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