Hindustan Institute of Technology and Science Chennai- 603103€¦ · [2] Wei Wu et. al., “A...
Transcript of Hindustan Institute of Technology and Science Chennai- 603103€¦ · [2] Wei Wu et. al., “A...
HYBRID ENERGY MANAGEMENT USING CYBER-PHYSICAL
CONTROLLER FOR REAL TIME EMS OF MICRO GRID
APPLICATION
PROJECT COMPLETION REPORT
Submitted to
Ministry of New and Renewable Energy
New Delhi, India
File No. 24/4/2014-SWES(R&D)
Principal Investigator(s) :Dr.A.K.Parvathy
Professor, EEE Department,
Hindustan Institute of Technology and Science
Co-Investigator(s): 1) R.Karthikeyan, Research Scholar
2) M.Maheswari, Assistant Professor (SG)
Hindustan Institute of Technology and Science
Chennai- 603103
January 2019
ii
ACKNOWLEDGEMENT
I wish to place on record my deep sense of gratitude to Ministry of New and
Renewable Energy (MNRE) for funding this R&D project and their worthy support and
guidance to make this project a successful one.
I express my sincere thanks to Dr. K. Balaraman, Director General, Mr. David
Solomon, Director and Division Head and Ms M.C. Lavanya, Scientist B, NIWE for their
inspiration and guidance in a large measure till the end of the project.
I express my thanks to Hindustan Institute of Technology and Science, for
providing a right platform to achieve my goal.
I would like to express my whole hearted gratitude to Dr. S. Gomathinayagam,
Former DG, NIWE, for his valuable guidance in helping me to reflect in the right perspective
and his continuous encouragement and support till the end of the project.
I owe my thanks to my Research scholar Mr. R. Karthikeyan and Research
fellow Ms. S. Priyadarshini who rendered their help in my endeavour.
Dr. A.K. PARVATHY
iii
Project Completion Report (PCR)
1. Title of the project: Hybrid Energy management using Cyber-Physical Controller
for real time EMS of Micro grid Application
2. Principal Investigator(s) and Co-Investigator(s): Dr. A.K. Parvathy, R. Karthikeyan,
M. Maheswari
i. Implementing Institution(s) and other collaborating Institution(s): Hindustan
Institute of Technology and Science
4. Date of commencement of Project: 31/1/15
5. Approved date of completion: 30/1/17
6. Actual date of completion: 31/12/18
7. Objectives of the Project:
(i) Development of new real time Energy Management System (EMS) to
improve energy utilization rate of solar and wind energy.
(ii) Design and implementation of cyber-physical controller based on improved
minority game algorithm for real –time EMS of micro grid.
(iii) Design and implementation of efficient battery management system so as to
keep the batteries within safe limits during charging and discharging and also
to prolong the life of batteries.
(iv) Specific Isolation if Fault occurs in any area rather switching off complete
area.
8. Output of the Project:
1. SOC meter and charge controller for battery- Design and implementation
2. Design and development of Novel Battery Management system
3. New Real time energy management system which improves the utilization of
wind and solar energy
(achieved vis-à-vis. originally planned in respect of:
i) Nature of Output: Material/Process/Product/Equipment/ Pilot scale
demonstration/Any other (Please Specify) Prototype development
ii) Performance specifications : Purity of materials, process details, product
specifications ( capacity, rating, efficiency, test results) Equipment
(performance features, capacity, bill of materials), Pilot production (capacity,
through put, yield, test results) : See Annexure I
iii) Details of engineering designs/ drawings (plans and sections) and prototype/
pilot/ full scale, specifications, etc. : See Annexure I
iv
8. Summary of the Project work, specially with respect to the project objectives and
proposed output.
As part of project, 5KW solar power plant (2KW+2KW+1KW) is installed and
connected to the real time load.
Forecasting of load is done and appliance priority is decided.
Design and implementation of Smart energy management system is completed.
Design and development of battery management system is completed
Integration of wind mill to the system and real time implementation completed.
9. Detailed progress report giving relevant information on work carried out, experimental
work, detailed analysis of results indicating contributions made towards increasing the
state of knowledge in the subject: Given in Annexure I
10. S&T benefits accrued:
i) Patents taken, if any Nil
ii) List of Research publications (a copy of the papers should be attached):
Sl. No
Authors Title of paper*
Name of
the Journal
Volume Pages Year
1 R.Karthikeyan,
A.K.Parvathy
Modelling of
Hierarchical Energy
Management System
for Micro-Smart Grid
Applications
International Journal
of Applied
Engineering
Research (IJAER)
10, No. 84 767-771 2015
2. R.Karthikeyan,
A.K.Parvathy
Real Time Energy
Optimization using
Cyber Physical
Controller for
Micro-Smart
Grid Applications
International Journal
of Control Theory
and Applications
9 (33) 145-156 2016
iii) List of Technical Documents prepared (a copy of the documents should be
attached): Research work was presented in 7 International conferences of which 3
conferences are IEEE conferences
iv) Manpower trained under the project:
a) Research Scientists/ Research Associates: 1 JRF and 2 M Tech students
b) No. of Ph.D. (s)produced: 1
c) Other Technical Personnel trained: 34 UG students were trained using the
project expertise
v) Awareness, training camps, etc. organized: Nil
11. Details of work which could not be completed ( if any):Nil
v
All the work modules are completed and the outputs obtained.
12. Suggestions on further work on the subject of research: 1.The hybrid energy
management using cyber physical controller can be extended further by integrating
other generators like DG sets, Biogas generators etc. 2. This work could be
implemented in on-grid systems.
13. Project Expenditure:
(Please enclose copies of year-wise audited UC and SoE as per MNRE format)
No Financial Position/
Budget Head
Amount
Sanctioned
Actual
Expenditure
Deviation if
any
% of Total
cost 1. Equipment 16,48,700 16,41,609
55.76
2. Consumables 1,50,000 1,20,368 4.08
3. Manpower 7,20,000 6,03,250 20.49
4. Travel 4,00,000 60,000 2.04
5. Consultancy 2,00,000 53,100 1.8
6. Contingencies Others, if any
1,64,300 1,53,430 5.21
7. Overhead Expenses 3,12,000 3,12,000 10.5
Total 35,95,000
29,43,757
14. EquipmentStatus:
Sl. Name of Year
of Purchase
Make/ Model
Cost (FE/ Rs)
Date of Installation
Utilizatio n Rate (%)
Remarks
regarding
Maintenance/
Breakdown
No Equipment
1. Laptop (2 No.s) 2015 Dell 73,000 15/4/15 70% Nil
2. Desktop (2 No.s) 2015 Dell 82,350 15/4/15 70% Nil
3. 5 kWSolar panel
with MPPT and
accessories
2015 Trina solar
Navsemi
Imax
4,23,710 18/11/15 70% 2 MPPTs broke
down
4 Battery (14
No.s)
2015 Marthon
Tubular lead
acid
2,41,969 18/11/15 60% 2 batteries were
replaced
5 Inverter 1200
VA
2015 Navsemi
solar
inverter
1,46,897 18/11/15
60% 2 inverters were
replaced
vi
viii
ANNEXURE I
ABSTRACT
Integration of various renewable energy sources like PV, wind etc. to the micro
grid, has paved way to the development of smart houses consisting of photovoltaic
panels, rechargeable batteries, external power supply and loads. The recent
development in micro-smart-grid technology has improved energy efficiency and
renewable energy utilization rate to serve local load with dispersed resources. In order
to achieve high utilization rate of solar-energy, we propose a cyber-physical controller
for real-time EMS of smart houses in this project. Based on physically sensed battery
profile, one cyber-physical controller enables their source to be shared between
different houses. The micro-smart-grid is modelled by decentralized multi-supplier
and multi-customer system. The proposed real-time EMS is formulated based on
optimization model for cluster of prosumers such that cooperative operation with
other prosumers in the neighbourhood achieves balanced distribution and hence
improved utilization rate. The experiment results show that the proposed EMS can
increase the solar energy utilization rate by 12.78% on average (up to 20%) when
compared to the independent prosumer optimization model.
ix
TABLE OF CONTENTS
CHAPTER
NO
TITLE
PAGE
NO
ABSTRACT viii
1 INTRODUCTION 1
1.1 LITERATURE REVIEW 2
2 BACKGROUND 3
3 MATHEMATICAL MODELLING 8
4 SIMULATION RESULTS 13
5 REAL –TIME EMS 28
5.1 HARDWARE DETAILS 30
5.2 MASTER AND SLAVE CONTROLLER 32
5.3 ENERGY METER ARRANGEMENT 35
5.4 ENERGY MANAGEMENT SYSTEM 36
5.5 BATTERY MANAGEMENT SYSTEM 37
5.6 SHARING OF LOADS 38
5.7 SHARING FLOW CHART 39
5.8 SHARING ALGORITHM 39
6 EXPERIMENTAL RESULTS 48
7 CONCLUSION
7.1 FUTURE SCOPE
52
REFERENCES 53
LIST OF PUBLICATIONS 54
1
CHAPTER 1
INTRODUCTION
The ever-increasing demands in energy supply exerted by the burgeoning
population, rapid urbanisation and demands of growing economy has caused serious
energy shortage and at times power outages. Renewable energy sources offer viable
option to address the energy security concerns of the country.
Microgrids which integrate renewable resources on the community level could
be an ideal solution. The proximity of power generation to micro grid consumptions
could result in improved power quality, lower power losses and higher reliability. To
address the challenges of energy and power shortage, intermittent energy sources, to
improve power system reliability, the following objectives are identified in this project.
The master controller which is a cyber-physical controller is designed:
(i) To improve the energy utilization rate of solar and wind energy and hence
decrease of dependency on power grid
(ii) Reduction of uncertainty about energy produced by renewable sources
(iii) Improve the battery operations by implementing efficient battery
management system.
The main motivation of this project is to reduce the utilization of the power grid
by decreasing the grid electricity load. The objective is to utilize as much energy as
possible from various renewable sources, including solar and wind. This will help in
reducing carbon emission, thereby contributing to ongoing efforts to limit the
accumulation of greenhouse gases.
The problems that are addressed from the supply side are:
1. Integration of various renewable energy sources like PV, wind etc. to the micro –
grid, to supplement the main grid.
2
2. Design and implementation of efficient battery management system so as to keep
the batteries within safe limits during charging and discharging and also to
prolong the life of batteries.
3. Development of new real time Energy Management System (EMS) to improve
energy utilization rate of solar and wind energy.
4. Design and implementation of cyber-physical controller for real –time EMS of
micro grid.
The problems that are addressed from the demand side are:
1. Schedule demands automatically so that less priority demands can be delayed so
that peak load can be kept low.
1.1 LITERATURE REVIEW
Applications of Cyber Physical Energy Systems (CPES) on smart grids and a
new proposal modeling of smart grids based on CPES approach is proposed in [1]. A
cyber-physical energy management system for smart houses so as reduce the utilization
of external electricity grid is the main motivation of this project [2]. Advanced methods
to estimate precisely the SoC in order to keep battery safely charged and discharged at a
suitable level and to prolong its life cycle is given in [3]. Universal state of charge
algorithm for batteries based on Open circuit voltage (OCV) is given in [4]. A
microcontroller based Power Management System (PMS) designed to operate a stand-
alone microgrid consisting of battery system, fuel cell and photovoltaic module is
studied [5].
Energy management system for building structures using a multi – agent
decision making control methodology for increasing energy utilization efficiency is
proposed in [6]. Cooperative energy management system for a cluster of household
prosumers used optimization model for a cluster of prosumers [7]. Energy sharing of
multiple PV prosumers with the assistance of public shared energy storage was
implemented using a hybrid approach [8].
3
CHAPTER 2
BACKGROUND
In this section we discuss the underlying physical system of smart houses. When
designing CPS, a practical approach is to consider three design layers, which include the
physical layer, the network/platform layer, and the software layer.
Fig. 2.1 Electrical Scheme of a cluster of prosumers
KEY:
- Access Point
4
Inverter- DC to AC Inverter
Switch- Computer Controlled Switch
Power line bundle
Communication Line
The systems require managing physical and cyber variables like battery life,
power flow, computing process, and network limitations. In order to adapt the future
energy systems to the new challenges, these systems should exhibit adaptive
performance such as flexibility, efficiency, sustainability, reliability, and security. This
performance can be obtained through a systematic embedding of cyber technologies
capable of monitoring, communicating, and controlling the evolving physical system.
The capabilities of monitoring, communicating and controlling physical energy system,
require to install smart meters, sensors and actuators. However, installing the most
advanced sensors, and actuators cannot be sufficient because the future system can be
flooded with data. Therefore, without an appropriate model of the system and scheme of
control, this approach would not help to operate the grid. For this reason, it is necessary
to develop a novel modeling methodology for this new type of energy systems. This
model should offer capabilities that include the impact of communication networks and
further cyber components, besides the relevant information of the physical system, in
terms of efficiency, sustainability, reliability, security, and stability. Cyber Physical
System methodologies have emerged as a new alternative in order to model and control
this class of smart energy systems.
The block presents a novel way of hybrid energy management system in a micro
grid. The objective of the proposed project is to increase the utility factor of the
renewable sources such as PV and Wind turbines attached to the system. Furthermore
the system can add other sources such as Fuel cells. The system further uses Cyber
Physical System for effective controlling of resources using the forecasting and further
matches the demand using the present renewable resources and utilizes the grid supply
in the needed conditions.
5
The system controls the energy resources through changing rules which are
dynamically modified according to the location characteristics and user situation. This
system further uses information convergence method it means the system not only uses
the information generated in the local system but also uses information generated in the
other domains.
When designing CPS, a practical approach is to consider three design layers,
which include the physical layer, the network/platform layer, and the software layer.
(i) The physical layer:
The physical layer represents physical components and their interactions, whose
behaviour is governed by Physical laws and is typically described in continuous time
Using ordinary differential equations.
The physical system consists of various blocks such as:
Energy Sources:
• Solar Panels
• Wind Turbines
• Grid Supply
Battery Management System:
BMU consist of a Battery charge controller to limit the rate at which electric
current is added. It prevents overcharging and may prevent against overvoltage, which
can reduce battery performance. It may also prevent completely draining ("deep
discharging") a battery, or perform controlled discharges, depending on the battery
technology, to protect battery life.
Lead acid batteries should be charged in three stages, which are constant-current
charge, topping charge and float charge. The constant-current charge applies the bulk of
the charge and takes up roughly half of the required charge time; the topping
charge continues at a lower charge current and provides saturation, and the float
charge compensates for the loss caused by self-discharge.
6
State of Charge (SOC) estimation is a fundamental challenge for battery use. The
SOC of a battery, which is used to describe its remaining capacity is a very important
parameter for a control strategy. As the SOC is an important parameter, which reflects
the battery performance, so accurate estimation of the SOC can not only protect battery,
prevent over discharge, and improve the battery life but also allow the application to
make rational control strategies to save energy.
The system is designed such that the total battery need for each domain is
separated in two equal half. Since charging and discharging of battery cannot be done on
the same time. We design an intelligent BMS such that it will monitor the SOC of the
both the battery and always try to maintain High SOC.
During the availability of the renewable energy the BMS charge the one half of
the battery and other part of the battery will start discharging until SOC goes low. This
type of simultaneous charging and discharging of battery helps in increasing the utility
factor of the battery and renewable resources.
Switching system in the Domain i.e. home
Each home is connected with a Computer controlled relay switch which helps in
connecting the supply with the load. The cyber physical controller decides the supply
need to be given to the load according to the demand factor in the system.
(ii) The network/platform layer:
The network/platform Layer represents the hardware side of CPS and includes the
Network architecture and computation platform that interact with the physical
components through sensors and actuators.
Communication Group consists of Sigsbee module, an Ethernet module and a WLAN
module. Thus the EMS can interact with various home appliances such as mobile phone,
Laptop.
7
(iii) Cyber Physical Control Unit:
The software layer represents the software components which are connected based on an
input/output model implying a notion of causality. The system uses optimization model
for cluster of prosumers for optimal utilization of available resources.
The distributed cyber physical control architecture helps in matching energy supply to
energy load at the sub transmission and distribution level. A hierarchical model of CPS
includes primary, secondary, tertiary levels. With a goal of ensuring frequency
regulation and optimal allocation of resources including renewable energy resources.
The energy monitoring and control unit plays an important role in monitoring the power
consumption and the power state, and controlling the relay. For example if a leakage of
electricity or an overvoltage occurs, the system can recognize the abnormal events and
react to them quickly so that we can implement a zero-risk building.
8
CHAPTER 3
MATHEMATICAL MODELLING
3.1 Independent prosumer optimization model:
The problem for optimal coordination of DERs has been defined as a mixed integer
linear programming (MILP) model that is composed by real variables x and binary
variables z and can be written in general as,
𝐦𝐢𝐧 𝒙, 𝒛 𝒔𝒖𝒃𝒋𝒆𝒄𝒕 𝒕𝒐.
𝒇(𝒙, 𝒛) = 𝒂𝑻𝒙 + 𝒃𝑻𝒛
𝑮(𝒙, 𝒛) = 𝒄
𝑯(𝒙, 𝒛) ≤ 𝒅
𝒙 ∈ 𝑹, 𝒛 ∈ [𝟎, 𝟏]
where, 𝑓(𝑥, 𝑧) is the objective function, c and d are scalar vectors, and 𝐺(𝑥, 𝑧), 𝐻(𝑥, 𝑧)
are linear combinations of the variables that define equalities and inequalities
constraints. Along this section, the mathematical formulation of the optimization
problem will be presented in detail. [7]
The variables used in the model are presented in Table I, one binary variable ( 𝑧 =
𝑧𝑙𝑜𝑎𝑑(𝑡) ) and real variables 𝑥 as shown in table. The parameters used in the model as
summarised in Table II.
The parameters 𝑆𝑜𝐶𝑚𝑎𝑥, 𝑆𝑜𝐶𝑚𝑖𝑛 and 𝜑𝑏𝑎𝑡are related to the battery bank.𝑆𝑜𝐶𝑚𝑎𝑥is
selected to allow the battery to be fully charged without overcharging and 𝑆𝑜𝐶𝑚𝑖𝑛is
chosen to limit the depth of discharge (DoD) accordingly with the recommendation of
the IEEE1561-2007 standard . 𝜑𝑏𝑎𝑡is a proportionality constant which relates the stored
energy in kWh with the SOC of the battery in percentage.
The model uses the index 𝑡 = {1,2 … . , 𝑇} for discrete time intervals of ∆𝑡 with 𝑇 as
time horizon for the optimization.
9
TABLE I
Variables of the local model
VARIABLE
DESCRIPTION
𝑷𝒈(𝒕) Power references of RES
𝑷𝒃𝒂𝒕(𝒕) Power of the battery
SoC(t) SoC of the battery
𝒛𝒍𝒐𝒂𝒅(𝒕) Connection of load
Cost Whole cost computed as objective function
TABLE 2
Parameters of the loacal model
PARAMETERS DESCRIPTION
𝑻 Time of scheduling
∆𝒕 Duration of interval
𝑷𝒍𝒐𝒂𝒅(𝒕) Average load consumption
𝑷𝒎𝒂𝒙(𝒕) Power max of RES
𝑪𝒍𝒐𝒂𝒅(𝒕) Penalty cost for load disconnection
𝑪𝒈(𝒕) Penalty cost for curtailed generation
𝑺𝒐𝑪𝒎𝒂𝒙 Maximum state of charge
𝑺𝒐𝑪𝒎𝒊𝒏 Minimum state of charge
𝝋𝒃𝒂𝒕 Coeffient of the ESS
ObjectiveFunction:
The objective function aims to minimize load disconnection of each household and
maximize the use of available generation
10
𝑚𝑖𝑛𝑥,𝑧𝑓(𝑥, 𝑧) = 𝑐𝑜𝑠𝑡
= ∑ 𝑃𝑙𝑜𝑎𝑑(𝑡)∆𝑡∗(1 − 𝑧𝑙𝑜𝑎𝑑(𝑡)) ∗ 𝐶𝑙𝑜𝑎𝑑(𝑡)
𝑇
𝑡=1
+ ∑[𝑃𝑚𝑎𝑥(𝑡)∆𝑡 − 𝑃𝑔(𝑡)∆𝑡] ∗ 𝐶𝑔(𝑡)
𝑇
𝑡=1
The first term corresponds to the cost associated to disconnect the load, which is equal
to 0 if the load is connected (𝑧𝑙𝑜𝑎𝑑 = 1). The second term is a penalisation for not using
the available power in the RES, computed as the difference between the predicted
available profile of energy and the power reference multiplied by the penalty cost of
curtailment.
Constraints:
1) Balance equation: The demand must be supplied by the sources and the storage
system.
𝑃𝑔(𝑡)∆𝑡 + 𝑃𝑏𝑎𝑡(𝑡)∆𝑡 = 𝑃𝑙𝑜𝑎𝑑(𝑡)∆𝑡 ∗ 𝑧𝑙𝑜𝑎𝑑(𝑡), ∀𝑡
2) Power Sources: The power reference for the power source at each 𝑡, 𝑝𝑔(𝑡), must be
less than or equal to the maximum power that can be provided for them 𝑃𝑚𝑎𝑥(𝑡), at
these 𝑡,
0 ≤ 𝑃𝑔(𝑡) ≤ 𝑃𝑚𝑎𝑥(𝑡), ∀𝑡
3) Energy Storage System: The SoC of a battery can be defined in terms of the power
as,
𝑆𝑜𝐶(𝑡) = 𝑆𝑜𝐶(𝑡 − 1) − 𝜑𝑏𝑎𝑡 ∗ [𝑃𝑏𝑎𝑡(𝑡)∆𝑡], ∀𝑡
11
When 𝑝𝑏𝑎𝑡(𝑡) is positive, the battery provides energy to the load (discharge mode) and
when is negative, it absorbs energy from the sources (charge mode).
𝑆𝑜𝐶 at each 𝑡 is bounded in the range,
𝑆𝑜𝐶𝑚𝑖𝑛 ≤ 𝑆𝑜𝐶(𝑡) ≤ 𝑆𝑜𝐶𝑚𝑎𝑥, ∀𝑡
3.2 Optimization model for cluster of prosumers
For the cluster of prosumers, the MILP model presented in the previous section is
adapted. In this case, the index 𝑖 = {1,2 … , 𝑛} is added to represent the i-th prosumer of
a cluster of n household prosumers.
Objective Function:
The objective function minimizes disconnection of the load, for the households
interconnected in the cluster, and maximises the available generation as,
∑ ∑ 𝑃𝑙𝑜𝑎𝑑(𝑖)
(𝑡)∆𝑡 ∗ (1 − 𝑧𝑙𝑜𝑎𝑑(𝑖) (𝑡)) ∗ 𝐶𝑙𝑜𝑎𝑑
𝑖 (𝑡) ∗ 𝜉(𝑖)
𝑛
𝑖=1
𝑇
𝑡=1
+ ∑ ∑[𝑃𝑚𝑎𝑥(𝑖) (𝑡)∆𝑡 − 𝑃𝑔
(𝑖)∆𝑡] ∗ 𝐶𝑔
(𝑖)(𝑡)
𝑛
𝑖=1
𝑇
𝑡=1
where𝜉(𝑖) has been included in order to penalise the load connection, and is defined as
𝜉(𝑖) = ∑ 𝑃𝑚𝑎𝑥(𝑖) (𝑡)
𝑇
𝑡=1
∑ 𝑃𝐿(𝑖)
𝑇
𝑡=1
⁄ (𝑡), ∀𝑖
The value of 𝜉(𝑖) will be bigger for the prosumer with larger ratio between generation
and consumption and the disconnection of its load will be highly penalised in the
optimization model.
Constraints
1)Balance equation: The energy balance should be fulfilled in the cluster of prosumers,
and it can be written as,
12
∑ 𝑃𝑔(𝑖)
∆𝑡 + ∑ 𝑃𝑏𝑎𝑡(𝑖) (𝑡)∆𝑡 = ∑ 𝑃𝑙𝑜𝑎𝑑
(𝑖)(𝑡)∆𝑡 ∗ 𝑧𝑙𝑜𝑎𝑑
(𝑖)(𝑡), ∀𝑡
𝑛
𝑖=1
𝑛
𝑖=1
𝑛
𝑖=1
2) Power sources : The scheduled generation reference can be generalised for the cluster
of prosumers as ,
0 ≤ 𝑃𝑔(1)(𝑡) ≤ 𝑃𝑚𝑎𝑥
(𝑖) (𝑡), ∀ 𝑖, 𝑡
3) Battery : The SoC can be rewritten as,
𝑆𝑜𝐶(𝑖)(𝑡) = 𝑆𝑜𝐶(𝑖)(𝑡 − 1) − 𝜑𝑏𝑎𝑡 ∗ ⌊𝑝𝑏𝑎𝑡(𝑖) (𝑡)∆𝑡⌋ ∀𝑖, 𝑡
Boundaries of the related variables can be written as,
𝑆𝑜𝐶𝑚𝑖𝑛 ≤ 𝑆𝑜𝐶(𝑖)(𝑡) ≤ 𝑆𝑜𝐶𝑚𝑎𝑥 , ∀𝑖, 𝑡
13
CHAPTER 4
SIMULATION RESULTS
Simulation of PV solar system and wind turbine is carried out using
MATLAB/SIMULINK software. Voltage, current and power waveforms are obtained
for Chennai region during different months. Hybrid system using battery sharing is
simulated and load sharing is validated.
Fig. 4.1 Simulation diagram of 1kW PV solar system
Fig. 4.2 Voltage waveform of Solar PV Vpv (109.6V)
14
Fig: 4.3 Current waveform Ipv(9.33A)
Fig 4.4 Solar power waveform (1001W)
Fig4.5 Load current (2A) and Load voltage (230.2V)
15
Fig: 4.6 Simulation diagram for PMSG wind turbine
The Simulink model of wind turbine driven permanent magnet synchronous generator
connected to the three phase load is given in Fig. 4.6. Wind is the input given to the
turbine, where the kinetic energy gets converted to mechanical energy, thus generator
torque Tm is produced. The wind turbine is designed with pitch control mechanism and
the simulation results are shown below.
16
Fig. 4.7 Standalone PMSG wind turbine for 1kW
Fig.4.7 depicts the standalone 1kW wind turbine, consists of rectifier, buck-boost
converter, batteries, inverter, step-up transformer and load.
17
Fig. 4.8 Input wind speed
Fig. 4.8 shows the input wind speed for the wind turbine. In general 3m/s is the
cut in speed, 25m/s is the cut out speed. Rated wind speed ranges in between 12-
25 m/s. Wind input is 6m/s.
Fig. 4.9 Wind velocity
Fig. 4.9 shows the change in wind speeds for different time period which is provided as
an input to the PMSG wind turbine.
Fig. 4.10 Wind turbine power characteristics
18
The above graph fig. 4.10 shows characteristic curve is drawn between the output
power and variable speed of PMSG wind turbine. The graph is obtained for zero degree
pitch angle.
Fig. 4.11 : Generator output current
Fig. 4.12 Generator output voltage
Fig. 4.13 Generator output power
The above graphs figures 4.11-4.13 shows generator output current, voltage and power
respectively. The generator gives the three phase output voltage around 100 volts, the
output current of 10A and power range of 900 W for particular wind speed (7m/s)
19
Fig. 4.14 Rectifier output voltage
Fig. 4.15 Converter output voltage (Vdc)
The figures 4.14-4.15 shows the rectifier output voltage and buck-boost converter
output voltage. The buck-boost converter will step up or stepdown the DC voltage
according to the battery needs. Here, 24 V battery is used.
Fig. 4.16 Inverter output voltage
20
The batteries are designed according to 1kW wind turbine, it works in
DC so rectifier is used which converters AC to DC. Then to match the voltage of
battery, buck –boost converter is designed that will give the controlled output with the
help of PWM controller. Buck boost converter gives the output of 26V to match the 24v
battery. Inverter gives the output voltage of 26 V by converting DC to AC. Figure no 11
shows the inverter output voltage.
Fig. 4.17 Load current
Fig 4.18 Load voltage
Figure 4.17-4.18 show the load current and load voltage in the range of 4A and 230 V
respectively.
21
Fig. 4.19 Average wind speed distribution for typical site in Chennai
Fig. 4.20 Month-wise wind turbine output in Matlab
The above graphs fig: 4.19-4.20 are showing the wind turbine output
according to the Chennai wind speed for every month .
0
2
4
6
8
10
12
Win
d S
peed
(m
/s)
Average wind speed Distribution for Typical site in Chennai
22
Fig 4.21 Hybrid system with battery sharing
23
Case1:
System1 working (SOC>40) and system2 working (SOC>40), Load1 takes the power
from system 1 and Load2 takes the power from system2 respectively.
Initial state of battery system 1 is 100%
Initial state of battery system 2 is 90%
Fig. 4.22 SOC of system1 and system 2
Fig. 4.23 Load 1 takes the power from system1 and load 2 takes the power
from system 2
24
Case2: System1 working (SOC>40) and system2 not working (SOC<40), Load1 not
using power from system 1 and so the required power for Load2 shared from system1
respectively.
Initial state of battery system 1 is 100%
Initial state of battery system 2 is 30%
Fig. 4.24 SOC of system1 and system 2
Fig4.25 Load 1 not using the power from system1 and system1 shares the
power to load2
25
Case 3: System1 not working (SOC<40) and system2 is working (SOC>40), Load2 not
using power from system 2 and so the required power for Load1 is shared from system2
respectively.
Initial state of battery system 1 is 35%
Initial state of battery system 2 is 100%
Fig. 4.26 SOC of system1 and system 2
Fig. 4.27 Load 2 not using the power from system2 and so system2 shares
the power to load 1
26
Case 4: System1 not working (SOC<40) and system2 is not working (SOC<40),Both
the loads are shutdown and it will be switched to EB
Initial state of battery system 1 is 35%
Initial state of battery system 2 is 38%
Fig. 4.28 SOC of system1 and system 2
Fig. 4.29 Load 1 and load 2 are shutdown since the system 1 and system 2 is
not working, so it may use the EB for power requirement
27
Inference: Sharing is successfully done in Case 2 and 3 by appropriately operating the
switches using Energy Management system.
Scenario considered for sharing: Source 1 is not working due to fault, Source 2 is not
connected to any load hence it supplies the loads of source 1. This enables sharing of
power generated between sources and improves utilization of solar energy generated and
reduces the dependency on utility grid. The energy saved due to sharing is 6485 W/day
which amounts to Rs 51.88 /day.
Fig. 4.30 Typical Daily Load Curves (29/5/18)
Fig. 4.31 Load curve during sharing
28
CHAPTER 5
REAL –TIME EMS
Fig. 5.1 Scheme of cluster of household prosumers
Micro grid consisting of multiple smart homes including smart appliances as
loads, local renewable resources such as small PV systems and/or micro-wind turbines,
and individual or shared ESS, running as a single controllable system is shown. We
considered a general smart microgrid which consists of a set of smart homes H, indexed
by h = {1, 2, …, H}. Each smart home is equipped with a set of smart home appliances
I, indexed by i={1, 2, …, I}. Each smart home has a rooftop PV system or micro-wind
turbine that are capable of harnessing energy. In addition, it is connected to the main
grid and to a shared ESS (orBusbar). In general, all smart homes in the microgrid can
received the power from any sources. The energy exchange between the smart homes is
controlled by an Energy Management System (EMS). The EMS controls the microgrid,
manages smart homes’ consumption, and distributes the shared energy storage. Smart
29
homes are connected to the main grid as a supplement. Main grid supply is utilised when
renewable energy production is unavailable, and when the ESS (Energy Storage System)
is empty or when the energy existing in the ESS is not scheduled. Furthermore, we
apply a set of time slot T as an optimization time interval, indexed by t ={1, 2, ..., T},
with T = 24 indicating the optimization time horizon and t = 1 hour indicating the time
slot duration, to minimize the total day-ahead energy cost of the costumers in the
microgrid. Weather forecasting gives 24 hours wind speed and solar irradiation data.
The energy demand in the residential area is able to fluctuate significantly between
different societies, we have considered in our project diverse category of smart homes
which differ in energy demand profile and number of home appliances used.
Fig: 5.2 Each Prosumer Architecture
30
Each prosumer utilizes a power transfer switch that is able to toggle the power source for the
home’s electrical panel between the grid and a DC→AC inverter connected to a battery array.
On-site solar panels or wind turbines connect to, and charge, the battery array. A smart gateway
server continuously monitors household consumption via an in-panel energy monitor and
renewable generation via current transducers, and the battery’s state of charge via voltage
sensors. The simplest way to measure energy consumption and generation is to wrap current
transducers (CT) around wires in the building’s electrical panel. Each Home is connected to a
slave controller and data are aggregated and passed to the master controller wirelessly
TABLE III: Details of Generation and Loads
Home
Number
Room No Renewable Energy
Connected
Storage
Capacity
Load
Connected
H1 PX104 Solar 1KW 24V, 200 AH 785W
H2 Renewable
energy Lab,
PX103
Solar 2KW 48V, 200 AH 1880W
H3 PX102,PX101 Solar 2Kw 48V,200 AH 1570W
H4 PX001 Wind 1KW 48V,200 AH 785W
Real time implementation of the project is carried out using 6 class room loads
simultaneously. The total capacity of the Renewable energy installed is 6kW (5kW solar
and 1kW of Micro wind System).
5.1 HARDWARE DETAILS
Total capacity of Renewable Energy installed is 6 kW. The single line diagrams
are given.
31
Fig. 5.3 Single Line Diagram of 2 kW Solar off-grid system
Fig. 5.4 Single Line Diagram of 1 kW Solar off-grid system
32
Fig. 5.5 Single Line Diagram of 1 kW Wind-hybrid system
5.2 MASTER AND SLAVE CONTROLLER
The project uses cloud based monitoring and control system which makes it
easier to access the system from anywhere in the world ( website address). The project
uses the Microchip 30 MIPS dsPIC30F Digital Signal Controller which helps to access
large amount of data from each node and process it. The system further uses MIPS
dsPIC30F at each prosumer node as slave unit along with LumisenseIoT board to
upload all the data such as power consumed, battery status , battery temperature,
voltage, current and status of sharing to the master controller through cellular
communication. Each IOT board is equipped withSIM900 GPRS modem to activate
internet connection also equipped with a controller to process all input UART data to
GPRS based online data. The data is collected from the Energy meter connected to the
each Prosumer Node.
33
F`ig. 5.6 Master controller of EMS Fig. 5.7 Slave controller of EMS
Fig:5.8 Master controller of BMS
Fig:5.9 slave controller of BMS
34
Fig: 5.10 voltage sensor for battery
Fig:5.11 Current sensor for battery
Fig. 5.12 Battery with BMS devices and Inverter system
35
5.3 ENERGY METER ARRANGEMENT
Fig. 5.13 Energy meter arrangement
The system uses indigenous made ICD meters for both AC and DC measuring
applications which enables us to get the real time data for monitoring purpose. ICD
make DC energy meter (DEM 9004F) is used for measuring the solar panel output to the
battery sources with the operating range of 0 – 800V DC and current range of 0 – 999.9
A. The meter has PC interface optically isolated RS485 communication provided by the
MODBUS – RTU which enables us to directly port the data in to system. Further each
prosumer is connected with ICD SEM 9510 (Single Phase Energy Meter) meter with the
36
wide range of input voltage and whole current operation (5A and 10A Ib with 600%
operating range). It has accuracy class 1.0 as per IS 13779, IEC 62053 and Accuracy
class 0.5 as per IS14697, IEC 62053 operated meters. The data collection is done
through optical port / IrDA with IEC 62056-21 protocol (standard) and through optical
port with DLMS (Optional) Isolated RS232 / RS485 with MODBUS RTU protocol.
5.4 ENERGY MANAGEMENT SYSTEM
Fig. 5.14 Screenshot of Energy Management System
Both AC and DC Meter connected in the system are directly capable of porting
data to the system and all the data are processed and uploaded to the cloud each minute
to get the realtime scheduling of the system. The system is totally controlled by the
cyber physical master controller and all the decisions for scheduling is done based on
the energy available in the storage system and based on the availability of the renewable
energy sources. The controller is designed to reduce the system dependency on the grid
and improve the utilisation of the renewable energy in order to reduce the electricity
bills and for quick recovery of the solar return.
37
5.5 BATTERY MANAGEMENT SYSTEM
Fig. 5.15 Screenshot of Battery Management System
The project further uses the ZigBee (cc2530) which is a true system on chip
(SoC) solution for IEEE 802.15.4 applications for collecting the panel temperature. The
solar panel temperature is collected to access the performance of solar panels at various
temperature to further study the performance of panels. Battery temperature is measured
by NTC sensor attached to cathode terminals of the battery to monitor the health of the
battery and the data of battery charging and discharging status. SOC of the system along
with temperature of the each battery is uploaded to the cloud system to monitor and
schedule the system.
38
5.6 SHARING OF LOADS
The sharing process in the system is done by connecting all the prosumers in a
AC busbar such that when excess energy is generated or unused can be sharing with
other prosumers.
Fig. 5.16 Connection diagram of energy meters
Each and every node of the system is protected by using MCB 20A. The sharing and
whole operation of the system can be performed by both automatic and manual control.
39
5.7 SHARING FLOW CHART
5.8 SHARING ALGORITHM
Fig. 5.17 Architecture of a Microgrid: Interconnected homes with
renewables and batteries.
40
Fig. 5.18 System block diagram
Energy Sharing:
The main challenge for the distributed energy sharing is to design the information
exchange among homes to decide the energy sharing home pairs. The design should
provide necessary information for homes while avoiding information overload. In the
energy sharing process, suppliers offer their surplus energy to consumers, then
consumers select the suppliers. The general steps for the energy sharing are as follows:
(i) Broadcasting Energy Difference: At the beginning of each time slot, homes
calculate the energy difference by subtracting predicted consumption from
predicted harvesting. The energy difference information should be known by
every home in the microgrid, no matter whether it belongs to a consumer or
supplier set. The reason is that consumers can get the knowledge of the amount
of surplus energy each supplier has, later on they can request specific amount of
energy from the suppliers. Suppliers can summarize the energy supply-demand
relationship in order to decide the energy selling price. To summarize, in current
phase, every home broadcasts energy difference in the microgrid and keeps
listening to others’ broadcasted messages.
(ii) On the supplier side, although the profit is same no matter which consumers are
selected, the energy transmission efficiency is different. The supplier can grant
the energy request from the consumers with less transmission loss.
41
The detailed energy sharing algorithm for suppliers is shown in Algorithm 1.
The supplier sorts the consumer set demand according to the energy
transmission efficiency in decreasing order (Line 1). Then it waits for an energy
sharing request from consumers. When there are incoming requests, it retrieves
requests based on ordered set D (Line 2). If the requested energy is less than
current surplus energy, it grants the amount of requested energy; otherwise it
grants the amount of surplus energy it has. The supplier sends a message to the
consumer to confirm the request, adds energy sharing instructions into a list, and
multicasts the remaining surplus energy to notify the consumers. It continues the
process until the current phase ends. Note the supplier still keeps listening to the
incoming requests even if it has granted all of its surplus energy, in this case, it
replies tothe consumer with the granted energy Eg equals 0. By doing this, the
consumers are able to receive the response and try the next supplier. In this way,
a deadlock on a specific supplier can be avoided. Note that when the suppliers
cannot share their energy due to too many suppliers in current time, they can
charge the battery using the surplus energy later.
Notations:
𝐸𝑖 (𝑛)- Difference between available energy and consumed Energy
𝐸𝑟 (𝑛)- – Requested sharing energy in window n
𝐸𝑔 (𝑛)- - Granted sharing energy in window
𝜂𝑖,𝑗 - Energy transmission efficiency from home i to j
Algorithm I
Input: Energy consumer set D, surplus energy 𝐸𝑖 (𝑛)
Output: Energy sharing instructions list L
1: Sort D by transmission efficiency;
2: Wait for incoming requests and list them by order in D
3: For incoming request,(assuming from home j with energy request of 𝐸𝑟 (𝑛))
do
42
4: if |𝐸𝑖 (𝑛)| ≥ 𝐸𝑟(𝑛)then
5: Granted Energy 𝐸𝑔(𝑛) = 𝐸𝑟 (𝑛)
6: else
7: Granted Energy 𝐸𝑔(𝑛) = 𝐸𝑖 (𝑛)
8: end if
9: Send message to home j to grant 𝐸𝑔(𝑛)energy
10: Add energy sharing instruction[𝑗, 𝐸𝑔(𝑛)]into list L
11: Updated𝐸𝑖(𝑛) = 𝐸𝑖(𝑛) − 𝐸𝑔(𝑛)
12: Multicast updated 𝐸𝑖(𝑛) to consumers
13: end for
14: go to Line 1 if current phase doesn’t end.
The detailed energy sharing algorithm for consumers is shown in
Algorithm 2.
The consumer sorts the supplier set S according to the energy sharing based on
renewable energy order (Line 1). Then it requests energy from the suppliers sequentially
(Line 2). The amount of requested energy is limited by the supplier’s surplus energy and
consumer’s energy shortage (Lines3-7).After sending out the energy request, the
consumer waits until it receives the response from the supplier (Lines 8-9). The supplier
may grant exactly the requested amount of energy, or less due to the competition from
other parallel consumers. Consumer updates its energy shortage, puts the energy sharing
instructions into a list and continues the process until its energy request is fulfilled or all
the suppliers are tried(Lines 10-13). Note that the consumer will also update the amount
of surplus energy once it receives the updated multicast message from suppliers.
Algorithm 2 Energy Sharing Algorithm for Consumer j
Input: Energy supplier set S with surplus energy, required energy𝐸𝑗(𝑛)
Output: Energy sharing instructions listL
1: Sort S by availability;
2: for home 𝑖 in sorted S do
43
3: if|𝐸𝑖(𝑛)| ≥ 𝐸𝑗(𝑛) 𝜂𝑖𝑗⁄ then
4: Requested Energy𝐸𝑟(𝑛) = 𝐸𝑗(𝑛) 𝜂𝑖𝑗⁄
5: else
6: Requested Energy 𝐸𝑟(𝑛) = updated 𝐸𝑖(𝑛)
7: end if
8: Send message to home 𝑖 for 𝐸𝑟(𝑛) energy
9: Wait and receive message from home 𝑖 for 𝐸𝑔(𝑛)amount of energy
10: Add energy sharing instruction [𝑖, 𝐸𝑔(𝑛)] into list L
11: Updated 𝐸𝑗(𝑛) = 𝐸𝑗(𝑛) − 𝐸𝑔(𝑛) ∗ 𝜂𝑖𝑗
12: if Updated 𝐸𝑗(𝑛) = 0 , break
13: end for
After the energy sharing home pairs are decided, homes will start energy
sharing. As we cannot control or account for the exact energy flow if two energy sharing
are executed at the same time on the shared bus, generally we have to use the shared bus
in a sequential way. The limitation on the transmission speed puts a cap on the amount
of energy shared within one time window. This introduces a challenge for scheduling
the energy sharing. If a fixed sharing sequence is adopted, homes at the end of the
sequence may be deprived of the chance of energy sharing due to the time limit, which
impacts the fairness and causes either starving or energy waste in these homes.
Energy Sharing:
With P2P energy sharing, several customers in a community Microgrid share the
connection to the main grid. This means that surplus PV power from a customer can be
consumed by another customer with excess consumption. In the P2P sharing
community, there are N customers, NB of which have individual PV battery systems
installed ( ⩽ NB N). Although the PV outputs in a community are likely to be similar
due to almost the same solar radiation, the net loads vary between prosumers, because of
the differences in load, kWp of PV systems and battery statuses. Therefore, it is possible
44
for the prosumers to share PV power with each other. The surplus PV power from
prosumers can also be traded with consumers who do not have PV systems. A third
party entity named “energy sharing coordinator (ESC)” coordinates between customers
and provides P2P sharing services, i.e. assuring the power balance and payment balance
Modelling of battery systems
The battery at a premises is modelled by a simplified linear expression. Assuming that
the charge and discharge power remain constant during a time slot, the stored energy of
the battery is described by fig 5.19.
Fig:5.19 battery system
For the P2P energy sharing, when the PV generation is higher than the load, the excess
PV energy is firstly used to supply for the neighbours who have excess consumption,
and the remaining PV power, if there is any, is used to charge the battery. In the
evening, when the load is higher than the PV generation, some of the load is met by
discharging his own battery, and the remaining load, if there is any, is met by the PV
power or battery energy from his neighbours who have excess PV/battery power. From a
single household’s point of view, using the P2P energy sharing is able to reduce the
amount of electricity sold to the grid as well as to reduce the amount of electricity
45
bought from the grid. The excess PV energy is sold to his neighbours and the excess
load is bought from his neighbours instead.
Fig:5.20 P2P energy sharing community
Average Energy production in 2KW system:
Mont
h Day
Hou
r
Beam
Irradianc
e
(W/m^2)
Diffuse
Irradiance
(W/m^2)
Ambient
Temperatur
e (C)
Cell
Temperatur
e (C)
DC
Array
Output
(W)
AC
System
Output
(W)
5 29 0 0 0 28.454 28.454 0 0
5 29 1 0 0 28.198 28.198 0 0
5 29 2 0 0 27.956 27.956 0 0
5 29 3 0 0 27.804 27.804 0 0
5 29 4 0 0 27.685 27.685 0 0
5 29 5 0 0 27.728 27.728 0 0
5 29 6 75 73 28.932 29.569 127.482 114.834
5 29 7 382 145 31.154 37.558 440.366 420.44
5 29 8 531 188 33.545 45.869 753.734 724.693
5 29 9 619 215 34.704 54.931 982.152 945.318
5 29 10 670 231 35.254 59.321
1133.81
3
1091.26
9
5 29 11 694 239 35.508 61.399
1197.48
3
1152.41
3
5 29 12 696 240 35.422 66.831
1140.42
7
1097.62
3
5 29 13 677 234 34.835 63.932
1043.96
1
1004.85
1
5 29 14 80 339 34.192 46.988 566.024 542.663
5 29 15 0 101 33.355 35.716 156.543 143.296
46
5 29 16 0 70 32.578 33.283 109.701 97.412
5 29 17 0 24 31.494 30.961 38.003 27.103
5 29 18 0 0 30.649 30.649 0 0
5 29 19 0 0 30.19 30.19 0 0
5 29 20 0 0 29.733 29.733 0 0
5 29 21 0 0 29.337 29.337 0 0
5 29 22 0 0 28.96 28.96 0 0
5 29 23 0 0 28.577 28.577 0 0
Fig:5.21 solar output of 2kW
Average Energy production in 1KW system:
Mont
h
Da
y
Hou
r
Beam
Irradianc
e
(W/m^2)
Diffuse
Irradianc
e
(W/m^2)
Ambient
Temperatur
e (C)
Cell
Temperatur
e (C)
DC
Array
Output
(W)
AC
System
Output
(W)
5 29 0 0 0 28.454 28.454 0 0
5 29 1 0 0 28.198 28.198 0 0
5 29 2 0 0 27.956 27.956 0 0
5 29 3 0 0 27.804 27.804 0 0
5 29 4 0 0 27.685 27.685 0 0
5 29 5 0 0 27.728 27.728 0 0
5 29 6 75 73 28.932 29.569 63.741 57.417
5 29 7 382 145 31.154 37.558
220.18
3 210.22
5 29 8 531 188 33.545 45.869
376.86
7
362.34
7
5 29 9 619 215 34.704 54.931
491.07
6
472.65
9
5 29 10 670 231 35.254 59.321 566.90 545.63
0
500
1000
1500
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Solar Output 2KW (W)
Solar Output (Wats)
47
7 4
5 29 11 694 239 35.508 61.399
598.74
1
576.20
7
5 29 12 696 240 35.422 66.831
570.21
3
548.81
2
5 29 13 677 234 34.835 63.932 521.98
502.42
5
5 29 14 80 339 34.192 46.988
283.01
2
271.33
1
5 29 15 0 101 33.355 35.716 78.272 71.648
5 29 16 0 70 32.578 33.283 54.85 48.706
5 29 17 0 24 31.494 30.961 19.002 13.551
5 29 18 0 0 30.649 30.649 0 0
5 29 19 0 0 30.19 30.19 0 0
5 29 20 0 0 29.733 29.733 0 0
5 29 21 0 0 29.337 29.337 0 0
5 29 22 0 0 28.96 28.96 0 0
5 29 23 0 0 28.577 28.577 0 0
Fig:5.22 Solar output of 1kW
Table: Energy savings
TOTAL
LOAD(Watts)
TOTAL
LOAD(K
Watts)
Cost
Saving
/day
TOTAL
LOAD(
Per Year
KW)
Cost
Saving
/year
Cost Saving
/month
CO2 Emission
Saving (kg)
11190 11.19 89.52 4084.35 32674.8 2722.9 3471.6975
10745 10.745 85.96 3921.925 31375.4 2614.616667 3333.63625
5380 5.38 43.04 1963.7 15709.6 1309.133333 1669.145
0
200
400
600
800
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Solar Output 1 KW(W)
AC System Output (W)
48
CHAPTER 6
EXPERIMENTAL RESULTS
The proposed strategy has been tested experimentally using the Real –time EMS
as discussed in chapter 5. The loads connected to different renewable energy sources are
given in Table III. The objective is to maximise the utilization rate of solar and wind
energy and minimise the dependence on utility grid.
Fig. 6.1 -6.2 shows the experimental results of independent operation of two
prosumers. Similarly Fig. 6.3-6.4 depicts the generation curve and load curve of three
prosumers. As we are dealing with solar PV systems and the loads being class room
loads of the University, the generation curve and load curve are more or less matching.
Considering the scenario that 1kW source is not working, 2 kW is in working condition:
1 kW consumer FN load is shed (or in this case connected to utility grid), while AN load
could be shared from 2 KW generation, (Since 2kW consumer AN have no load
demand).
This sharing phenomenon saves 2.9 kWhr which amounts to Rs 23.2/half day.
The same phenomenon could be carried out for other generation sources as well. The
experimental set up for this project can facilitate sharing among all four generation
sources and connected load.
The experiment results show that the proposed EMS can increase the solar
energy utilization rate by 12.78% on average.
49
Fig. 6.1 Individual generation curves for prosumer 1 and 2
Fig. 6.2 Individual load consumption curves for prosumer 1 and 2
0 5 10 15 20 250
200
400
600
800
1000
1200
1400
Load Duration
Po
wer
Gen
era
tio
n (
kW
)
1 kW Solar
2 kW Solar
0 5 10 15 20 250
200
400
600
800
1000
1200
1400
1600
1800
2000
Load Duration
Po
wer
Co
nsu
mp
tio
n (
kW
)
Consumer 1
Consumer 2
50
Fig. 6.3 Individual generation curves for prosumer 1,2 and 3
Fig. 6.4Individual load consumption curves for prosumer 1,2 and 3
0 5 10 15 20 250
200
400
600
800
1000
1200
1400
1600
1800
2000
Load Duration
Po
wer
Gen
era
tio
n (
kW
)
1 kW Solar
2 kW Solar
2 kW Solar
0 5 10 15 20 250
200
400
600
800
1000
1200
1400
1600
1800
2000
Load Duration
Po
wer
Co
nsu
mp
tio
n (
kW
)
Consumer 1
Consumer 2
Consumer 3
51
Fig. 6.5 Generation curve when 1 kW source is not working
Fig 6.6 Experimental results with central scheduling
0 5 10 15 20 250
200
400
600
800
1000
1200
1400
Load Duration
Po
wer
Gen
era
tio
n (
kW
)
1 kW Solar
2 kW Solar
0 5 10 15 20 250
200
400
600
800
1000
1200
1400
1600
1800
2000
Load Duration
Po
wer
Co
nsu
mp
tio
n (
kW
)
Consumer 1
Consumer 2
Load
Shedding
Load Sharing
52
CHAPTER 7
CONCLUSION
The proposed central EMS enhances the performance of the cluster of household
prosumers compared with their independent operation. This improvement can be
reflected in better load profile and improved usage of renewable energy sources. The
collaborative behaviour such as load sharing and stored energy balance between
distributed energy storage sources can be easily addressed by the given algorithm and a
central EMS which is operated by a cyber physical controller. Cooperative operation
between household prosumers is a feasible option in order to achieve independence from
the utility grid while increasing the reliability and usage of local micro-grid.
The cyber physical system enables to adapt the future energy systems to the new
challenges, exhibiting adaptive performance such as flexibility, efficiency,
sustainability, reliability, and security.
7.1 FUTURE SCOPE
Improved strategies to enable more prosumers to participate in the cooperative
network could be a future work. The hybrid energy management using cyber physical
controller could be extended further by integrating other generators like DG sets, biogas
generators etc. This work could be implemented in on-grid systems.
53
REFERENCES
[1] Macana C.A. et. al., “A Survey on Cyber Physical Energy Systems and their
Applications on Smart Grids” IEEE PES Conference on Innovative Smart Grid
Technologies 2011, pp. 1-7
[2] Wei Wu et. al., “A Real-time Cyber-physical Energy Management System for Smart
Houses” IEEE PES Conference on Innovative Smart Grid Technologies Asia 2011, pp.
1-8.
[3] K.L.Man et.al. , “ Towards a Hybrid Approach to SoC Estimation for a Smart
Battery Management System (BMS) and Battery Supported Cyber-Physical Systems
(CPS)” IEEE Conference on 2nd Baltic Congress on Future Internet Communications
2012, pp.113-116
[4] B.Xiao, Y. Shi, L. He, "A Universal State-of-Charge Algorithm for Batteries," 47th
ACM/IEEE Design Automation Conference (DAC), 2010.
[5] Bruno Belvedere et. al. , “A Microcontroller-Based Power Management System
for Standalone MicrogridsWith Hybrid Power Supply” IEEE Transactions on
Sustainable Energy, vol. 3, no. 3, July 2012, pp. 422-431
[6] PengZaoet. al. , “An Energy Management System for Building Structures Using a
Multi-Agent Decision-Making Control Methodology”, IEEE Industrial Applications
Society Annual Meeting 2010, pp. l -8.
[7] Adriana C.Luna et.al. ,” Cooperative Energy Management for a Cluster of
Household Prosumers”, IEEE Trans. On Consumer Electronics, 62(3), 2016,235-242.
[8] Nian Liu et.al., “ Energy sharing Provider for PV prosumer clusters: A hybrid
approach using Stochastic Programming and Stackelberg Game”, IEEE Trans. on
Industrial Electronics, Jan 2018.
54
LIST OF PUBLICATIONS
JOURNAL PUBLICATION
1.R.Karthikeyan and A.K.Parvathy, “Study of PV Panels and Analysis of Various
MPPT Techniques” Journal of Theoretical and Applied Information Technology,
Scopus Indexed Vol. 68, No.2 ,ISSN: 1992-8645, October 2014.
2.R.Karthikeyan, Dr.A.K.Parvathy“Modelling of Hierarchical Energy Management
System for Micro-Smart Grid Applications”,International Journal of Applied
Engineering Research (IJAER) , pp.767-771,Volume 10, Number 84 (2015).
3.R.Karthikeyan and A.K.Parvathy,”Real Time Energy Optimization using Cyber
Physical Controller for Micro-Smart Grid Applications”, International Journal of
Control Theory and Applications, 9 (33), 2016, pp.145-156.
CONFERENCE PUBLICATION
4.R. Karthikeyan,Dr.A.K.Parvathy, A ZIGBEE Based Home automation system for
hybrid energy management system., Proceedings of International Conference on
Computing, Communication and Energy Systems-2015 (CESCON-2015) 15-17 April
2015, Held on Hindustan Institute of Science and Technology,Chennai.
5.R. Karthikeyan,Dr.A.K.Parvathy, Hierarchical Energy Management System For
Smart Building., Proceedings ofInternational Conference on Development in
Engineering Research-2015 (ICDER-2015) 21st June 2015, Organized by
International Association of Engineering and Technology for Skill Development.(Best
Presentation Award)
6.R. Karthikeyan, Dr.A.K.Parvathy, Hierarchical energy management Technique for
Micro Smart Grid Applications., Proceedings of International Conference on
Modelling, Simulation and Control-2015 (ICMSC-2015) 15-16 October 2015, Held on
Karpagam College and Engineering, Coimbatore.
7.R. Karthikeyan, Dr.A.K.Parvathy, Peak Load Reduction in Micro Smart Grid Using
Non Intrusive Monitoring and Hierarchical Load Scheduling., Proceedings of
IEEEInternational Conference on Smart Sensors and Systems (IC SSS-2015) 21-
23 December 2015, Held on M. S. Ramaiah Institute of Technology, Bengaluru.
8.R. Karthikeyan, Dr.A.K.Parvathy, Real Time Energy Optimization using Cyber
Physical Controller for Micro-Smart Grid Applications., Proceedings of 5th IEEE
International Conference on Recent Trends on Information Technology 8-9 April
2016, Held on Madras Institute of Technology, Chennai.
55
9.A.K.Parvathy, Integration of Solar-Wind Hybrid System to Micro grid with Demand
Side Management, presented in 2nd UK India bilateral workshop on Sustainable Energy
and Smart Grid , 14-15 July 2016, Double Tree by Hilton,Leeds, UK
10.R.Karthikeyan, DrA.K.Parvathy, “Distributed Demand Side Management System
Using Cyber Physical Controller for Micro Smart Grids ", presented in Indian Institute
of Technology, Tirupati (IITTP) International Conference on " Sustainable Energy
Technologies for Smart and Clean Cities (SETS&CC-2016) , July 27-29, 2016