenta-se a análise de um sistema de
baterias conectado a um sistema fotovoltaico residencial e os resultados
são comparados a uma estratégia de demand response para o
deslocamento da carga. Para a larga escala, simula-se a integração de uma
instalação de baterias no sistema de energia de uma ilha e os resultados
mostram um nível muito menor de curtailments de energia renovável. Em
ambas as situações, usaram-se os custos projetados das tecnologias para
avaliar o seu desempenho técnico-económico.
As células de eletrólise de óxido sólido (Solid Oxide Electrolysis Cells –
SOECs) foram analisadas para um processo power-to-gas mediante o
upgrading de biogás. Esta tecnologia constitui o segundo tipo de
armazenamento eletroquímico de energia que foi estudado. O processo de
upgrading serve para aumentar o conteúdo de metano no biogás,
convertendo diretamente o dióxido de carbono contido nele, por meio de
vi
eletrólise a alta temperatura e metanação. A eficiência de conversão de
energia deste processo mostrou-se superior a 80%. Contudo, a viabilidade
económica deste conceito depende maioritariamente do custo da
eletricidade, do custo dos componentes principais e do preço do gás
natural.
Palavras Chave
Armazenamento de Energia Elétrica, Armazenamento de Energia em
Baterias, Baterias de Iões de Lítio, Baterias de Fluxo de Vanádio, Power-
to-Gas, Eletrólise, Upgrading de Biogás
vii
Preface
This doctoral thesis was completed at the Center for Innovation,
Technology and Policy Research (IN+), Instituto Superior Técnico
(Lisbon), and at the Heat and Power Technology (HPT) Division,
Department of Energy Technology (EGI), KTH Royal Institute of
Technology (Stockholm). The PhD was conducted in the framework of the
SELECT+ Doctoral Program (Environomical Pathways for Sustainable
Energy Services), financed by the EACEA (Education, Audiovisual and
Culture Executive Agency). The main supervisors are Prof. Carlos Augusto
Santos Silva (IST) and Andrew Read Martin (KTH). The supervision team
also includes a Co-supervisor, Prof. Massimo Santarelli, affiliated
professor at KTH. The research at IN+ focused mainly on the economic
assessment of battery energy storage solutions for small- and large-scale
applications. At HPT the evaluation of the techno-economic performance
of power-to-gas pathways was performed.
The topic of electrical energy storage is addressed both in general terms
and in connection with some practical applications, highlighting the
potential and the barriers of the analyzed solutions. The main motivation
of such work is to provide a reasonable estimation of the economic
performance that can be expected by storage systems in a medium-term
scenario. The outcomes of the investigation aim at quantifying the progress
in terms of efficiency and cost reduction that storage systems should
achieve before being considered economically viable as low-risk
investments.
The present document is a paper-compilation thesis. The thesis
presents a synthesis of the motivation, the methodology and the key-
findings published in the form of research papers in several international
journals. These papers can be found in Appendix B with the same order as
they are referred to in the dissertation.
Stockholm, October 2018
Guido Lorenzi
viii
ix
List of appended publications
This thesis is based upon the following appended scientific articles. The
contribution of each author is summarized after the paper list.
Paper I
Guido Lorenzi, Carlos Augusto Santos Silva (2015), “Techno-
economic comparison of storage vs. demand response strategies in
distributed generation systems”, Proceeding of the 12th International
Conference on the European Energy Market (Lisbon 20th – 22nd May 2015).
Paper II
Guido Lorenzi, Carlos Augusto Santos Silva (2016), “Comparing
demand response and battery storage to optimize self-consumption in PV
systems”, Applied Energy, 2016, 180, 524-535.
Paper III
Guido Lorenzi, Ricardo da Silva Vieira, Carlos Augusto Santos Silva
and Andrew Martin (2018), “Techno-economic analysis of utility-scale
energy storage in island settings”, submitted to Journal of Energy Storage
Paper IV
Guido Lorenzi, Andrea Lanzini, and Massimo Santarelli (2015),
“Digester gas upgrading to synthetic natural gas in solid oxide
electrolysis cells”, Energy & Fuels, 2015, 29 (3), 1641-1652
Paper V
Guido Lorenzi, Andrea Lanzini, Massimo Santarelli and Andrew
Martin (2017), “Exergo-economic analysis of direct biogas upgrading
process to synthetic natural gas via integrated high-temperature
electrolysis and methanation”, Energy, 2017, 141, 1524-1537
x
Other publications not included
Paper A
Guido Lorenzi, Andrea Lanzini, Massimo Santarelli, Carlos Augusto
Santos Silva (2015), “Thermoeconomic analysis of a biogas upgrading
process in solid oxide electrolysis cells (SOEC)”, Proceeding of ECOS 2015
- The 28th International Conference on Efficiency, Cost, Optimization,
Simulation and Environmental Impact of Energy Systems (Pau 29th June
– 3rd July 2015).
Paper B
Lucio De Fusco, Guido Lorenzi, Hervé Jeanmart (2016), “Insight into
electric utility business models for high-share renewables and storage
integration”, Proceeding of the 13th International Conference on the
European Energy Market (Porto 6th – 9th June 2016).
Paper C
Guido Lorenzi, Patrícia Baptista (2018), “Promotion of renewable
energy sources in the Portuguese transport sector: A Scenario analysis”,
Journal of Cleaner Production, 2018, 186, 918-932.
Paper D
Guido Lorenzi, Marco Gorgoroni, Carlos Augusto Santos Silva, Massimo
Santarelli (2018), “Life Cycle Assessment of Biogas Upgrading Routes”,
10th International Conference on Applied Energy 2018 (Hong Kong 22nd –
25th August 2018)
Paper E
Guido Lorenzi, Leopoldo Mignini, Baldassarre Venezia, Carlos
Augusto Santos Silva, Massimo Santarelli (2018), “Integration of High‐Temperature Electrolysis in an HVO Production Process Using Waste
Vegetable Oil”, 10th International Conference on Applied Energy 2018
(Hong Kong 22nd – 25th August 2018)
xi
Contributions to the appended papers
In Paper I and II the model optimization and the choice of the
technologies to compare were developed by the author of this thesis. The
same author was also in charge of the manuscript drafting and of the
presentation of the results. Paper I was publicly presented to the 12th
International Conference on the European Energy Market, held in Lisbon
on May 20th-22nd 2015. Co-author C. Silva was involved in the definition of
the topic and of the methodology, and in the data collection. He also proof-
read the papers and participated in the review process.
In Paper III the development of the unit commitment model, its
implementation and the data collection were performed by the author of
this thesis. Similarly, the manuscript was also written by the author of this
thesis. Co-author Ricardo Vieira was involved in the definition of the
assumptions to describe the energy system and provided useful insights for
the environmental analysis of the power system. Co-author C. Silva
supervised the definition of the model for storage. Co-author Andrew
Martin contributed to highlight the relevance and the originality of the
selected case study.
In Paper IV the model development and implementation, as well as the
analyses, were performed by the author of this thesis. Similarly, all sections
of the paper were written by the author of this thesis. Co-author A. Lanzini
was involved in the study by providing relevant literature, useful
suggestions for the model definition, and by participating in the review
process. Co-author M. Santarelli gave his support along the study by
providing feedback on the structure of the paper and the comparison with
other technologies.
In Paper V the model was adapted from Paper IV and complemented
with the design of the heat exchange network that was necessary to satisfy
the thermal needs of the plant with the heat produced by the process itself.
The calculations for the exergo-economic analysis were done by the author
of this thesis. Similarly, the estimations of the benefits deriving from the
accounting of the saved CO2 and from the economic exploitation of oxygen
were performed by the same author. All sections of the paper were written
by the author of this thesis. Co-author A. Lanzini was involved in the
determination of the cost of the components, especially the sulfur polisher.
Co-author M. Santarelli supported the work with his expertise in exergo-
economic analysis. Co-author A. Martin gave useful insights concerning
xii
the structure of the paper, by improving its clarity and readability. All the
co-authors participated in the discussion during the paper review to
formulate responses to the reviewers.
xiii
Acknowledgements
I would like to thank my supervisors for their inspiring support during
the course of my PhD. In particular, prof. Carlos Silva, who was with me
since the first day of this adventure, has always encouraged me to pursue
my research ideas and has taught me that people are more important than
deadlines. The precious guidance and the fundamental contribution to the
review of the doctoral dissertation of prof. A. Martin are gratefully
acknowledged, as well as the support that I received from prof. M.
Santarelli through his scientific supervision.
I would like to acknowledge the Erasmus Mundus Joint Doctoral
Programme SELECT+ and the VULCANO Project that have financed the
research that I conducted. I am really grateful to Chamindie that has
always done her best to help me overcome the numerous difficulties of a
double degree program. I would also like to acknowledge the Innoenergy
PhD School for their financial support and for all the opportunities of
professional growth that they offered me.
I want to express my sincere gratitude to Ana Barbosa. Her support in
practical and administrative issues, and her positive attitude helped me get
through the numerous daily problems that were continuously popping up
during my first months in Portugal.
A heartfelt thanks goes to my whole family because during these four
years spent away from home they have never stopped providing me their
full practical and moral support. Un grazie speciale ai miei genitori che mi
han sempre sostenuto nelle difficoltà di questo percorso e si sono sempre
spesi per aiutarmi coi miei problemi, malgrado la lontanza. A mio fratello
Giorgio che mi ha spinto a dare sempre il massimo. Ai miei nonni che mi
hanno sempre pensato in questo periodo di vita all’estero e a cui, quasi
certamente, sono mancato di più. Un ricordo particolare per lo zio Hassan
che avrebbe sfruttato volentieri la possibilità di venirmi a trovare nelle mie
nuove città.
A special thought is reserved to the friends of the ACcidenti group that
helped me keep a bond with my home town, giving me a reason to go back
home for their weddings.
I would like to thank my friend Marco most profoundly. He was with
me in Lisbon when this wonderful story began and shared with me some
unforgettable experiences and an amazing piece of life. A cherish memory
goes also to Cecy and Elisa.
xiv
Thanks to Lucio for his friendship and for the important collaborations
that we have had during these years of PhD.
Thanks to Sérgio and Teresa that have been perfect flat mates in the
charming Lisbon.
Thanks to Damiano and Andrea for their friendship was one of the
reasons why I never suffered too much the Swedish winter.
Thanks to the friends of the Missione Italiana a Stoccolma and
especially to don Furio that has made me feel at home. Thanks also to
Giovanni and Silvia who opened the doors of their house to me, and
everybody knows how expensive the houses in Stockholm are.
Thanks to my colleagues at IN+ and in particular to João, Nathália,
José, Patrícia, Felix and Paulo for their cheerfulness and their fundamental
language support.
Thanks to my colleagues at EGI that made my year at KTH joyful and
pleasant. Thanks to Monica, José, Mauricio, Rafael, Monika, Justin, Jorge,
Tobias and Hanna.
Thanks to the STEPS research group at Polito and especially to Andrea,
Marta, Davide and Domenico for their insightful scientific suggestions.
Finally, an infinite thanks to Irene who has shared every day of
enthusiasm, joy and satisfaction, but also every moment of sadness,
downheartedness and difficulty that I have had during this incredible
experience. Without you this journey would have been much more solitary
and tasteless.
xv
Nomenclature
Abbreviations
AC Alternate Current
BEC Bare Erected Cost
BES Battery Energy Storage
CAES Compressed Air Energy Storage
CAPEX Capital Expenditure
CC Cycle Cost
CEPCI Chemical Engineering Plant Cost Index
CF Capacity Factor
DC Direct Current
DoE Department of Energy
DHW Domestic Hot Water
DPB Discounted Payback
DR Demand Response
DSO Distribution System Operator
ERSE Entidade Reguladora dos Serviços Energéticos (Energy Services
Regulation Entity)
FIT Feed-in Tariff
HP High Pressure
GHG Greenhouse Gas
INV Investment
LHV Lower Heating Value
LiB Lithium-ion Battery
LP Low Pressure
MS Membrane Separation
NG Natural Gas
NO STO System without any storage device or flexibility measure
NPV Net Present Value
O&M Operation and Maintenance
OPEX Operational Expenditure
P2G Power-to-Gas
PHES Pumped Hydroelectric Energy Storage
PSA Pressure Swing Adsorption
PV Photovoltaic
ROI Return on Investment
SNG Synthetic Natural Gas
SOEC Solid Oxide Fuel Cell
SP Specific Productivity
TASC Total As Spent Cost
xvi
TES Thermal Energy Storage
TPC Total Plant Cost
TSO Transmission System Operator
USD United States Dollar
VFB Vanadium Flow Battery
W2E Waste-to-Energy
WS Water Scrubbing
Latin Symbols
�̇� Exergy Flow [W] C Cost [€] d Discount Rate [-] E Energy [J] G Gibbs Free Energy [J] F Faraday Constant [C/mol] H Enthalpy [J]
�̇� Enthalpy Flow [W] L Methane Loss [-] �̇� Mass Flow Rate [kg/s] n Number of electrons involved in the reaction [-] N Plant Lifetime [y] P Price [€] S Entropy [J/K] T Temperature [K] V Voltage [V] W Electrical Power [W] Y Molar Fraction [-] Z Exergo-Economic Cost [€/s]
Greek Symbols
Δ Delta/Difference η Efficiency [-]
𝜃𝑠 Sulfur Coverage [-] Ф Thermal Power [W]
Subscripts
cap Capacity ch Charge CoE Co-Electrolysis disch Discharge el Electric elect Electrolysis ex Exergy exec Exergo-Economic F Fuel
xvii
in Inlet out Outlet OVN Overnight pl Plant RG Raw Gas sh,in Shifted-in sh,out Shifted-out sep Separation sp Specific STO Storage TN Thermoneutral
xviii
xix
Contents
Abstract......................................................................................... i
Sammanfattning ......................................................................... iii
Resumo ........................................................................................ v
Preface ....................................................................................... vii
List of appended publications ................................................... ix
Other publications not included ................................................. x
Contributions to the appended papers ..................................... xi
Acknowledgements .................................................................. xiii
Nomenclature ............................................................................ xv
Contents .................................................................................... xix
1. Introduction ............................................................................. 1
1.1. Objectives and Methodology ............................................................... 6
1.2. Research questions .............................................................................. 7
1.3. Limitations ............................................................................................. 8
1.4. Structure of the thesis .......................................................................... 9
2. Context and literature review................................................ 11
2.1. Classification of energy storage pathways ...................................... 12
Battery Energy Storage ........................................................................... 14
Electrolysis cells for chemical energy storage ......................................... 16
2.2. Battery energy storage: literature review ......................................... 21
Battery storage in residential PV systems ............................................... 21
Battery energy storage in island settings ................................................ 24
xx
2.3. Power-to-gas: literature review ......................................................... 26
Carbon sources for Power-to-gas............................................................ 27
Biogas upgrading .................................................................................... 28
2.4. Summary of the main contributions to the state of the art ............. 31
3. Optimization of electricity dispatch with battery storage
systems (Papers I, II and III) ..................................................... 33
3.1. Battery energy storage vs. demand response in residential PV
plants (Papers I and II) ................................................................................ 33
Preliminary study: Boolean logic approach (Paper I)............................... 34
In-depth study: optimization approach (Paper II) ..................................... 39
Assumptions for storage and demand response operations ................... 42
Energy analysis results ........................................................................... 45
Economic analysis results ....................................................................... 46
3.2. Battery energy storage to promote renewable integration and grid
support: the case of Terceira (Paper III) ................................................... 49
Case study description and main assumptions ....................................... 50
Energy and environmental results ........................................................... 56
Investment analysis ................................................................................. 57
Sensitivity analysis .................................................................................. 60
4. Techno-economic analysis of a biogas upgrading process
via electrolysis and methanation (Papers IV and V) ............... 63
4.1. Process description ............................................................................ 63
Plant layout ............................................................................................. 63
Sulfur effects on SOEC nickel catalyst .................................................... 66
4.2. Energy and exergo-economic analyses............................................ 67
Energy and exergy analyses ................................................................... 67
Assumptions for the economic analysis .................................................. 68
Exergo-economic analysis ...................................................................... 69
4.3. Energy analysis results (Paper IV) .................................................... 70
Comparison with other biogas upgrading routes ..................................... 71
4.4. Exergo-economic analysis results (Paper V) ................................... 75
xxi
Exergetic results ..................................................................................... 75
Economic results ..................................................................................... 77
Exergo-economic results ......................................................................... 79
Economic valorization of the by-products ................................................ 82
Sensitivity analysis .................................................................................. 84
4.5. Economic profitability with net present value ................................. 88
5. Conclusions ........................................................................... 93
5.1. Future Work ......................................................................................... 96
References ................................................................................. 99
Appendix A – Exergo-Economic Analysis ............................. 113
Appendix B - Publications ...................................................... 115
xxii
INTRODUCTION | 1
1. Introduction
The paradigm shift in electricity generation from fossil fuel plants to
renewable alternatives that has taken place during the first two decades of
the third millennium faces challenges in the way electricity is managed.
Increases in renewable energy have been very visible in the European
Union, where the supply of electrical energy from renewable origin
doubled between 2004 and 2016 – passing from 467 to 960 TWh – while
the total electricity generation remained stable at around 3250 TWh [1].
Additionally, in 2016 the annual wind production (311 TWh) almost
reached the same level provided by hydroelectric sources (350 TWh), and
solar PV supply increased more than tenfold up to 110 TWh in the period
2004-2016 [1]. The electricity supply of wind and solar is heavily
influenced by sudden changes in weather conditions (e.g. wind speed and
direction, or cloud cover for solar plants). Therefore, the variable nature of
these sources has introduced more uncertainty on the supply side of the
electricity dispatch, jeopardizing the instantaneous balance between
electricity supply and demand [2,3]. Moreover, the oversupply during
production peaks can lead to energy generation curtailments for lack of
demand, resulting in a waste of low-carbon energy. All these issues can be
handled with the introduction of energy storage systems that are able to
provide a buffer for the fluctuations of electrical energy supply.
The oldest and most widespread technology for electrical energy
storage is pumped hydroelectric energy storage (PHES), a technique that
exploits the reversible conversion of gravitational potential to electrical
energy via water flow between reservoirs at different altitudes. PHES is
mainly used for bulk energy storage, that is the storage of considerable
quantities of energy that can be gradually released when the demand is
high. So far, more than 96% of the storage facilities by installed capacity
operate with this technology [4]. However, the construction of a reservoir
is a capital-intensive and complex process that requires, besides the
selection of an appropriate site, a long approval procedure due to its
considerable impact on the surrounding environment [5]. These
constraints limit the expansion potential of pumped hydro and, for this
2 | INTRODUCTION
reason, it cannot represent the only solution for energy storage. Therefore,
research on alternative storage mechanisms that can help integrate
variable renewable sources and provide bulk storage in the energy system
has received a growing interest.
Storage pathways are typically classified according to the form into
which the energy is stored, as shown in Figure 1. The rate at which energy
can be stored or retrieved is a critical aspect of the particular storage
pathway. Figure 2 (a variant of the Ragone plot [6]) shows this relationship
for the various storage methods. Among the technologies included in
Figure 1, those that can perform similar tasks to PHES (i.e. bulk storage)
are compressed air energy storage (CAES), batteries and flow batteries,
and electrolyzers. Generally, the minimum charge or discharge time at full
power of all these technologies is one hour, which allows for the storage of
considerable quantities of energy. A closer look at the related storage
pathways reveals that electrochemical devices are at the core of all of these
technologies, excluding compressed air storage. On the other hand,
flywheels, capacitors and supercapacitors are options that address
primarily the issues and challenges related to the use of storage for short-
time-scale ancillary services (e.g. frequency regulation and voltage control)
[7]. The application of thermal energy storage (TES) solutions for
electricity management are mainly three: heat storage in form of sensible
heat (usually molten salts) to allow night operation of concentrating solar
plants [4]; production of ice (latent heat) or chilling of water (sensible heat)
to smooth the air conditioning load of residential and industrial premises
[8]; and use of low-cost electricity to produce and store heat in order to
reduce the electrical peak load for space heating and domestic hot water
supply in residential and commercial premises [9–11].
Another critical aspect of linking renewables-based electricity supply to
energy storage is its context. The European Union (EU) serves as a useful
example, since it is one of the few geo-political entities that have set clear
and ambitious goals in terms of decarbonization in the use of energy
connected to human activities. The ambition to “become the world number
one in renewable energy” has been clearly stated by the European
Commission [12]. Among the various energy storage pathways analyzed in
the Strategic Energy Technology plan [13], electrochemical energy storage
technologies are explicitly mentioned among those that are strongly
promoted for a transition towards low-carbon energy systems. The number
of projects carried out worldwide in electrochemical storage (almost
INTRODUCTION | 3
exclusively with batteries) is three times larger than those for PHES. Total
installed capacity is growing fast (from 0.1 to 1.5 GW between 2011 and
2017) [4], although the cumulative power is still one order of magnitude
lower than PHES.
In general, batteries are characterized by modularity, and they have a
reduced impact on the ecosystem compared to the construction of a
reservoir, provided that their disposal is properly managed. This can favor
their global diffusion, also in view of the fact that they can accompany the
growth of renewable installed capacity both for small-scale (at building
level to reduce locally the mismatch between supply and demand) and
large-scale facilities (at utility level to reduce the curtailments of renewable
energy and the risk of supply failure). In particular, battery technologies
are suitable to improve the profitability of residential renewable plants
under grid-parity conditions [14], favoring self-consumption and reducing
the exposure to high electricity prices [15]. This translates in huge
opportunities for market growth [16,17], especially if the energy exported
to the grid is poorly compensated. However, other options are available to
achieve similar results in terms of savings, exploiting the difference in
electricity prices under time-of-use rates. These alternatives go under the
name of demand response (DR) and consist in deferring part of the load to
hours when the electricity price is lower. Quantifying the utility and
convenience of the approach(es) from an end user perspective is
paramount towards finding optimal solutions for small-scale systems.
4 | INTRODUCTION
CLASSIFICATION OF ELECTRICAL
ENERGY STORAGE PATHWAYS
Mechanical ChemicalElectro
chemicalThermal
· Pumped hydro (PHES)
· Compressed air (CAES)
· Flywheels
· Electrolyzers (Hydrogen and Synthetic Natural Gas)
· Batteries · Flow
batteries
· Sensible heat storage
· Latent heat storage
Electrical
· Capacitors· Supercapaci
tors
Figure 1: Classification of the electrical energy storage pathways. Adapted from [7,8].
TES
PHES
Chemical Energy Storage
(Hydrogen, Synthetic Natural Gas,
Higher Hydrocarbons)
Ch
arge
/Dis
char
ge T
ime
Power
1 kW 10 kW 100 kW 1 MW 10 MW 100 MW 1 GW
1 m
s1
s1
min
1 h
ou
r1
day
FOCUS OF THIS WORK
CAES
Flywheels
(Super)
Capacitors
Battery Energy Storage
Figure 2: Map of the energy storage technologies.
In contrast, the presence of multi-MW renewable plants has introduced
additional variability in the electricity supply promoting the need for large-
INTRODUCTION | 5
scale energy storage solutions [18]. Batteries appear to have a large
potential because they do not present significant constraints on the
geographical location, unlike PHES and in-ground CAES [8]. The use of
battery technologies has been investigated in island settings, as part of so-
called hybrid systems [19,20], to mitigate the effects that fluctuations of
electricity supply entail on micro or mini electrical grids. Although some
studies already examined the economic viability of multi-MW battery
facilities in remote energy systems [21], none of them compared two
different technologies in terms of investment profitability.
In addition to the previously mentioned issues, high penetrations of
variable renewables in the power system are likely to produce large
curtailments of renewable electricity due to a lack of demand or
transmission capacity [22–25]. The need for additional flexibility to
accommodate large quantities of energy for a relatively long time is offered
by a second electrochemically-linked pathway, namely chemical energy
storage through electrolysis devices. Chemical energy storage (also known
as power-to-chemicals [26]) refers to the processes and technologies that
use electrical energy for the synthesis of chemical compounds that can be
easily stored and released at a later time. The energy density of the
produced substances is higher 1 than that of other bulk storage options
(pumped hydro, compressed air and batteries) and, for this reason, can
facilitate the storage and the transport of very high quantities of energy
[7,27]. This route is called power-to-gas (P2G)2 if the final product is in
gaseous form. The production of hydrogen through water electrolysis and
of synthetic natural gas through CO or CO2 methanation are the most
studied processes in this category of storage [28]. Then, the produced gas
could be injected into the natural gas grid, acting both as reservoir and as
distribution system for the stored energy. Hydrogen can be directly
produced in an electrolyzer and then used for electricity production in a
fuel cell, in a so-called power-to-power process, like what happens in
pumped hydro and battery energy storage. However, this process is
characterized by a low round-trip efficiency (around 30%) in comparison
with the other competing technologies (generally higher than 70%), and
this constitutes the main obstacle for its adoption [7]. On the other hand,
1 Typical values for energy density per unit volume are: PHES 1-2 Wh/l, CAES 2-6 Wh/l, lithium-ion 200-500 Wh/l, lead-acid 50-90 Wh/l, vanadium flow batteries 25-35 Wh/l, hydrogen 500-3000 Wh/l, other synthetic fuels 500-10000 Wh/l [8]. 2 Power-to-gas can be abbreviated as P2G or PtG. In this thesis the first acronym is preferred.
6 | INTRODUCTION
the synthesis of hydrocarbons requires a substantial feed of carbon
compounds, mainly carbon monoxide and carbon dioxide, which can be
obtained from the processing of another renewable resource, namely
biomass [29,30]. Moreover, if the produced fuels were used for
applications that do not include the reconversion into electricity, this could
be a way to decarbonize other energy sectors. The intersection between the
challenges of energy storage and transport and their possible synergies
could favor reduction of fossil fuel dependence. The promotion of system
integration has been defined as the “core of the low-carbon transition” in a
roundtable organized by the European Commission on these topics [31].
Addressing all of these challenges – i.e. mismatch between supply and
demand, and curtailment reduction – requires significant efforts in
technology development and in integration with existing facilities and
infrastructures. Electrochemical storage pathways would necessitate
considerable investments because their diffusion is currently quite limited.
However, they can be positively affected by other present technological
trends, as well as be integrated with other practices for the reduction of the
energy sector’s environmental impact. Particularly, the price of batteries is
likely to drop thanks to the mass production of electric vehicles [32],
although supply chain risks of materials like Cobalt and Nickel could
reverse the trend, while chemical energy storage could be coupled with
biomass conversion processes to increase their yields [33]. Therefore,
techno-economic analyses are a critical first step towards identifying
efficient combinations of technologies and process layouts.
1.1. Objectives and Methodology
The overall aim of this thesis is to provide specific knowledge on the
opportunities for the adoption of electrochemical energy storage systems
in the transition towards a low-carbon energy framework. The goal is to
assess if these technologies can take a cost-effective advantage of the excess
electrical energy supply that variable renewables are likely to cause.
Specific objectives are linked to case studies examining the following two
lines of investigation:
· Assessment of the profitability of battery storage at residential and
utility scales
· Evaluation of the techno-economic conditions that can make
electrolysis a viable alternative for power-to-gas biogas upgrading
INTRODUCTION | 7
This work intends to show that different electrochemically-linked
energy storage pathways can co-exist in high RES shares energy systems.
The rigorous presentation of opportunities and obstacles for the
conversion of electrical energy into chemical energy, for different scales
and addressing diverse challenges of a low-carbon energy supply,
constitutes the main general novelty of this investigation.
The method of investigation of this thesis focuses on the determination
of the techno-economic performance of a restricted set of storage options,
as identified through a literature study (Chapter 2). Techno-economic
evaluations are based on functional system models taken in the perspective
of mid-term energy planning, meaning that most of the assumptions are
realistic for 2030. Then, the obtained results are compared to alternative
solutions. Various methodological approaches are employed for the
specific case studies in order to fill the gaps pinpointed in the literature
review (Chapter 2):
· Optimization algorithms for modeling electricity dispatch and
evaluating operating and fixed costs for (i) single-family home
equipped with PV panels and buffering systems and (ii) electricity
supply of islands featuring large-scale battery storage.
· Physical and chemical modeling of electrolysis and methanation
processes for analyzing biogas upgrading, accompanied by exergo-
economic analyses.
1.2. Research questions
In alignment with the aim and objectives listed above, a number of
research questions were defined, grouped according to the two main
objectives. The following list contains the main research questions as
related to the appended papers:
Assessment of the profitability of battery storage at residential and utility
scales
1. What is the most cost-effective way of reducing the mismatch
between energy production and demand in a small-scale
residential PV system: battery storage or demand response
through management of the domestic hot water load? (Papers I-
II)
8 | INTRODUCTION
2. What are the benefits that large-scale battery energy storage
systems can create in island settings in terms of reduction of
curtailed energy and fuel cost savings? Do these benefits justify an
investment decision according to common investment evaluation
methods (net present value, payback time and return on
investment)? (Paper III)
Evaluation of the techno-economic conditions that can make electrolysis
a viable alternative for biogas upgrading
3. What is the energy performance of a plant that can upgrade biogas
into synthetic natural gas via high-temperature electrolysis and
methanation, and what is the configuration that produces the best
results? (Paper IV)
4. Is biogas upgrading through electrolysis and methanation
economically profitable? Which are the factors (e.g. cost of
components, cost of utilities, price of natural gas) that have the
most influence on its profitability? (Paper V)
1.3. Limitations
The virtual impossibility to address the numerous applications of all the
energy storage pathways forced the analysis to be limited to some specific
aspects of the integration of variable renewable sources by means of
storage systems. The choice of the geographical context, i.e. the European
Union, influenced the selection of relevant technologies and case studies,
which were defined based on the pertinence to the topic, the possibility to
provide original knowledge and the availability of data. A detailed
description of the state of the art and of the gaps in the literature that led
to the selection of the case studies is reported in Chapter 2. As described in
the Introduction, electrochemical energy storage technologies are
considered strategic by the European Union for the transition towards low-
carbon energy systems [13]. On the other hand, other technologies could
represent a technically viable alternative to electrochemical storage. For
instance, compressed air energy storage (CAES), both in the in-ground and
modular variants is theoretically suitable to defer large quantities of
electrical energy. However, efficiencies lower than 70% [8] and the need
for an appropriate geographical location (especially for in-ground plants)
restrain the CAES expansion potential. These reasons, together with the
INTRODUCTION | 9
limited presence of CAES facilities in Europe3, explain why this technology
is not part of this investigation.
Furthermore, the focus of this dissertation is on storage processes and
technologies having a charge and discharge time framework in the order of
hours. Therefore, this excludes from the scope of this work all the energy
storage technologies that deal with short-time issues and are designed with
high power and low energy capacity (e.g. flywheels, capacitors and
supercapacitors).
Thermal energy storage can be used for power applications mainly in
concentrating solar power plants, in order to allow such plants to produce
continuously also during nighttime [34]. However, this application is
specific for a particular power technology, which is not investigated in this
work. Alternatively, thermal storage solutions can be adopted in buildings
or industrial processes where there is a consumption of electrical energy to
satisfy thermal needs [35]. Although this does not constitute the main
focus of the thesis, this kind of storage has been studied as a form of
demand response in small-scale PV systems, as an alternative to battery
storage (see Chapter 3).
In addition to the production of synthetic natural gas via electrolysis
and methanation, liquid fuels can also be obtained by using products
resulting from electrolysis processes [36]. In particular, the conversion of
CO or CO2 into liquid hydrocarbons, through the addition of hydrogen
(Fischer-Tropsch synthesis) has been investigated as a pathway for
biomass utilization [37,38]. However, these processes would entail higher
thermodynamic losses [39,40] and for this reason they have been
neglected in the dissertation.
1.4. Structure of the thesis
This thesis intends to offer an organized and coherent framework in
which the author’s publications serve as building blocks. The collection of
papers published throughout the course of the doctoral studies is
contained in Appendix B.
Chapter 2 contains the general context of the thesis and the literature
review that highlights the previous works and the gaps that are addressed
3 Two plants are operating in Europe, one in Germany (Kraftwerk Huntorf) and one in Switzerland (ALACAES demonstration plant). Some more plants are present in the US [4].
10 | INTRODUCTION
in this dissertation. It also contains a section with the summary of the main
contributions that the appended papers brought to the state of the art.
Chapter 3 includes the studies about battery energy storage. Section 3.1
reports the techno-economic comparison between battery storage in
alternative to demand response for the integration of a small-scale
residential PV plant in Lisbon (Portugal). Section 3.2 presents the
investment analysis of a large-scale battery plant on the island of Terceira
(Azores), comparing two different technologies (i.e. lithium-ion and
vanadium flow batteries) in terms of net present value and return on
investment.
Chapter 4 describes a power-to-gas process that uses electricity to
upgrade biogas to synthetic natural gas, via electrolysis and methanation.
It contains the description of the process, the layout of the analyzed plant
and presents its energy and economic performance.
Chapter 5 collects the main conclusions of the thesis and the possible
future works to expand the knowledge on the integration of storage
technologies in energy systems.
Appendix A contains a more detailed description of the exergo-
economic analysis methodology, which was used to quantify the costs of
the biogas upgrading plant.
CONTEXT AND LITERATURE REVIEW | 11
2. Context and literature review
The growing environmental concern connected to the use of non-
renewable resources promoted the massive growth of intermittent
renewable energy sources over the last 20 years. This has profoundly
changed the electricity dispatch paradigm that was in place before. Not
only the generation is distributed and the flow is bi-directional, but the
uncertainty has affected also the supply side, in addition to the demand
side. The challenge to increase the penetration of variable renewable
sources, can be won by introducing flexibility measures [41] that can
mitigate the variability attributable to intermittent sources.
Among the European countries, Germany has experienced a
tremendous surge in RES capacity, especially wind and solar (Figure 3),
and this has caused profound changes in the way electrical energy is
managed. In 2016 the renewable content of the German electricity was 32%
of a total supply of more than 590 TWh [1]. Other countries in Europe (e.g.
Austria, Sweden, Portugal) have higher shares of renewables, but they are
characterized by a lower absolute electricity demand.
Figure 3: Evolution of RES in Germany between 2000 and 2016 [42,43].
12 | CONTEXT AND LITERATURE REVIEW
The case of Germany is relevant because it is the country with the
largest power system in Europe and, in spite of the interconnections with
ten countries and a considerable hydro-pumping capacity (6.8 GW in 2016
[44]), has been experiencing problems in integrating the massive amount
of intermittent renewable energy produced by the 56 GW installed power
of wind farms (onshore and offshore) and 43 GW peak power of solar
panels (2017 figures [45]). The most updated data show that the curtailed
renewable energy reached its maximum in 2015 with 4.7 TWh and
decreased in 2016 to 3.7 TWh [46]. The same situation is present, though
with lower figures, in other countries that have to integrate large quantities
of wind energy, such as Ireland, UK and Spain [47]. This is mainly due to
constraints in the transmission lines that connect neighboring countries or
different parts of the same country. The presence of storage facilities in the
areas where renewable plants are more present reduces the need for
investments in new transmission capacity, which are usually more difficult
to realize because they affect a larger area and can encounter the
opposition of the public opinion [48].
2.1. Classification of energy storage pathways
Given the complexity of the electricity generation system, the roles that
storage can play within this technical environment are various. They are
mainly: power quality support, bridging power and energy management4
[49]. They span between multiple time and energy scales (Figure 4) in
order to fulfil the multiplicity of services that electricity generation and
dispatch require.
Power quality support involves the issues related to the minimization of
the deviations between the instantaneous values of frequency, current and
voltage, and their target values or wave forms [50]. The characteristics of
the storage systems that operate in this range are very fast response time
(ms) and discharge time up to approximately 15 minutes [49].
Bridging power is an expression used to indicate the role of storage in
guaranteeing the reliability of the power supply or the possibility to restart
the operation in case of black-out. Some of the services included in this
4 In this context the expression “energy management” characterizes the energy storage processes that offer a continuous energy discharge for a prolonged time [49].
CONTEXT AND LITERATURE REVIEW | 13
group are contingency reserve (spinning and non-spinning), load following
and forecast uncertainty [51].
Energy management encompasses all those actions that are aimed at
using energy for time intervals longer than 30 minutes to perform mainly
renewable energy arbitrage, peak shaving and seasonal storage.
LEGEND
- Frequency stabilization
- Voltage control
- Reactive power control
- Contingency reserve
- Load following
- Forecast uncertainty
- Peak shaving
- Renewable energy arbitrage
- Seasonal storage
Power Quality Support
Bridging Power
Energy Management
Ch
arge
/Dis
char
ge T
ime
Power
1 kW 10 kW 100 kW 1 MW 10 MW 100 MW 1 GW
1 m
s1
s1
min
1 h
ou
r1
day
Figure 4: Applications of energy storage systems in the electric grid. The bottom side of each rectangle indicates the response time, while the top border is the maximum discharge time at full power (adapted from [49,51,52]).
The most common categorization of energy storage solutions is done
with reference to the form into which electrical energy is stored. Several
papers attempt to organize the knowledge around these topics [7,8,49,53]
and they extensively describe the different technologies for energy storage.
A possible taxonomy is the following:
- Electrochemical energy storage systems
- Mechanical energy storage systems
14 | CONTEXT AND LITERATURE REVIEW
- Electrical energy storage
- Thermal energy storage
As explained in the Introduction, only electrochemically-linked energy
storage (namely batteries and electrolyzers) is analyzed in detail, while
more information on the other types can be found in the literature
[7,8,49,53].
Battery Energy Storage
Battery energy storage includes those devices that enable the reversible
conversion between electrical and chemical energy. Batteries and flow
batteries constitute the main categories of devices that belong to this
storage option [53]. Both of them can be employed for bridging power and
energy management [49,54].
A battery system is composed by electrochemical cells that generate
electricity at a particular voltage level or store electricity into chemical form
through reversible electrochemical reactions. The main cell components
are the positive and negative electrodes, where the chemical energy is
stored, and the electrolyte, that is responsible for the flow of ions that
closes the circuit (Figure 5) [8]. If the energy is stored in two redox couples
(electrolyte) dissolved in a solution, and the electrodes have mainly the
function of electrons collectors, then the electrochemical device is called
flow battery. This technology shares the same working principles of the
other batteries, but the main difference resides in the fact that the
electrolyte is stored in a separate tank and it is continuously pumped in the
half chamber where the reaction takes place (Figure 6). This configuration
allows the power and the capacity to be decoupled since the power is
proportional to the number of cells that collect the current, while the
capacity is related to the size of the electrolyte tank.
The distinction among batteries is usually made according to the
electrode/electrolyte materials. The main technologies that are considered
for energy storage are:
- Lead-acid batteries
- Sodium-sulfur batteries
- Lithium-ion batteries
CONTEXT AND LITERATURE REVIEW | 15
a)
ELECTROLYTE
ELECTRODE 1
(ANODE)
ELECTRODE 2
(CATHODE)
LOADe- e-
ions
b)
ELECTROLYTE
ELECTRODE 1
(CATHODE)
ELECTRODE 2
(ANODE)
POWERe- e-
ions
Figure 5: Operation of a storage system during a) charge and b) discharge.
Lead-acid batteries (cathode material PbO2, anode material Pb,
electrolyte sulfuric acid) are the most mature and widely used battery
technology. They are mainly used in conventional cars because of their low
cost and satisfactory round-trip efficiency, while their adoption in utility-
scale applications is limited by their low cycle life (< 2000
charging/discharging cycles) [8,49].
The sodium-sulfur technology (electrode materials molten sodium and
molten sulfur, electrolyte in β-Al2O3) features high energy density and low
idle discharge, although the working temperature is around 600 K, which
increases the operating costs.
Lithium-ion batteries (cathode in Li-oxide, anode in graphite,
electrolyte non-aqueous solution of Li-salts) are the most versatile
technology because of their fast response time (ms), high energy density
(200-500 Wh/L) and high cycle efficiency (up to 97%) [8].
Flow batteries are instead categorized in redox flow batteries and
hybrid flow batteries. The latter have one of the substances participating to
the electrochemical reaction at the solid state. In this thesis only redox flow
batteries are considered because they are the most technologically mature
[55,56] and they have a good potential for large-scale storage applications
[57].
16 | CONTEXT AND LITERATURE REVIEW
V2+
V3+E
lecctr
od
e
ElectrolyteElectrolyte
V4+
V5+M
em
bra
ne
Ele
ctr
od
e
Load
e- e -
H+
Figure 6: Working principle of a Vanadium Flow Battery (discharging phase).
An extensive description of the working principle, the reactions, the
materials and the applications for vanadium flow batteries can be found in
Doetsch and Burfeind [58]. Several articles focus on the optimization of the
design and on the search for new materials [59–61], while only a limited
number on their integration in energy systems [57,62].
Electrolysis cells for chemical energy storage
Chemical energy storage embodies the processes in which electrical
energy is used to produce chemical compounds via electrolysis. The high
energy density of the produced substances is the main reason why they are
used as storage media. The most studied options for storage applications
are: hydrogen, methane and liquid fuels. Other non-hydrocarbons are
alcohols and ammonia [7].
Hydrogen is obtained through water electrolysis, while the other fuels
require the addition of carbon, which is usually available in the form of
CO2. The options to produce hydrocarbons are methanation of syngas and
Fischer-Tropsch synthesis if the resulting compound is methane [63] or a
liquid hydrocarbon [38,64], respectively.
Gaseous compounds can be used directly in fuel cells to produce
electricity or they can be injected into the natural gas grid, which serves as
CONTEXT AND LITERATURE REVIEW | 17
storage capacity. Hydrogen can be tolerated by the gas infrastructure up to
10% on volume basis. However, the presence of a compressed natural gas
filling station or of specific gas turbines and engines restrain this value
under 2% [65]. The synthesis of synthetic natural gas solves the problem
of compatibility with the distribution infrastructure and with the end-use
appliances.
Electrolysis of water is the key process to connect power systems with
the production of gaseous fuels. The main technologies that are available
for this process are: alkaline cells, proton exchange membrane (PEM) cells,
and solid oxide electrolysis cells (SOECs). The first two options have an
operating temperature around 60-100°C and they are referred to as low-
temperature electrolysis. The third one instead works at 600-1000°C and
is considered high-temperature electrolysis. Many review papers describe
the main features of these three technologies, which are summarized in
Table 1. All of them are already commercially available, however, alkaline
cells technology is the most widespread, while solid oxide electrolysis cells
have only been used in research projects so far [66].
The working principle, the materials and the advantages of SOECs are
described in e.g. Laguna-Bercero [67]. In the study the author explains that
the required electrical power is reduced with increasing temperature, since
the Joule effect of the flowing current produces heat that is used in the
water splitting process. The minimum electrical energy contribution for
the electrolysis reaction is expressed through the Gibbs free energy (Δ𝐺),
while 𝑇Δ𝑆 is the heat demand:
Δ𝐺 = Δ𝐻 − 𝑇Δ𝑆 ( 1 )
where Δ𝐻 is the enthalpy change, ΔS is the entropy change and 𝑇 is the
temperature expressed in Kelvin. In Figure 7 the behavior of Δ𝐺 is reported
as a function of temperature and it is observable that the electric need has
a decreasing trend. Thereby, high-temperature operation can decrease the
hydrogen production cost, because a larger portion of the energy is
obtained through Joule effect. This is the reason why these devices tend to
be operated at the thermoneutral potential (𝑉𝑇𝑁 – equation ( 2 )), that is
the condition at which the heat produced by the cell electrical resistance is
equal to the thermal need of the electrolysis reaction ( Δ𝐻𝑓 is the total
energy required for the electrolysis, 𝑛 is the number of electrons involved
in the reaction, i.e. 2 for water electrolysis, and 𝐹 is the Faraday constant).
18 | CONTEXT AND LITERATURE REVIEW
𝑉𝑇𝑁 =Δ𝐻𝑓
𝑛𝐹 ( 2 )
Under this condition the efficiency of the cell 𝜂𝑆𝑂𝐸𝐶 (equation ( 3 )) is 100%
because all the energy flow for the electrolysis (Δ�̇�) is supplied as electrical
power (𝑊𝑒𝑙) and the external heat need (Φ) is zero.
𝜂𝑆𝑂𝐸𝐶 =Δ�̇�
𝑊𝑒𝑙 + Φ ( 3 )
Figure 7: Specific energy demand for water and steam electrolysis. Adapted from [68].
In addition, high-temperature electrolysis devices, and especially
SOECs, are the only ones that can perform co-electrolysis of water and
carbon dioxide (i.e. reactions ( 4 ) and ( 5 )) where water and CO2 are both
split on the cells’ active surface. Such a possibility enables the conversion
of CO2 into hydrocarbons, e.g. methane, and is the main reason why this
technology was selected to be studied for power-to-gas applications.
𝐻2𝑂 → 𝐻2 +1
2𝑂2 ( 4 )
𝐶𝑂2 → 𝐶𝑂 +1
2𝑂2 ( 5 )
CONTEXT AND LITERATURE REVIEW | 19
All the considerations on electrical and thermal needs, thermoneutral
voltage and SOEC efficiency are valid for co-electrolysis. The energy
required for the electrolysis of CO2 is reported in Figure 8.
Figure 8: Specific energy demand for H2O and CO2 electrolysis. Adapted from [69].
ΔH
ΔH
20 | CONTEXT AND LITERATURE REVIEW
Table 1: Comparison of the main technologies for electrolysis [66,70].
Alkaline PEM SOEC
Electrolyte Alkaline
solution of KOH or NaOH
Polymeric membrane
(Nafion)
Yittria Stabilized
Zirconia (YSZ) Temperature [°C]
60-80 50-100 600-1000
Voltage [V] 1.8-2.4 1.7-2.0 0.9-1.4 Current density [A/cm2]
0.2-0.3 1-2 0.4-1
Catalyst material
Ni, Co, Fe Pt, Ir Ni
Charge carrier OH- H+ O2- Efficiency [%] 45-70 40-80 70-100 System life span [years]
20-30 20-30 10-20
Operation range [% of nominal capacity]
20-100 5-100 d.n.a.*
Current investment cost [€/kW]
600-2600 400-900 5000-10000
Projected investment cost in 2030 [€/kW]
400-900 300-1300 400-1000
*data not available
CONTEXT AND LITERATURE REVIEW | 21
2.2. Battery energy storage: literature review
The size of variable renewable energy source plants ranges from few kW
to several MW. The small-scale plants are often referred to as distributed
energy sources because they are directly connected to the distribution grid.
These units do not have constant electricity generation profiles and their
widespread presence is likely to introduce problems for the energy
management to maintain the reliability of the service. Therefore, storage
systems should be adopted to mitigate the effect of the mismatch between
supply and demand. The fact that variable renewables are suitable for
small- and large-scale applications, suggests that also storage systems
should follow this classification.
Battery energy storage has acquired an increasingly important role in
electricity management mainly because of its adaptability to small- and
large-scale applications to reduce the negative effects introduced by wind
and solar for distributed and centralized applications. Moreover, the mass
production of electric vehicles has triggered enormous efforts in research
and development that caused the price of batteries to drop dramatically
after 2010. According to a study conducted by McKinsey and Company
[71], between 2010 and 2016 the average price of a battery pack passed
from 1000 to 227 $/kWh. This trend has benefitted also from the
application of battery for stationary application as it is proved by the very
high number of installations around the world [4].
Small- and large-scale battery energy storage have received a growing
attention as evinced by the number of scientific publications (6-fold
increase between 2010 and 2017)5 and industrial investments [72]. For this
reason, both of these opportunities were assessed under a techno-
economic perspective in relevant case studies as described in Chapter 3.
The next two sections present the state of the art and the research gaps that
oriented the investigation.
Battery storage in residential PV systems
The progressive phase-out or reduction of the incentives for solar
photovoltaic (PV) in different European states has changed the economic
paradigm of small distributed generation systems. The significant decrease
in the cost of PV panels and the increasing retail price of electricity has
5 Based on a search in Scopus [203] with the query “battery energy storage”.
22 | CONTEXT AND LITERATURE REVIEW
made the concept of grid parity true in many European countries [14,73].
This means that, when the PV plant is working, the cost of generating
electricity is equivalent or lower to buying electricity to the grid. Thus, in
those places where it is more favorable to install a PV system to supply
electricity, the economic feasibility should be guaranteed by a high level of
self-consumption and this explains the reduction of direct subsidization
schemes (i.e. feed-in-tariffs).
The power supplied by residential solar PV installations is a function of
the instantaneous solar irradiation that reaches the panels, which is in turn
affected by numerous factors concerning the season, the time of the day
and the meteorological conditions. Therefore, such installations suffer
from the problem of mismatch between production and demand, because
the peak in demand often occurs when the PV production is low (e.g. early
morning and late night). The approaches that can be adopted to mitigate
the discrepancy between the production and the demand patters are
mainly two: (i) equipping the PV installation with a storage means, i.e.
adjusting the production to the demand; or (ii) setting up a load deferral
strategy capable of adapting the demand profile to the renewable energy
availability. The latter option is usually referred to as demand response
(DR). The presence of a flexibility measure can also be used to consume
more energy at times when the price is lower, like in time-of-use tariffs (i.e.
dual tariff, where the energy has a peak and an off-peak price).
A number of studies considered the integration of storage devices with
renewable plants in grid connected systems. Gitizadeh and Fakharzadegan
[74] suggests a procedure for the sizing of batteries that favors their
optimal scheduling. Dufo-López and Bernal-Agustín [75] proposes a
methodology for the profitability assessment of battery storage systems to
reduce consumers’ electricity bills when variable purchase prices of
electricity are in place. This study explicitly considers lead-acid and
lithium-ion batteries, but does not include any renewables-based power
plant. Ratnam et al. [76] quantifies the operational savings deriving from
the addition of a battery system to a residential PV system with time-of-use
electricity prices, without considering the battery technology and the
equipment cost. Mulder et al. [77] studies how the increasing price of
electricity along the years impacts on the economic attractiveness of
different sizes of the storage. Nottrott et al. [78] calculates the net present
value for the investment of different sizes of lithium-ion batteries, in
connection with their ability to reduce the peak load consumption of a large
CONTEXT AND LITERATURE REVIEW | 23
consumer. Assuming the conditions present in 2011, the study concludes
that only a decrease in the system cost around 55% would improve the
profitability of batteries to an acceptable level.
On the other hand, demand response has been studied primarily from
the perspective of distribution system operators [79,80], with the aim of
reducing dispatch cost and increasing power quality performance. Demand
response strategies at residential level encompass alternative options
which can be subdivided into postponable appliances and buffered
appliances. The first category includes all those appliances (e.g. washing
machines, tumble dryers and dishwashers) that offer flexibility since their
operating cycle can be shifted within a time interval defined by the user.
The second group involves appliances that have a built-in buffer in which
energy is stored (e.g. electric water heaters with a tank). The flexibility
potential of each appliance is determined by the size of the buffer (e.g. the
water storage tank) [81]. The role of voluntary demand response on a
residential aggregate electrical load, under a dual-tariff pricing structure,
is assessed in Muratori et al. [82]. The study finds that DR contributes to
increasing the load level during the night, but creates an even steeper load
peak during the transition between the peak and the off-peak rate,
concluding that the local optimum does not coincide with the global one.
Residential demand response constitutes the focus of a Belgian pilot
project with the scope of quantifying the flexibility potential of domestic
appliances [81,83]. The project defined the potential to modulate the
power consumption of a household as a function of the time of day for
different typologies of appliances. Water heating was found to offer the
highest savings in terms of energy bills since the domestic hot water load
is larger than that of postponable appliances [83].
A few attempts to study the effect of the combination of battery storage
and demand response have been reported. Wang et al. [84] optimizes the
operation of a stand-alone hybrid renewable energy system that includes
PV panels, a wind turbine, a diesel generator, and a battery pack in order
to minimize the operating cost. The possibility to defer loads of particular
postponable appliances (dishwashers and tumble dryers) is also
considered in the study. Huang et. al [85] proposes an optimization
algorithm that minimizes the expenditure for electricity of a house using
both household appliances and the battery of an electric vehicle as
flexibility measures.
24 | CONTEXT AND LITERATURE REVIEW
None of the previous studies performed a comparison between the
operational savings obtained with one flexibility measure against those
obtained with the other, as applied to the same system. Also, the equipment
cost has never been taken into account when quantifying the profitability
of the two options in an energy system directly connected to a renewable
energy plant. Therefore, one part of the present study (Chapter 3) contains
a direct techno-economic comparison between the integration of batteries
and the implementation of a demand response strategy to a small-scale PV
plant. This constitutes the main original contribution of the study. The two
options are ranked according to their potential to reduce the electricity bill,
with limited adjustments to the original equipment.
Battery energy storage in island settings
Battery energy storage systems can provide support to the operation of
the electrical grid by facilitating the integration of intermittent renewable
generators, which are characterized by steep variations in their production
patterns [86]. This in the main reason why grid operators (TSOs and
DSOs)6 have started investigating the possibility of employing batteries to
compensate the fluctuations introduced by utility-scale wind and solar
power plants in order to avoid congestion issues. Their variable behavior
can lead the grid operator to curtail the energy coming from renewable
plants if the demand is not high enough [87]. The problems connected to
intermittent renewable plants are more apparent in remote systems, e.g.
islands, with isolated mini-grids (with tens or hundreds MW of installed
power) [87]. The effects deriving from a significant number of RES
generators are amplified by the small size of the network, hence, the
occurrence of curtailments is higher. Therefore, islands represent a
framework where the economic and environmental benefits of storage
systems are more visible. The replacement of fossil fuels for electricity
generation with less polluting technologies in island settings has been the
object of some review articles (e.g. [88,89]) that affirmed that hybrid
energy systems (i.e. where the integration of renewable and storage
technologies with conventional thermoelectric plants occurs), can improve
the reliability and the environmental impact of the electricity generation
system [90]. In these contexts, energy storage systems are considered
necessary to reach renewable shares above 50% of the total electrical
6 TSO: Transmission System Operator. DSO: Distribution System Operator.
CONTEXT AND LITERATURE REVIEW | 25
energy consumption [88]. Among the available technological options,
batteries appear the preferred alternative if pumped hydro solutions
cannot be implemented [89]. For this reason, economic and environmetal
performance of utility-scale battery energy storage in micro-grids has been
analyzed in several articles [21,91]. These studies considered lithium-ion
batteries as the most suitable technology to support the management of the
electricity grid. In fact, the tasks that battery systems can provide in
support of grid operation range from reserve capacity [92,93] to energy
arbitrage [94]. Although most of the studies analyzed lithium-ion batteries
as an option for energy storage, other alternatives might also be cost-
effective, e.g. vanadium flow batteries (VFB). Vanadium flow batteries are
an interesting technology for large-scale energy storage since they have
long cycle lives (up to 20,000 charge/discharge cycles) [7] and they allow
for the decoupling of the design of power and energy specifications [58].
Moreover, this type of flow batteries has to keep an operating temperature
between -5°C and 40°C to maintain the stability of the electrolyte, making
it suitable for warm climates [95]. Some studies investigated the behavior
of VFB both in connection with small-scale solar PV systems [96,97], and
for utility-scale energy storage [98,99]. However, there is room for
additional studies that assess the profitability of these kinds of solutions in
comparison with other storage technologies in large-scale applications.
A number of research institutes and associations have studied how to
improve the environmental performance of energy systems in islands. In
particular, the National Renewable Energy Laboratory (NREL) analyzed
the possibility to integrate wind and solar plants to decarbonize the power
system of the Hawaiian archipelago [100,101]. The International
Renewable Energy Agency (IRENA) dedicated some studies on the
development of renewable energy and storage in island settings [102,103].
Moreover, several research efforts have been conducted to favor the
integration of renewables in the electricity generation sector in the
archipelago of the Azores [104,105], which is politically an autonomous
region of Portugal.
In this thesis, the investments that characterize the installation of
lithium-ion and vanadium flow batteries in island settings are compared.
The degradation of the battery performance along the years is also
considered in the economic analysis, an aspect that until now has been
omitted in a comparison between large-scale storage projects in islands.
These elements constitute the main original contributions of the study.
26 | CONTEXT AND LITERATURE REVIEW
2.3. Power-to-gas: literature review
The key concept of power-to-gas lies in the possibility to transform the
excess of electricity production – which cannot be dispatched because of a
lack of demand – into a gaseous fuel that could be more easily stored,
transported and used. The obtained gas can be re-converted into electricity
or used for other purposes. The main strength of this approach is the
simplification of the electrical energy management, which would pass from
a paradigm in which an imbalance between production and demand is not
permitted – that is, the two quantities have always to be equal to one
another – to one in which the energy can be produced at one point in time
and easily used at another. Moreover, the high energy density of gaseous
fuels permits a higher quantity of energy to be transported and this has a
crucial role in countries were the centers of consumption are distant from
the bulk of the intermittent generation [21].
The idea of power-to-gas was born in the framework of the hydrogen
economy, that had the objective to utilize hydrogen as the preferred vector
for energy transformation. This is confirmed by Geithner [106] in a review
study of the P2G pilot projects. Until 20127, the majority of the projects
involving electrolysis with low-cost renewable energy were focused on the
production of hydrogen, both for mobility and for electricity generation
applications. Around the first years of the 2010s, a growing interest
towards the production of synthetic fuels through electrolysis of water and
CO2 can be noticed. The study produced by Graves et al. [107] presents the
perspectives and the technologies of CO2-recycling to produce non-fossil
hydrocarbons. Special relevance is given to the pathways that include the
electrolysis of water and CO2 in high-temperature cells and the subsequent
methanation of the produced syngas, to obtain a methane-rich gas.
During the last 7-8 years, the number of studies and pilot projects
involving P2G showed a marked increase. A part of them still focuses on
the production of hydrogen [23,108], while others have broadened the
perspective through the analysis of power-to-methane and power-to-liquid
processes, where the final product is synthetic methane or a liquid fuel
produced mainly via Fischer-Tropsch synthesis [24,37], and the hydrogen
needed for the reactions comes from the electrolysis of water. A new
7 G. Geithner published her paper [106] at the end of 2012 and, therefore, that is considered the time limitation of the study.
CONTEXT AND LITERATURE REVIEW | 27
expression, electrofuels, has been coined [70,109] to indicate the crucial
role that electricity plays in their synthesis.
Carbon sources for Power-to-gas
The availability of cheap or zero-cost electricity is one of the main
factors influencing the economic viability of P2G systems. However, the
production of synthetic hydrocarbons also requires the presence of a
source of carbon to recycle, which is relatively cheap and concentrated. The
possibilities for carbon collection, especially in the form of CO2, are
thoroughly presented in Reiter and Lindorfer [110]. In particular, the
processes are:
· CO2 from combustion in power plants
· CO2 as a by-product in production processes
o CO2 with biologic origin (biogenic)
o CO2 from industrial processes8
· CO2 from the air
The concentration of CO2 influences the ease and the profitability of
recycling carbon. Most of the exhaust streams deriving from the listed
routes have a concentration below 15% vol.; however, in biotechnological
processes (e.g. bioethanol production, biogas upgrading and biomass
fermentation) the streams can reach concentrations higher than 95%. CO2
recovery from combustion offers room for improvement with respect to
capture efficiency and energy intensiveness, which increase the fuel
consumption. This performance handicap can be quantified in terms of
additional emissions of CO2 per quantity of captured CO2. This so-called
CO2 penalty ranges between 142 and 615 gCO2/kgCO2,captured according to the
fuel (natural gas, coal or solid biomass) and the capture technology (Post-
combustion, Pre-combustion, Oxy-combustion) [110]. Fermentation and
biogas upgrading through CO2 separation are able to yield a purge gas
(considered a by-product) with very high CO2 concentrations, that can be
used with minor further treatments and, consequently, very little
8 The biggest anthropologic single source of CO2 is a synthetic liquid fuel production plant (operated by SASOL in Secunda, South Africa) that performs a coal-to-liquid process. However, the number of these plants is very limited in the world and they are not present in Europe. Therefore, they are not explicitly considered.
28 | CONTEXT AND LITERATURE REVIEW
additional energy or monetary expenditures9. Refineries, cement factories,
and steelworks are large CO2 emitters and with appropriate technologies
they could become suitable for carbon capture. The capture efficiency
spans from 59 to 90%, and the CO2 penalty lies in the range 116-487
gCO2/kgCO2,captured.
Finally, the presence of CO2 in the ambient air makes it theoretically
possible to isolate this compound for fuel synthesis. Nevertheless, the
process, usually chemical absorption, is expensive because of the very low
concentration in the atmosphere (around 400 ppm) [111], leading to
energy requirements in the order of 1.5–2.5 kWh per kgCO2 and specific cost
around 150 – 320 € per kgCO2. In spite of the great potential of this concept,
the techno-economic conditions appear prohibitive for a massive diffusion.
Biogas upgrading
Among the possible carbon sources for power-to-gas, the biomass-
related ones appear to be the most cost-effective, as pointed out in Section
2.3.1. In particular, biogas upgrade can be considered one of the most
convenient routes because raw biogas contains a substantial quantity of
CO2 that can be converted into hydrocarbons. Furthermore, biogas is
usually obtained from residues treatment processes, which already
contribute to reduce pollution and to minimize landfilled waste. Therefore,
coupling biogas upgrading processes with power-to-gas applications
results in a carbon-negative environmental impact.
Prior to the diffusion of the power-to-gas concept, biogas had already
gained large attention as a renewable fuel. Biogas is obtained through the
anaerobic digestion of residual organic matter, such as sewage sludge,
manure and agricultural waste. This gaseous mixture is principally
composed of methane (50-70% vol.) and carbon dioxide (30-50% vol.)
[112]. In Europe the production of this energy carrier started to be
incentivized at the beginning of the 21st century in several European
countries such as Germany, Sweden, Austria and Italy [113]. At the
beginning it was mainly used in boilers or CHP plants because the
relatively high content of non-combustible components limited its use as
fuel. Therefore, the removal of CO2 is necessary to turn it into a natural gas
substitute.
9 The only process that might be required is the drying of the CO2-rich stream.
CONTEXT AND LITERATURE REVIEW | 29
Several processes have been developed to perform this operation and
they all focus only on CO2 separation, without paying much attention to its
re-utilization. The most important ones are water scrubbing, membrane
separation and pressure swing adsorption.
Water scrubbing (WS) is the most widespread technology both for the
cleaning and the upgrading of biogas and it exploits the different solubility
that carbon dioxide and methane have in water, according to Henry’s law
[114]. The process takes place in a scrubbing column where the contact
between water and biogas is favored by properly shaped internal packing
elements (i.e. Raschig rings, Berl saddles, pall rings, Lessig rings) [115].
The CO2-rich water is then sent to a stripping column in which an air
stream removes the CO2 and regenerates the water stream.
Membrane separation (MS) processes remove CO2 and other pollutants
from biogas thanks to the selective permeability of porous media. The
materials used for the membranes are usually polymers (e.g. cellulose
acetate and polyimide) [116]. The cleaning of sulfur compounds and water
should be done prior to CO2 separation, in order to prevent corrosion from
degrading the porous medium [114], unless the membrane is specifically
treated to withstand an aggressive environment [117].
Pressure swing adsorption (PSA) is based on the different affinity that
the chemical species present in the biogas have with an adsorbent material,
usually zeolites or activate carbon. In this process, the pressurized biogas
enters a vessel filled with a high surface area material that selectively
adsorbs the compounds that need to be separated, while methane can be
collected from the top. When the adsorbent pack reaches saturation, the
internal pressure drops, the gases desorb and can be removed [118]. The
desorbed gases usually contain a non-negligible amount of methane and
they should be recycled to increase the methane recovery. H2S is
recommended to be removed before the PSA takes place since it is strongly
retained by the absorbent material and interferes with the CO2 separation
[114].
The methane content in the upgraded biogas is usually higher than 95%
in all the aforementioned technologies, while the separated CO2 is usually
vented, without being further utilized.
The conversion of CO2 into synthetic fuels has been extensively
investigated in connection with high-temperature electrolysis devices (e.g.,
solid oxide electrolysis cell (SOEC)) [69,119,120]. The findings allowed
techno-economic analyses of power-to-gas processes using pure CO2
30 | CONTEXT AND LITERATURE REVIEW
streams to be developed [121,122]. These processes included the injection
of pure CO2 into the solid oxide electrolyzer where syngas was produced.
Subsequently, a series of methanation reactors, placed downstream from
the SOEC, transformed hydrogen and carbon monoxide into methane. An
available commercial catalytic process that can be used to obtain this result
is named TREMP (“Topsøe Recycle Energy-efficient Methanation
Process”) [123]. This process allows the methane content in the product
gas (called biomethane or synthetic natural gas) to reach a molar fraction
greater than 90%, with a consequent improvement of the net calorific
value. The resulting product gas has to be compatible with the
specifications required for grid injection: gas gravity, Wobbe Index and
content of non-hydrocarbons. The conditioning is generally performed by
the admixture of a heavier compound (e.g. propane or nitrogen).
The high CO2 content present in biogas (around 40%) that is usually
lost in separation-based upgrading processes could be used as carbon
source for the synthesis of hydrocarbons. Based on this consideration,
Collet et al. [124] analyzed the possibility of upgrading biogas using excess
electricity in a life cycle assessment perspective. However, the processes
that are described in the paper only consider the production of hydrogen
and the subsequent reaction with biogas, instead of feeding biogas directly
to the electrolyzer. The direct upgrading of biogas through co-electrolysis
of water and carbon dioxide avoids the separation of CO2 from CH4 that,
instead, characterizes all the other upgrading techniques. In addition to
that, electrolysis produces high-purity oxygen that is a high value added
substance, with a relevant economic value. The simulation of an upgrading
process where biogas is directly injected in the electrolyzer, its exergy and
economic analyses, as well as the comparison with other upgrading
techniques, constitute the original contributions of this thesis to the state
of the art.
CONTEXT AND LITERATURE REVIEW | 31
2.4. Summary of the main contributions to the state of the
art
The present thesis is structured as a compilation of published articles
and each of them added a piece of knowledge to the state of the art. In the
previous sections a literature review is performed to identify the research
gaps. The original contributions of this dissertation were stated at the end
of sections 2.2.1, 2.2.2, and 2.3.2, and Table 2 links them to the appended
papers.
Table 2: Summary of main contributions of the papers appended in this thesis work
Papers I and II
Research Topic Integration of residential PV systems with cost-effective
flexibility measures
Contributions to the state of
the art
· Comparison between the monetary savings produced
by two buffering options (battery storage or demand
response) separately applied on the same small-scale
grid-connected residential system powered with a PV
plant.
· Quantification of the equipment cost to implement
demand response
Paper III
Research Topic Integration of large-scale battery systems in island
settings
Contributions to the state of
the art
· Comparison between a lithium-ion and a vanadium
flow battery investment in island settings
· Incorporation of battery degradation issues in the
investment analysis of the two large-scale battery
technologies
32 | CONTEXT AND LITERATURE REVIEW
Table 2 continued
Papers IV and V
Research Topic
Thermodynamic and techno-economic analyses of a
biogas upgrading process via electrolysis and
methanation
Contributions to the state of
the art
· Thermodynamic model of the studied process for
biogas upgrading with the calculation of energy and
exergy efficiency of the process for different
configurations
· Comparison of the analyzed plant with other
techniques for biogas upgrading
· Determination of the cost of synthetic natural gas
resulting from the biogas upgrading process through
electrolysis and methanation
· Quantification of the benefits deriving from the
valorization of by-products (e.g. oxygen)
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 33
3. Optimization of electricity dispatch with battery
storage systems (Papers I, II and III)
In this chapter, two possible applications of battery energy storage are
presented. One addresses the reduction of the mismatch between
electricity production and consumption in a residential system equipped
with a small-scale PV plant. The other concerns the installation of a large-
scale battery storage system to integrate large-scale renewable plants in an
island setting. Lithium-ion batteries have been considered in both the
cases due to their versatility, and they have been compared with other
alternatives (i.e. lead-acid and vanadium flow batteries) on the basis of the
economic benefits that they produce. For both the applications the
simulations have been performed using Matlab and its routines.
3.1. Battery energy storage vs. demand response in
residential PV plants (Papers I and II)
The installation of distributed and intermittent RES in residential
contexts, where the electricity load profiles are driven by end-user
behavior, leads to a mismatch between demand and production of
electricity. The solution to this issue is expected to reduce the quantity of
electrical energy exchanged with the grid with positive repercussions on
the electricity costs borne by the end users. Two approaches can be used to
face this challenge: adapting the production profile to the demand (energy
storage), or shaping the demand profile in order for it to be more similar
to the production schedule (demand response). For the purpose of this
study, demand response entails only the alteration of the electric load
connected to domestic hot water (DHW) supply, on the basis of time-of-
use electricity prices and renewable energy availability. The study is
focused on a Portuguese household that is connected to the electricity grid
and buys electricity in a dual-tariff regime.
In Section 3.1.1, key qualitative results are reported based on a
preliminary study. This investigation addressed the reduction of the
mismatch based on the allocation of the surplus of production according to
34 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
a Boolean logic approach, following an “if-then-else” rule with two
mutually exclusive options. Lead-acid batteries were chosen as storage
means, due to their low cost and technological maturity.
Section 3.1.2 presents a linear programming optimization algorithm
that was developed to minimize the cost of the system in a constrained
environment, by taking into account the cost contribution of the
equipment. This method was chosen because all the components of the
physical energy model were assumed to have a linear behavior and because
it guaranteed the existence of an optimal solution. Therefore, the use of
more complex techniques was considered unnecessary.
Lead-acid and lithium-ion batteries were compared in terms of cycle
cost [94]. Lithium-ion appeared to be the most convenient option. Demand
response was implemented through a smart-electric-boiler control, which
is able to heat the water when the electricity is cheaper.
Preliminary study: Boolean logic approach (Paper I)
In the following pages the preliminary techno-economic analysis of a
household with small renewable plants and buffering options is performed.
Two weeks, representative for different periods of the year, are considered
in order to simulate different production levels of the renewables. The
storage plant is constituted by a deck of batteries, while the demand
response consists in the possibility of deferring part of the load.
System layout and general assumptions
The data for the renewable production are obtained by the
meteorological station located in the main campus (Alameda) of Instituto
Superior Técnico (IST - Lisbon) and they have a 1-hour resolution [125].
The main information about the PV panels, wind turbine and battery pack
is reported in Table 3, while the schematic representation of the system is
illustrated in Figure 9. The renewable installations and the batteries
reproduce the micro-generation unit present in the IST campus.
The load profile was derived from the typical consumption patterns for
residential customers available on the Portuguese TSO website [126]. For
the demand response situation, the maximum hourly load that was allowed
to be deferred was equal to 500Wh, equivalent to the minimum load value.
The load could be shifted continuously, as for hot-water boilers and electric
heaters. The dual tariff for electricity is characterized by a peak price of 16
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 35
c€/kWh – active from 8 am to 10 pm – and an off-peak price of 9 c€/kWh
for the remaining hours. These prices are the same for selling and for
purchasing electricity.
Table 3: Summary of the equipment specifications [127].
Equipment
type
Name Number of
units
Total
Power/Capacity
Wind Turbine Southwest Wind-
Power Air 40 1 400 W
PV
modules LDK 235Wp 6 1410 W
Battery
system Autosil EC3 240Ah/5h 3 720Ah/5h
Inverter SMA Sunny Island SI
2224 1 230 V; 2200 W
Figure 9: Layout of the renewable generators, storage system and load [127].
The control logic for the two cases is reported in Figure 10 and Figure 11.
The operating parameters of the batteries are shown in Table 4.
36 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
IFRES POWER
> LOAD POWER
IFSTORED ENERGY
< MAX. STORED
ENERGY
IFSTORAGE POWER
> UNFULFILLED
LOAD
THEN ELSE
INJECT INTO STORAGE
INJECT INTO GRID
WITHDRAW FROM
STORAGE
WITHDRAW FROM GRID
THEN ELSE THEN ELSE
Figure 10: Control logic for the storage case [128].
For the demand response case the residual shifted load, i.e. the load that
has still to be processed at the end of the day, is executed at the midnight
of the new day, by buying electricity from the grid at a constant power of
500W for the number of hours necessary to complete the residual load.
This expedient simulates the reallocation of the loads into the off-peak
hours.
Table 4: Operating parameters of the batteries.
Maximum depth of discharge (DODmax) 80%
Capacity (kWh) 17.28
Maximum charge/discharge power [W] 2200
Charging efficiency (ηch) 0.9
Discharging efficiency (ηdisch) 0.9
Inverter efficiency (ηinv) 0.95
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 37
IFRES POWER
> LOAD POWER
IFSURPLUS POWER
< MAX. SHIFTABLE
LOAD
IFDEFICIT POWER
< MAX. SHIFTABLE
LOAD
THEN ELSE
FEED LOAD FEED LOAD + INJECT INTO
GRIDSHIFT LOAD
SHIFT LOAD + WITHDRAW FROM GRID
THEN ELSE THEN ELSE
Figure 11: Control logic for the demand response case [128].
38 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Qualitative results
The simulations showed that the electricity supply from the renewable
energy generators is below 40% of the weekly load both for winter and
summer conditions. This entails that the majority of the production is
already in phase with the consumption and that the surpluses are very
limited, especially in winter. This fact does not favor the presence of the
batteries, which are seldom charged. Also demand response does not have
a large impact because there are few occasions to reabsorb the shifted load.
For this reason, the system was also studied with a larger PV plant having
a size double than the previous one. The results in this case are favorable
to demand response because this technique is able to reduce the
consumption during peak hours, spreading it over the off-peak ones,
independently of the availability of surplus PV electricity. Instead,
batteries’ operation is restrained by the energy previously stored, which is
directly proportional to the excess of solar energy. In addition to that, the
electricity is sold and purchased at the same price and this fact does not
promote self-consumption. Moreover, batteries are penalized by the round
trip efficiency, which further reduces the amount of stored energy that is
reallocated for consumption.
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 39
In-depth study: optimization approach (Paper II)
The approach adopted in the previous sections is characterized by a set
of limitations that need to be addressed to enlarge the validity of the
results. Firstly, in the preliminary study the strategy for the dispatch of
energy had not been optimized considering the different prices of
electricity. Secondly, the cost of the equipment to mitigate mismatch had
not been estimated. Therefore, a more detailed comparison should be
performed, as described in the next sections.
Case study description
The study was conducted on a typical Portuguese single-family dwelling
in the area of Lisbon with 3 occupants [129], 4 rooms and 3.45 kW of
contracted power, equipped with an electric water boiler. The household
was assumed to be connected to a PV plant.
The load profiles were obtained from the measurement campaign
performed in the framework of the project “Net zero energy school -
Reaching the community”, where residential buildings were characterized
according to their electricity consumption [130].
Four weeks (one per each season) were selected and the load
distribution between peak and off-peak hours is represented in Table 5.
Table 5: Load distribution for the four representative weeks (adapted from [94]).
Winter Spring Summer Autumn
Weekly load consumption [kWh] 111 124 138 123
Peak load share in dual tariff 58% 55% 59% 55%
Off-peak load share in dual tariff 42% 45% 41% 45%
The purchase price of electricity was defined according to a dual-tariff
weekly cycle [131,132], including VAT (23% in Portugal). The selling price
was equal to 90% of the average market price (on the Day Ahead Market)
of the previous month [133] (Table 6).
The solar irradiation data were obtained from the same weather station
that was used for the preliminary study, with the same resolution. These
values were used to determine the electricity production of the PV panels
that provide electricity to the house, using the three-parameter model
described in Crispim et al. [134]. The calibration of such model was done
40 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
with the data reported on the datasheet of the panels (polycrystalline LDK
250P-20 [135]). The three-parameter model leads to satisfactory
estimations10 of the PV production for polycrystalline panels, if datasheet
values are used for its calibration [136]. The analysis was carried out with
three different sizes of PV, namely 1.5, 3 and 4.5 kWp, in order to correlate
the PV installation size with the economic profitability of the proposed
flexibility measures.
Table 6: Selling and purchase prices of electricity. Adapted from [94].
[€/kWh] Purchase price - weekdays 7 a.m. – 0 a.m. 0.234 0 a.m. – 7 a.m. 0.123 Purchase price - Saturday 10 a.m. – 1 p.m. / 6 p.m. – 10 p.m. (winter time); 9 a.m. – 2 p.m. / 8 p.m. – 10 p.m. (summer time)
0.234
Remaining hours 0.123 Purchase price - Sunday All day 0.123 Selling price Winter 0.0427 Spring 0.0388 Summer 0.0493 Autumn 0.0467
The presence of a feed-in-tariff (FIT) scheme in the past led the PV
system to be often oversized in order to maximize the energy production.
Oversized systems are likely to produce a substantial excess of electricity
during low-consumption hours. If this surplus is not used locally, it must
be transferred to the grid, though being poorly compensated (90% of the
wholesale market price for electricity).
Dispatch optimization
The design of the algorithm for the optimal dispatch has been oriented
to reach the cost minimization, which constitutes the objective function
(Equation ( 6 ) for battery storage and ( 7 ) for demand response,
respectively). The decision variables are the quantities of energy crossing
10 Relative error below 5% [136].
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 41
the system borders (corresponding to purchased and sold electricity), and
those relative to the buffering option (charged/discharged energy for
batteries - 𝐸𝑐ℎ /𝐸𝑑𝑖𝑠𝑐ℎ - or energy shifted in/out for demand response -
𝐸𝑠ℎ,𝑖𝑛/𝐸𝑠ℎ,𝑜𝑢𝑡11). The upper and lower boundaries of the previous variables
were established in accordance with the contracted power and the
technical specification of the equipment. Figure 12 shows a block structure
of the system under evaluation.
𝑂𝐶𝐵𝐸𝑆 = ∑((𝑃𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑑𝐸𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑑,𝑖 − 𝑃𝑠𝑜𝑙𝑑𝐸𝑠𝑜𝑙𝑑,𝑖)
24
𝑖=1
+ (𝐶𝐵𝐸𝑆𝐸𝑐ℎ,𝑖))
( 6 )
𝑂𝐶𝐷𝑅 = ∑ ((𝑃𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑑𝐸𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑑,𝑖 − 𝑃𝑠𝑜𝑙𝑑𝐸𝑠𝑜𝑙𝑑,𝑖)
24
𝑖=1
+ (𝐶𝐷𝑅𝐸𝑠ℎ,𝑖𝑛 + 𝐶𝐷𝑅𝐸𝑠ℎ,𝑜𝑢𝑡))
( 7 )
The objective function (OC – operation cost12) contains two groups of
terms: the first one, includes the value of the electricity purchased
(𝐸𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑑) and sold (𝐸𝑠𝑜𝑙𝑑) with the relative prices (𝑃𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑑 and 𝑃𝑠𝑜𝑙𝑑);
while the second one corresponds to the cost of using the equipment (𝐶𝐵𝐸𝑆
for batteries or 𝐶𝐷𝑅 for demand response) and it considers the cost related
to the actual use of the system. Linear programming does not allow the cost
of the equipment to be inputted as a lump sum – which is how the payment
transaction is usually conducted –; therefore, the investment cost should
be spread over the operation of the components, in a levelized-cost-like
approach.
11 The energy shifted-in represents the additional load that is executed when the conditions for consumption are favorable, while the energy shifted-out is the load deferred when the conditions are unfavorable. 12 The expression operation cost groups all the cost components that determine the operation of the system.
42 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Figure 12: Schematic structure of the system with the indication of the energy flows. The buffering option is the battery system or the water tank connected to the boiler controller.
For storage and load deferral systems this concept is named cycle cost
(CC) [137], and it is referred to the number of charge cycles for BES and of
load shifting cycles for DR ( 8 ).
𝐶𝐶 =𝐸𝑞𝑢𝑖𝑝𝑚𝑒𝑛𝑡 𝐶𝑜𝑠𝑡
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑦𝑐𝑙𝑒𝑠 𝑖𝑛 𝑙𝑖𝑓𝑒 ( 8 )
It can be also referred to each kWh of energy that is handled by the
system and in this case it is called specific cost (𝐶𝑠𝑝) ( 9 ).
𝐶𝑠𝑝 = 𝐶𝐶 ∙1
𝐸𝑛𝑒𝑟𝑔𝑦 𝑝𝑒𝑟 𝑐𝑦𝑐𝑙𝑒 ( 9 )
For the case of storage, the specific cost is assigned totally to the
charging process. On the other hand, the cost of DR is attributed both to
the energy shifted in and out. The detailed description of the algorithm is
available in Paper II.
Assumptions for storage and demand response operations
The choice of the battery technology was made thanks to a comparison
between the cycle cost of lead-acid and lithium-ion batteries. Two
equivalent options were simulated and lithium-ion batteries were found to
have the lowest cycle cost (Table 7). The option that was selected for the
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 43
storage system is represented by the lithium-ion battery Tesla Powerwall13,
manufactured by Tesla Motors© and launched on the US market in early
April 2015. The battery is coupled with an inverter capable of handling the
bidirectional power flow. In this case, the inverter is Fronius Symo Hybrid
5.0-3-S [138], which is considered optimized for the chosen battery, due to
the collaboration between the two companies [139]. However, the
computed battery cost appears very high in order to allow the profitability
of the storage system. Therefore, the cost for the storage system has been
decreased by 65%, resulting into a capacity cost lower than 300
€/kWhinstalled, which reflects an optimistic learning curve for residential
storage systems [140].
Table 7: Operating and economic parameters for the two storage options. Adapted from [94].
Li-ion Lead-acid Specs Capacity [kWh] 6.4 6.2 Maximum depth of discharge (DODmax)
80% 60%
Roundtrip efficiency (ηRT) 0.92 0.81 Expected Life [cycle number] 3500 800
Cost (€) Battery deck 2730 [141] 14 1486 [142] Battery inverter 2208 [143] 2241 [144] Installation (5% of the total) 247 186 Total Cost of Storage 5185 3913 Cycle Cost (CC) [€/cycle] 1.48 4.89 Specific cost (Csp) [€/kWhcharged] 0.289 1.317 Capacity Cost [€/kWhinstalled] 810 632
The presence of an electric boiler for domestic hot water supply and the
nature of the DHW load makes it suitable for a load deferral strategy
[104,145]. The load connected to the consumption of domestic hot water,
which has been estimated equal to 4.19 kWh/day15, corresponds to a daily
13 In this work the first model of Tesla Powerwall is considered. By now, the only option available has 13.5 kWh of usable capacity (at DoD = 100%) and a round-trip efficiency of 90% [141]. 14 The conversion factor is €/$ = 1.1. 15 Domestic hot water is intended as the drinking water that must be supplied at a temperature of at least 45ºC [204]. However, to exclude the proliferation of legionella bacteria in the tank the minimum temperature should be 60 ºC. The grid water temperature is considered equal to 15 ºC.
44 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
consumption of 40 l/person [146]. This value has been incremented by 5%
to account for the heat loss through the pipes and the tank shell, resulting
in a value of 4.40 kWh/day. The cost of a smart boiler controller has been
set at 200 € after a market analysis of the product that could perform the
task of shifting the DHW load [94]. The functional life has been considered
equal to 10 years.
The European eco-labeling directive [147] was used to model the energy
load profile for DHW (Figure 13). The variations of the proposed DHW load
profile are buffered by the presence of a storage tank, and this justifies the
use of a constant profile.
The installation of a so-called “smart boiler” or of a controller that turns
a traditional electric boiler into a smart one is necessary for the
implementation of a DR strategy. A smart boiler is an appliance that can
shift in time the heating of water according to the instantaneous electricity
price, without compromising the consumers’ needs.
Figure 13: DHW daily load profile.
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 45
Energy analysis results
The generation of electricity of the PV system was simulated for 4
weeks, one per season, and three sizes of the installation (1.5, 3 and 4.5
kWp). In the majority of the cases the weekly renewable production is not
enough to quantitatively satisfy the weekly electrical consumption of the
house. The share of solar energy over the total load spans from 29% (in
January with 1.5 kWp) to 126% (in July with 4.5 kWp). The renewable
production is likely to surpass the load consumption for particular periods
of the year (e.g.: spring and summer) only in the case of 4.5 kWp. However,
even if the weekly production is lower than the demand, the instantaneous
renewable production can be higher than the corresponding load, and the
resulting energy surplus has to be handled.
In Table 8, the comparison of the two strategies for enhanced self-
consumption (BES and DR) shows that, with the increase of the installed
PV power, the percentage of the renewable production which is sent to the
grid increases more for DR than for storage. This means that the exchange
of electricity with the grid is more intense when DR is adopted, because the
buffer capacity of DR (4.4 kWh/day) is lower than that of the batteries (5.12
kWh/day). The highest percentage of self-consumed (i.e. charged) energy
over the total surplus energy appears in the case of 3.0 kWp, because for
larger installed capacity the battery size acts as the limiting factor for
batteries to uptake more electricity (Figure 14).
Table 8: Summary of the energy flows for the considered week in summer for the cases of PV production without buffering strategies (PV-only), with battery storage (BES) and with demand response (DR) (adapted from [94]).
PV
only BES DR
PV only
BES DR PV
only BES DR
PV peak power [kW] 1.5 3.0 4.5
RES production [kWh]
58 116 174
Load consumption [kWh]
138 138 138
Import from grid [kWh/week]
92 83 80 80 52 57 76 48 55
Export to grid [kWh/week]
12 2 0 59 24 36 113 80 91
46 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Figure 14: Percentage of stored energy over the total renewable production.
Economic analysis results
The different weekly savings16 for DR and BES are shown in Figure 15.
The implementation of demand response measures appears more
convenient in all the cases because of the considerably lower specific cost
(𝐶𝐷𝑅 = 0.62 vs. 𝐶𝐵𝐸𝑆 = 10.1 c€/kWh). In other words, every kWh of surplus
electricity can be used either to heat up water for future demand with a
price of 1.24 c€/kWh (2𝐶𝐷𝑅); or it can be stored with a cost of 10.1 c€/kWh
(𝐶𝐵𝐸𝑆), of which only 88% is available to cover the load due to the round-
trip and inverter efficiencies (𝜂𝑖𝑛𝑣𝑒𝑟𝑡𝑒𝑟 = 95%).
If each week is considered to be representative for an entire season (13
weeks); then, the annual costs and savings can be roughly estimated.
Referring the annual savings to the installed PV capacity, the configuration
with 3 kWp appears the one which gives the best results for storage (25
€/kWp/year), while the one with 1.5 kWp produces the highest specific
savings for demand response (102 €/kWp/year). The total annual savings
for DR reach a maximum in the 4.5 kWp configuration, but the specific
savings are the highest for the smallest installation size (Table 9).
The annual equipment cost is calculated by dividing the initial cost of
the equipment by the functional life in years (see Section 3.1.3 for the
16 Difference between the cost for electricity in the PV-only case and the cost in presence of a flexibility measure.
0%
5%
10%
15%
20%
25%
30%
35%
WINTER SPRING SUMMER AUTUMN
Sto
red
En
erg
y /
RE
S p
rod
uc
tio
n
RES Production 4.5 kWp RES Production 3 kWp RES Production 1.5 kWp
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 47
equipment cost definition). The lower specific cost of demand response
allows the system to purchase energy during off-peak hours to cover part
of the peak load. For this reason, demand response strategies reach higher
relative savings for smaller PV systems. The cost of storing energy is higher
than the difference between peak and off-peak electricity, hence it is not
convenient to store off-peak electricity, restricting the storable energy to
the overproduction. This means that the new installations of PV systems
for self-generation, which are generally small compared to the contracted
power would benefit from demand response, while storage solutions will
be more interesting for oversized systems installed when the feed-in-tariff
supporting scheme was in place. This conclusion is valid if the house is
equipped with an electric boiler, because the application of automatic
demand response strategies to other appliances could present additional
challenges due to the interference with people’s behavior [81].
Table 9: Total and specific annual savings on the electricity bill for the considered cases (adapted from [94]).
1.5 kWp 3.0 kWp 4.5 kWp
BES DR BES DR BES DR
Annual PV-only cost [€] 879 879 729 729 606 606
Annual electricity cost [€] 800 706 472 522 321 407
Annual equipment cost [€] 181 20 181 20 181 20
Annual savings [€] -102 153 75 187 103 179
% of PV-only cost [%] -12 17 10 26 17 30
Specific annual savings [€/kWp]
-68 102 25 62 23 40
48 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Figure 15: Percentage weekly cost savings compared to the condition without flexibility measures (PV-only) for different PV peak powers. The columns delimited by thicker borders and filled with a texture represent the weekly savings when the storage option is concerned, the solid-fill columns refer to demand response (adapted from [94]).
0%
5%
10%
15%
20%
25%
30%
35%
40%
WINTER SPRING SUMMER AUTUMN
% w
eekly
savin
gs
RES Production 4.5 kWp RES Production 3 kWp RES Production 1.5 kWp BES
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 49
3.2. Battery energy storage to promote renewable
integration and grid support: the case of Terceira (Paper
III)
Islands’ energy needs are usually fulfilled through the importation of
fossil fuels, both for electricity generation and for transport. Therefore,
some studies [88,89] address the issue of replacing fossil fuels with less
polluting alternatives.
In the archipelago of the Azores, an autonomous region of Portugal
located 1,500 km away from the mainland, the local utility (Eletricidade
dos Açores – EDA) has been actively committed to increase the quota of
electricity produced by renewable energy sources, especially in the biggest
islands (i.e. São Miguel and Terceira). The island of Terceira17 was chosen
as case study since the renewable installed power on the islands has been
growing considerably in the period 20o8-2018 with the installation of two
wind farms (12.6 MW in total), a geothermal plant (3.5 MW) and a waste
to energy plant (2.3 MW) [148,149]. Moreover, the investment plan for the
period 2018-2022 states that in Terceira the investments in fossil-fueled
electric power plants will come to an end in 2020, while those in renewable
capacity expansion are supposed to rise, with more than 9.5 M€ in 2020
and almost 14 M€ in 2021 [150].
The need for storage to enable higher RES shares in the electricity mix
of isolated energy systems is pointed out in Section 2.2.2 and, among the
possible storage technologies, battery systems are able to perform different
tasks to support the grid. In addition to this, their modularity and their
site-independence (unlike pumped hydro and in-ground compressed air)
make them suitable for different applications and geographical settings, as
thoroughly described in Castillo and Gayme [92], in which grid-scale
battery applications are reviewed in connection with their services to the
electrical system. In this study, batteries are utilized to implement
renewable energy time shifting (also called renewable energy arbitrage)
[151] and to provide reserve power [93].
Lithium-ion (LiB) and vanadium flow batteries (VFB) are the
technologies considered for this study. According to the database of the US
DoE [4], LiB is the technology in stationary energy storage with the highest
17 Terceira is the second most populated of the Azores archipelago with nearly 56,000 inhabitants (2014) [205].
50 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
number of storage projects (665), 68 of which can be classified as very
large-scale (>10 MW or 10 MWh). On the other hand, the long cycle life
(up to 20,000 cycles) [7] and the decoupling between energy and power
levels makes VFBs a promising option for large-scale storage [58]. They are
suited for mild climates – like the one present in the Azores – because the
operating temperature should fall in the range -5°C – 40 °C in order to
avoid precipitation in the electrolyte solution [95].
The assessment of the economic performance of a battery storage
facility to reduce RES curtailment and limit the use of thermoelectric
power plants is the main aim of the study. The objective is to model
accurately the unit commitment of the generation system in Terceira and
test the viability of the investment connected to energy storage, according
to common evaluation criteria (i.e. net present value, payback time, return
on investment). Different cost scenarios for LiB and VFB are considered,
in order to deal with the uncertainty that will affect the cost reduction trend
for battery systems in the coming years. This study simulates the behavior
of the system for a whole year and not only for typical days, including also
an estimation of the battery degradation in the economic evaluation.
Case study description and main assumptions
The structure of the market for electricity generation in Terceira is
monopolistic and the price of electricity is set by the Portuguese Energy
Services Regulator (ERSE). The island’s electrical energy production for
the years 2011-2017 is reported in Table 1, in which a decreasing trend in
electricity production can be observed and it is mainly due to the effects of
the economic crisis. On the other hand, the share attributable to renewable
energy sources is growing, with clear implications on the electricity
management. The time horizon for a storage facility to be built has been
assumed between 10 and 15 years, because of the long decision-to-
implementation time that characterizes these projects. Therefore, the
investment analysis is performed starting from the year 2030. The year
2013 is taken as a reference, because it is the year with the most recent and
sufficiently detailed data about the electricity production (i.e. with a half-
hour resolution and the indication of the power produced by each
generator).
The electricity production in 2030 is assumed to be 5% higher than the
corresponding value in 2013, due to the growing electrification trend
taking place both in the residential [152] and transport [105] sectors. The
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 51
shape of the electricity profile reproduces that of 2013 and the profile of
wind generation in 2013 is assumed to be sufficiently representative of the
weather conditions of the island for the following years, according to the
approach presented in Varone and Ferrari [24].
Table 10: Electricity production by source and population for the island of Terceira (2011-2017) [149,153].
GWh 2011 2012 2013 2014 2015 2016 2017
Thermoelectric plants 194 181 172 168 168 156 144
Wind 19.0 29.2 34.0 34.6 31.3 33.3 31.3
Waste-to-Energy 0 0 0 0 0 8.5 8.7
Hydro 1.5 2.4 2.8 0.5 0 0 0
Geothermal 0 0 0 0 0 0 9.8
Total RES 20.5 31.7 36.8 35.1 31.3 41.8 49.8
Electricity Production 214 212 209 203 200 198 194
RES Share [%] 9.6 14.8 17.2 16.4 14.6 19.5 25.6
Population (103 ) 56.6 56.6 56.6 56.4 56.1 56.0 56.6
The structure of the generation system by type of plant in 2030 and the
operating assumptions are shown in Table 11. The cost function of each
thermoelectric generator is extrapolated from the data of specific
consumption, reported by the system operator (Equation ( 10 )) [154]. The
dispatch optimization algorithm that is used to simulate the unit
commitment in Terceira is extensively described in Paper III.
𝐶𝐻𝑃,𝑖 = (𝐺𝐻𝑅,𝑖 ∙ 𝑊𝑒𝑙,𝑖 + 𝐺𝑁𝐿,𝑖) ∙ 𝐶𝐹,𝑖 ( 10 )
𝐶𝐻𝑃 represents the hourly production cost of the ith committed thermal
generator, 𝐺𝐻𝑅 is the incremental heat rate, 𝐺𝑁𝐿 is the no-load
consumption and 𝐶𝐹 is the fuel cost.
52 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Table 11: Operating assumptions for the electricity generators in Terceira.
Type of plant
Maximum power [MW]
Operating conditions
Notes
Diesel Engines
12.1 According to the dispatch optimization
Fuel Oil Generators
49 According to the dispatch optimization
Geothermal 7
Continuously at 75% of the maximum power for 11 months (1 is for maintenance)
Not enough data on real behavior. Assumption in accordance with IEA [155]. Rated power value higher than current installed capacity because of planned capacity expansion.
Waste-to-Energy
2.3
Continuously at 80% of the maximum power for 11 months (1 is for maintenance)
In the first year of operation the availability factor was lower than 50%. These plants can work up to 8000 hours per year [156].
Wind 12.6 According to wind availability profile
2013 used as base year.
In electricity generation systems the spinning reserve18 should be able
to counterbalance the loss of the largest generator, without introducing
excessive fluctuations in the power flows or frequency level, thus, ensuring
the continuity of the supply in case of sudden outage [21,157]. In this study,
the spinning reserve was assumed equal to half of the power injected by the
largest online generator. This hypothesis was adopted to tune the model so
that it matches baseline data. Figure 16 shows the relative difference
between the results of the model and the baseline data in terms of
operating cost for electricity generation, and GHG emissions. One week for
18 Spinning Reserve: “Generation capacity that is online but unloaded and that can rapidly respond to compensate for generation or transmission outages” [157].
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 53
each month has been sampled and used for the purpose of validation. The
baseline data values are obtained by calculating the result of the cost
function defined in Equation ( 10 ) applied to the values of power of each
generator, made available by the system operator. It can be observed that
the difference is always below 2.5%. This outcome is satisfactory for the
purpose of this study because the uncertainty introduced by the model is
much lower than the uncertainty on the battery price, that will be
addressed in Section 3.2.4.
a)
b)
Figure 16: Percentage difference between model results and baseline data for a) GHG emissions and b) operating cost.
The parameters for the techno-economic analysis of the two battery
plants are reported in Table 12.
-1.5%
-1.0%
-0.5%
0.0%
0.5%
1.0%
% Δ
GH
G e
mis
sio
ns
-2.5%
-2.0%
-1.5%
-1.0%
-0.5%
0.0%
0.5%
1.0%
% Δ
op
era
tin
g c
ost
54 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Table 12: Technical parameters and costs for lithium-ion (LiB) and vanadium flow batteries (VFB)
LiB Ref. VFB Ref.
Capacity MWh 20 – 30 - 20 - 30 -
Maximum Charge/Discharge Power
MW 10 - 10 -
Depth of Discharge % 0.9 [158] 1 [159]
Round-trip efficiency % 95 [160] 85 [7]
Efficiency (end of life) % 80 o.a.19 75 o.a.
Capacity Loss (end of life)
% 20 [161] 10 o.a.
Cycle life 5000-10000
[8] >10000 [8]
Capacity Cost (CAPEX) €/kWh 296 [32] 243 [32]
Operation and
Maintenance (OPEX)20 €/year
3% 𝐼𝑁𝑉𝑂𝑉𝑁
o.a. 5%
𝐼𝑁𝑉𝑂𝑉𝑁 o.a.
Investment life years 20 [91] 20 [91]
Discount rate % 3 [162]21 3 [162]
The estimation of the cash flows for the NPV analysis and the
quantification of the Return on Investment (ROI) are performed according
to Equations ( 11 ) and ( 12 ).
𝑁𝑃𝑉 = ∑ (𝐺𝑎𝑖𝑛𝑆𝑇𝑂 − 𝐶𝑆𝑇𝑂,𝑂&𝑀
(1 + 𝑑)𝑦)
𝑁𝑌
𝑦=1
− 𝐼𝑁𝑉𝑂𝑉𝑁 ( 11 )
𝑅𝑂𝐼 =𝑆𝑎𝑣𝑒 − 𝐶𝑆𝑇𝑂,𝑂&𝑀
𝐼𝑁𝑉𝑂𝑉𝑁
( 12 )
The positive cash flows are represented by the annual savings on the
cost for electricity generation due to the storage installation (𝐺𝑎𝑖𝑛𝑆𝑇𝑂). The
negative ones are due to the operation and maintenance expenditures
attributable to the storage plant (𝐶𝑆𝑇𝑂,𝑂&𝑀). 𝑁𝑌 is the life of the project, 𝑑 is
19 Own assumption. 20 The O&M cost for LiB is lower than that for VFB due to the moving parts needed for the electrolyte flow in the second option. 21 This value corresponds to the average of the cost of capital in the period 2008-2016 that the local utility has declared in their annual report (Relatório e Contas 2016 [162]).
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 55
discount rate and 𝐼𝑁𝑉𝑂𝑉𝑁 is the overnight initial investment. The
discounted payback is calculated by equating the NPV to zero.
The ROI is the percentage of the initial investment that is recovered
each year, on average, along all the project. 𝑆𝑎𝑣𝑒 is the average yearly saving
throughout the life of the project.
The performance of the batteries suffers from degradation due to cycle
and calendar life [163–165] and this appears to have always been neglected
in similar studies. The performance loss of the flow batteries is linked to
the mechanical degradation of the membrane between the two electrodes,
which can have implications on efficiency and useful capacity [96]. This
phenomenon affects also the economic benefits produced by the storage
facility. In order to quantify this effect, a series of simulations with
degraded values of efficiencies and useful capacities was performed and the
annual savings under these new conditions were computed. Then, the
positive cash flow was assumed to decrease linearly between these two
values.
56 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Energy and environmental results
The reduction of curtailed energy and of fossil-generated electricity are
considered as the indicators to quantify the contributions of energy storage
to the grid integration of renewables. For the island of Terceira, lithium-
ion and vanadium flow batteries are simulated in two different
configurations, in order to minimize the curtailed energy: with a rated
capacity of 20 MWh (Case A) and of 30 MWh (Case B).
The reduction of the thermoelectric energy supply and the related GHG
emissions for the effect of the battery storage facility is remarkable (Table
13). The thermal production is reduced by 9% compared to the case without
storage (NO STO), as well as the curtailment of wind energy, while the
GHG emissions experience a slightly lower decrease (around 7%) because
of the effect of the emissions deriving from the waste-to-energy (W2E)
plant.
Wind production has a lower dispatch priority, because the renewable
energy coming from plants that would suffer more from being cut out (e.g.
geothermal and W2E) is privileged. Although the majority of the curtailed
energy comes from wind, the operation of the geothermal and waste-to-
energy plants can also be affected by the constraints about spinning reserve
when the demand is low22. The presence of the batteries, thanks to their
possibility to provide spinning reserve, allows the system operator to
disconnect also one of the two fuel-oil generators, so that all the energy can
be dispatched. This option considerably reduces the surplus, which passes
from 13.5 to less than 1 GWh per year. The losses of the storage cycle
slightly increase the total energy production, but this is totally
compensated by the large reduction in curtailed energy.
22 The geothermal and waste-to-energy plants can be forced to reduce their energy production because a fossil plant has to run to guarantee the minimum reserve power.
OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III) | 57
Table 13: Energy and environmental benefits deriving from the installation of a storage facility.
NO STO LiB A LiB B VFB A VFB B
Electrical Energy Supply [GWh/y]
220.5
Thermoelectric Plants [GWh/y]
132.1 119.6 119.6 120.0 119.6
Geothermal + W2E [GWh/y]
56.0 56.8 56.8 56.8 56.8
Wind [GWh/y] 32.3 44.3 44.7 44.4 44.8
Fossil Share 60% 54% 54% 54% 54%
BES Discharged Energy [GWh/y]
4.30 4.60 4.11 4.38
Fuel Consumption [1000 m3]
31.1 28.2 28.3 28.5 28.3
Percentage Reduction 9.4% 9.2% 8.6% 9.0%
Energy Surplus [GWh/y] 13.5 0.7 0.3 0.6 0.2 Total Production [GWh/y]
220.5 221.4 221.4 221.8 221.4
Total Emissions [ktCO2,eq] 104.0 96.6 96.3 96.9 96.5
Percentage Reduction 7.1% 7.4% 6.8% 7.2%
Investment analysis
The profitability of the investment is assessed using NPV and ROI, as
mentioned in Section 3.2.1. The values of the positive and negative cash
flows are reported in Table 14. The flow battery option is favored by the
lower specific capital expenditure, which produces a faster recovery of the
invested capital and a higher NPV after 20 years. This happens in spite of
the worse round-trip efficiency and the higher maintenance costs for VFB.
Under these conditions, all the projects have a discounted payback time
shorter than 10 years (Figure 17 and Figure 18), which means that they
generate positive economic value for more than half of their considered
lifetime and this can be an acceptable condition for the project realization.
In the system under analysis the storage facility with 20 MWh of
capacity produces better financial results for both the technologies,
because most of the capacity remains underused in summertime due to low
wind production. Therefore, the integration of a solar PV plant appears as
a favorable complement in order to improve the economic viability of the
project, owing to its peak production in summer months.
58 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Table 14: Overnight investment, positive cash flows for the first and last year, and operation and maintenance cost in the four analyzed cases.
LiB A LiB B VFB A VFB B
Investment (𝐼𝑁𝑉𝑂𝑉𝑁) – negative cash flow [M€]
6.2 9.3 5.1 7.6
Savings first year – positive cash flow [M€]
1.50 1.56 1.45 1.50
Savings last year – positive cash flow [M€]
1.37 1.44 1.39 1.41
O&M annual cost – negative cash flow [M€]
0.19 0.28 0.25 0.38
In addition to the Net Present Value and the payback time, the Return
on Investment (ROI) can be calculated for each project as reported in Table
15. In any scenario, the values are very high, which is a good indicator of
the economic benefits of the project.
Table 15: Return on Investment after the first year (ROI first year) and Return on Investment as defined in Eq. ( 12 ) (ROI average).
ROI (first year) ROI (average)
LiB A 21.3% 20.2%
LiB B 13.8% 13.2%
VFB A 23.5% 22.9%
VFB B 14.7% 14.1%
OP
TIM
IZA
TIO
N O
F E
LE
CT
RIC
ITY
DIS
PA
TC
H W
ITH
BA
TT
ER
Y S
TO
RA
GE
SY
ST
EM
S (P
AP
ER
S I,
II AN
D III) | 5
9
Figure 17: Net Present Value along the life of a lithium-ion battery system for two values of capacity.
Figure 18: Net Present Value along the life of a vanadium flow battery system for two values of capacity.
-15
-10
-5
0
5
10
15
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
NP
V [
M€]
Time [y]
LiB A
LiB B
-10
-5
0
5
10
15
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20NP
V [
M€]
Time [y]
VFB A
VFB B
60 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
Sensitivity analysis
Vanadium flow batteries show better economic performance than
lithium-ion batteries, however, this is strongly influenced by the
assumption that their CAPEX is 20% lower than the corresponding value
for LiB. A dramatic reduction of battery system costs is predicted by several
studies [166–168], but high uncertainty characterizes the extent of this
reduction. Therefore, given that the base-case cost is already quite low, the
base-case CAPEX was doubled in order to cover a larger part of the cost
range proposed by Few et al. [166] for remote systems. The analysis was
performed for both the technologies considering only the 20 MWh capacity
case. In addition, the OPEX was sensibly increased take into account the
uncertainty on this factor, and the results are reported in Table 16. In all
the cases, vanadium flow batteries have higher ratios between the net
present value after 20 years and the capital invested (NPV20/INVOVN)
compared to lithium-ion batteries. However, the impact of a double capital
cost on the profitability is remarkable, with an NPV20 that is approximately
half of the base case. Doubling the OPEX has a slightly lower impact on the
net present value, which for both technologies is visibly lower than the base
case, but remains approximately 1.5 times higher than the initial
investment.
OP
TIM
IZA
TIO
N O
F E
LE
CT
RIC
ITY
DIS
PA
TC
H W
ITH
BA
TT
ER
Y S
TO
RA
GE
SY
ST
EM
S (P
AP
ER
S I,
II AN
D III) | 6
1
Table 16: NPV and discounted payback (DPB) for increased CAPEX and OPEX referred to LiB and VFB (the subscript 0 indicates the base-case conditions; the subscripts 10 and 20 indicate the number of years after which the NPV is calculated).
LiB 20MWh VFB 20MWh
INVOVN [M€]
DPB [y]
NPV10
[M€] NPV20 [M€]
NPV20/ INVOVN
INVOVN [M€]
DPB [y]
NPV10
[M€] NPV20 [M€]
NPV20/ INVOVN
CAPEX0 - OPEX0
6.18 6 4.79 12.54 203% 5.09 5 5.02 12.33 242%
1.5*CAPEX0 - OPEX0
9.27 9 1.70 9.45 102% 7.64 8 2.47 9.78 128%
2*CAPEX0 - OPEX0
12.36 12 -1.39 6.36 51% 10.18 11 -0.08 7.24 71%
CAPEX0 - 1.5*OPEX0
6.18 6 4.00 11.16 180% 5.09 6 3.93 10.44 205%
CAPEX0 - 2*OPEX0
6.18 7 3.21 9.78 158% 5.09 7 2.84 8.54 168%
1.5*CAPEX0 - 1.5*OPEX0
9.27 9 0.91 8.07 87% 7.64 9 1.38 7.89 103%
2*CAPEX0 - 2*OPEX0
12.36 15 -2.97 3.60 29% 10.18 14 -2.25 3.45 34%
62 | OPTIMIZATION OF ELECTRICITY DISPATCH WITH BATTERY STORAGE SYSTEMS (PAPERS I, II AND III)
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 63
4. Techno-economic analysis of a biogas
upgrading process via electrolysis and
methanation (Papers IV and V)
This chapter presents a process that enables the upgrading of biogas to
synthetic natural gas through its direct injection in an electrolyzer. The
concept is based on a US patent [169]. The information contained in the
patent was used to model the components of the plant and its operating
conditions. A detailed simulation model allowed the energy efficiency of
the process to be calculated, which constituted the basis for the comparison
with other competitive options. The economic evaluation was conducted
through the exergo-economic analysis methodology, supplemented by a
sensitivity analysis and net present value determination for a further
examination of the investment viability of the concept.
4.1. Process description
The following pages contain the description of the biogas upgrading
process via electrolysis and methanation. In addition, a specific
explanation of the passivation effect that sulfur has on the SOEC catalyst is
provided because this is a specific feature of the modeling of co-electrolysis
in presence of methane.
Plant layout
The biogas upgrading system comprises three main sections: reactant
conditioning, electrolysis, and methanation (Figure 19). The modeled plant
is fed with liquid water and biogas. The latter is assumed to have an average
molar composition of 60% CH4 and 40% CO2 on a volume basis. All the
reactants are supplied at 20°C and atmospheric pressure. The feed streams
have to be heated up to be suitable for the operation in the electrolyzer.
This operation can be realized by using external sources of heat (e.g.
electricity or combustion processes) or recovering the heat made available
by other processes in the plant (i.e. cooling of electrolysis products and
64 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
methanation). The electrolysis reactions (see Section 2.1.2) take place in
solid oxide electrolysis cells (SOECs), which represent the component with
the highest consumption of electricity, which is supplied in form of direct
current. This component can operate at different pressures, ranging from
1 to approximately 30 bar. The reactants enter the SOEC with a
temperature of 800°C, while the electrolytic reactions are considered to
take place at 850°C. The temperature increase from 800 to 850°C is
accomplished inside the stack through the Joule effect produced by the
current flow. This fact allows the plant to reach a complete thermal
integration, meaning that all the heating processes are realized through the
utilization of recovered heat. The network of heat exchangers that connects
the hot and cold flows, according to the principles of the Pinch analysis
[170], is depicted in Figure 19. The energy performance of the plant is
simulated for two different pressure levels of the electrolyzer and for two
different values of sulfur coverage of the nickel catalyst (see Section 4.1.2).
TE
CH
NO
-EC
ON
OM
IC A
NA
LY
SIS
OF
A B
IOG
AS
UP
GR
AD
ING
PR
OC
ES
S V
IA E
LE
CT
RO
LY
SIS
AN
D M
ET
HA
NA
TIO
N (P
AP
ER
S IV
AN
D V
) | 65
Figure 19: Plant layout of the electrochemical biogas upgrading process. The cleanup unit is optional and it is present only if a controlled quantity of sulfur is fed to the SOEC stack (Adapted from [171]).
66 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
Sulfur effects on SOEC nickel catalyst
The high-temperature electrolysis reactions (Equations ( 4 ) and ( 5 ) –
Section 2.1.2) occur on the solid oxide cell’s cathode, which is usually made
of Yttria-stabilized Zirconia (YSZ) with nickel particles dispersed on top,
acting as catalyst for the electrolytic conversion.
When the stack of cells is fed with a gaseous mixture containing
significant quantities of methane, the nickel promotes the reaction of
steam reforming ( 13 ). This reaction is exothermic and it is
thermodynamically favored at low pressure and high temperature. Hence,
the reduction of the methane content in the inlet gas is inversely
proportional to the operating pressure of the stack, creating also a non-
uniform temperature distribution, if the produced heat is not properly
managed through an effective cooling system.
CH4 + H2O → 3H2 + CO ( 13 )
Such methane depletion in the biogas could be inhibited by co-feeding
a controlled amount of H2S to the inlet stream, since sulfur covers the
active sites of the catalyst and reduces the rates of the surface reactions
[172]. Biogas usually contains sulfur in the form of H2S, which is the
product of the reactions involving the organic sulfur compounds present in
the feedstock to the anaerobic digester. Hydrogen sulfide should be
completely removed from biogas in a cleanup unit [173] to avoid
combustion problems; however, in this case a small quantity could be kept
to inhibit steam reforming. This phenomenon is called sulfur passivation
effect and it can be simulated by multiplying the reaction rate by a
correction coefficient of (1 − 𝜃𝑠 )3, where 𝜃𝑠 is the sulfur coverage23 [174].
The catalyst active sites for heterogeneous catalysis that favor the steam
reforming reaction were found to be separate from those promoting the
electro-catalytic reactions [175], thus it seems that sulfur prefers to interact
with those of the first type. Therefore, the presence of sulfur inhibits the
decomposition of methane more than the electrochemical conversion of
water and carbon dioxide. However, the electrochemical reactions are
partially penalized by the reduction in the number of the active sites, the
effect of which can be modeled as an increase in cell resistance.
23 The coverage expresses the ratio between the number of covered active sites and the total number of active sites for a catalyst.
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 67
Although H2S can have a beneficial effect for the electrolyzer, it is surely
detrimental in the methanation reaction, since it inhibits the methane
formation reaction (the reverse of steam reforming). Therefore, an
additional cleanup unit (also called sulfur polisher) has to be placed
between the two sections to ensure that the gas entering the series of
methanation reactors is sulfur-free.
4.2. Energy and exergo-economic analyses
The determination of the energy performance (mass and energy
balances) is done through the implementation of the plant model in the
chemical engineering software Aspen Plus [176], that allows one to
quantify the mass and energy streams involved. The techno-economic
evaluation is based on the functional model and it is developed with the
methodology of the exergo-economic analysis, a brief description of which
is reported in the next sections and in Appendix A. This methodology
integrates an exergetic evaluation with an economic analysis of the process,
where the economic value is linked with the exergy content of each stream.
In other words, the economic value assigned to all the mass and energy
streams is based on the amount of exergy and capital that is needed to
produce each specific stream. Therefore, this approach provides a way to
distribute the production cost among the different outputs of a multi-
product process, as the one under evaluation. Such analysis supplies
information that traditional economic analysis based on NPV or ROI are
not able to capture, because they assess the profitability of a project for
different market conditions. Instead, they fail to provide a complete
indication of the techno-economic modifications that can be implemented
to improve the profitability of the process.
Energy and exergy analyses
The energy analysis includes the calculation of all the enthalpy values
associated to the mass streams and the quantification of the electrical and
thermal needs of the different components of the process, in order to define
the efficiency of the process. This step is complemented by the
computation of the physical (Equation ( 14 )) and chemical exergy
(Equation ( 15 )) of all the mass and energy streams of the plant (Figure
19). The calculations are done following the approach described in
Gandiglio et al. [177]. The dead state condition (that is the zero exergy
68 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
state) are T0 = 298.15 K, p0 = 1.0 atm, and the environment composition is:
N2 75.67%, O2 20.35%, H2O 3.12%, CO2 0.03%, other 0.83% [178].
𝑏𝑝ℎ − 𝑏𝑝ℎ0 = (ℎ − ℎ0) − 𝑇0 ∙ (𝑠 − 𝑠0) ( 14 )
𝑏𝑐ℎ = ∑ 𝑦𝑖 ∙ 𝑏𝑐ℎ,𝑖
𝑖
+ 𝑅𝑇0 ∑ 𝑦𝑖 ∙ ln (𝑦𝑖)
𝑖
( 15 )
𝑏𝑝ℎ is the physical exergy, 𝑏𝑐ℎ is the chemical exergy, ℎ is the enthalpy (all
expressed in J kg-1) and 𝑠 is the entropy (in kJ kg-1 K-1), 𝑦𝑖 is the molar
fraction of the ith component of the mixture and 𝑅 is the universal gas
constant (8.31 J mol-1 K-1).
Moreover, a resource and a product are defined for each unit of the process
(see Appendix A), the behavior of which is described with particular
performance indices, which are:
· Exergy efficiency (𝜂𝑒𝑥,𝑖) – ratio between the product and the resource
of the ith component.
· Exergetic factor (𝑓𝑖 ) – ratio between the resource utilized by the ith
component and the total input exergy given to the plant as the external
resource.
Assumptions for the economic analysis
The economic evaluation is influenced by the different specific cost
between an SOEC operating at high pressure and one operating at low
pressure. The value of 4,000 €/m2 was considered for the SOEC stack
operating at low pressure, while 6,000 €/m2 when it operates at high
pressure. These estimations are similar to those proposed by Mogensen
[179] and slightly higher than those found in Giglio et al. [180]. When a
controlled quantity of sulfur is fed to the electrolyzer, a high-temperature
polisher filled with ZnO is used to yield a sulfur-free syngas to the series of
methanators. The cost of this component is related to the cathodic outflow
of the cells [36]. The methanation reactors are considered to have a cost of
400 €/kWSNG [63].
The costs of pressure changers, heat exchangers and vessels are derived
from Turton et al. [34] and they are expressed in 2001 USD. The values are
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 69
updated to 2014 USD via the chemical engineering plant cost index (CEPCI
- equal to 576.1 in 2014 and to 397 in 2001). The results are converted into
Euros through a fixed exchange rate equal to 1.1 $/€. In addition to the
purchase cost of the physical components (bare erected cost – BEC), the
cost estimation incorporates also all the sub-components that are
indispensable for the connection and the control of the various units (such
as piping and instrumentation), as well as the costs for the design and
construction of the plant. Their sum represents the Total Plant Cost (TPC).
The conversion to TASC (Total As Spent Cost) is done by applying the
correction factors described by Giglio et al. [180] 24 , according to the
methodology developed by the National Energy Technology Laboratory
(NETL) for energy conversion plants [181]. On top of this, the operation
and maintenance costs are added, in this case they are estimated as 4% of
the total plant cost.
Exergo-economic analysis
The exergo-economic analysis requires the computation of the exergo-
economic costs of the components (Z) and those of the inlet mass and
energy flows. These quantities are useful to define two indices of
performance that contribute to detail the behavior of the plant. In
particular:
· The cost of exergy destruction (CD), defined as the fuel cost
entering a component, multiplied by the rate of exergy destruction
of the same unit
· The exergo-economic factor (𝑓𝑒𝑥𝑒𝑐), defined as the ratio between
the exergo-economic cost and the sum of exergo-economic and
exergy destruction costs of each component (𝑍
𝑍+𝐶𝐷)
The exergo-economic factor determines to which extent the presence of a
component is responsible for the stream-cost increase. If 𝑓𝑒𝑥𝑒𝑐 is close to 1,
the majority of the cost increase of a stream is due to the equipment cost;
otherwise, the cost of exergy destruction is predominant.
24 The TPC is 30% higher than the bare erected cost (BEC) and the TASC is 30% higher than TPC.
70 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
4.3. Energy analysis results (Paper IV)
The upgrading process, the layout of which is shown in Figure 19, was
simulated in four different operating conditions (Table 17), combining two
pressure levels of the electrolyzer and two sulfur coverage values. For all
the cases, the inlet and the outlet energy and mass flows were computed,
using as design variable the current absorbed by the electrolyzer (10,000
A). The energy efficiency of the plant (𝜂𝑝𝑙 ) was obtained as defined in
equation ( 16 ).
𝜂𝑝𝑙 =𝐿𝐻𝑉𝑆𝑁𝐺 ∙ �̇�𝑆𝑁𝐺𝑜𝑢𝑡
𝐿𝐻𝑉𝐶𝐻4�̇�𝐶𝐻4𝑖𝑛
+ 𝑊𝑒𝑙,𝑝𝑙
( 16 )
Table 17: Main distinctions among the four cases analyzed in the study.
Case A Case B Case C Case D
SOEC Pressure Low
(ca 1 atm) Low
(ca 1 atm) High
(ca 30 bar) High
(ca 30 bar)
Sulfur Coverage (𝜃𝑠)
0% 95% 0% 95%
In all the four cases the efficiency is quite high (above 70%), however,
Case A is characterized by a much lower value compared to the others
(Table 18). This is due to the higher electrical consumption in the pre-
methanation compressor and in the SOEC stack. The compression of the
gaseous stream entering the series of methanation reactors is more energy-
intensive than the pressurization of the liquid water before it enters the
SOEC. Moreover, the endothermic reaction of steam reforming is favored
by the thermodynamic conditions present in Case A and this entails that
all the inlet methane is converted into carbon monoxide and hydrogen,
with an additional thermal need to maintain the temperature constantly at
850 ºC. Therefore, a larger part of the electrical energy is used to keep the
temperature constant and is not used for the electrolysis. This fact is
reflected by the lower oxygen production rate (Table 18).
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 71
Table 18:Global mass and energy flows for the plant in Figure 19.
Quantity Unit A B C D
Total AC electric need [kW] 15.5 15.9 14.2 14.3
Pumping Power [kW] 2.2 2.2 0.59 0.60
Stack Power [kW] 13.2 13.6 13.7 13.7
Upgraded Gas (SNG) [kg/h] 1.39 1.82 1.81 1.89
Methane inlet [kg/h] 0.82 1.08 1.07 1.12
Oxygen outlet [kg/h] 2.17 2.87 2.84 2.97
LHVSNG [MJ/kg] 49.2 49.4 49.4 49.4
SNG chemical energy flow
[kW] 19.0 25.0 24.8 25.9
Plant efficiency (ηpl) [%] 71 81 86 87
Comparison with other biogas upgrading routes
The process that has been described so far represents a tentative one for
coupling biogas upgrading with an efficient use of low-carbon and low-cost
electricity. However, there are other possibilities to remove CO2 from
biogas in order to improve its quality. The principal options are: water
scrubbing (WS), pressure swing adsorption (PSA) and membrane
separation (MS); the description and the specific energy consumption of
which can be found in Pertl et al. [182]. The efficiency and the specific
productivity (SP) of each process can be defined as in Equations ( 17 ) and
( 18 ).
𝜂𝑠𝑒𝑝 =𝐿𝐻𝑉𝑆𝑁𝐺 ∙ �̇�𝑆𝑁𝐺,𝑜𝑢𝑡
(𝐿𝐻𝑉𝐶𝐻4 ∙ �̇�𝐶𝐻4𝑖𝑛 + 𝑊𝑒𝑙 + 𝐸𝑒𝑥𝑡𝑟𝑎)
( 17 )
𝑆𝑃 = 𝑌𝐶𝐻4(
100 − 𝐿𝐶𝐻4
100) ( 18 )
where 𝐸𝑒𝑥𝑡𝑟𝑎 includes the additional energy flow required to obtain a grid-
suitable SNG, YCH4 [Nm3SNG/Nm3
RG] is the methane molar fraction in the
biogas and 𝐿𝐶𝐻4 is the percentage of methane lost during the process. For
instance, PSA produces an SNG that needs to be mixed with propane in
order to be suitable for grid injection and the propane consumption has
72 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
been estimated to be 0.013 [Nm3/Nm3RG], while the electrical energy
requirement is 0.053 [MJ/Nm3RG] [182].
The results for each technology are summarized in Table 19, assuming
that the outlet SNG has a methane content of 97% in volume25.
Table 19: Efficiencies and specific productivities for Water Scrubbing, Pressure Swing Adsorption and Membrane Separation (adapted from [183]).
PSA Efficiency (𝜂𝑠𝑒𝑝) 92.9 [%]
Specific productivity 0.576 [Nm3SNG/Nm3
RG]
WS Efficiency (𝜂𝑠𝑒𝑝) 94.2 [%]
Specific productivity 0.591 [Nm3SNG/Nm3
RG]
MS Efficiency (𝜂𝑠𝑒𝑝) 86.6 [%]
Specific productivity 0.564 [Nm3SNG/Nm3
RG]
The electrochemical upgrading process that has been analyzed in this
study is a one-step process, because the biogas is fed directly to the SOEC
and the product gas of the electrolysis undergoes methanation. The sealed
structure of the SOEC and of the methanation reactors drastically reduces
the carbon dioxide slip from the plant, which was neglected. Therefore, the
carbon dioxide present in the biogas is completely transformed into
methane throughout the SOEC and the methanation section. Moreover, CO
is almost absent after the water removal step and this implies that the
conversion of CO2 into CO, due to the water gas shift reaction, is negligible
in the methanation process.
However, in order for the one-step process to be comparable with the
other commercial options, a two-step process is considered. Namely, CO2
must be firstly removed from the digester gas and then a mixed H2O/CO2
stream is fed to the integrated SOEC-methanation plant (with an efficiency
– 𝜂𝐶𝑜𝐸 – of 81.4%26). A scheme of the two-step process is depicted in Figure
20. The two-step process efficiency (𝜂𝑡𝑠) is defined in Equation ( 19 ) and
this quantity is affected by the efficiency of the CO2 separation technique
(𝜂𝑠𝑒𝑝).
25 The productivity values refer to a molar composition of biogas equal to 60% CH4 and 40% CO2 vol. The lower heating value of methane is assumed equal to 35 MJ/Nm3 for the efficiency calculations. 26 The H2O/CO2 integrated SOEC-methanation plant was extensively studied in Giglio et al. [180], on the basis of a previous study [206].
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 73
𝜂𝑡𝑠 =(𝐿𝐻𝑉𝑆𝑁𝐺 ∙ �̇�𝑆𝑁𝐺,𝑜𝑢𝑡)𝑠𝑒𝑝 + (𝐿𝐻𝑉𝑆𝑁𝐺 ∙ �̇�𝑆𝑁𝐺,𝑜𝑢𝑡)𝑒𝑙𝑒𝑐𝑡
(𝐿𝐻𝑉𝐶𝐻4 ∙ 𝑚𝐶𝐻4𝑖𝑛 + 𝑊𝑒𝑙 + 𝐸𝑒𝑥𝑡𝑟𝑎)𝑠𝑒𝑝 +(𝐿𝐻𝑉𝑆𝑁𝐺 ∙ �̇�𝑆𝑁𝐺,𝑜𝑢𝑡)𝑒𝑙𝑒𝑐𝑡
𝜂𝐶𝑜𝐸
( 19 )
The efficiencies are clearly higher for the conventional methodologies
(Table 20). However, these methodologies do not recycle CO2, which is the
reason why they are characterized by much lower SNG yields. The coupling
with the electrolytic upgrade increases considerably the specific
productivity values, with a lower CO2 release. The atmospheric one-step
process has a lower energy efficiency than the two-step routes, but entails
a reduced plant complexity, since the CO2 flow is directly fed to the SOECs.
CO-ELECTROLYSIS +
METHANATION
BIOGAS:
0.6 CH4
0.4 CO2
0.4 CO2
0.6 CH4
0.4 CH4
H2O
CO2 SEPARATION
H2O
Wel1
Wel2
CONTROL VOLUME
Figure 20: Scheme of the two-step upgrading process with the indication of the mass and energy flows (adapted from [183])
The comparison between the different techniques shows that SNG
production from biogas can become a carbon negative process, able to
convert all the carbon present in organic waste into usable fuel.
74 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
Table 20: Summary of the efficiencies and specific productivities for all the digester gas upgrading routes (adapted from [183]).
Process Action(s) Efficiency Specific
productivity
[%] [Nm3
SNG/ Nm3
RG] Pressure Swing Adsorption (PSA)
CO2 separation 92.9 0.576
Water Scrubbing (WS)
CO2 separation 94.2 0.591
Membrane Separation (MS)
CO2 separation 86.6 0.564
PSA + H2O/CO2 CoE (Two steps)
CO2 separation + H2O/CO2 co-electrolysis
88.0 0.976
WS + H2O/CO2 CoE (Two steps)
CO2 separation + H2O/CO2 co-electrolysis
90.7 0.991
MS + H2O/CO2 CoE (Two steps)
CO2 separation + H2O/CO2 co-electrolysis
84.4 0.964
Digester gas CoE (One step) – atmospheric
H2O/CO2 co-electrolysis
71.0-81.0 1.0
Digester gas CoE (One step) – pressurized
H2O/CO2 co-electrolysis
86.0-87.0 1.0
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 75
4.4. Exergo-economic analysis results (Paper V)
The energy performance analysis oriented the subsequent economic
assessment, inasmuch as only the configurations that appeared more
relevant were further considered. Therefore, the Cases A and D described
in the previous sections were disregarded either because the efficiency was
much lower than the other configurations (Case A) or because the
efficiency improvement due to the steam reforming inhibition performed
by H2S was not high enough to justify the presence of a sulfur cleanup unit
downstream from the SOEC (Case D). Secondly, in order to reduce the
specific cost of the plant components (i.e. compressor, pumps, heat
exchangers and reactors), the plant size had to be increased. Therefore, the
direct current (DC) power absorbed by the electrolyzer that was used as
size variable was assumed equal to 5 MW. In the following sections the
results of the three methodological steps of the exergo-economic analysis
(see Appendix A), namely exergetic, economic and integrated exergo-
economic analyses are reported [177].
Exergetic results
The exergy efficiency of the plant in the two analyzed configurations
(Cases B and C) is calculated according to Equation ( 20 ) and it is equal to
83 and 87%, respectively.
𝜂𝑒𝑥,𝑝𝑙 =�̇�𝑆𝑁𝐺
�̇�𝑏𝑖𝑜𝑔𝑎𝑠 + 𝑊𝑒𝑙
( 20 )
The components that treat a large quantity of exergy (i.e. with a high
exergetic factor - fi27) have exergetic efficiencies greater than 90%, which
means that these components have very good performance and the margin
to improve the exergy management is very limited (Table 21).
The lowest exergy efficiency belongs to the low-temperature heat
exchangers, since there is a quite large temperature difference between the
two fluids involved in each heat exchanger. This is due to the large
availability of excess heat from the overall process, which may be used to
restrain the heat exchange area and, in turn, the equipment cost. The
exergy lost in these heat exchangers could be reduced, but it would imply a
27 See Section 4.2.1 for its definition.
76 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
higher investment cost for the plant, without a real benefit in terms of
efficiency, since the thermal needs of the plant are fully satisfied.
Table 21: Exergetic indices of performance for the main components of the low-pressure (LP) and high-pressure (HP) configurations ordered by decreasing exergetic factor (fi).
LP configuration
Component
Exergy
efficiency,
ηex,i
Rate of exergy
destroyed [kW]
Exergetic
factor, fi
Methanator 1 >99% 115.0 210%*
Cleanup unit 99% 56.8 89%
Methanator 2 >99% 45.9 87%
Drum 99% 86.5 84%
SOEC 90% 504.3 42%
Pre-methanation
Compressor 53% 375.8 7%
Oxygen Flow 0% 123.5 1%
Biogas Compressor 76% 18.1 1%
HP configuration
Methanator 1 99% 173 192%*
Methanator 2 >99% 46 90%
Drum >99% 42 87%
SOEC 91% 445 46%
Biogas Compressor 72% 52 1.7%
Oxygen Flow 0% 154 1%
Pre-methanation
Compressor 76% 5 ~0%
* 𝑓𝑖 values higher than 100% appear because some components are placed in a
recirculation loop in the methanation section (Figure 19), where the total exergy entering the
component is higher than the exergy entering the whole plant
The highest rates of exergy destruction are found in the SOEC, Pre-
methanation Compressor, Methanator 1 and Methanator 2, although some
of them have high efficiencies (i.e. SOEC and Methanator 1). However, they
process a large quantity of exergy and, therefore, they substantially impact
the exergy balance. Despite this, the satisfactory energy performance
makes them very expensive to improve and, thus, they are considered
optimized.
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 77
Moreover, the oxygen flow is responsible for the loss of a good quantity
of exergy if it is completely vented. Instead, it may be further exploited
because it has high purity and, if the SOEC is pressurized, it is available at
about 30 bar. In particular, in the latter case it represents 10% of the total
dissipated exergy. This, together with its relatively high pressure, makes it
more appealing for a further economic use than the product of the low-
pressure electrolysis, because the conditioning treatment to make it
suitable for sale quality is more expensive.
Finally, lower consumption of the auxiliary components, higher overall
exergetic efficiency and better thermodynamic quality of the outlet oxygen
stream suggest that the high-pressure configuration should be preferred
over the low-pressure one.
Economic results
Following the assumptions and the methodology described in Section
4.2.2, the total costs of the main components of the plant 28 have been
estimated and they are reported in Table 22. The equipment cost
contributes significantly to the cost of the products of the plant (see Section
4.4.3).
Another item influencing the economic performance of the plant is the
cost of the utilities, i.e. electricity, water and biogas. The electricity price
includes only the distribution and transmission cost (calculated according
to the parameters provided by the Italian TSO [184]). The plant size is quite
large for this kind of application and it requires a substantial biogas feed,
which is unlikely to be provided by the anaerobic digestion of agricultural
sources since these are usually small-to-medium scale plants (typically
below 500 - 600 Sm3/h of biogas production29). Instead, the biogas is
assumed to be available from a large wastewater treatment plant or a
landfill, with a concentration of H2S of 500 ppm. This value is
representative of biogas streams with considerable sulfur loads, while the
majority of the plants shows values around 100 ppm [173]. The biogas
produced in such plants can even be considered to have a zero cost, since
it comes as a by-product of the water sanitation process (or waste
28 The cost of the heat exchangers is not included, but it is available in [171] as well as the procedure for the estimation of the cleanup cost. 29 A biogas plant for electricity production with a rated electric power of 1 MW and an average electrical efficiency of 30% requires a biogas feed of 600 Sm3/h. Such size is usually considered medium to large scale [207].
78 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
management, in the case of landfills) [185]. Nonetheless, a cost is
associated to the inlet biogas feed since it has to be collected, stored,
pipelined across the plant and finally treated to remove contaminants. The
components of the biogas cost are, thence, ascribable to the production and
management of raw biogas and to contaminant cleanup. The cost of raw
biogas per standard cubic meter (at 15°C and 1 atm) is 0.09 €/Sm3 [39],
while the cost for pre-upgrading cleanup is calculated according to the
assumption made by Papadias et al. [186], thus resulting in a cost of 0.04
€/Sm3. The total biogas cost is therefore 0.13 €/Sm3.
Table 22: Total cost of the main components for the low- and high-pressure configurations. (adapted from [171]).
Components (Low Pressure)
Total Cost [k€]
% of total
Components (High Pressure)
Total Cost [k€]
% of total
Ref.
Biogas Compressor
155 1.8 Biogas Compressor
731 11.1 [187]
SOEC 2881* 37.1 SOEC 3058* 46.2 [188]
Cleanup unit 134 1.6 [186]
Pre-methanation Compressor
2375 27.8 Pre-methanation Compressor
56 0.8 [187]
Recirculation Compressor
14 0.2 Recirculation Compressor
6 0.1 [187]
Methanator 1 1213 14.2 Methanator 1 995 15.0 [63]
Methanator 2 301 3.5 Methanator 2 410 6.2 [63]
Methanator 3 38 0.4 Methanator 3 389 5.9 [63]
Total 8244 Total 6613
* includes the replacement of the stack after 10 years
The assumptions for the economic assessment and the specific costs of
the utilities are summarized in Table 23. A high discount rate (i.e. 20%) is
considered because of the risk premium that this kind of plant bears, due
to the limited commercial availability.
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 79
Table 23: Economic and financial parameters, and costs related to the utilities (adapted from [171]).
Economic and financial parameters
Utilities
Capacity factor, CF 70% Biogas price [€/Sm3] 0.13
O&M, fraction of the TPC 4% Electricity price [c€/kWh]
1.1
Capital recovery factor, CRF 0.21 Biogas price [c€/kWh] 2.54
Discount Rate, d 20% Demineralized Water [€/m3]
2.1
Plant Lifetime [years], NY 20 SOEC module lifetime [years] 10
Exergo-economic results
The value of the exergo-economic factor discriminates if the cost
increase is mainly attributable to the equipment cost or to the low
efficiency of the component. For the electrolyzer this value is very close to
1, which means that the component cost has a high impact on the cost of
the outlet streams. Pressure changers have relatively high exergo-
economic factors, because of their small-to-medium size, compared to
those used in large chemical plants (Table 24). On the other hand, a further
increase in plant size would be problematic since it would not favor the
modularity and the possibility to exploit the biogas produced in smaller
plants.
80 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
Table 24: Exergo-economic indices of performance for the main components of the low-pressure (LP) and high-pressure (HP) configurations (adapted from [171]).
LP configuration
Component
Cost of destroyed Exergy, CD [€/h]
Exergo-economic equipment cost, Z [€/h]
Exergo-economic factor, fexec
Methanator 1 5.69 53.52 90% Cleanup unit 2.08 5.40 72% Methanator 2 2.39 13.27 85% Drum 4.82 0.43 8% SOEC 5.55 128.31 96% Pre-methanation Compressor
4.13 105.79 96%
Biogas Compressor 0.20 6.91 97%
HP configuration
Methanator 1 6.94 45.10 87% Methanator 2 1.97 18.57 90% Drum 2.03 0.43 17% SOEC 4.90 136.18 97% Biogas Compressor 0.57 32.55 98% Pre-methanation Compressor
0.06 2.47 98%
The difference between the exergo-economic costs of SNG in the two
configurations (5.62 vs. 4.87 c€/kWh for LP and HP respectively) is
created mainly downstream from the SOEC, where the influence of the
compression of the cathodic gas, in terms of energy and capital
requirement is substantially higher in the LP case. The cleanup unit also
contributes to increase the exergo-economic cost of the final product.
However, the latter component is strictly necessary to avoid problems in
the operation of the subsequent methanation section.
The exergo-economic cost of SNG in both cases is broken down into the
different component in Table 25.
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 81
Table 25: Breakdown of the SNG thermo-economic cost for the two configurations (adapted from [171]).
Low-
pressure High-
pressure
Exergo-economic cost [c€/kWhexergy] 5.62 4.87
Equipment 64.1% 60.0%
Utilities 35.9% 40.0%
- Electricity 11.6% 12.2%
- Biogas 23.8% 27.2%
- Water 0.5% 0.6%
Energy cost [c€/kWhLHV] 5.77 5.04
Energy cost [c€/kWhHHV] 5.19 4.53
The equipment cost occupies the predominant share. In the LP case, it
is higher both in relative and in absolute terms, meaning that the negative
effect that sulfur has on the electrochemical performance, leading to a
larger active area of SOEC ( 𝐴𝐴𝑇𝑀~1.4 ∗ 𝐴𝑃𝑅𝐸𝑆 ) for the same electrical
consumption, is not counterbalanced by the lower cost of the equipment
operating at nearly atmospheric pressure. Therefore, with the assumptions
adopted in this study the pressurized configuration dominates the
atmospheric one both in terms of energy efficiency and in terms of
economic convenience.
The biogas cost represents approximately 25% of the total cost, while
electricity slightly more than 10%, since the specific cost is very low (1.1
€/kWh).
The specific cost of SNG reported in Table 25 can be compared with the
cost estimated for biogas upgrading through water scrubbing. This value
has been found to fall in the interval 2.8-5.8 c€/kWhLHV for a cost of biogas
up to 0.17 €/Sm3 [115]. Therefore, with a biogas cost equal to 0.13 €/Sm3
the resulting SNG cost would be 4.89 c€/kWhLHV. This estimate considers
a higher cost of electricity compared to the present analysis (18.7 vs. 1.1
c€/kWh), however, for water scrubbing the electricity cost represents 14%
of the SNG cost. Therefore, if the scrubbing process used low-cost
electricity, the SNG would result in 4.2 c€/kWhLHV, which is 15% lower
than the cost of the electrolytic upgrading.
82 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
Economic valorization of the by-products
In addition to synthetic natural gas, the plant produces high-purity
oxygen, an electrolysis by-product with a good market potential. The main
uses of oxygen are: steel industry, healthcare and gasification processes.
The most common way of producing oxygen is through cryogenic air
separation, which is generally energy-intensive and thereby the price of
electricity has a considerable impact on the oxygen production cost. The
energy consumption for large oxygen plants is reported to be around 300
– 325 kWh/tO2 and the final price can span between 20 and 50 $/t,
according to the size of the plant [189].
In the present case, the oxygen coming from the SOECs is cooled down
to preheat the mixture of biogas and water and then it is compressed up to
150 bar. The specific cost of the oxygen treatment components is carried
out considering the formulas defined in Turton et al. [187]. In the low-
pressure case, the higher pressure increase influences both the quantity of
electrical energy that is needed, and the sizing of the oxygen compressor,
resulting in a much higher cost of the oxygen. The cost per unit mass of the
oxygen is 54.6 €/tO2 for the LP configuration and 15.2 €/tO2 in the HP one.
Assuming a sale price of 50 $/tO2 (45.5 €/tO2), only the HP case would be
profitable, with an hourly gain of 32 €/h (0.89 c€/s) corresponding to 6.7%
reduction on the SNG thermo-economic cost. The latter would decrease to
4.54 c€/kWhexergy. However, these estimates do not consider the cost for
bottling and transportation, hence, it would be much better if the oxygen
was used onsite.
The electrochemical biogas upgrading prevents the CO2 from being
released, because it is recycled into a fuel. For this reason, the saved CO2
could be accounted as an avoided emission and, it might be eligible to be
traded in the carbon market. Three possible scenarios for the carbon price
are considered to calculate the reduction of exergo-economic cost of SNG
[190]. The results are summarized in Table 26.
The maximum saving per unit of exergy is around 8% of the SNG
exergo-economic cost. However, it is likely that the impact of cost
reduction will be lower, due to the several political and economic obstacles
which can prevent the market from reaching high CO2 prices in the next
decades. Despite this, the participation to the carbon market could be an
option to encourage the recycle of carbon dioxide in digestion processes,
which are already considered carbon neutral.
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 83
Table 26: Economic gains for avoided emissions in different scenarios of CO2 price for the period 2022-2050 (adapted from [171]).
CO2 Price
[€/ton] CO2 Gain [c€/s]
CO2 Gain
[c€/kWhexergy SNG]
LP HP LP HP
Low Price 23 0.45 0.49 0.17 0.17
Medium
Price 38 0.75 0.81 0.28 0.29
High Price 55 1.09 1.17 0.41 0.41
If the exergo-economic cost of the biomethane obtained with the
electrochemical upgrading process in the high-pressure case (Table 25) is
referred to its higher heating value, this results in a value of 4.53
c€/kWhHHV. This value is higher than the price of natural gas for medium-
sized consumers 30 both in 2013 (~4.1 c€/kWhHHV) and in 2015 (~3.5
c€/kWhHHV). However, the inclusion of oxygen sale and CO2 allowances
can bring the production cost closer to the natural gas price. The SNG
exergo-economic cost could be lowered up to 15%. This is equivalent to
3.84 c€/kWhHHV, which is still higher than the 2015 natural gas price, but
lower than the 2013 one (Figure 21).
30 Average EU-28 price for medium-sized industrial consumers (100 000 GJ < Consumption < 1 000 000 GJ) [208]. In 2013 the natural gas price reached its maximum and for this reason this year has been taken as reference.
84 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
Figure 21: SNG unit exergo-economic cost reduction for different options of by-product valorization with the indication of the percentage saving. The black markers illustrate the results in case of combined oxygen sale and CO2 credits. The horizontal lines report the price of natural gas for medium-size industrial consumers (average EU28) per unit of exergy. Note that the scale is on the y axis is partial (adapted from [171]).
Sensitivity analysis
Electricity has a fundamental role in power-to-gas applications and,
since its price affects the profitability of the whole process, a sensitivity
analysis is warranted. The minimum price has been considered zero, while
the maximum price is assumed to be equal to 4 c€/kWh, which is in the
range of the LCOE of capital intensive technologies such as coal and
nuclear [191]. The taxes and levies are excluded, since the electricity used
for power-to-gas applications could be exempted from taxation. Between
the electricity price and the exergo-economic cost of the synthetic natural
gas there is a linear relation with an increase of 0.59 c€/kWhSNG (LP) and
0.54 c€/kWhSNG (HP) per c€/kWhel, as shown in Figure 22. The percentage
variation of the SNG unit exergo-economic cost is steeper for the
pressurized case since the weight of the electricity cost on the total cost is
higher.
3.5
3.6
3.7
3.8
3.9
4.0
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
SNG cost O2 sale CO2 low CO2 medium CO2 high
2013 NG price
2015 NG price
6.7%3.5%
5.8%8.5%
SN
G u
nit
ex
erg
o-e
co
no
mic
co
st
[c€
/kW
h]
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 85
Figure 22: Sensitivity analysis of the SNG exergo-economic cost by varying the electricity price. The red marker indicates the base case. The labels indicate the costs expressed in c€/kWhexergy for the low- and high-pressure conditions, respectively. Adapted from [171].
Another important parameter which has a large impact on the
evaluation of the profitability of the plant is the capacity factor, that is the
equivalent number of hours of plant operation at full power in a year.
Theoretically, the plant should be operating when the electricity is
available at a low price. If power-to-gas was used exclusively to reduce the
curtailment of solar energy, the storage plant would work 26% of the time
at full load and 7% at partial load [192]. Varone and Ferrari [24] simulates
a situation in which the installed wind power in Germany is increased by 8
times and the electricity demand is reduced to 89% of the 2012 value. In
this scenario the number of surplus hours would be 53% of the total. These
two facts combined suggest that it is not easy to reach high capacity factor
values. Therefore, a capacity factor of 60 - 70% can be assumed as a target
for power-to-gas plants to achieve. The results of the sensitivity for
different capacity factors are presented in Figure 23. In this case the
slightly steeper variation occurs for the atmospheric case, since less
working hours mean less efficient use of the machinery, the cost of which
is predominant compared to the cost of utilities.
-20%
-10%
0%
10%
20%
30%
40%
0 0.01 0.02 0.03 0.04
Δ%
SN
G c
ost
Electricity price [€/kWh]
HP
LP
4.97 – 4.27
5.62 – 4.87
7.33 – 6.42
6.74 – 5.88
6.15 – 5.34
86 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
Figure 23: Sensitivity analysis of the SNG exergo-economic cost for different capacity factors. The red marker indicates the base case. The labels indicate the costs expressed in c€/kWhexergy for the low- and high-pressure conditions, respectively. Adapted from [171].
Since the project was considered very risky and with a high operational
uncertainty, a high discount rate was used for the evaluation of the project.
However, generally for energy project evaluation much lower values are
applied (3-5%). For this reason, a sensitivity analysis on this parameter was
done to estimate the effect of the investment risk on the exergo-economic
cost in the two cases. A considerable reduction is observable in Figure 24,
which causes the exergo-economic cost to fall below the 2015 price of
natural gas (3.81 €/kWhexergy) used in Section 4.4.4.
-20%
-10%
0%
10%
20%
30%
0.5 0.6 0.7 0.8 0.9
Δ%
SN
G c
os
t
Capacity Factor
HP
LP
6.21 – 4.87
7.04 – 6.04
5.17 – 4.51 4.99 – 4.36
4.83 – 4.22
5.62 – 4.87
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 87
Figure 24: Sensitivity analysis of the SNG exergo-economic cost for different discount rates. The red marker indicates the base case. The labels indicate the costs expressed in c€/kWhexergy for the low- and high-pressure conditions, respectively.
-50%
-40%
-30%
-20%
-10%
0%
0 5 10 15 20
Δ%
SN
G c
os
t
Discount rate [%]
HP
LP
3.43 – 3.13
3.64 – 3.30 4.24 – 3.77
4.90 – 4.30
5.62 – 4.87
88 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
4.5. Economic profitability with net present value
The methodology and the results illustrated in the previous sections are
focused on the determination of the production cost of SNG using exergy
as allocation criterion. However, an economic analysis using more
traditional methodologies can be beneficial to get an idea of the
profitability that such an investment could have under different
technological and market conditions. Therefore, the net present value
(NPV) of the biogas upgrading plant previously described in this chapter is
calculated, using the economic and financial parameters reported in
Section 4.2.2. Its formulation is given in Equation ( 21 ).
The positive annual cash flow is the revenue deriving from the sale of
SNG ( 𝑅𝑒𝑣𝑆𝑁𝐺 ), while the negative annual cash flows are the costs of
electricity, biogas production, demineralized water (grouped into
feedstock costs 𝐶𝑓𝑒𝑒𝑑𝑠𝑡𝑜𝑐𝑘 ) and operation and maintenance ( 𝐶𝑂&𝑀 ). The
overnight investment (𝐼𝑁𝑉𝑂𝑉𝑁) is partially discounted because the SOEC
pack is supposed to be replaced after 10 years (Table 23). The price of SNG
was set at 80 €/MWh which is a value in line with the incentive package
recently approved by the Italian government [193]. The results for the low-
and high-pressure configurations are shown in Figure 25, using a discount
rate of 20% like the one considered for the exergo-economic analysis.
𝑁𝑃𝑉 = ∑ (𝑅𝑒𝑣𝑆𝑁𝐺 − 𝐶𝑓𝑒𝑒𝑑𝑠𝑡𝑜𝑐𝑘 − 𝐶𝑂&𝑀
(1 + 𝑑)𝑦 )
𝑁𝑌
𝑦=1
− ∑ (𝐼𝑁𝑉𝑂𝑉𝑁
(1 + 𝑑)𝑦)
𝑁𝑌
𝑦=0
( 21 )
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 89
Figure 25: NPV comparison between the high- and low-pressure configurations of the plant for biogas upgrade.
However, the discount rate commonly used for the economic and
financial analysis of energy projects is much lower than the value of 20%
used in Section 4.4.5. For this reason, from now on a discount rate of 5%
will be considered. Some other factors that influence the profitability of the
investment are the cost of electricity, the specific cost of the electrolyzer
and the price that the SNG producer is going to be paid. The results are
shown only for the pressurized configuration since it is the most profitable
one (Figure 26). The baseline assumptions are: 30 €/MWh for the
electricity cost, 60 €/MWh as SNG price31, 6 k€/m2 of active area for SOEC
(CAPEX)32. The cost of electricity is slightly higher than that assumed so
far since this value is closer to the minimum price of electricity in Europe33.
The results varying the aforementioned parameters are reported in Figure
26. The highest impact on the profitability of the investment is attributable
31 This price is takes into account both the price that SNG would have on the natural gas market and the value of the incentive provided in the form of a subsidy for each MWh of produced gas. 32 6 k€/m2 is equivalent to approximately 470 €/kW for pressurized SOECs. 33 Sweden is the European country with the lowest price of electricity for large consumers (>150 GWh/year). These consumers paid 38 €/MWh (excluding all the taxes) in 2017. The minimum electricity price for consumers with a level of consumption comparable to the upgrading plant (between 20 and 70 GWh/year) is in Norway and it is equal to 40 €/MWh (in Germany 45.9, while in Sweden 47.1 €/MWh) [209].
-12
-10
-8
-6
-4
-2
0
2
4
6
8
0 4 8 12 16 20
NP
V [
M€
]
Time [y]
High Pressure
Low pressure
90 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
to the price of synthetic natural gas, while the cost of the electrolyzer has
the lowest impact because it constitutes just a part of the whole plant.
TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V) | 91
(a)
(b)
(c)
Figure 26: Net present value for different (a) electricity costs, (b) SNG price and (c) SOEC CAPEX.
-5
0
5
10
15
20
25
Electricity cost
NP
V [
M€
]
0 €/MWh11 €/MWh20 €/MWh30 €/MWh50 €/MWh
-10
-5
0
5
10
15
20
25
SNG price
NP
V [
M€
]
40 €/MWh
50 €/MWh
60 €/MWh
70 €/MWh
80 €/MWh
-6
-4
-2
0
2
4
6
8
SOEC CAPEX
NP
V [
M€
]
6 k€/m2
9 k€/m2
12 k€/m2
15 k€/m2
18 k€/m2
92 | TECHNO-ECONOMIC ANALYSIS OF A BIOGAS UPGRADING PROCESS VIA ELECTROLYSIS AND METHANATION (PAPERS IV AND V)
CONCLUSIONS | 93
5. Conclusions
In this dissertation an investigation of the techno-economic potential
for different electrochemical energy storage solutions has been presented,
covering different applications which have the potential to play a
significant role in future energy systems. In particular, small-scale and
large-scale battery energy storage was studied in order to mitigate the
uncertainty that variable renewables-based electricity supply brings to the
management of electricity. On the other hand, power-to-gas was
considered as a way to limit the curtailments of low-carbon energy by
transforming it into synthetic natural gas. The techno-economic evaluation
of the aforementioned processes represented the main research objective
of the thesis, that translated into four main research questions.
The first one focused on the most cost-effective strategy to reduce the
mismatch between energy production and demand in a small-scale
residential PV system. Batteries and demand response were compared in
terms of yearly savings for residential consumers. It turned out that
demand response produces better savings on the electricity bill, because
the cost of the equipment is much lower. Conversely, the implementation
of an automatic load shifting strategy, as the one addressed in this work, is
limited to cases in which thermal needs (e.g. domestic hot water or heating)
are directly satisfied with electricity. The present cost34 of the compound
battery+inverter (both for lithium-ion and for lead-acid) appears too high
to achieve a positive economic result. On the other hand, with a projected
cost that is 65% lower than the market price, the investment turns
convenient, although less attractive than DR implementation.
The second research question investigated the energy and economic
benefits of a large-scale battery energy storage plant in island settings,
adopting lithium-ion or vanadium flow batteries alternatively. The results
of the unit commitment model showed that both the technologies
considerably reduce the consumption of imported fossil fuels for electricity
generation. They have satisfactory performance in terms of NPV (>200%
34 The study was published in 2016, therefore the cost is referred to this year.
94 | CONCLUSIONS
after 20 years) and ROI (>20%), meaning that they are a meaningful choice
to promote renewables. The comparison reveals that vanadium flow
batteries produce slightly better results than lithium-ion because they are
assumed to have a lower capacity expenditure. In contrast, vanadium flow
batteries have a lower cycle efficiency, meaning that they waste more
energy in the storage process, but their lower projected cost compensates
for their poorer energy performance.
The third research question was related to the energy performance
analysis of biogas upgrading through high-temperature electrolysis and
methanation. The main outcome of the process is synthetic natural gas, but
also pure oxygen and excess heat are available as by-products. The
configuration with the electrolyzer operating at high pressure (around 30
bar) was the one characterized by the highest efficiency, even though
pressurizing the stack of SOECs increases the complexity and the cost. The
comparison with other techniques to upgrade biogas via carbon dioxide
separation revealed that the electrochemical upgrade reaches higher yields
for the same inlet biogas, even though it requires higher electrical inputs.
The profitability analysis of such process was part of the fourth research
question. Considering a low electricity price and a favorable estimation of
the price per ton of CO2 on the emission trading market, the production
cost of upgraded biogas was comparable with the price that medium
industrial consumers had to pay for natural gas in 2013, when the oil and
natural gas prices reached their maxima. Although, this analysis is referred
to a convenient future scenario, it shows that a viable pathway for CO2
recycle and synthetic fuel production is profitable, provided that a high
degree of process integration is adopted. Moreover, methanation-based
processes are exothermic and this constitutes a good basis for their further
integration with district heating or cooling network.
The adoption of electrochemical energy storage processes and
technologies proved to be able to mitigate some of the problems caused by
the increasing presence of renewables in the electricity generation mix.
Cost-wise the energy and environmental medium-term benefits have been
demonstrated for different storage applications and this validates the
efforts of a large part of the scientific and industrial community for the
improvement of such pathways. The main reasons that are restraining the
massive diffusion of storage solutions are: the still limited shares of
intermittent renewable energies in most of the world countries and the
high investment cost. In just a few of the advanced countries the share of
CONCLUSIONS | 95
variable renewable energy in power generation is above 20%, which is a
value for which the already present pumped hydroelectric plants and the
high degree of interconnection can handle the variability of renewables
[194]. Among the countries with higher share of renewables in the power
system, some of them, e.g. Portugal, have been able to incorporate large
quantities of renewable energies without curtailments, because of the high
number of hydro pumped plants [47].
The not-yet-competitive cost of storage solutions is an issue that has
been confirmed by all the case studies reported in this thesis, where the
economic evaluations have been performed based on projected costs. Both
batteries and electrolyzers produce good economic results if their capital
expenditures are strongly reduced compared with the current values and if
the price of fossil fuels is higher than the present level. The analysis of
small- and large-scale battery systems highlighted their high efficiency (up
to 95%), meaning that less energy is lost during the storage cycles in
comparison with other options, e.g. pumped hydro energy storage (PHES)
and compressed air energy storage (CAES)35.
The exergy analysis of the power-to-gas route through biogas upgrading
showed that the conversion efficiency of such a process can reach values
higher than 80%. This is favored by the use of the heat produced by the
process, which is in excess compared to the need of the process itself. This
fact favors the further integration with processes that have thermal needs
to satisfy (e.g. the anaerobic digester or the nearby buildings). On the other
hand, the economic evaluation revealed that, albeit with low-cost
electricity, a low price of natural gas, as the one in 2015, hinders
irremediably the profitability of the production of synthetic natural gas.
However, the combination of higher natural gas price, economic
exploitation of the by-products and participation to the carbon allowance
market can make this process competitive.
The main environmental advantage of renewable gaseous fuels lies in
the possibility to recycle residues and waste, which is in accordance with
the targets contained in the Circular Economy Package approved by the
European Parliament. In particular, 65% of municipal solid waste should
be recycled by 2035 and only a maximum of 10% could be landfilled, of
which the biodegradable fraction, useful for biogas production, will have to
be either collected separately or recycled at home through composting
35 The round trip efficiency of PHES is in the range 65-85% [7], while for CAES it can reach 70% [8].
96 | CONCLUSIONS
[195]. The integration of the disposal process with the fuel synthesis can
solve two problems at the same time, namely waste treatment and
decarbonization of large energy consumption sectors (e.g. transport), by
reducing their pressure on the environment. These strategies are
functional to follow the long-term roadmap that the European Commission
defined to promote the reduction of greenhouse gas emissions [196]. It is
true that the EU emissions account only for 9% of all the emissions related
to fossil-fuel combustion, cement production and gas flaring in the world
[197]; however, the fulfilment of the goals that have been established to
reduce the dependence on fossil fuels and to lower environmental impacts
of human activities, can be an inspiring example for others to follow.
Finally, the connection between the power and the transport sectors is
becoming more and more evident both for the diffusion of electric mobility
solutions (not only cars but also scooters and, to some extent buses [198]),
and for the rising interest in electrofuel production. Electric mobility seems
more suitable for urban contexts because it does not produce local
emissions and the issues connected to limited mileage can be solved
through a dense network of charging opportunities. On the other hand, for
long-distance travel and heavy-duty vehicles the higher energy density of
fuels is a feature that appears difficult to be replaced by electric vehicles.
Therefore, electrochemical energy storage solutions seem to have a large
potential to promote the development of greener and more sustainable
consumption patterns.
5.1. Future Work
This doctoral thesis started when the topic of energy storage was
considered a challenge that energy systems were going to face in a near
future, though this future was still quite far to come. After only four years
the interest in this topic has grown exponentially, especially driven by the
dramatic reduction in the cost of solar PV, wind, and battery systems. This
is proved by the huge amount of peer-reviewed researches published on
the topic of energy storage. They increased exponentially passing from
1500 in 2010 to 7249 in 201736 and this testifies the importance that the
scientific community is attributing to the topic. These favorable conditions
have triggered the growth of renewable power plants in many countries.
36 These numbers are based on a search in Scopus among the peer-reviewed papers and book chapters with the keyword “energy storage”.
CONCLUSIONS | 97
That being said, some of the problems connected to the integration of
variable renewables are still open and the possibility for the study of
interesting applications is unprecedented. Thereby, the development of
ecologic and cost-effective solutions to accompany the transition in the
direction of an energy production paradigm less dependent on fossil fuels
is one of the priorities of the current innovation trends.
In general, isolated or remote systems should be regarded as test beds
for innovative solutions. The exploitation of local energy resources is vital
to reach a higher degree of sustainability of the local energy system. Many
islands can count on high solar irradiation levels, on wind blowing for most
of the year, and on wave and tidal energy. All these resources are
intermittent and should be complemented with storage systems to fully
exploit their decarbonization potential for the generation system.
Therefore, additional efforts should be put into the design of effective
pathways that can promote the use of local resources, also through the
actual implementation of pilot projects to discover the challenges related
to the peculiarities of each case.
For power-to-gas processes using biogas, the main need is the scale-up
from the proof-of-concept stage to a more reliable system, which is able to
work in real-time conditions. Some projects of this kind are already
running [199], but most of them use alkaline electrolyzers, while an
additional advantage could be brought by the use of high-temperature
electrolyzers. The studies presented in this thesis are the result of a precise
of an accurate modeling effort, which would greatly benefit from an
experimental campaign to validate and fine-tune it. Moreover, market
mechanisms to make the use of surplus electricity profitable should be
studied, as a means to reward the activity of the processes that are able to
act as buffers during hours of extra-production. Some initial attempts have
been done as in Larscheid et al. [200], but more efforts are needed to define
how to efficiently couple the electricity and the natural gas markets to allow
power-to-gas to be profitable.
Finally, the integration of power system and transportation should be
further investigated both for islands and for continental settings. Strategies
to favor the rational management of mobility demand (i.e. vehicle and ride
sharing, carpooling, and public transport optimization) should be coupled
with a sustainable procurement of the necessary energy resources. In other
words, the systemic reduction of the energy consumption in transport is as
important as finding new low-environmental-impact energy carriers. In
98 | CONCLUSIONS
particular, the implementation of power-to-gas routes should be
considered as a possibility to pare back the quantity of organic waste that
have to be landfilled. Together with this, an analysis of the potential raw
materials locally available is important to define the extent up to which
local resources can serve the aim of decarbonization. This step is complex
because it needs an accurate inventory of crops, agricultural and industrial
residues and other material streams, the data of which are usually not
easily available. The choice of isolated systems facilitates the definition of
the boundaries and, therefore, can provide useful insights on the further
developments of the interrelation between energy storage and
transportation services.
REFERENCES| 99
References
[1] Eurostat. Eurostat Shares 2018. http://ec.europa.eu/eurostat/web/energy/data/shares (accessed January 28, 2018).
[2] Holttinen H, Milligan M, Ela E, Menemenlis N, Dobschinski J, Rawn B, et al. Methodologies to determine operating reserves due to increased wind power. IEEE Trans Sustain Energy 2012;3:713–23. doi:10.1109/TSTE.2012.2208207.
[3] Rejc ŽB, Čepin M. Estimating the additional operating reserve in power systems with installed renewable energy sources. Int J Electr Power Energy Syst 2014;62:654–64. doi:10.1016/j.ijepes.2014.05.019.
[4] DoE. Global Energy Storage Database 2018. [5] Rogner H-H, Aguilera RF, Archer C, Bertani R, Bhattacharya, S.C.,
Dusseault MB. Chapter7 - Energy resources and potentials. Glob. Energy Assessment—Toward a Sustain. Futur., Cambridge, UK and New York, NY, USA and the International Institute for applied systems analysis, Laxenburg, Austria. Cambridge University Press; 2012, p. 423–512.
[6] Christen T, Carlen MW. Theory of ragone plots. J Power Sources 2000;91:210–6. doi:10.1016/S0378-7753(00)00474-2.
[7] Aneke M, Wang M. Energy storage technologies and real life applications – A state of the art review. Appl Energy 2016;179:350–77. doi:10.1016/j.apenergy.2016.06.097.
[8] Luo X, Wang J, Dooner M, Clarke J. Overview of current development in electrical energy storage technologies and the application potential in power system operation. Appl Energy 2015;137:511–36. doi:10.1016/j.apenergy.2014.09.081.
[9] Kepplinger P, Huber G, Petrasch J. Autonomous optimal control for demand side management with resistive domestic hot water heaters using linear optimization. Energy Build 2015;100:50–5. doi:10.1016/j.enbuild.2014.12.016.
[10] Kepplinger P, Huber G, Petrasch J. Field testing of demand side management via autonomous optimal control of a domestic hot water heater. Energy Build 2016;127:730–5. doi:10.1016/j.enbuild.2016.06.021.
[11] Kim Y, Norford LK. Optimal use of thermal energy storage resources in commercial buildings through price-based demand response considering distribution network operation. Appl Energy 2017;193:308–24. doi:10.1016/j.apenergy.2017.02.046.
[12] European Commission. The European Union leading in renewables. 2015. [13] European Commission. The Strategic Energy Technology (SET) Plan.
Publications Office of the European Union,; 2017. doi:10.2777/48982. [14] Shah V, Booream-Phelps J. Crossing the Chasm. Dtsch Bank – Mark Res
2015:185.
100 | REFERENCES
doi:http://www.climatecouncil.org.au/uploads/ed4518226c655546cc529390c7cd4a8f.pdf.
[15] European Comission. Best practices on Renewable Energy Self-consumption. J Chem Inf Model 2015;53:1689–99. doi:10.1017/CBO9781107415324.004.
[16] TechNavio. Global Residential Solar Energy Storage Market 2017-2021, Technavio (2017). 2017.
[17] Ranaweera I, Midtgård OM, Korpås M. Distributed control scheme for residential battery energy storage units coupled with PV systems. Renew Energy 2017;113:1099–110. doi:10.1016/j.renene.2017.06.084.
[18] O’Dwyer C, Ryan L, Flynn D. Efficient large-scale energy storage dispatch: challenges in future high renewables systems. IEEE Trans Power Syst 2017;32:3439–50. doi:10.1109/TPWRS.2017.2656245.
[19] Lopes AS, Castro R, Silva C. Contributions to the Design of a Water Pumped Storage System in an Isolated Power System with High Penetration of Wind Energy. Technol Innov Cyber-Physical Syst 2016;470:378–86. doi:10.1007/978-3-319-31165-4.
[20] Ma T, Yang H, Lu L. A feasibility study of a stand-alone hybrid solar-wind-battery system for a remote island. Appl Energy 2014;121:149–58. doi:10.1016/j.apenergy.2014.01.090.
[21] Ellison JF, Rashkin LJ, Serio J, Byrne RH. The benefits of grid-scale storage on Oahu. J Energy Storage 2018;15:336–44. doi:10.1016/j.est.2017.12.009.
[22] Guandalini G, Campanari S, Romano MC. Power-to-gas plants and gas turbines for improved wind energy dispatchability: Energy and economic assessment. Appl Energy 2015;147:117–30. doi:10.1016/j.apenergy.2015.02.055.
[23] Lyseng B, Niet T, English J, Keller V, Palmer-Wilson K, Robertson B, et al. System-level power-to-gas energy storage for high penetrations of variable renewables. Int J Hydrogen Energy 2018;43:1966–79. doi:10.1016/j.ijhydene.2017.11.162.
[24] Varone A, Ferrari M. Power to liquid and power to gas: An option for the German Energiewende. Renew Sustain Energy Rev 2015;45:207–18. doi:10.1016/j.rser.2015.01.049.
[25] Mazza A, Bompard E, Chicco G. Applications of power to gas technologies in emerging electrical systems. Renew Sustain Energy Rev 2018;92:794–806. doi:10.1016/j.rser.2018.04.072.
[26] Peduzzi E, Boissonnet G, Maréchal F. Screening methodology for biorefinery and power to chemicals concepts. Comput Aided Chem Eng 2016;38:1893–8. doi:10.1016/B978-0-444-63428-3.50320-9.
[27] Gutiérrez-Martín F, Rodríguez-Antón LM. Power-to-SNG technology for energy storage at large scales. Int J Hydrogen Energy 2016;41:19290–303. doi:10.1016/j.ijhydene.2016.07.097.
[28] Blanco H, Faaij A. A review at the role of storage in energy systems with a focus on Power to Gas and long-term storage. Renew Sustain Energy Rev 2018;81:1049–86. doi:10.1016/j.rser.2017.07.062.
[29] Clausen LR. Energy efficient thermochemical conversion of very wet biomass to biofuels by integration of steam drying, steam electrolysis and gasification. Energy 2017;125:327–36. doi:10.1016/j.energy.2017.02.132.
REFERENCES| 101
[30] Qian H, Zhu W, Liu C, Lu X, Kontogeorgis GM, Gani R. Exergy efficiency based design and analysis of utilization pathways of biomasses. vol. 43. Elsevier Masson SAS; 2018. doi:10.1016/B978-0-444-64235-6.50150-9.
[31] Turk D. Energy storage and system integration – an international perspective. In: IEA, editor. Sect. Integr. Support. by Energy Storage Hydrog. High Lev. Roundtable, 2018.
[32] Schmidt O, Hawkes A, Gambhir A, Staffell I. The future cost of electrical energy storage based on experience rates. Nat Energy 2017;6:17110. doi:10.1038/nenergy.2017.110.
[33] Petrakopoulou F, Sanz-Bermejo J, Dufour J, Romero M. Exergetic analysis of hybrid power plants with biomass and photovoltaics coupled with a solid-oxide electrolysis system. Energy 2016;94:304–15. doi:10.1016/j.energy.2015.10.118.
[34] Liu M, Steven Tay NH, Bell S, Belusko M, Jacob R, Will G, et al. Review on concentrating solar power plants and new developments in high temperature thermal energy storage technologies. Renew Sustain Energy Rev 2016;53:1411–32. doi:10.1016/j.rser.2015.09.026.
[35] Sarbu I, Sebarchievici C. Thermal Energy Storage. 2017. doi:10.1016/B978-0-12-811662-3.00004-9.
[36] Buttler A, Spliethoff H. Current status of water electrolysis for energy storage, grid balancing and sector coupling via power-to-gas and power-to-liquids: A review. Renew Sustain Energy Rev 2018;82:2440–54. doi:10.1016/j.rser.2017.09.003.
[37] Pozzo M, Lanzini A, Santarelli M. Enhanced biomass-to-liquid (BTL) conversion process through high temperature co-electrolysis in a solid oxide electrolysis cell (SOEC). Fuel 2015;145:39–49. doi:10.1016/j.fuel.2014.12.066.
[38] Ail SS, Dasappa S. Biomass to liquid transportation fuel via Fischer Tropsch synthesis - Technology review and current scenario. Renew Sustain Energy Rev 2016;58:267–86. doi:10.1016/j.rser.2015.12.143.
[39] Herz G, Reichelt E, Jahn M. Techno-economic analysis of a co-electrolysis-based synthesis process for the production of hydrocarbons. Appl Energy 2018;215:309–20. doi:10.1016/j.apenergy.2018.02.007.
[40] Becker WL, Braun RJ, Penev M, Melaina M. Production of Fischer-Tropsch liquid fuels from high temperature solid oxide co-electrolysis units. Energy 2012;47:99–115. doi:10.1016/j.energy.2012.08.047.
[41] Lund PD, Lindgren J, Mikkola J, Salpakari J. Review of energy system flexibility measures to enable high levels of variable renewable electricity. Renew Sustain Energy Rev 2015;45:785–807. doi:10.1016/j.rser.2015.01.057.
[42] Ehlers N. The German energy transition & integration of renewable energy 2017.
[43] Eurostat. Infrastructure - electricity - annual data [nrg_113a] 2018. https://ec.europa.eu/eurostat/statistics-explained/index.php/Energy_from_renewable_sources (accessed August 30, 2018).
[44] IHA. Hydropower in Germany 2017. https://www.hydropower.org/country-profiles/germany (accessed April 25, 2018).
102 | REFERENCES
[45] Net installed electricity generation capacity in Germany. Fraunhofer ISE 2018. https://www.energy-charts.de/power_inst.htm (accessed April 4, 2018).
[46] Federal Network Agency. Monitoring report 2017. 2017. [47] WindEurope. WindEurope views on curtailment of wind power and its link
to the priority of dispatch. 2016. [48] Buijs P, Bekaert D, Cole S, Van Hertem D, Belmans R. Transmission
investment problems in Europe: Going beyond standard solutions. Energy Policy 2011;39:1794–801. doi:10.1016/j.enpol.2011.01.012.
[49] Akinyele DO, Rayudu RK. Review of energy storage technologies for sustainable power networks. Sustain Energy Technol Assessments 2014;8:74–91. doi:10.1016/j.seta.2014.07.004.
[50] Bollen MHJ. What is power quality? Electr Power Syst Res 2003;66:5–14. doi:10.1016/S0378-7796(03)00067-1.
[51] Denholm P, Ela E, Kirby B, Milligan M. The Role of Energy Storage with Renewable Electricity Generation. 2010. doi:69.
[52] Grid Scale Energy Storage Systems 2018. http://www.mpoweruk.com/grid_storage.htm (accessed March 15, 2018).
[53] Guney MS, Tepe Y. Classification and assessment of energy storage systems. Renew Sustain Energy Rev 2017;75:1187–97. doi:10.1016/j.rser.2016.11.102.
[54] Yang Y, Bremner S, Menictas C, Kay M. Battery energy storage system size determination in renewable energy systems: A review. Renew Sustain Energy Rev 2018;91:109–25. doi:10.1016/j.rser.2018.03.047.
[55] Divya KC, Østergaard J. Battery energy storage technology for power systems-An overview. Electr Power Syst Res 2009;79:511–20. doi:10.1016/j.epsr.2008.09.017.
[56] Chen H, Cong TN, Yang W, Tan C, Li Y, Ding Y. Progress in electrical energy storage system: A critical review. Prog Nat Sci 2009;19:291–312. doi:10.1016/j.pnsc.2008.07.014.
[57] Arenas LF, Ponce de León C, Walsh FC. Engineering aspects of the design, construction and performance of modular redox flow batteries for energy storage. J Energy Storage 2017;11:119–53. doi:10.1016/j.est.2017.02.007.
[58] Doetsch C, Burfeind J. Vanadium Redox Flow Batteries. Storing Energy, Elsevier Inc.; 2016, p. 227–46. doi:10.1016/B978-0-12-803440-8/00012-9.
[59] Cao L, Skyllas-Kazacos M, Menictas C, Noack J. A review of electrolyte additives and impurities in vanadium redox flow batteries. J Energy Chem 2018;0:1–23. doi:10.1016/j.jechem.2018.04.007.
[60] Choi C, Kim S, Kim R, Choi Y, Kim S, Jung H young, et al. A review of vanadium electrolytes for vanadium redox flow batteries. Renew Sustain Energy Rev 2017;69:263–74. doi:10.1016/j.rser.2016.11.188.
[61] Liu T, Li X, Zhang H, Chen J. Progress on the electrode materials towards vanadium flow batteries (VFBs) with improved power density. J Energy Chem 2018. doi:10.1016/j.jechem.2018.07.003.
[62] Vanýsek P, Novák V. Redox flow batteries as the means for energy storage. J Energy Storage 2017;13:435–41. doi:10.1016/j.est.2017.07.028.
[63] Götz M, Lefebvre J, Mörs F, McDaniel Koch A, Graf F, Bajohr S, et al. Renewable Power-to-Gas: A technological and economic review. Renew
REFERENCES| 103
Energy 2016;85:1371–90. doi:10.1016/j.renene.2015.07.066. [64] Rauch R, Kiennemann A, Sauciuc A. Fischer-Tropsch Synthesis to Biofuels
(BtL Process). vol. #volume#. © 2013 Elsevier B.V. All rights reserved.; 2013. doi:10.1016/B978-0-444-56330-9.00012-7.
[65] Judd R, Pinchbeck D. Hydrogen admixture to the natural gas grid. Compend Hydrog Energy 2016:165–92. doi:10.1016/B978-1-78242-364-5.00008-7.
[66] Wang L, Pérez-Fortes M, Madi H, Diethelm S, herle J Van, Maréchal F. Optimal design of solid-oxide electrolyzer based power-to-methane systems: A comprehensive comparison between steam electrolysis and co-electrolysis. Appl Energy 2018;211:1060–79. doi:10.1016/j.apenergy.2017.11.050.
[67] Laguna-Bercero MA. Recent advances in high temperature electrolysis using solid oxide fuel cells: A review. J Power Sources 2012;203:4–16. doi:10.1016/j.jpowsour.2011.12.019.
[68] Mingyi L, Bo Y, Jingming X, Jing C. Thermodynamic analysis of the efficiency of high-temperature steam electrolysis system for hydrogen production. J Power Sources 2008;177:493–9. doi:10.1016/j.jpowsour.2007.11.019.
[69] Sun X, Chen M, Jensen SH, Ebbesen SD, Graves C, Mogensen M. Thermodynamic analysis of synthetic hydrocarbon fuel production in pressurized solid oxide electrolysis cells. Int J Hydrogen Energy 2012;37:17101–10. doi:10.1016/j.ijhydene.2012.08.125.
[70] Brynolf S, Taljegard M, Grahn M, Hansson J. Electrofuels for the transport sector: A review of production costs. Renew Sustain Energy Rev 2018;81:1887–905. doi:10.1016/j.rser.2017.05.288.
[71] Knupfer SM, Hensley R, Hertzke P, Schaufuss P, Laverty N, Kramer N. Electrifying Insights: How Automakers Can Drive Electrified Vehicle Sales and Profitability. 2017.
[72] EASE – Delta-ee. European Market Monitor on Energy Storage - Second Edition (EMMES 2.0). 2018.
[73] Briano J, Báez M, Morales R. Photovoltaic Grid Parity Monitor - Residential. Berlin: 2015.
[74] Gitizadeh M, Fakharzadegan H. Battery capacity determination with respect to optimized energy dispatch schedule in grid-connected photovoltaic (PV) systems. Energy 2014;65:665–74. doi:10.1016/j.energy.2013.12.018.
[75] Dufo-López R, Bernal-Agustín JL. Techno-economic analysis of grid-connected battery storage. Energy Convers Manag 2015;91:394–404. doi:10.1016/j.enconman.2014.12.038.
[76] Ratnam EL, Weller SR, Kellett CM. Scheduling residential battery storage with solar PV: Assessing the benefits of net metering. Appl Energy 2015;155:881–91. doi:10.1016/j.apenergy.2015.06.061.
[77] Mulder G, Six D, Claessens B, Broes T, Omar N, Mierlo J Van. The dimensioning of PV-battery systems depending on the incentive and selling price conditions. Appl Energy 2013;111:1126–35. doi:10.1016/j.apenergy.2013.03.059.
[78] Nottrott A, Kleissl J, Washom B. Energy dispatch schedule optimization and cost benefit analysis for grid-connected, photovoltaic-battery storage
104 | REFERENCES
systems. Renew Energy 2013;55:230–40. doi:10.1016/j.renene.2012.12.036.
[79] Gutiérrez-Alcaraz G, Tovar-Hernández JH, Lu C-N. Effects of demand response programs on distribution system operation. Int J Electr Power Energy Syst 2016;74:230–7. doi:10.1016/j.ijepes.2015.07.018.
[80] Koliou E, Bartusch C, Picciariello A, Eklund T, Söder L, Hakvoort RA. Quantifying distribution-system operators’ economic incentives to promote residential demand response. Util Policy 2015;35:28–40. doi:10.1016/j.jup.2015.07.001.
[81] D’hulst R, Labeeuw W, Beusen B, Claessens S, Deconinck G, Vanthournout K. Demand response flexibility and flexibility potential of residential smart appliances: Experiences from large pilot test in Belgium. Appl Energy 2015;155:79–90. doi:10.1016/j.apenergy.2015.05.101.
[82] Muratori M, Schuelke-Leech BA, Rizzoni G. Role of residential demand response in modern electricity markets. Renew Sustain Energy Rev 2014;33:546–53. doi:10.1016/j.rser.2014.02.027.
[83] Vanthournout K, Dupont B, Foubert W, Stuckens C, Claessens S. An automated residential demand response pilot experiment, based on day-ahead dynamic pricing. Appl Energy 2015;155:195–203. doi:10.1016/j.apenergy.2015.05.100.
[84] Wang X, Palazoglu A, El-Farra NH. Operational optimization and demand response of hybrid renewable energy systems. Appl Energy 2015;143:324–35. doi:10.1016/j.apenergy.2015.01.004.
[85] Huang Y, Tian H, Wang L. Demand response for home energy management system. Int J Electr Power Energy Syst 2015;73:448–55. doi:10.1016/j.ijepes.2015.05.032.
[86] Zame KK, Brehm CA, Nitica AT, Richard CL, Schweitzer GD. Smart grid and energy storage: Policy recommendations. Renew Sustain Energy Rev 2018;82:1646–54. doi:10.1016/j.rser.2017.07.011.
[87] Song J, Oh SD, Yoo Y, Seo SH, Paek I, Song Y, et al. System design and policy suggestion for reducing electricity curtailment in renewable power systems for remote islands. Appl Energy 2018;225:195–208. doi:10.1016/j.apenergy.2018.04.131.
[88] Neves D, Silva CA, Connors S. Design and implementation of hybrid renewable energy systems on micro-communities: A review on case studies. Renew Sustain Energy Rev 2014;31:935–46. doi:10.1016/j.rser.2013.12.047.
[89] Kuang Y, Zhang Y, Zhou B, Li C, Cao Y, Li L, et al. A review of renewable energy utilization in islands. Renew Sustain Energy Rev 2016;59:504–13. doi:10.1016/j.rser.2016.01.014.
[90] Raju L, Morais AA, Rathnakumar R, Ponnivalavan S, Thavam LD. Micro-grid grid outage management using multi-agent systems. Proc - 2017 2nd Int Conf Recent Trends Challenges Comput Model ICRTCCM 2017 2017;117:363–8. doi:10.1109/ICRTCCM.2017.21.
[91] Stenzel P, Schreiber A, Marx J, Wulf C, Schreieder M, Stephan L. Environmental Impacts of Electricity Generation for Graciosa Island , Azores Location of the Azores and Graciosa Island Graciosa Island : J Energy Storage 2017;15:292–303. doi:10.1016/j.est.2017.12.002.
[92] Castillo A, Gayme DF. Grid-scale energy storage applications in renewable
REFERENCES| 105
energy integration: A survey. Energy Convers Manag 2014;87:885–94. doi:10.1016/j.enconman.2014.07.063.
[93] Lin Y, Johnson JX, Mathieu JL. Emissions impacts of using energy storage for power system reserves. Appl Energy 2016;168:444–56. doi:10.1016/j.apenergy.2016.01.061.
[94] Lorenzi G, Silva CAS. Comparing demand response and battery storage to optimize self-consumption in PV systems. Appl Energy 2016;180. doi:10.1016/j.apenergy.2016.07.103.
[95] Zhang C, Zhao TS, Xu Q, An L, Zhao G. Effects of operating temperature on the performance of vanadium redox flow batteries. Appl Energy 2015;155:349–53. doi:10.1016/j.apenergy.2015.06.002.
[96] López-Vizcaíno R, Mena E, Millán M, Rodrigo MA, Lobato J. Performance of a vanadium redox flow battery for the storage of electricity produced in photovoltaic solar panels. Renew Energy 2017;114:1123–33. doi:10.1016/j.renene.2017.07.118.
[97] Sadeghi S. Study using the flow battery in combination with solar panels and solid oxide fuel cell for power generation. Sol Energy 2018;170:732–40. doi:10.1016/j.solener.2018.05.091.
[98] Alotto P, Guarnieri M, Moro F. Redox flow batteries for the storage of renewable energy: A review. Renew Sustain Energy Rev 2014;29:325–35. doi:10.1016/j.rser.2013.08.001.
[99] Minke C, Kunz U, Turek T. Techno-economic assessment of novel vanadium redox flow batteries with large-area cells. J Power Sources 2017;361:105–14. doi:10.1016/j.jpowsour.2017.06.066.
[100] Eber K, Corbus D. Hawaii Solar Integration Study. 2013. [101] Woodford D. Oahu Wind Integration and Transmission Study. 2011. [102] Komor P, Glassmaire J. Electricity Storage and Renewables for Island
Power. 2012. [103] Sioshansi FP. National Energy Roadmaps for Islands. 2013. [104] Neves D, Silva CA. Optimal electricity dispatch on isolated mini-grids using
a demand response strategy for thermal storage backup with genetic algorithms. Energy 2015;82:436–45. doi:10.1016/j.energy.2015.01.054.
[105] Pina A, Baptista P, Silva C, Ferrão P. Energy reduction potential from the shift to electric vehicles: The Flores island case study. Energy Policy 2014;67:37–47. doi:10.1016/j.enpol.2013.07.120.
[106] Gahleitner G. Hydrogen from renewable electricity: An international review of power-to-gas pilot plants for stationary applications. Int J Hydrogen Energy 2013;38:2039–61. doi:10.1016/j.ijhydene.2012.12.010.
[107] Graves C, Ebbesen SD, Mogensen M, Lackner KS. Sustainable hydrocarbon fuels by recycling CO2 and H2O with renewable or nuclear energy. Renew Sustain Energy Rev 2011;15:1–23. doi:10.1016/j.rser.2010.07.014.
[108] Gondal IA, Masood SA, Khan R. Green hydrogen production potential for developing a hydrogen economy in Pakistan. Int J Hydrogen Energy 2018;43:6011–39. doi:10.1016/j.ijhydene.2018.01.113.
[109] Ridjan I, Mathiesen BV, Connolly D. Terminology used for renewable liquid and gaseous fuels based on the conversion of electricity: A review. J Clean Prod 2016;112:3709–20. doi:10.1016/j.jclepro.2015.05.117.
[110] Reiter G, Lindorfer J. Evaluating CO2 sources for power-to-gas applications – A case study for Austria. J CO2 Util 2015;10:40–9.
106 | REFERENCES
doi:10.1016/j.jcou.2015.03.003. [111] Elfving J, Bajamundi C, Kauppinen J, Sainio T. Modelling of equilibrium
working capacity of PSA, TSA and TVSA processes for CO2adsorption under direct air capture conditions. J CO2 Util 2017;22:270–7. doi:10.1016/j.jcou.2017.10.010.
[112] Ryckebosch E, Drouillon M, Vervaeren H. Techniques for transformation of biogas to biomethane. Biomass and Bioenergy 2011;35:1633–45. doi:10.1016/j.biombioe.2011.02.033.
[113] Vagonyte E. Biogas & Biomethane in Europe. 2011. doi:247896520. [114] Angelidaki I, Treu L, Tsapekos P, Luo G, Campanaro S, Wenzel H, et al.
Biogas upgrading and utilization: Current status and perspectives. Biotechnol Adv 2018:0–1. doi:10.1016/j.biotechadv.2018.01.011.
[115] Rotunno P, Lanzini A, Leone P. Energy and economic analysis of a water scrubbing based biogas upgrading process for biomethane injection into the gas grid or use as transportation fuel. Renew Energy 2017;102:417–32. doi:10.1016/j.renene.2016.10.062.
[116] Baker RW. Membrane Technology and Applications. third. Wiley; 2012. doi:10.1002/0471238961.1305130202011105.a01.
[117] Scholz M, Melin T, Wessling M. Transforming biogas into biomethane using membrane technology. Renew Sustain Energy Rev 2013;17:199–212. doi:10.1016/j.rser.2012.08.009.
[118] Augelletti R, Conti M, Annesini MC. Pressure swing adsorption for biogas upgrading. A new process configuration for the separation of biomethane and carbon dioxide. J Clean Prod 2017;140:1390–8. doi:10.1016/j.jclepro.2016.10.013.
[119] Graves C, Ebbesen SD, Mogensen M. Co-electrolysis of CO2 and H2O in solid oxide cells: Performance and durability. Solid State Ionics 2011;192:398–403. doi:10.1016/j.ssi.2010.06.014.
[120] Li W, Wang H, Shi Y, Cai N. Performance and methane production characteristics of H2O- CO2co-electrolysis in solid oxide electrolysis cells. Int J Hydrogen Energy 2013;38:11104–9. doi:10.1016/j.ijhydene.2013.01.008.
[121] Giglio E, Lanzini A, Santarelli M, Leone P. Synthetic natural gas via integrated high-temperature electrolysis and methanation: Part I—Energy performance. J Energy Storage 2015;1:22–37. doi:10.1016/j.est.2015.04.002.
[122] De Saint Jean M, Baurens P, Bouallou C, Couturier K. Economic assessment of a power-to-substitute-natural-gas process including high-temperature steam electrolysis. Int J Hydrogen Energy 2015;40:6487–500. doi:10.1016/j.ijhydene.2015.03.066.
[123] From solid fuels to substitute natural gas (SNG) using TREMPTM n.d. [124] Collet P, Flottes E, Favre A, Raynal L, Pierre H, Capela S, et al. Techno-
economic and Life Cycle Assessment of methane production via biogas upgrading and power to gas technology. Appl Energy 2017;192:282–95. doi:10.1016/j.apenergy.2016.08.181.
[125] Irradiation Data for Lisbon n.d. http://www.wunderground.com/personal-weather-station/dashboard?ID=ILISBONL3#history/s20150219/e20150226/mweek (accessed June 26, 2015).
REFERENCES| 107
[126] REN, Perfis de Consumo n.d. http://www.mercado.ren.pt/PT/Electr/InfoMercado/Consumo/Paginas/PerfisConsumo.aspx.
[127] Paleta R, Pina A, Santos Silva CA. Polygeneration Energy Container: Designing and Testing Energy Services for Remote Developing Communities. IEEE Trans Sustain Energy 2014;5:1–1. doi:10.1109/TSTE.2014.2308017.
[128] Lorenzi G, Silva CAS. Techno-economic comparison of storage vs . demand response strategies in distributed generation systems. 12th Int. Conf. Eur. Energy Mark., 2015.
[129] Pordata. Average Number of People per Household 2016. http://www.pordata.pt/en/Europe/Average+number+of+persons+per+household-1613.
[130] Pombeiro H, Pina A, Silva C. Analyzing Residential Electricity Consumption Patterns Based on Consumer’ s Segmentation. Procs. 1st Int’l Work. Inf. Technol. Energy Appl., n.d.
[131] ERSE (Entitade Regulatoria dos Serviços Energéticos). Caracterização da procura de energia elétrica em 2012 (Document in Portuguese). 2013.
[132] ERSE (Entitade Regulatoria dos Serviços Energéticos). Preços De Referência No Mercado Liberalizado De Energia Elétrica E Gás Natural (Document in Portuguese). 2015.
[133] Ministry of Environment Land Management and Energy. Decreed Law n.o 153/2014 - 20 October 2014. 2014.
[134] Crispim J, Carreira M, Rui C. Validation of photovoltaic electrical models against manufacturers data and experimental results. POWERENG 2007 - Int. Conf. Power Eng. - Energy Electr. Drives Proc., 2007, p. 556–61. doi:10.1109/POWERENG.2007.4380161.
[135] LDK 250P-20 n.d. http://www.ldksolar.com/uploadfiles/down/LDK_215P_250P_20_EN_V1_12_120229.pdf (accessed June 27, 2016).
[136] Dolara A, Leva S, Manzolini G. Comparison of different physical models for PV power output prediction. Sol Energy 2015;119:83–99. doi:10.1016/j.solener.2015.06.017.
[137] Mark D. Hopkins, Pahwa A, Easton T. Intelligent Dispatch for Distributed Renewable Resources. IEEE Trans Smart Grid 2012;3:1047–54.
[138] Fronius Symo Hybrid 5.0-3-S Datasheet n.d. https://www.fronius.com/cps/rde/xchg/SID-450DE065-4091E8DA/fronius_international/hs.xsl/83_35331_ENG_HTML.htm#.Vj-6LLerTIU (accessed June 27, 2016).
[139] Collaboration between Tesla and Fronius n.d. http://www.fronius.com/cps/rde/xbcr/SID-9AAC2BC5-F7233319/fronius_international/SE_MAIL_Newsletter_ROW_May_2015_579409_snapshot.pdf (accessed June 27, 2016).
[140] Kittner N, Lill F, Kammen DM. Energy storage deployment and innovation for the clean energy transition. Nat Energy 2017;2:1–6. doi:10.1038/nenergy.2017.125.
[141] TESLA MOTORS. Tesla Powerwall. Energy Storage a Sustain Home 2016. https://www.teslamotors.com/powerwall?redirect=no (accessed June 27, 2016).
108 | REFERENCES
[142] Wholesalesolar n.d. http://www.wholesalesolar.com/9905042/concorde-sun-xtender/batteries/concorde-sun-xtender-pvx-2120l-agm-battery (accessed June 27, 2016).
[143] Fronius Symo Hybrid 5.0-3-S Price n.d. http://www.europe-solarstore.com/solar-inverters/fronius/fronius-symo-hybrid/fronius-symo-hybrid-5-0-3-s.html (accessed June 27, 2016).
[144] SMA Sunny Island 3.0M Price n.d. http://www.mg-solar-shop.de/solar-inverter/sma-solar-inverter/SMA-Sunny-Island-SI-3-0-M-11-with-SRC-20.html (accessed June 17, 2016).
[145] Ruelens F, Claessens BJ, Vandael S, Iacovella S, Vingerhoets P, Belmans R. Demand response of a heterogeneous cluster of electric water heaters using batch reinforcement learning. 2014 Power Syst Comput Conf 2014:1–7. doi:10.1109/PSCC.2014.7038106.
[146] Fuentes E, Arce L, Salom J. A review of domestic hot water consumption profiles for application in systems and buildings energy performance analysis. Renew Sustain Energy Rev 2018;81:1530–47. doi:10.1016/j.rser.2017.05.229.
[147] European Commision. Supplement to the European Parliament directive regarding energy labelling of water heaters. 2013:77.
[148] Abrantes J. Avaliação técnica e económica da aplicação de sistemas Waste to Energy no tratamento de resíduos urbanos em aglomerados de média e pequena dimensão. Instituto Superior Técnico, 2016.
[149] EDA - Historical data. EDA 2018. http://www.eda.pt/Mediateca/Publicacoes/Producao/Paginas/Produção-de-Energia-Elétrica.aspx (accessed February 15, 2018).
[150] EDA. Multi-annual strategic plan and budget for 2018. 2017. [151] Wilson IAG, Barbour E, Ketelaer T, Kuckshinrichs W. An analysis of
storage revenues from the time-shifting of electrical energy in Germany and Great Britain from 2010 to 2016. J Energy Storage 2018;17:446–56. doi:10.1016/j.est.2018.04.005.
[152] Neves D, Silva CA. Modeling the impact of integrating solar thermal systems and heat pumps for domestic hot water in electric systems - The case study of Corvo Island. Renew Energy 2014;72:113–24. doi:10.1016/j.renene.2014.06.046.
[153] Pordata - Portuguese Statistics. Resident Population in Portugal 2017. www.pordata.pt (accessed January 20, 2017).
[154] EDA. Relatório Técnico 2007. 2008. [155] IEA. Renewable Energy Essentials: Geothermal. 2010. [156] European Commission. Reference Document on the Best Available
Techniques for Waste Incineration. 2006. doi:10.1002/0470012668.ch5. [157] Association ES. Spinning Reserve n.d. http://energystorage.org/energy-
storage/technology-applications/spinning-reserve (accessed March 20, 2018).
[158] Jaiswal A. Lithium-ion battery based renewable energy solution for off-grid electricity: A techno-economic analysis. Renew Sustain Energy Rev 2017;72:922–34. doi:10.1016/j.rser.2017.01.049.
[159] Li MJ, Zhao W, Chen X, Tao WQ. Economic analysis of a new class of vanadium redox-flow battery for medium- and large-scale energy storage in commercial applications with renewable energy. Appl Therm Eng
REFERENCES| 109
2017;114:802–14. doi:10.1016/j.applthermaleng.2016.11.156. [160] Abdon A, Zhang X, Parra D, Patel MK, Bauer C, Worlitschek J. Techno-
economic and environmental assessment of stationary electricity storage technologies for different time scales. Energy 2017;139:1173–87. doi:10.1016/j.energy.2017.07.097.
[161] Wang D, Coignard J, Zeng T, Zhang C, Saxena S. Quantifying electric vehicle battery degradation from driving vs. vehicle-to-grid services. J Power Sources 2016;332:193–203. doi:10.1016/j.jpowsour.2016.09.116.
[162] EDA. Relatório e Contas. 2016. [163] Käbitz S, Gerschler JB, Ecker M, Yurdagel Y, Emmermacher B, André D, et
al. Cycle and calendar life study of a graphite|LiNi1/3Mn1/3Co1/3O2Li-ion high energy system. Part A: Full cell characterization. J Power Sources 2013;239:572–83. doi:10.1016/j.jpowsour.2013.03.045.
[164] Sarasketa-Zabala E, Martinez-Laserna E, Berecibar M, Gandiaga I, Rodriguez-Martinez LM, Villarreal I. Realistic lifetime prediction approach for Li-ion batteries. Appl Energy 2016;162:839–52. doi:10.1016/j.apenergy.2015.10.115.
[165] Petit M, Prada E, Sauvant-Moynot V. Development of an empirical aging model for Li-ion batteries and application to assess the impact of Vehicle-to-Grid strategies on battery lifetime. Appl Energy 2016;172:398–407. doi:10.1016/j.apenergy.2016.03.119.
[166] Few S, Schmidt O, Offer GJ, Brandon N, Nelson J, Gambhir A. Prospective improvements in cost and cycle life of off-grid lithium-ion battery packs: An analysis informed by expert elicitations. Energy Policy 2018;114:578–90. doi:10.1016/j.enpol.2017.12.033.
[167] Nykvist B, Nilsson M. Rapidly falling costs of battery packs for electric vehicles. Nat Clim Chang 2015;5:329–32. doi:10.1038/nclimate2564.
[168] Berckmans G, Messagie M, Smekens J, Omar N, Vanhaverbeke L, Mierlo J Van. Cost projection of state of the art lithium-ion batteries for electric vehicles up to 2030. Energies 2017;10. doi:10.3390/en10091314.
[169] Hansen JB. Process for converting biogas to a gas rich in methane. WO2012003849 A1, 2012.
[170] Kemp I. Pinch Analysis and Process Integration. 2nd ed. Oxford: 2007. [171] Lorenzi G, Lanzini A, Santarelli M, Martin A. Exergo-economic analysis of
a direct biogas upgrading process to synthetic natural gas via integrated high-temperature electrolysis and methanation. Energy 2017;141. doi:10.1016/j.energy.2017.11.080.
[172] Rostrup-Nielsen J. Sulfur-passivated nickel catalysts for carbon-free steam reforming of methane. J Catal 1984;85:31–43. doi:10.1016/0021-9517(84)90107-6.
[173] Lanzini A, Madi H, Chiodo V, Papurello D, Maisano S, Santarelli M, et al. Dealing with fuel contaminants in biogas-fed solid oxide fuel cell (SOFC) and molten carbonate fuel cell (MCFC) plants: Degradation of catalytic and electro-catalytic active surfaces and related gas purification methods. Prog Energy Combust Sci 2017;61:150–88. doi:10.1016/j.pecs.2017.04.002.
[174] Rostrup-Nielsen JR. Catalytic Steam Reforming. 1984. [175] Rostrup-Nielsen JR, Hansen JB, Helveg S, Christiansen N, Jannasch a.-K.
Sites for catalysis and electrochemistry in solid oxide fuel cell (SOFC) anode. Appl Phys A 2006;85:427–30. doi:10.1007/s00339-006-3702-1.
110 | REFERENCES
[176] AspenTech Products n.d. https://www.aspentech.com/products/aspen-plus.aspx (accessed January 1, 2017).
[177] Gandiglio M, Lanzini A, Leone P, Santarelli M, Borchiellini R. Thermoeconomic analysis of large solid oxide fuel cell plants: Atmospheric vs. pressurized performance. Energy 2013;55:142–55. doi:10.1016/j.energy.2013.03.059.
[178] Moran, Michael J, Shapiro HN. Fundamentals of Engineering Thermodynamics. 5th ed. John Wiley & Sons, Inc; 2006.
[179] Mogensen MB. Sustainable fuels from renewable energy. Seminar activity in Politecnico di Torino, Torino, Italy, 21/05/2013 2013.
[180] Giglio E, Lanzini A, Santarelli M, Leone P. Synthetic natural gas via integrated high-temperature electrolysis and methanation: Part II-Economic analysis. J Energy Storage 2015;2:64–79. doi:10.1016/j.est.2015.06.004.
[181] Cost Estimation Methodology for NETL Assessments of Power Plant Performance. 2011.
[182] Pertl A, Mostbauer P, Obersteiner G. Climate balance of biogas upgrading systems. Waste Manag 2010;30:92–9. doi:10.1016/j.wasman.2009.08.011.
[183] Lorenzi G, Lanzini A, Santarelli M. Digester gas upgrading to synthetic natural gas in solid oxide electrolysis cells. Energy and Fuels 2015;29. doi:10.1021/ef5023779.
[184] Distribution components of the electricity tariff. Ital Auth Electr Energy, Gas Water Syst 2016. http://www.autorita.energia.it/it/elettricita/distr.htm (accessed July 18, 2016).
[185] Curletti F, Gandiglio M, Lanzini A, Santarelli M, Maréchal F. Large size biogas-fed Solid Oxide Fuel Cell power plants with carbon dioxide management: Technical and economic optimization. J Power Sources 2015;294:669–90. doi:10.1016/j.jpowsour.2015.06.091.
[186] Papadias DD, Ahmed S, Kumar R. Fuel quality issues with biogas energy - An economic analysis for a stationary fuel cell system. Energy 2012;44:257–77. doi:10.1016/j.energy.2012.06.031.
[187] Turton R. Analysis, Synthesis and Design of Chemical Processes. Third Edit. Prentice Hall; n.d.
[188] Jensen SH, Larsen PH, Mogensen M. Hydrogen and synthetic fuel production from renewable energy sources. Int J Hydrogen Energy 2007;32:3253–7. doi:10.1016/j.ijhydene.2007.04.042.
[189] Fu Q, Mabilat C, Zahid M, Brisse A, Gautier L. Syngas production via high-temperature steam/CO2 co-electrolysis: an economic assessment. Energy Environ Sci 2010;3:1382. doi:10.1039/c0ee00092b.
[190] Luckow P. Spring 2016 National Carbon Dioxide Price Forecast. 2016. [191] Projected Costs of Generating Electricity - 2015 Edition. 2015.
doi:http://dx.doi.org/10.1787/cost_electricity-2015-en. [192] Estermann T, Newborough M, Sterner M. Power-to-gas systems for
absorbing excess solar power in electricity distribution networks. Int J Hydrogen Energy 2016:1–10. doi:10.1016/j.ijhydene.2016.05.278.
[193] GSE. Procedure applicative DM 2 maro 2018 (document in Italian). 2018. [194] IEA. Renewables 2017. 2017.
REFERENCES| 111
[195] European Parliament. Circular economy: More recycling of household waste, less landfilling 2018. http://www.europarl.europa.eu/news/en/press-room/20180411IPR01518/circular-economy-more-recycling-of-household-waste-less-landfilling (accessed July 25, 2018).
[196] European Commission. 2050 low-carbon economy 2018. https://ec.europa.eu/clima/policies/strategies/2050_en (accessed July 25, 2018).
[197] Boden TA., Marland G., Andres RJ. National CO2 Emissions from Fossil-Fuel Burning, Cement Manufacture, and Gas Flaring: 1751-2014. 2017. doi:10.3334/CDIAC/00001_V2017.
[198] Lajunen A. Lifecycle costs and charging requirements of electric buses with different charging methods. J Clean Prod 2018;172:56–67. doi:10.1016/j.jclepro.2017.10.066.
[199] Bailera M, Lisbona P, Romeo LM, Espatolero S. Power to Gas projects review: Lab, pilot and demo plants for storing renewable energy and CO2. Renew Sustain Energy Rev 2017;69:292–312. doi:10.1016/j.rser.2016.11.130.
[200] Larscheid P, Lück L, Moser A. Potential of new business models for grid integrated water electrolysis. Renew Energy 2018;125:599–608. doi:10.1016/j.renene.2018.02.074.
[201] Valero A, Lozano M, Munoz M. A general theory of exergy saving, parts I, II and III. Comput Eng Energy Syst 1986;3:1–22.
[202] Tsatsaronis G, Winhold M. Exergoeconomic analysis and evaluation of energy-conversion plants — I. A new general methodology. Energy 1985;10:69–80. doi:10.1016/0360-5442(85)90020-9.
[203] Scopus n.d. www.scopus.com. [204] Decreto-Lei n.o 118/2013 (Decree-Law on the improvements of building
energy performance - Document in Portuguese) n.d. [205] SREA. Azores Statistics 2017. http://srea.azores.gov.pt/ (accessed January
20, 2018). [206] Hansen JB. Production of Sustainable Fuels by Means of Solid Oxide
Electrolysis. ECS Trans. Vol. 35, issue 1, 2011, p. 2941–8. [207] Reichhalter H, Bozzo A, Dal Savio S, Guerra T. Analisi energetica ,
ambientale ed economica di impianti a biogas in Provincia di Bolzano - Relazione conclusiva [document in Italian]. 2011.
[208] Eurostat. Gas prices for non-household consumers - bi-annual data 2017. [209] Eurostat. Electricity price for non-household consumers [nrg_pc_205]
2018.
112 | REFERENCES
APPENDIX A – EXERGO-ECONOMIC ANALYSIS| 113
Appendix A – Exergo-Economic Analysis
The exergo-economic analysis (EEA, also known as thermo-economic
analysis) is a methodology developed by Losano, Valero and Munoz [201]
which has the aim of attributing both an exergetic and an economic value
to each substance, according to its thermodynamic value.
The key concept is the exergetic cost which is the exergy that must be
spent to obtain a certain product. It is fundamentally different from the
exergy content of the same product as explained in [201]: “[…] if we were
to produce certain chemical substances from their elements, their
exergetic cost would be very high due to the inefficiency of current
processes. However, if these some substances are used as fuel, we cannot
get more than their exergy, and the maximum possible saving will be
given by the lost exergy which can be recovered in the process”. However,
the production cost does not determine the value of a substance, which is
linked with the purpose for which they are used.
The EEA methodology provides the tools to determine the cost of each
stream in a process, based on the exergy content and cost of the resources
and the cost for the components of the process. It is structured as an
integrated exergy and economic analysis which is based on the definition
of a productive structure for each subcomponent of the process under
analysis (Figure 27). Resources and Products can be mass or energy
streams as well as the difference of two streams. The definition of Resource
and Product is done through the definition of the Incidence Matrix (A)
which contains the information about the productive structure of the
process and specifies the exergetic cost balance [201].
Sub-comp.Resource Product
Figure 27: Productive structure of a generic subcomponent.
114 | APPENDIX A – EXERGO-ECONOMIC ANALYSIS
The described procedure is characterized by the following steps [202]:
a) Detailed exergy balance of the plant for the calculation of the
Resource and Product values, with reference to a dead state (zero-
exergy state). This step includes the determination of the exergetic
costs and the definition of the incidence matrix (A).
b) Comprehensive annual cost estimation for each component,
including investment, operation and maintenance and other
overheads. Definition of the exergo economic costs of the
components of the process and of the exergy flows entering the
process (all expressed in €/s). Each of these costs is a component
of the vector 𝑍.
c) Integration of the first two steps as shown in Equation ( A1 ).
𝐴 ∗ 𝐶 = 𝑍 ( A1 )
where C is the cost vector (€/s) containing the exergo-economic costs of
each mass/energy stream. The incidence matrix (step a) is coupled with
the cost estimation (step b) and the solution of the system (step c) provides
the cost distribution along the process, highlighting if the cost increase of
a stream is due to the exergy loss or to the capital expenditure for the
specific component.
APPENDIX B - PUBLICATIONS| 115
Appendix B - Publications
Paper I: “Techno-economic comparison of storage vs. demand response
strategies in distributed generation systems”, Proceeding of
the 12th International Conference on the European Energy
Market (Lisbon 2015).
Paper II: “Comparing demand response and battery storage to optimize
self-consumption in PV systems”, Applied Energy, 2016, 180,
524-535.
Paper III: “Techno-economic analysis of utility-scale energy storage in
island settings”, (Submitted to Journal of Energy Storage)
Paper IV: “Digester Gas Upgrading to Synthetic Natural Gas in Solid
Oxide Electrolysis Cells”, Energy & Fuels, 2015 29 (3), 1641-
1652
Paper V: “Exergo-economic analysis of direct biogas upgrading process
to synthetic natural gas via integrated high-temperature
electrolysis and methanation”, Energy, 2017 141, 1524-1537