Submitted to the Faculty of INHA UNIVERSITY In Partial ... · CONTROL STRATEGIES FOR DGS TO IMPROVE...

저 시-비 리-동 조건 경허락 2.0 한민

는 아래 조건 르는 경 에 한하여 게

l 저 물 복제, 포, 전송, 전시, 공연 송할 수 습니다.

l 차적 저 물 성할 수 습니다.

다 과 같 조건 라야 합니다:

l 하는, 저 물 나 포 경 , 저 물에 적 허락조건 확하게 나타내어야 합니다.

l 저 터 허가를 러한 조건들 적 지 않습니다.

저 에 른 리는 내 에 하여 향 지 않습니다.

것 허락규약(Legal Code) 해하 쉽게 약한 것 니다.

Disclaimer

저 시. 하는 원저 를 시하여야 합니다.

비 리. 하는 저 물 리 적 할 수 없습니다.

동 조건 경허락. 하가 저 물 개 , 형 또는 가공했 경에는, 저 물과 동 한 허락조건하에서만 포할 수 습니다.

http://creativecommons.org/licenses/by-nc-sa/2.0/kr/legalcode

http://creativecommons.org/licenses/by-nc-sa/2.0/kr/

CONTROL STRATEGIES FOR DGS TO IMPROVE

OPERATION PERFORMANCE OF MICROGRID

by

Loc Nguyen Khanh

A Dissertation

Submitted to the Faculty of

INHA UNIVERSITY

In Partial Fulfillment of the Requirements

for the Degree of

DOCTOR OF PHILOSOPHY

Graduate School of Electrical Engineering

August 2011

ACKNOWLEDGEMENT

I wish to express my sincere gratitude and appreciation to my Advisor

Professor Dr. Dong-Jun Won for his invaluable guidance, advice and

financial support throughout the course of this work. Without his help, this

work would not be possible and I am grateful to him for his supportive and

friendly attitude. I would also like to express my sincere thanks to Dr. Seon-

Ju Ahn for his valuable and outstanding advice. I would also like to thank the

members of my laboratory for their kind helps throughout the PhD course at

INHA University. I would also like to thank the members of my committee

for reviewing the dissertation and attending my defense: Prof. Dr. Kim

Yeong Seok, Prof. Dr. Lee Bok Hee, Prof. Dr. Song Seung Ho, and Prof. Dr.

Chung Il Yop. Their advice and patience are appreciated.

Lastly, I would like to thank my family for their continuous support and

encouragement, and especially my wife, Mrs. Tran Thi Hong Nga, for her

unconditional love, support, and patience.

ABSTRACT

This dissertation presents a control strategy for distributed generators

including controllable and uncontrollable sources to improve operation

performance of micgrogrids.

In order to achieve the control strategy for controllable DGs, the

frequency and active power responses according to the droop characteristic

have been investigated to show advantages and disadvantages of different

microgrid configurations and different control modes. The analysis is based

on three conditions of operations, i.e. grid-connected mode, islanded mode,

and transition mode, and four possible configurations of the microgrid

namely the FFC-series-, FFC-parallel-, UPC-series-, and UPC-parallel-

configuration.

According to the analysis, control strategy for controllable DGs is

proposed to take the advantages and overcome the disadvantages of each

configuration. Namely, the control strategy for controllable DGs is to operate

the microgrid as constant load from the utility viewpoint, minimize the

frequency change in the transition mode, and control the frequency

unchanged during the islanded mode.

Uncontrollable sources like PV and Wind turbine coupled with

controllable sources like Fuelcell in a hybrid system can become controllable.

The operation algorithm for hybrid sources, i.e. the PV-PEMFC hybrid

source, based on the unit-power-control mode and feeder-flow-control mode

ii

is proposed to operate in the microgrid with a high efficiency, maximize the

solar energy, and optimize the PEMFC operational efficiency.

Overall control strategy for microgrid with presence of controllable

DGs and the PV-PEMFC hybrid system is finally implemented in a multi-

feeder microgrid in EMTDC /PSCAD. The simulation results show

advantages and improvements of proposed method over conventional control

method.

KEYWORDS:

Microgrid, microgrid control, distributed generation, microsource,

energy management system, control strategy, hybrid system, droop

characteristic, unit power control, feeder flow control mode, inverter

control.

STUDENT NUMBER: 2207-2231

iii

CONTENTS

Abstracts .......................................................................................................... i

Contents ........................................................................................................ iii

Abbreviation ................................................................................................. vii

Index of Figures and Tables ........................................................................... ix

Chapter I: Introduction ............................................................................. 1

1.1 Motivations and Purposes ................................................................... 1

1.2 Highlights and Major Contributions ................................................... 4

1.3 Dissertation Structure ......................................................................... 6

Chapter II: Microgrid Concept and Control ............................................. 9

2.1 Overview ............................................................................................ 9

2.2 Microgrid Management and Control .................................................. 10

2.3 Microsource Controller ...................................................................... 14

2.3.1 The Two Power Control Modes ......................................................... 18

2.3.2 Droop Control through Active and Reactive Power ........................... 20

Chapter III: Analysis of P and f Responses According to Microgrid

Configurations ............................................................................................. 31

3.1 Overview ........................................................................................... 31

3.2 Active Power and Frequency Response in a Microgrid ..................... 33

3.2.1 Active Power and Frequency Response in Transition Mode .............. 33

iv

3.2.2 Active Power and Frequency Response in Grid Connected Mode ..... 34

3.2.3 Active Power and Frequency Response in Islanded Mode ................. 42

3.2.4 Comparison of the Different Microgrid Configurations ..................... 48

Chapter IV: Control Strategies for Controllable DGs in Microgrid ........ 53

4.1 Overview ........................................................................................... 53

4.2 Control strategy for Microgrid in the Grid-connected Mode .............. 54

4.2.1 Control Strategy for Single-Feeder Microgrid ................................... 54

4.2.2 Control Strategy for Multi-Feeder Microgrid ..................................... 63

4.3 Control Strategy for Microgrid in the Islanded Mode ........................ 65

4.4 Simulation Studies and Results .......................................................... 67

4.4.1 Test System and Simulation Scenarios .............................................. 67

4.4.2 Simulation Results in the Grid-Connected Mode ............................... 69

4.4.3 Simulation Results in the Transition Mode ........................................ 73

Chapter V: Control Strategies for a Hybrid Source Connected to

Microgrid ................................................................................................. 77

5.1 Overview ........................................................................................... 77

5.2 Hybrid System Description ................................................................ 78

5.2.1 Structure of Grid Connected Hybrid System ...................................... 78

5.2.2 PV Array Model ................................................................................ 79

5.2.3 PEMFC Model ................................................................................... 81

5.2.4 Maximum Power Point Tracking Control .......................................... 82

5.3 Control Algorithm of the Hybrid System ........................................... 85

v

5.3.1 Control Strategy for the Hybrid System in the UPC mode ................. 85

5.3.2 Overall Control Strategy for the Hybrid System .............................. 92

5.4 Simulation Studies and Results .......................................................... 97

5.4.1 Simulation Results in the Case without Hysteresis ............................ 97

5.4.2 Improving Operation Performance by Using Hysteresis .................... 99

5.4.3 Discussion ........................................................................................ 100

Chapter VI: Case Study .......................................................................... 105

6.1 Introduction to Case Study ............................................................... 105

6.1.1 System Configuration ...................................................................... 106

6.1.2 System Parameter ............................................................................ 106

6.1.3 Simulation Scenarios ....................................................................... 107

6.2 Results and Discussion .................................................................... 108

6.2.1 Simulation Results without Hybrid Source ...................................... 108

6.2.2 Simulation Results with the PV-FC Hybrid Source ......................... 117

6.2.3 Simulation Results with Multiple Feeder Microgrid ........................ 132

6.2.4 Proposed control method versus frequency restoration control scheme

in island mode .................................................................................. 137

Chapter VII: Conclusions and Future Extensions .................................. 141

7.1 Conclusions ..................................................................................... 141

7.2 Future Extensions ............................................................................ 143

Bibliography .............................................................................................. 145

Appendix A: Frequency and Active Power Responses in the Islanded

Mode ........................................................................................................... 155

vi

A.1 PSCAD Model of Three-DG Microgrid ........................................... 155

A.2 Simulation Results ........................................................................... 157

Appendix B: Case Study ........................................................................... 161

B.1 PSCAD Model of Whole System ..................................................... 161

B.2 PSCAD Model of Photovoltaic ........................................................ 162

B.3 PSCAD Model of MPPT Algorithm of PV ...................................... 163

B.4 PSCAD Model of PEMFC ............................................................... 165

B.5 PSCAD Model of Buck-Boost DC/DCs and Controllers ................. 169

B.6 PSCAD Model Inverter Controller .................................................. 170

vii

ABBREVIATION

BESS Battery energy storage system

CANMET Canada Centre for Mineral and Energy Technology

CERTS Consortium for Electric Reliability Technology Solutions

CHP Combined heat and power

CSI Current source inverter

CV Constant voltage

DER Distributed energy resources

DG Distributed generation

DGs Distributed generators

DMS Distribution management system

ED Economic dispatch

EMS Energy management system

EMTDC Electromagnetic transients including DC

FFC Feeder flow control

HB Hybrid

IC Internal combustion

INC Incremental conductance

LCs Load controllers

LD Load demand

MCs Microsource controllers

viii

MGCCs Microgrid central controllers

MGT Micro-gas turbine

MPPT Maximum power point tracking

NEDO New Energy and Industrial Technology Development

Organization

P&O Perturbation and observation

PCC Point of common coupling

PEMFC Proton exchange membrane fuel cell

PSCAD Power system computer aided design

PV Photovoltaic

PWM Pulse width modulation

RE Renewable energy

SC Super capacitor

SS Static switch

UC Unit commitment

UPC Unit power control

VSI Voltage source inverter

ix

INDEX OF FIGURES AND TABLES

FIGURES

Fig. 1.1 Dissertation structure ..................................................................... 7

Fig. 2.1 Microgrid architecture diagram ...................................................... 9

Fig. 2.2 Microgrid control architecture ....................................................... 12

Fig. 2.3 Microgrid energy management system .......................................... 13

Fig. 2.4 Inverter-based microsource diagram ............................................. 16

Fig. 2.5 Power-control mode of a DG ........................................................ 17

Fig. 2.6 Inverter-based system .................................................................... 19

Fig. 2.7 Voltage vs. Reactive power droop ................................................. 21

Fig. 2.8 Diagram of P vs. f droop control block ......................................... 25

Fig. 2.9 Power vs. Frequency droop ........................................................... 26

Fig. 2.10 Diagram of FL vs. f droop control block ....................................... 27

Fig. 2.11 Control block diagram to enforce the output limit ......................... 28

Fig. 2.12 Microgrid configuration with two Sources .................................... 29

Fig. 2.13 Feeder flow vs. Frequency ............................................................ 29

Fig. 2.14 Control block diagram to enforce the output limit ......................... 30

Fig. 3.1 Different configurations of microgrid ........................................... 32

Fig. 3.2 FL vs. f characteristic in transition mode for series–FFC

configuration ................................................................................. 34

Fig. 3.3 P vs. f characteristic in transition mode for series–UPC

configuration ................................................................................. 36

x

Fig. 3.4 FL vs. f characteristic in transition mode for parallel-FFC

configuration. ................................................................................ 37

Fig. 3.5 FL vs. f characteristic in transition mode for parallel-UPC

configuration ................................................................................. 39

Fig. 3.6 FL vs. f characteristic in islanded mode for series-FFC

configuration when load LDi increases ......................................... 42

Fig. 3.7 P vs. f characteristic in islanded mode for series-UPC

configuration and parallel-UPC configuration ............................... 45

Fig. 3.8 FL versus f characteristic in islanded mode for parallel-FFC

configuration ................................................................................. 46

Fig. 4.1 Grid-connected Microgrid with Multiple DGs .............................. 54

Fig. 4.2 Feeder Flow vs. Frequency Droop (DG1 & DG3: FFC mode) ...... 57

Fig. 4.3 Algorithm of the control-mode change of UPC-mode DGs ........... 63

Fig. 4.4 Microgrid with m feeders .............................................................. 64

Fig. 4.5 Power sharing for an islanded microgrid ....................................... 65

Fig. 4.6 Power sharing for a multi-feeder microgrid in islanded mode ....... 66

Fig. 4.7 The Grid-connected Microgrid with Three DGs ............................ 68

Fig. 4.8 Simulation results for the conventional method ............................ 70

Fig. 4.9 The simulation results for the proposed method ............................ 72

Fig. 4.10 The frequency change in the transition mode ................................ 74

Fig. 5.1 Grid Connected PV-FC Hybrid System ........................................ 79

Fig. 5.2 P&O MPPT Algorithm.................................................................. 83

Fig. 5.3 Buck-Boost converter and control ................................................. 84

Fig. 5.4 Operation Strategy of Hybrid Source in the UPC .......................... 86

Fig. 5.5 Control Algorithm Diagram in UPC mode .................................... 90

xi

Fig. 5.6 Hysteresis Control Scheme for PMSref Control ............................... 91

Fig. 5.7 Overall Operating Strategy for the Grid Connected Hybrid System93

Fig. 5.8 Simulation result without hysteresis .............................................. 99

Fig. 5.9 Improving operation performance by using hysteresis ................ 102

Fig. 6.1 System Configuration .................................................................. 105

Fig. 6.2 Active power response as LD1 changes ...................................... 108

Fig. 6.3 Voltage and frequency response in grid-connected mode ........... 109

Fig. 6.4 Frequency change due to islanding ............................................. 110

Fig. 6.5 Active power response as LD1 changes ...................................... 110

Fig. 6.6 Voltage and frequency response in grid-connected mode ........... 111

Fig. 6.7 Frequency change due to islanding ............................................. 112

Fig. 6.8 Active power response as LD1 changes during island mode ....... 113

Fig. 6.9 System voltage ............................................................................ 113

Fig. 6.10 System frequency ........................................................................ 114

Fig. 6.11 Active power response as LD1 changes during island mode ....... 115

Fig. 6.12 DG2 control mode ....................................................................... 115

Fig. 6.13 System Voltage and frequency .................................................... 116

Fig. 6.14 Load demand and feeder flow in front of HB source ................... 118

Fig. 6.15 Power response in case without control strategy ......................... 119

Fig. 6.16 System voltage and frequency ..................................................... 120


Fig. 6.18 Power response in case with control strategy .............................. 121

Fig. 6.19 The control mode of DG2 and hybrid source .............................. 122



xii

Fig. 6.22 Power response in case without control strategy ......................... 125

Fig. 6.23 The hybrid source control mode .................................................. 126



Fig. 6.26 Power response in case with control strategy .............................. 129

Fig. 6.27 The control mode of DG2 and hybrid source .............................. 130


Fig. 6.29 Active power responses of feeder 1 ............................................. 133

Fig. 6.30 Active power responses of feeder 2 ............................................. 133

Fig. 6.31 Power from the main grid ............................................................ 134

Fig. 6.32 Active power responses of feeder 2 in island mode ..................... 135

Fig. 6.33 Active power responses of feeder 1 in island mode ..................... 135

Fig. 6.34 Feeder flow in front of DG2, Pf1_DG2 ....................................... 136

Fig. 6.35 Microgrid frequency .................................................................... 136

Fig. 6.36 Active power responses in case of frequency restoration method 138

Fig. 6.37 Microgrid frequency in case of frequency restoration method .... 138

Fig. 6.38 Active power responses in case of proposed method ................... 139

Fig. 6.39 Microgrid frequency in case of proposed method ....................... 139

Fig. A.1 Three-DG microgrid in series configuration ................................ 155

Fig. A.2 Three-DG microgrid in parallel configuration ............................. 156

Fig. A.3 Power sharing of three DGs in series-FFC configuration ............ 157

Fig. A.4 Frequency response in series-FFC configuration......................... 157

Fig. A.5 Power sharing of three DGs series-UPC configuration ............... 158

Fig. A.6 Frequency response in series-UPC configuration ........................ 158

Fig. A.7 Power sharing of three DGs in parallel-FFC configuration ......... 159

xiii

Fig. A.8 Frequency response in parallel-FFC configuration ...................... 159

Fig. A.9 Power sharing of three DGs in parallel-UPC configuration ......... 160

Fig. A.10 Frequency response in parallel-UPC configuration ..................... 160

xiv

TABLES

Table 2.1 Microsource Types and Typical Capability ................................ 14

Table 2.2 Typical Line Parameters ............................................................. 24

Table 3.1 Frequency and Active Power Change in Transition Mode .......... 40

Table 3.2 Frequency and Active Power Change in Islanded Mode ............. 48

Table 3.3 Comparison of Four Configurations of Microgrid ...................... 50

Table 3.4 Simplified Comparison of Four Configurations of Microgrid ..... 51

Table 4.1 The System Parameters ............................................................... 68

Table. 5.1 The Hybrid System Parameters ................................................... 97

Table. 6.1 Scenarios without Hybrid Source .............................................. 107

Table. 6.2 Scenarios with Hybrid Source ................................................... 107

Table. 6.3 Scenarios with Multiple Feeder Microgrid ................................ 107

Table. 6.4 Proposed Method and Frequency Restoration Control Scheme 108

Chapter I:

INTRODUCTION

1.1. Motivations and Purposes

Conventional power systems are currently facing the challenges of

handling increasing electricity demand levels together with environmental

consciousness and market deregulation. This has resulted in two main

trends developed in power system recently. The first is a wide utilization of

renewable energy (RE) sources along with combined heat and power (CHP)

systems. The second is decentralization of power generation, so called

distributed generation [1, 2]. The global warming causes many challenges

to human being and it has become a hot issue than ever. In addition, the

earth’s natural resources are becoming exhausted. Therefore, the use of

renewable energy sources in power system is exploiting rapidly. There is

universal agreement that by the end of this century the majority of our

electrical energy will be supplied from RE sources. Generators powered

from renewable energy sources (except large-scale hydro and large

offshore and onshore wind farms) are typically much smaller than the

fossil fuelled and nuclear powered generators. Small generators cannot be

connected to the transmission system due to the cost of high voltage

transformers and switchgear. Also, the transmission system is often a long

way away as the geographical location of the generator is constrained by

the geographical availability of the resource. Small generators must

therefore be connected to the distribution network. Such generators are

CHAPTER I

2

known as distributed generators or microsources. They are in range of

kilowatts to megawatts.

The electricity supply is more and more based on distributed

generators (DGs). Presently, they are only injecting the available active

power into the interconnected network. At higher penetration levels they

might jeopardize the stability of the network. In a more sophisticated

approach, they should participate in network operation to guarantee a

sustainable and secure electricity supply. Distributed generation also has

the potential to increase system reliability and power quality due to the

decentralization of supply. Increase in reliability levels can be obtained if

DG is allowed to operate autonomously in transient conditions, namely

when the distribution system operation is disturbed upstream in the grid.

The DGs are connected to the grid through an electronics device such

as DC-DC converter, DC-AC inverter etc. Power electronics interface has

many advantages such as easy and flexibility in operation due to digital

controls; faster dynamic response compared to the electro mechanical

converters; lower acoustic noise when compared to electromagnetic

controllers, relays and contactors; high efficiency due to low losses in the

thyristors; long life and reduced/minimal maintenance due to the absence

of mechanical wear; control equipments using thyristors are compact in

size.

Such power grid trends make the future grid more environmentally

friendly, more reliable, more intelligent, higher power quality, as well as

enhanced the operational efficiencies. These trends also lead to new grid

concepts such as “SmartGrid”, “Intelligrid”, and “Microgrid”. As the

INTRODUCTION

3

penetration of distributed generation (DG) increases at the distribution

level, managing these systems effectively becomes increasingly

challenging. The microgrid is one effectively proposed way to manage

these systems. Microgrids are also seen as one of the cornerstones of the

future smartgrids [3].

The interconnection of small modular generation system (PV, fuel

cell, micro-turbines, small wind generator) and storage devices to Low

Voltage (LV) distribution grids will lead to a new energy system paradigm,

usually referred as the Microgrid [4-9]. Microgrid is generally defined as a

low to medium voltage distribution networks (e.g. a small urban area, a

shopping center, or an industrial park) comprising various controllable

DGs, such as micro-turbines, fuel cells, etc., uncontrollable DGs like

photovoltaic, together with storage devices, i.e. flywheels, energy

capacitors and batteries, and controllable loads that can operate either

interconnected to the main distribution network or in an islanded mode [9-

13]. The microgrid can be thought as a controllable cell of the power

system from the utility view point. For example, microgrid could be

controlled as a single dispatchable load which can respond in seconds to

meet the requirements of the transmission system. The microgrid can also

be designed to meet special needs of the customers such as local reliability

enhancement, feeder losses reduction, local voltage support, voltage sag

correction, increase of the efficiency through use of waste heat etc [4].

Microgrid is a new type of the power system. Many efforts have been

being put into it by different organizations as well as countries.

Considerable researches on microgrids currently carried out are 1) the

CHAPTER I

4

Consortium for Electric Reliability Technology Solutions (CERTS) by the

National Science Foundation Industry/University Cooperative Research

Center, the U.S., 2) the “MICROGRID” and the “MORE MICROGRIDS”

projects funded by the European Commission, 3) the three projects by the

New Energy and Industrial Technology Development Organization

(NEDO), Japan, and 4) the microgrid project by CANMET Energy

Technology Center, Varrenes, Canada. The technique is still not mature

nowadays. A lot of works have to be done until it can be put in the market.

It is still under the research and experimental stage [9].

In addition, the increase of distributed energy resources (DERs)

including intermittent renewable resources will pose many challenges, such

as protection, control, and operational cost [14], for the future power grid

operators. Especially for the distribution infrastructure and the system

operator, new flow patterns caused by DERs may require changes to the

protection and control strategies, voltage and Var management, and overall

enforcement of distribution grid infrastructure.

Under the circumstances, this dissertation will focus on the operation

and control of DGs including controllable and uncontrollable sources to

meet the various requirements of the Microgrid.

1.2. Highlights and Major Contributions:

The dissertation will focus on the active power and frequency responses in

the microgrid. The DGs output power and/or microgrid frequency can be

affected by three cases of operation: i) load variation during the grid-

connected mode, ii) load variation during islanded mode, and iii)

disconnection from the main grid.

INTRODUCTION

5

i) During the grid-connected operation mode, the local load variation can

be compensated by the DGs in the microgrid to keep the power from the

main grid constant and thus the microgrid becomes a controllable load

from the utility view point.

ii) During islanded mode, the frequency should be maintained constant

due to the load variation.

iii) During transition mode, the microgrid frequency and DGs’ outputs are

changed due to the loss of power from the utility grid. In this mode, the

local frequency and DGs’ outputs should be maintained within

predetermined limits.

In addition, different control mode of DGs and different

configuration of microgrid will lead to different changes in the microgrid

frequency and DGs’ outputs. Therefore, a proper configuration as well as

control strategies should be achieved to satisfy the above requirements.

The contributions in this dissertation will deal with such problems and can

be summarized as follows:

1) Investigate the active power and frequency response in microgrid

2) Suggestion of a proper configuration for microgrid in terms of the

structure and the DGs’ control mode.

3) Proposed control strategies for a microgrid, in grid connected and

islanded mode, to

– Improve power quality

– Reduce the loss of power

– Enhance operation efficiency

4) Proposed control strategies for hybrid source in a microgrid to

CHAPTER I

6

– Enhance generation performance

– Enhance microgrid stability

– Operate hybrid source more economically (PV-MPPT…)

5) Microgrid operation with presence of hybrid source

1.3. Dissertation Structure

While the main subject of this dissertation is analysis of droop control

scheme for different microgrid configurations and control strategies for

DGs in the microgrid, it will be necessary to look in some details at the

issues surrounding the microgrid too. To this end, Chapter II will be

devoted to the microgrid concept and control including unit power control

mode, feeder flow control mode, and droop characteristics to control the

voltage and frequency as well as power output. The droop characteristics

and power sharing among DGs in the microgrid and their advantages,

disadvantages are investigated in Chapter III. In Chapter IV, a control

strategy for controllable DGs in microgrid is proposed to take the

advantages and overcome disadvantages of different configurations and

control modes pointed out in Chapter III. Chapter V deals with operation

and control issues of hybrid system when it connects to the microgrid, a

control strategy is also proposed. The operation schemes mentioned in

Chapter IV and V are independent therefore, the two operation strategies

are integrated to a multiple feeder microgrid and the overall simulation

results as well as discussions are presented in Chapter VI. Finally, the

conclusions and future extensions are devoted in Chapter VII. The

structure of the dissertation can be summarized as in Fig. 1.1.

INTRODUCTION

7

Fig. 1.1 Dissertation structure

Chapter 5: Control Strategy for a Hybrid Source Connected to Microgrid

Chapter 2: Microgrid Concept and Control

Chapter 3: Analysis of P and f Responses in Microgrid

Chapter 1: Introduction

Chapter 4: Control Strategy for Controllable DGs in Microgrid

Chapter 7: Conclusions and Future Extensions

Chapter 6: Case Study

CHAPTER I

8

Chapter II:

MICROGRID CONCEPT AND CONTROL

2.1. Overview

Fig. 2.1 Microgrid Architecture Diagram [15]

In the context of increasing in distributed generation sources including

renewable and CHP systems, the microgrid concept is an advanced

approach for adding value to distributed energy resources by aggregating

them into autonomous grids that provide high levels of efficiency, security

DG

A

DG

B

DG

C

DG

DStatic Switch

Sensitive Loads

Non Sensitive

Utility Grid PCC

CHAPTER II

10

and controllability. The Microgrid concept is also enable integration of an

unlimited quantity of DGs into the electricity. The concepts of microgrid

were firstly proposed by Robert H. Lasseter in the CERTS microgrid

project. Basic microgrid architecture is shown in Fig. 2.1. This consists of

a group of radial feeders, which could be part of a distribution system or

building’s electrical system. The feeders are connected to the utility grid at

a single point called point of common coupling (PCC) [4]. Some feeders

(Feeder A-C) have sensitive loads, which require local generation. The

non-critical load feeders do not have any local generation. Feeders A-C can

autonomously island from the utility grid using the static switch.

The CERTS microgrid has two critical components, the static switch

and the microsource. The static switch is able to autonomously island the

microgrid from disturbances such as faults and power quality event. During

the islanded mode, the sensitive loads are supported by the local generation.

The power can be seamlessly balanced by microsource using a power

versus frequency droop controller. When the disturbances are eliminated,

the reconnection of the microgrid is achieved autonomously by the static

switch. The synchronization is achieved by using the frequency difference

between the microgrid and the main grid to match frequency and phase

angles at the connection point [16].

2.2. Microgrid Management and Control

Microgrid controllers have responsibilities to ensure that [15]:

(1) Microsources work properly at predefined operating point of slightly

different from predefined operating point but still satisfy the

operating limits;


11

(2) Active and reactive powers are transferred according to necessity of

the microgrids and/or the distribution system;

(3) Disconnection and reconnection processes are conducted seamlessly;

(4) Market participation is optimized by optimizing production of local

microsources and power exchanges with the utility;

(5) Heat utilization for local installation is optimized;

(6) Sensitive loads, such as medical equipment and computer servers are

supplied uninterruptedly;

(7) In case of general failure, the microgrid is able to operate through

black-start; and

(8) Energy storage systems can support the microgrid and increase the

system reliability and efficiency.

Based on the above responsibilities and the controller coordination,

the microgrid controls can be classified as local controls, centralized

controls, and decentralized controls.

Local controls are the basic category of microgrid control. The

measured data for local controllers are local voltages and currents [4, 15].

Local controllers are aimed to control operating point of the microsources

and their power-electronic interfaces without communication systems, to

ensure peer-to-peer and plug-and-play function of microsources, to

seamlessly connect to or disconnect from the distribution network when

needed [17].

Centralized controls, that will be used in this dissertation, base on

hierarchical controls as shown in Fig. 2.2, including three levels of control,

Distribution Management System (DMS), Microgrid Central Controllers

CHAPTER II

12

Fig. 2.2 Microgrid Control Architecture

(MGCCs), and local controllers consisting of Microsource Controllers

(MCs) and Load Controllers (LCs) [18-20]. The MCs, presented in details

below, in centralized controls have similar principle as the Local

Controllers. LCs are installed at the controllable load and commonly used

for demand side management. The MGCC has main responsibility of

optimizing the microgrid operation, MCs and LCs follow the orders of

MGCC during grid-connected mode and have autonomy to perform their

own control during islanded mode [18]. DMS has responsibility to manage


13

the operation of medium and low voltage areas in which more than one

microgrid may exist.

Decentralized controls have similar description to the centralized

control and can be explained based on Fig. 2.2. The main responsibility of

the decentralized controls is given to the MCs that compete to maximize

their production in order to satisfy the demand and probably provide the

maximum possible export to the grid taking into account current market

prices [18, 21-23].

Fig. 2.3 Microgrid Energy Management System

In order to maximize the benefit of microgrid and minimize the

global energy cost, an Energy Management System (EMS) supported by a

communication infrastructure has to be built. The EMS uses the

information on the local electrical and heat demands, weather, electricity

RES Conventional generation

Controllable loads

Uninterruptible loads

EMS

Generation Demand

Batteries, Super capacitors, Flywheel

Energy Storage

Scheduling Energy Storage

Load ForecastingLoad

Planning Generation Forecasting

Optimization of Generation

CHAPTER II

14

price, fuel cost, power quality requirements, demand side management

requests, congestion levels, etc. The information exchanged between the

EMS and the local devices are summarized in Fig. 2.3. The key functions

of the EMS are as follows [24-26]:

- to provide the individual power and voltage set point for each

microsource controller;

- to insure that heat and electrical loads are met;

- to insure that the microgrid satisfies operational contracts with the

bulk system;

- to minimize emissions and system losses;

- to maximize the operational efficiency of the microsources;

- to provide logic and control for islanding and reconnecting the

microgrid during events.

2.3. Microsource Control

TABLE 2.1: MICROSOURCE TYPES AND TYPICAL CAPABILITY

Microsource Type Capability Range

Internal combustion engines 10kW~10MW Mini- to small- size combustion

turbines 0.5~50MW

Micro turbines 20~50MW Fuel Cells 1kW~10MW

Photovoltaic systems 5W~5MW Wind turbines 30W~10MW


15

Microsource controls need to insure that new microsources can be added to

the system without modification of existing equipment, set-points can be

independently chosen, the microgrid can connect to or isolate itself from

the grid in a rapid and seamless fashion, reactive and active power can be

independently controlled, and can meet the dynamic needs of the loads.

Each microsource controller must autonomously respond effectively to

system changes without requiring data from the loads, the static switch or

other sources [4, 6].

Microsource comprises a wide range of prime mover technologies,

such as internal combustion (IC) engines, gas turbines, micro turbines,

photovoltaic, fuel cells, and wind power. The suitable generation

technologies for microgrid are shown in Table 2.1 [27]. Among those DGs,

the renewable sources like PV and Wind turbine are uncontrollable sources

if tracking maximum power point. They can be coupled with other

controllable source as a hybrid system to improve the operational

performance and make them become controllable [28-32]. In this

dissertation those controllable sources will be coordinated with each other.

The stand-alone uncontrollable sources will be considered as negative

loads.

Basically, there are two classes of microsources; one is a DC source,

such as fuel cells, and battery storage, the other is a high frequency AC

source such as the micro turbine, which needs to be rectified. In both cases,

the resulting DC voltage is converted to an acceptable AC source using a

voltage source inverter [4]. In addition, most of microsource technologies

that can be installed in a microgrid are not suitable for direct connection to

CHAPTER II

16

the electrical network due to the characteristics of the energy produced.

Therefore, power electronic interfaces (DC/AC or AC/DC/AC) are

required. Inverter control is thus the main concern in microgrid operation

[7].

Generally, inverters can be classified as line-commutated and self-

commutated inverters as shown in Fig. 2.4 [33]. The line-commutated

inverters are not used in DG interfacing because it needs an extensive filter

due to low order harmonics, and it is not capable to operate in islanded

mode since it relies on the distribution system waveform for

communication. Most inverter interfaces for DG, however, utilize self-

Inverters

Self-Commutated Inverters

Voltage Source Inverters

Current Source Inverters

Current-Control Scheme

Voltage-Control Scheme

Line-Commutated Inverters

Fig. 2.4 Classification of Inverters


17

commutated inverter because the high switching frequency can be easily

filtered, and unlike the line-commutated inverter, the total distortion of the

self-commutated inverters does not increase in proportionality to the

number of installed units due to phase cancellations. The self-commutated

inverter can be divided into two main types based on nature of the DG link:

current source inverter (CSI) and voltage source inverter (VSI). The CSI is

equipped with a large inductor on the DC side and act as a current source.

On the other hand, the VSI is normally equipped with a large capacitor on

the DC side to act as a voltage source. VSIs are typically used in DG

interfacing since DGs resemble voltage sources more than current sources.

According to the above expressions, in this study, the VSIs will be used to

interface DGs with grid. The voltage source inverter-based microsource

diagram is shown in Fig. 2.5.

Inverter Controller

To

n XF

CF

X

Gate Pulses

VDC vabc(t)

eabc, iinv,abc

iabc

E

Filter InductorVSI

DC Storage

Micro-source

From Grid

Loca

l Fee

der

Fig. 2.5 Inverter-based Microsource Diagram

CHAPTER II

18

Fig. 2.5 shows a block diagram of the power-electronic-interface

microsource that is coupled to a microgrid through a voltage-source

inverter. The inverter is connected to a DC voltage generated by the

microsource and converts the input DC voltage to a three-phase voltage

with a desired frequency, voltage magnitude and phase angle at the output

terminals. If the generation at the microsource site is retrofitted into the

facility, a common transformer arrangement is delta-wye (grounded) [34-

36]. This arrangement is typically chosen to provide isolation for the utility

from ground faults in a microsource system’s facility, and to supply a

ground source for that facility. I does not cause an over voltage for a

ground fault in the microgrid site.

2.3.1 Power Control Modes

Unit Power Control Mode:

In the UPC mode, the DG output power is regulated at a constant level

(PDGref). In order to control the DG output, the voltage (V) at the

interconnection point and the current (I) injected by the DG are measured

as shown in Fig. 2.6(a). The active power injected by the DG (PDG) is

calculated from the measured voltage and current, and then fed back to the

inverter controller.

When the microgrid is connected to the main grid, DG regulates its

output to a constant power regardless of the load variation. If the load

demand is changed anywhere in the microgrid, the extra power will be

compensated by the main grid. On the other hand, when the microgrid

disconnects from the main grid with respect to the islanded mode, DGs


19

must follow the load demand accurately.

In numerous studies, a power versus frequency (P–f) droop control, that will be discussed later on, has been adopted for DG power-sharing methods [9, 12, 37-39]. This control uses the frequency of the microgrid as a common signal among the DGs to balance the active power generation of the system [12]. P–f droop-based power controllers have proven to be robust and adaptive to variation in the power system

(a)

(b)

Load

V

PDG

PDGref

Grid

DG

Static Switch

Controller

I

V

PDG

FLLineref

Grid

DG

Static Switch

IFeeder

FLLine

LoadController

Fig. 2.6 Power-control mode of a DG: (a) Unit output power control (UPC), (b) Feeder flow control

CHAPTER II

20

operational conditions, such as frequency- and/or voltage-dependent loads and system losses [12, 38].

Feeder Flow Control Mode:

The objective of the FFC mode is to control the active power flow in the

feeder where the unit is installed at a desired value (FLLineref). In this mode,

the DGs regulate the voltage magnitude at the connection point and the

power flow in the feeder at connection point (FLLine). The feeder current

(IFeeder) and voltage at the connection point (V) are measured in order to

calculate the power as shown in Fig. 2.6(b).

During the grid-connected mode, extra load demands are picked up

by the DGs, and power supplied from the main grid remains unchanged

regardless of the load variation within the microgrid. Thus the microgrid

looks like a controllable load from the utility view point. On the other hand,

in the islanded mode when the microgrid is disconnected from the main

grid, the feeder flow versus frequency (FL–f) droop characteristic is used

to share the load demand [4, 16].

Derivation of droop characteristics used in the FFC mode and the

UPC mode will be discussed below.

2.3.2 Droop Control through Active and Reactive Power

The power injecting to the feeder by the microsource, as represented in Fig.

2.7, is described as follows [40, 41]:


21

Fig. 2.7 Inverter-based system (a) Simplified Interface of Inverter System

(b) Phasor diagram.

( ) ( )

**

2

2.1

j

j

jj

V EP jQ S VI VZ

V EeVZe

V VEe eZ Z

δ

θ

θ δθ

−

+

⎛ ⎞−+ = = = ⎜ ⎟

⎝ ⎠⎛ ⎞−

= ⎜ ⎟⎝ ⎠

= −

Thus, the active and reactive powers flowing into line are

( ) ( )

( ) ( )

2

2

cos cos 2.2

sin sin 2.3

V VEPZ Z

V VEQZ Z

θ θ δ

θ θ δ

= − +

= − +

(a)

Inverter

V∠0 E∠-δ

Z=R+jX

P, Q

I∠-Φ

V

E

-δ

-φ

I RI

jXI

(b)

CHAPTER II

22

With jZ Ze R jXθ= = + , (2.2) and (2.3) are rewritten as

( ) ( )

( ) ( )

2 2

2 2

cos sin 2.4

sin cos 2.5

VP R V E XER X

VQ RE X V ER X

δ δ

δ δ

= − +⎡ ⎤⎣ ⎦+

= − + −⎡ ⎤⎣ ⎦+

Or

( )

( )

sin 2.6

cos 2.7

PX QRVE

PR QXV EV

δ

δ

−=

+− =

For overhead line X R , this means that R may be neglected. Equations

(2.4) and (2.5) then become

( )

( ) ( )

sin 2.8

cos 2.9

VEPXVQ V EX

δ

δ

=

= −

E is the voltage magnitude at the point of common coupling.

If the power angle δ is also small (less than 10 degrees [42]), then

sin and cos 1δ δ δ . Equations (2.6) and (2.7) then become


23

( )

( )

2.10

2.11

PXVE

QXV EV

δ

− =

Equation (2.10) and equation (2.11) show that the power angle depends

pre-dominantly on P, whereas the voltage difference depends pre-

dominantly on Q. In other words, the angle δ can be controlled by

regulating P, whereas the inverter voltage V is controllable through Q.

Control of the frequency dynamically controls the power angle and, thus,

the real power flow. As a result, by adjusting P and Q independently,

frequency and amplitude of the grid voltage are determined. In other words,

it is possible to independently control the real and reactive power. These

conclusions form the basis for well-known frequency and voltage droop

regulation through respectively active and reactive power:

( ) ( )( ) ( )

0 0

1 0 0

2.12

2.13P

q

f f k P P

E E k Q Q

− = − −

− = − −

f0 and E0 are rated frequency and grid voltage respectively, and P0 and Q0

are the (momentary) set points for active and reactive power of the inverter.

The droop characteristics shown in equation (2.12) and equation

(2.13) are for the conventional power system where the active power

relates with the frequency and reactive power relates with the voltage. On

the contrary, the medium voltage line has mixed parameters and the low

voltage line is even predominantly resistive as shown in Table 2.2 [43].

CHAPTER II

24

The characteristic of low voltage line causes reversed droop characteristics

where the active power is related with the voltage and the reactive power is

related with the frequency [43, 44]. However, it is also stated in [43] that

the droops used in the conventional grid can be effectively used in the low

voltage level due to their “indirect operation”. Therefore, the control

strategy of conventional grid can be down scaled to the low voltage level

without any restrictions.

TABLE 2.2:

TYPICAL LINE PARAMETERS

Type of line R Ω/km

X Ω/km

R X

Low voltage line 0.642 0.083 7.7

Medium voltage line 0.161 0.190 0.85

High voltage line 0.06 0.191 0.31

Reactive Power (Q) versus Voltage (V) Droop:

Integration of large number of microsources into a microgrid could

experience voltage and/or reactive power oscillations. Therefore, instead of

basic unity power factor controls, voltage regulation is necessary for local

reliability and stability [6, 42].

Voltage control must insure that there are no large circulating

reactive currents between sources. Unlike a large power system, the

impedance between microsources in a microgrid is small. With small errors

in voltage set points, the circulating current can exceed the ratings of the


25

microsources. This can be prevented by a voltage versus reactive power

droop control, Fig. 2.8, which is described in equation (2.13).

Fig. 2.8 Voltage vs. Reactive Power Droop

Fig. 2.8 illustrates the basic function of a controller that operates

based on the voltage versus reactive power droop. When the reactive power

generated by the microsource becomes more capacitive, the local voltage

set point is reduced. Conversely, as Q becomes more inductive, the voltage

set point is increased. The reactive power limit Qmax is a function of the

volt-ampere VA rating of the inverter and the real power of the prime

mover P as follows:

( ) ( )22 2max 2.14Q VA P= −

Unit Power (P) versus Frequency (f) Droop:

When the microgrid is connected to the utility grid, loads receive power

both from the grid and from local microsources, depending on the

Qmax -Qmax

Qcapacitive Qinductive

V

CHAPTER II

26

customer’s situation. With loss of the grid due to voltage drops, faults,

blackouts etc. the microgrid smoothly transfer to island operation.

When regulating the output power, corresponding to UPC mode, each

microsource has a constant negative slope droop on the P, f plane. With the

disconnection from the grid, the microsources will participate in sharing

the local loads according to P versus f droop characteristic as in equation

(2.12). The droop characteristics described in equation (2.12) can be

rewritten as follows:

( ) ( )0 0 2.15Uf f K P P− = − −

Where, KU is droop constant of the UPC-mode DG.

The control diagram of equation (2.15) is expressed in Fig. 2.9. The

error of active power is then sent to the PI controller to control the output

power to the predetermined value, P0.

Consider two microsources with the power set points P01, P02 for two

units, as shown in Fig. 2.10. This is amount of power injected by each

source when connected to the utility grid, at system frequency.

f0 -1/K

PI

f P0

P

+ _

_ +

+

Fig. 2.9 Diagram of P vs. f Droop control block


27

Fig. 2.10 Power vs. Frequency Droop

If the system transfers to island when importing power from the grid,

then the generation needs to increase power to balance power in the island.

The new operating point will be at a frequency that is lower than the

nominal value. In this case both sources have increased their power output

(P1’ and P2’). If the system transfer to island when exporting power to the

grid, then the new frequency will be higher, corresponding to a lower

power output from the sources (P1” and P2”).

The characteristics shown in Fig. 2.10 are steady state characteristics.

Each source has a fixed slope in the region where the unit is operating

within its power range. The slope becomes vertical as soon as any limit is

reached. The droop is the locus where the steady state points are

constrained to come to rest, but during dynamics the trajectory will deviate

from the characteristic.

f

P

fo

Exporting to Grid

Importing from Grid

P1’

f ’

f”

P2’P1” P2”

P01 P02

CHAPTER II

28

Feeder Flow (FL) versus Frequency (f) Droop:

When regulating the feeder flow FL, the FL vs. f can be derived from (2.12)

as follows:

( ) ( )0 0 2.16Ff f K FL FL− = −

Where KF is the FFC droop constant.

The control diagram of equation (2.16) is expressed in Fig. 2.11. The

error of feeder flow is then sent to the PI controller to regulate the feeder

flow to the predetermined value, FL0.

When regulating FL the relative location of loads and source is

important [6]. Fig. 2.12 shows two possible microgrid configurations,

series and parallel. Fig. 2.13 shows the set points FL01 and FL02 for the two

units when connected to the utility system.

f0 1/KF PI

f FL0

FL

+ _

_ +

+

Fig. 2.11 Diagram of FL vs. f Droop control block


29

Fig. 2.13 Feeder Flow vs. Frequency

f

fo

Exporting to Grid

Importing from Grid

f ’

f” FL01 FL02

FL

LD

FL1

Utility Grid

LD

FL2

LD

FL1

Utility Grid

LD

FL2

(a) Series Configuration

(b) Parallel Configuration

DG1 DG2

DG1

DG2

Fig. 2.12 Microgrid Configuration with Two Sources

CHAPTER II

30

When in series configuration, FL01 is the grid flow. The microgrid is

exporting power to the grid, since flow is negative. When the system

transfers to the island, the flow reaches zero and the frequency increases

(squares).

In the parallel configuration, the grid flow is the algebraic sum of the

two flows. Since |FL02|>|FL01| the microgrid is importing power from the

grid. Fig. 2.13 shows that in island mode FL01=-FL02 and the frequency is

reduced (triangles).

In both the unit power control mode and the feeder flow control

mode DG output is limited to its generation band by using the control

block shown in Fig. 2.14.

f0 -1/K

PI

f P0

P

+ _

_ +

+

Fig. 2.14 Control block diagram to enforce the output limit

Pmax +

_

Kis

0

Pmin+

Kis

0

P

_

+

+ +

errPmax

errPmin

Chapter III:

ANALYSIS OF ACTIVE POWER AND FREQUENCY

RESPONSE IN MICROGRID

3.1. Overview

This considering multiple-DG microgrid connects to the main grid at a

point of common coupling (PCC) via a static switch (SS). DGs in the

microgrid can connect to the feeder either in series or parallel. Each DG

can also operate in unit power control (UPC) mode or feeder flow control

(FFC) mode. In this chapter responses active power and frequency in the

microgrid with different configurations (parallel/series), different control

modes (UPC/FFC), and different operation modes (grid-

connected/islanded) is investigated. Four possible configurations, as shown

in Fig. 3.1, are considered. The analysis shows the power sharing among

DGs in the microgrid according to the droop characteristic as well as the

frequency changes according to the load change and mode change.

Additionally, advantages and disadvantages of each configuration are

achieved.

The DGs connect to the microgrid in series or parallel and form the

microgrid in different configurations. Four possible configurations that will

be investigated, as shown in Fig. 3.1, are:

- Configuration 1: Series-FFC, the DGs are connected in series and the

DGs’ control modes are FFC.

CHAPTER III

32

Fig. 3.1. Different configurations of microgrid: (a) Configuration 1: Series –

FFC, (b) Configuration 2: Series – UPC, (c) Configuration 3: Parallel – FFC, (d)

Configuration 4: Parallel – UPC

- Configuration 2: Series-UPC, the DGs are connected in series and

DGs’ control modes are UPC.

FL1

FFC

…

FLPCC

P1

(c)

1

Main Grid

LD1

FL2

FFC

P2

2 LD2

FLn

FFC

Pn

n LDn

FL1

UP

…

FLPCC

P1

(d)

1

Main Grid

LD1

FL2

UP

P2

2LD2

FLn

UP

Pn

nLDn

(a) (b)

1

LD1FLPCC

LD2

FFC

FL2 FLn

UPC UPC UPC

LDn

MainGrid

LD1FLPCC

LD2

FL2 FLn

LDn

2

FFC

n

FFC

1 2 n Main Grid

ANALYSIS OF ACTIVE POWER AND FREQUENCY RESPONSE IN MICROGRID

33

- Configuration 3: Parallel-FFC, the DGs are connected in parallel and

DGs’ control modes are FFC.

- Configuration 4: Parallel-UPC, the DGs are connected in parallel and

DGs’ control modes are UPC.

The two control modes, UPC and FFC, as well as droop

characteristics were basically presented in chapter II.

3.2. Active Power and Frequency Response in a Microgrid

Both UPC mode and FFC mode were investigated in [4, 9, 12, 16, 31, 32,

37-39]. The active power and frequency responses were also presented in

[4, 6, 42] where the two-DG microgrid was mentioned. In addition, the

change in frequency and active power was investigated in three different

conditions, condition of load change during the grid-connected mode,

disconnection from the main grid, and load change during the islanded

mode. A comparison was also achieved according to the analysis [45].

However, the references only consider the active power and frequency

responses in case of either one- or two-DG microgrid. In analysis of [45],

the droop coefficient of UPC mode (|KU|) and FFC mode (|KF|) are chosen

the same. Therefore, the conclusion was then applicable to particular cases.

In this section, an analysis of frequency and power changes for

multiple-DG microgrid is investigated with arbitrary droop coefficients

[46]. The four possible configuration of microgrid, Fig. 3.1, considered.

The analysis is based on three typical conditions: (A) Disconnection from

the main grid, (B) Load change during grid-connected mode, and (C) Load

change during the islanded mode.

CHAPTER III

34

3.2.1 Active power and frequency response during transition mode

In this section the change of active power and frequency when the

microgrid disconnects from the main grid is considered. Before the

disconnection, the system frequency is equal to nominal value, DGs’ output

is Pi0 and feeder flow is FLi

0, i=1÷n. When the microgird disconnects from

the utility grid, the loss of power from/to the main grid may cause the

change in the local frequency and DGs’ outputs. The change of frequency

and active power are changed according to the droop characteristic.

Configuration 1: Series–FFC

Fig. 3.2. FL vs. f characteristic in transition mode for series–FFC

configuration

In this configuration all DGs work in the FFC mode, Fig. 3.1(a). The

feeder flow vs. frequency droops of DGs are depicted in Fig. 3.2. The

change in feeder flow, frequency and thus the DGs’ output depends on the

droop characteristic and initial conditions. The DGs are connected in series

therefore the first feeder flow FL10 is the power from/to the main grid.

f 1

f 0

FL11=0 FL1

0 FL21

f

FLn1 FLn

0 FL20

DG1 DG2 DGn

f f


35

After the disconnection, FL1 become zero due to non-power from the main

grid then we have:

( )1 11 0 3.1PCCFL FL= =

From (2.16) and (3.1) we have:

( )

( )

0 1 01

22

* 3.2

,..., 3.3

PCC

nn

f f f KF FLf fFL FL

KF KF

Δ = − =

Δ ΔΔ = − Δ = −

And the changes in DGs’ output are as follows:

( )1 1

1 1 2

3.4...

n n

n n n

P FLP FL FL

P FL FL

− −

Δ = −Δ⎧⎪Δ = −Δ + Δ⎪⎨⎪⎪Δ = −Δ + Δ⎩

From (3.2)-(3.4) we have:

( )

01

1

*3.51 1

PCC

ii i

f KF FL

P fKF KF+

⎧Δ =⎪

⎛ ⎞⎨Δ = Δ −⎜ ⎟⎪

⎝ ⎠⎩

Equation (3.5) shows the frequency and DGs’ output change when the

microgrid disconnects from the main grid for configuration 1.

Configuration 2: Series–UPC

CHAPTER III

36

Fig. 3.3. P vs. f characteristic in transition mode for series–UPC

configuration

In this configuration all DGs work in the UPC mode, Fig. 3.1(b). In grid-

connected mode the load change is compensated by the main grid, the DGs

regulate the output power to the reference value. When microgrid is

isolated, all the DGs participate in sharing the load according to the active

power vs. frequency droop characteristic as shown in Fig. 3.3. After the

disconnection, the total power change of DGs equal to the loss of power

from the main grid, therefore:

( )0

1

3.6n

PCC ii

FL P=

= Δ∑

From (2.15) and (3.6) we also have:

f 1 f 0

P10 P1

1 P20

f f

P21

DG1 DG2 DGn Pn

0

f

Pn1


37

( )

0

1

13.7

PCCn

i i

Ui

i

FLf

KUfP

KU

=

⎧Δ = −⎪⎪⎪⎨⎪ Δ⎪Δ = −⎪⎩

∑

Configuration 3: Parallel–FFC

Fig. 3.4. FL vs. f characteristic in the transition mode for the parallel-FFC

configuration.

In this configuration, the DGs are connected in parallel, all DGs work in

FFC mode as depicted in Fig. 3.1(c). Before the disconnection,

( )0 0 0 01 2 ... 3.8n PCCFL FL FL FL+ + + =

Unlike configuration 1, after disconnection the FL11 is not equal to zero,

however, the sum of all feeder flow is equal to zero:

( )1 2 ... 0 3.9nFL FL FL+ + + =

f 2 f 1

FL10FL1

1 FL21

f f

FLn1

f

FLn0 FL2

0

DG1 DG2 DGn

CHAPTER III

38

From (2.16) we can also have:

( )

01 1

1

02 2

2

0

3.10...

n nn

fFL FLKF

fFL FLKF

fFL FLKF

Δ⎧ = −⎪⎪

Δ⎪= −⎪

⎨⎪⎪

Δ⎪ = −⎪⎩

From (3.8), (3.9) and (3.10) we have:

( )

0

1

11

3.11

PCC n

i i

ii

f FL

KFfP

KF

=

⎧Δ =⎪⎪⎪⎨⎪ ΔΔ =⎪⎪⎩

∑

Configuration 4: Parallel–UPC

In this parallel configuration, all DGs work in UPC mode. If the

disconnection from the main grid happens, the DGs output will change

according to the droop characteristic shown in Fig. 3.5 to meet the load

demands. From droop characteristic will also have:


39

Fig. 3.5. FL vs. f characteristic in transition mode for parallel-UPC

configuration

( )

01 1 1

1

02 2 2

2

0

1

13.12

...1

n n nn

P P P fKU

P P P fKU

P P P fKU

⎧Δ = − = − Δ⎪⎪⎪Δ = − = − Δ⎪⎨⎪⎪⎪Δ = − = − Δ⎪⎩

( )01 2 ... 3.13n PCCP P P FLΔ + Δ + + Δ =

From (3.12) and (3.13) we have:

( )

0

1

13.14

PCCn

i i

ii

FLf

KUfP

KU

=

⎧Δ = −⎪⎪⎪⎨⎪ Δ⎪Δ = −⎪⎩

∑

f 1

f 0

P10 P1

1 P20

f f

P21

DG1 DG2 DGn Pn

0

f

Pn1

CHAPTER III

40

The above four configurations have been analyzed in term of the frequency

and active power change due to the disconnection from the main grid.

From (3.5), (3.7), (3.11) and (3.14) the change of frequency and active

power can be summarized as shown in Table 3.1.

TABLE 3.1:

FREQUENCY AND ACTIVE POWER CHANGE IN TRANSITION MODE

Δf ΔP

Config. 1 01 * PCCf KU FLΔ = −

1

1 1i

i i

P fKU KU +

⎛ ⎞Δ = −Δ −⎜ ⎟

⎝ ⎠

Config. 2 &

Config. 4

0

1

1PCC

n

i i

FLf

KU=

Δ = −

∑ 1

ii

P fKU

Δ = −Δ

Config. 3

0

1

1PCC

n

i i

FLf

KU=

Δ =

∑ 1

ii

P fKU

Δ = −Δ

3.2.2 Active power and frequency response during grid-connected

operation mode

The frequency remains unchanged in all cases during grid-connected

operation mode.

Configuration 1: Series–FFC

If the load changes, the DG output will change to match the load and

regulate the feeder flow unchanged. The response of DG output starts from


41

the nearest DG to the furthest DG (DG1) from the load. For example, when

load demand LDi increases, DGi increase its output until its maximum,

then the next DG (DGi-1) increases until its limit, and so on. If DG1 output

reaches its maximum, the variation in load will be matched by the main

grid, hence the feeder flow will be changed and the microgrid is no longer

a constant load from the main grid view point.

Configuration 2: Series–UPC

All DGs are in the UPC control mode therefore their output powers are

regulated to be unchanged. As a result, the power comes from/to the main

grid will change according to the load variations and the microgrid is a

non-dispatchable entity from the utility point of view.

Configuration 3: Parallel–FFC

Every change in load demand is picked up by the DG in the same feeder.

Once the DG output reaches to its limit, the rest of load variation will be

matched by the main grid, thus the feeder flow at PCC is non-constant and

the microgrid is no longer a constant load from the utility view point. In

this configuration, the change in load of each feeder does not affect to

other feeders operation. In brief, the feeder flow at PCC remains

unchanged until one of the DGs reaches to their limit.

Configuration 4: Parallel–UPC

The power sharing in this configuration is same to the configuration 2, all

DGs’ output powers are regulated unchanged and the load variations are

matched by the main grid. Therefore, the feeder flow at PCC point is

CHAPTER III

42

always changed according to the load change and the microgrid becomes a

non-constant load from the main grid view point.

3.2.3 Active power and frequency response in islanded operation mode

In this section, the islanded operation mode is considered to analyze the

frequency and active power responses. The four configurations will also be

investigated as occurring load variations.

In islanded mode, there is no power from the main grid therefore all

DGs have to participate in sharing the load demand. If the load changes,

each of the different configurations will lead to the different changes in

frequency and DGs’ output.

Configuration 1: Series-FFC

Fig. 3.6. FL vs. f characteristic in islanded mode for series-FFC

configuration when load LDi increases

f(1)

0

f(2)

FLi(1,2)

f f

FLn(1) FLn

(2)

f

DG1 DG2÷i DG(i+1)÷n

(1)

(2)

(1)

(2)

(1): LDi increases & DG1 does not reaches to its limit P1<P1max

(2): P1=P1max, LDi keep increasing ΔLDi

Δf


43

In this configuration, if load demand increases/decreases, the nearest

DG will increase/decrease its output to match the load change and regulate

the feeder flow unchanged as long as the nearest DG’s output reaches its

limit, then the next DG will compensate the load change. In other words,

the load change is matched by from the nearest DG to the furthest DG

(DG1). If the DG1 reaches its limit, the power will be shared by the

remained DGs, as a result, the frequency and feeder flow will change

according to the FL-f droop characteristic as shown in Fig. 3.6.

When the frequency starts to decrease from Fig. 3.6 we have:

( )1 2

11

... 03.15

,...,

i

i ni n

FL FL FLf fFL FL

KF KF++

Δ = Δ = = Δ =⎧⎪

Δ Δ⎨Δ = − Δ = −⎪⎩

( )

1 1

1 1 2

1

...3.16

0...

0

n n

n n n

i i i

i

P FLP FL FL

P FL FLP

P

− −

+ + +

Δ = −Δ⎧⎪Δ = −Δ + Δ⎪⎪⎪Δ = −Δ + Δ⎨⎪Δ =⎪⎪⎪Δ =⎩

Equations (3.15) and equation (3.16) deduce:

CHAPTER III

44

( )1 1

1 3.17n

kk i

P fKF= +

Δ =Δ∑

On the other hand,

( )1

3.18n

k ik

P LD=

Δ =Δ∑

From (3.17) and equation (3.18) we have:

( )1 3.19i if KF LD+Δ = Δ

( )

( )( )

1

0; 1...3.201 1 ; 1...

k

kk k

P k i

P f k i nKF KF +

Δ = =⎧⎪

⎛ ⎞⎨Δ = Δ − = +⎜ ⎟⎪

⎝ ⎠⎩

Configuration 2 & 4: Series-UPC & Parallel-UPC

The frequency and power responses in both configuration 2 (series-UPC)

and configuration 4 (parallel-UPC) are the same for islanding operation

mode. All DGs work in UPC mode therefore all DGs participate in sharing

the load demand according to P vs. f droop characteristic, Fig. 11.


45

Fig. 3.7. P vs. f characteristic in islanded mode for series-UPC

configuration and parallel-UPC configuration

The frequency and DGs’ output is changed whenever the load changes.

According to the droop characteristic shown in Fig. 3.7 we have:

( )1

3.21n

ii

LD P=

Δ = Δ∑

( )0 3.22Ui i i

i

fP P PKUΔ

Δ = − = −

Equations (3.21) and equation (3.22) deduce:

( )

1

3.231n

i i

LDf

KU=

ΔΔ = −

∑

f (2)

f (1)

P1(1) Pi

(1)

f f

Pi(2)

(DG1) (DGi) (DGn)

Pn(1)

f

Pn(2) P1

(2)

CHAPTER III

46

Configuration 3: Parallel-FFC

Fig. 3.8. FL versus f characteristic in islanded mode for parallel-FFC

configuration

In this configuration, all DGs play the same role to regulate its own feeder

flow unchanged as long as its output power does not reach the limit. If the

DG’s output reaches its limit, the remained DGs will participate in sharing

the load.

Fig. 3.8 shows the droop characteristic of all DGs when load change

in LD1. Any change in load demand LD1, the load variation is picked up

by DG1 first therefore FL1 and frequency remain constant, corresponding

to f(1). If DG1 output reaches its limit, the load will be shared by remained

DGs (DG2÷DGn) thus the frequency and feeder flow will be changed

according to FL vs. f droop characteristic, corresponding to f(2). As seen in

Fig. 3.8, the frequency is determined by the droop characteristics of other

DGs except DG1. Therefore we have:

f(1)

0

f(2)

FL21FL2

2

f f

FLn1 FLn

2

f

(DG1) (DG2) (DGn)

FL12FL1

1


47

( )

( )

1 1

1 2

3.24, 2...

... 0 3.25

ii

n

FL LDfFL i n

KF

FL FL FL

Δ = Δ⎧⎪

Δ⎨Δ = − =⎪⎩Δ + Δ + + Δ =

We also have:

( )3.26i ii

fP FLKFΔ

Δ = −Δ =

From equations (3.24), (3.25) and (3.26) we have:

( )

( )

1

1

3.271

3.28

n

i i

ii

LDf

KF

fPKF

=

ΔΔ =

ΔΔ =

∑

From the analysis of four configurations in islanded operation mode

we can also summarize the change of frequency and DGs’ output as shown

in Table 3.2.

It is noted that, in configuration 2 and configuration 4 any change in

load will lead to the change in frequency. However, in configuration 3, the

frequency only changes if one of the DGs output reaches its limit. Besides,

in configuration 1, the frequency will change if the outputs of all the DGs

(DG1÷DGi) in front of the load change (LDi) reach their limit. Therefore,

CHAPTER III

48

the reserve before the change in frequency in configuration 1 is higher than

other configurations.

3.2.4 Comparison of the Different Microgrid Configurations

TABLE 3.2:

FREQUENCY AND ACTIVE POWER CHANGE IN ISLANDED MODE

Δf ΔP

Config. 1 1i if KF LD+Δ = Δ

0; 1...kP k iΔ = =

1

1 1 ;

1...

kk k

P fKF KF

k i n+

⎛ ⎞Δ = Δ −⎜ ⎟

⎝ ⎠= +

Config.

2 & 4 1

1n

i i

LDf

KU=

ΔΔ = −

∑

Ui

i

fPKUΔ

Δ = −

Config. 3 1

1

1n

i i

LDf

KF=

ΔΔ =

∑

ii

fPKFΔ

Δ =

According to the above analysis the comparison of both cases are presented

in Table 3.3 and simplified in Table 3.4, where the notation “X” means less

advantage or negative and notation “√” means more advantage or positive.

Table 3.4 shows that configuration 1 is more advantage than other

configurations in grid-connected mode and islanded mode. However, in


49

transition mode this configuration has more drawbacks in terms of

frequency and DGs’ output changes. Besides, if the DGs operate in the

UPC mode with respect to configuration 2 and configuration 4, it shows

negative performance comparing to other configurations in both grid-

connected mode and islanded mode. Table 3.4 shows that the configuration

3 has more advantage over other configurations except the reserved power

in the islanded mode. The comparison also shows that the FFC mode

presents a better performance in both cases parallel and series

configuration.

From the analysis it is suggested that the first DG of each feeder

should work in FFC mode to regulate the microgrid as a constant load from

the utility viewpoint.

In this chapter, the microgrid with four possible configurations is

considered. The changes of the frequency and active power of each

configuration due to the different control mode of DG (UPC/FFC) and

different operation modes (grid-connected / islanded / transition) were

investigated. The study shows that: 1) in both parallel and series

configurations, the FFC mode has more advantages over the UPC mode in

terms of frequency change and active power reserve; 2) in islanded

operation mode, the configuration of series–FFC has more reserve to

regulate frequency unchanged and therefore it has more advantage over

other configurations. Otherwise, in grid-connected mode and transition

mode, the parallel–FFC configuration is better.

Simulation results show the active power and frequency responses in

islanded mode of a three-DG microgrid are shown in Appendix A.

CHAPTER III

50


51

TABLE 3.4:

SIMPLIFIED COMPARISON OF FOUR CONFIGURATIONS OF MICROGRID

Configuration Config. 1 Config. 2&4

Config. 3

Transition mode

Δf x √ √ ΔP x √ √

Grid-Connected

f √ √ √ FL √ x x

Islanded f √ x x

reserve √ x x

Each configuration has advantages and disadvantages as well.

According to these results, a control strategy is proposed in Chapter IV to

overcome the disadvantages and inherit the advantages as well.

CHAPTER III

52

Chapter IV:

CONTROL STRATEGY FOR CONTROLLABLE DGS IN MICROGRID

4.1. Overview

As mentioned earlier, the microgrid can either connect to the main grid or

work autonomously with respect to the grid-connected operation mode or

the islanded operation mode, respectively. In the grid-connected operation

mode, the microgrid is connected to the main grid at the point of common

coupling (PCC) to deliver power to the load. During the transition mode

(from the grid-connected mode to the islanded mode or vice versa) the

frequency changes due to the loss of power from/to the main grid. Thereby,

in the grid-connected mode, the higher the power that comes from/to the

main grid, the larger the frequency changes that occurs due to

disconnection or reconnection. Additionally in some cases, e.g. the contract

between the utility operator and distributor, the microgrid needs to

minimize the change of power flow between the microgrid and the main

grid [7, 47]. To achieve this goal, a multiple-FFC configuration and related

power sharing method was proposed in [6, 45], in which the first DG that

is connected to PCC was always operated in the FFC mode to regulate the

power from/to the main grid to remain unchanged. However, this

conventional method has limitations in terms that, if the first FFC-mode

DG output reaches its limit, the variation in load will be matched by power

coming from the main grid and hence the feeder flow no longer remains

CHAPTER IV

54

unchanged.

In order to overcome the above limitations and the disadvantages

indicated in Chapter III, as well as to take over the advantages of each

configuration, a control strategy is proposed in this chapter to operate the

microgrid based on the two control modes of DG, the UPC mode, the FFC

mode, and droop characteristic. The mixed configuration with the first DGs

in the FFC mode is used hereafter to derive the control scheme.

4.2. Control strategy for Microgrid in the Grid-connected Mode

4.2.1. Control Strategy for Single-Feeder Microgrid

Fig. 4.1. Grid-connected Microgrid with Multiple DGs

This section presents a method to operate DGs in the microgrid as depicted

in Fig. 4.1. The studied microgrid includes small scale DGs which are in

UPC- or FFC-mode, and the control mode of the first DG (connecting to

MainGrid

LD1 LD4FLPCC

P2

LD2 LD3

FFC FFCUPC UPCFLPCC

ref P20 P4

0FL3ref

LD1max LD4

maxLD2max LD3

max

P1 P3 P4

1 2 3 4

P1max P3

maxP2max P4

max


55

the PPC) is the FFC. Before discussing the proposed power sharing method,

the key points of the conventional approach is stated.

a. Conventional Control Strategy

In general, the power generated by each DG is determined by its control

mode (UPC/FFC) and the droop characteristic [6, 16, 45]. In the grid-

connected operation mode, the features of the microgrid can be

summarized as follows:

- The frequency is equal to the main grid frequency (e.g. 60Hz).

- UPC mode DGs output remains unchanged.

- Any variation in load is matched by the FFC-mode DGs and by the

power that coming from the main grid.

- The first DG (DG1) regulates the feeder flow at the PCC point

(FLPCC) to a constant by changing its output to match the load

demand. FLPCC remains constant, and thus the microgrid can be a

constant load from the main grid view point until the DG1 output

reaches its limits. If the variation of loads exceeds the capacity of the

DG1, the microgrid becomes a non-constant load from the main grid

view point.

In the system depicted in Fig. 4.1, DG1 and DG3 are in the FFC

mode whereas DG2 and DG4 are in the UPC mode. If the load demand

LD3 and/or LD4 increases, DG3 output will increase in order to

compensate the extra power, whereas the UPC-mode DGs (DG2 and DG4)

output remains unchanged. If DG3 output reaches its maximum, DG1 will

increase its output to supply the load instead of DG3 and regulate FLPCC to

CHAPTER IV

56

a constant (i.e. FLPCC0). If DG1 output reaches its maximum, the extra

power will be provided by the main grid and hence the feeder flow at the

PCC point FLPCC will be increased. Therefore, the microgrid is no longer a

constant load from the utility view point.

If the main grid power is lost because of IEEE 1547 events, e.g.

voltage sags, faults, blackout, etc., the microgrid can autonomously

transfer to island operation [48]. In the islanded mode, the microgrid

frequency is controlled by the DGs based on the power versus frequency

(P-f) droop characteristic. If the system transfers to island when importing

power from the grid, then the DGs needs to increase the power output to

balance the power in the island. The new operating point will be at a

frequency that is lower than the nominal value. On the contrary, if the

system transfers to island when exporting power to the grid, then the new

frequency will be higher than the nominal value. The feeder flow vs.

frequency droop characteristic for FFC-mode DGs is shown in Fig. 4.2.

DG1 and DG3 are in the FFC mode and the feeder flow vs. frequency

droops for DG1 and DG3 are shown in Fig. 3a and Fig. 3b, respectively. In

the grid-connected operation mode, the microgrid frequency is equal to the

main grid frequency of (f0), the DG1 feeder flow is the power coming from

the main grid FLPCC0. If the operation mode changes to the islanded mode,

the frequency of the microgrid is changed to f ' due to the loss of power

from/to the main grid (FLPCC becomes zero). The frequency change of the

microgrid during the transition mode can be calculated as shown in

equation (2.16):


57

( )0 ' 01 * 4.1F

PCCf f f K FLΔ = − =

Where, K1F is the droop constant of DG1. Equation (4.1) shows that the

frequency deviation depends on grid flow (FLPCC0). Therefore, the change

of frequency in the transition mode can be out of its limits if |FLPCC0| is

high enough.

From the above expression it is seen that, in the conventional power

sharing method, the microgrid can be a non-constant load from the utility

view point and the frequency change during a transition mode can be out of

its limits.

Fig. 4.2. Feeder Flow vs. Frequency Droop (DG1 & DG3: FFC mode).

b. Proposed Control Strategy

From the above analysis, it is clear that the conventional power sharing

method has limitations, especially during heavy or light load conditions, in

f 0

FL10=FLPCC

0

f

FL30 FL3’

(a) DG1 droop (b) DG3FL1

’=FLPCC’=0

f

Grid-connected mode Islanded mode

f

0

CHAPTER IV

58

terms that the microgrid can be a non-constant load from the utility view

point and the frequency change during the transition mode can be out of its

limits. To overcome such disadvantages, a new power sharing method is

proposed in which a proper feeder flow reference is determined and the

algorithm of changing the DG control mode is presented.

In order for the microgrid to be a constant load from the main grid

view point regardless of the load conditions, the reference feeder flow of

the DG1 FLPCCref must be increased high enough. However, if the feeder

flow at the PCC point is very high, the microgrid frequency can be beyond

its limits during the transition mode. The higher the feeder flow at the PCC

(FLPCC0), the larger the change of frequency during transition mode.

Therefore, to minimize the change of frequency in the transition mode, the

reference value of DG1 feeder flow should be minimized. The proposed

power sharing method will deal with those two conflicting objectives, and

overcome the disadvantages of the conventional method. In a microgrid the UPC-mode DGs output power is regulated to a

constant, the FFC-mode DGs compensate for any change in load demand

and maintain the feeder flow to be unchanged and equal to a reference

power. The feeder flow at the PCC FLPCC is calculated as follows:

( )0 4.2PCC i ji FFC j UPC

FL LD P P∈ ∈

⎛ ⎞= − +⎜ ⎟

⎝ ⎠∑ ∑

Where:

1

n

ii

LD LD=

=∑ : Microgrid load demand


59

iP : The output of the ith DG which is in the FFC mode.

0jP : The reference output power of the jth DG which is in the

UPC mode.

The first DG is in the FFC mode, and the feeder flow will be equal to

the reference value (i.e., FLPCC = FLPCCref), if the FFC-mode DGs outputs

Pi are within their limits. Otherwise, the extra power is compensated by the

main grid, and hence the FLPCC becomes non-constant. In order for FLPCC

to remain unchanged, the feeder flow reference (FLPCCref) must be high

enough so that the outputs of FFC-mode DGs do not reach their limits at

the peak load. Therefore, from equation (4.2) we have:

( )max max 0 4.3refPCC i j

i FFC j UPCFL LD P P

∈ ∈

⎛ ⎞≥ − +⎜ ⎟

⎝ ⎠∑ ∑

Where:

max max

1

n

ii

LD LD=

=∑

Pimax : The maximum capacity of the ith DG

If FLPCCref satisfies the inequality (4.3) the feeder flow will not

change even if the load reaches its peak value.

Meanwhile, it is seen from equation (4.1) that the change of

frequency (Δf), during the transition mode, is directly proportional to

FLPCC. Thus, in order to reduce the frequency variation during the

transition mode, FLPCCref should be minimized while satisfying equation

CHAPTER IV

60

(4.3), as follows:

( )max max 0 4.4refPCC i j

i FFC j UPCFL LD P P

∈ ∈

⎛ ⎞= − +⎜ ⎟

⎝ ⎠∑ ∑

In equation (4.4), FLPCCref can be reduced further by increasing the power

output reference of UPC-mode DGs (i.e., Pj0). If the power references of

UPC-mode DGs are set to their maximum (Pjmax), FLPCC

ref can be

minimized:

( ) ( )max max maxmin 4.5refPCC i j

i FFC j UPC

FL LD P P∈ ∈

⎛ ⎞= − +⎜ ⎟

⎝ ⎠∑ ∑

Equation (4.5) means that the feeder flow can be minimized if all DGs,

including the UPC-mode DGs, increase the output powers to their

maximum when load demands reach the peak point. It is noted that the

condition set in equation (4.5) corresponds to the positive value of FLPCC

and thus the direction of power is from the main grid to the microgrid. In

case of the reversed direction the approach is similar. Even though the

power references of the UPC-mode DGs are not normally set at their

maximum, and the load demand is also usually not the peak value,

equation (4.5) suggests that the two aforementioned conflicting objectives

can be satisfied by changing the control mode of UPC-mode DGs if

necessary. In other words, when the load is heavy and other FFC-mode

DGs reach their maximum limits, we can maintain the FLPCC constant by


61

changing the control mode of UPC-mode DGs to FFC. The principle of

control mode change will be explained by using the sample microgrid

system shown in Fig. 4.1. The microgrid comprises four DGs; two DGs

(DG1 and DG3) are always in the FFC mode whereas the other two DGs

(DG2 and DG4) are in the UPC mode.

The outputs of DG2 and DG4 are regulated unchanged, 02P and 0

4P ,

respectively, while the change in the load is matched by the two FFC-mode

DGs (DG1 and DG3). Any variation in load demands LD1 and LD2 is

firstly compensated by DG1. However, if DG1 output reaches its limit, the

control mode of DG2 will be changed from the UPC to FFC, and hence the

output of DG2 can be increased more to match the load demand. As a

result, the feeder flow at PCC will not change when DG1 output reaches its

limit. The condition of DG2 mode change is that the DG1 output reaches

its maximum.

In a similar manner, the change of load demands LD3 and LD4 are

firstly matched by DG3. However, once DG3 output reaches its maximum,

DG1 will then match the load demand. In the other words, the change of

load demand is compensated for by the FFC-mode DGs, in order from the

nearest DG to the furthest DG (DG1). When the DG1 output reaches its

maximum, the control modes of the UPC-DGs are changed to FFC. For

example, when DG1 output is maximized, the control mode of DG2 is

changed to FFC mode, and DG2 participates in sharing the load together

with DG1. Similarly, if the load keeps increasing and the outputs of DG1

and DG2 reach their limits, the control mode of DG4 will be changed to

FFC mode. It can be seen that, the condition of changing the control mode

CHAPTER IV

62

of DG2 and DG4 is that the DG1 output reaches its maximum. However,

DG4 only changes its control mode when DG2 is in the FFC mode. In

other words, the UPC-mode DG will change its control mode to FFC if

DG1 output reaches its maximum P1max and the UPC-mode DG in front of

DG4 is in FFC mode, e.g. the UPC-mode DG in front of DG4 is DG2. DG2

has no UPC-mode DG in front of itself so the condition that causes it to

change its mode is only the status of DG1 output.

Changing the control mode of UPC-mode DGs to FFC allows the

DGs to operate at maximum capacity, and hence the load can be shared

more by the UPC-mode DGs. It can be seen from equation (4.5) that,

changing the control mode of UPC-mode DGs allows the feeder flow

reference at PCC ( refPCCFL ) to be minimized, and it is set as follows:

( )max max max 4.6refPCC i j

i FFC j UPC

FL LD P P∈ ∈

⎛ ⎞= − +⎜ ⎟

⎝ ⎠∑ ∑

Equation (4.6) means that, the feeder flow at PCC can be minimized

and always remained unchanged although the load reaches maximum, if

the control modes of UPC-mode DGs are changed to FFC.

The algorithm of the control-mode change, as depicted in Fig. 4.3,

shows that the DG will not change its control mode to FFC if either the

front UPC-mode DG is in the UPC mode or DG1 output does not reach its

maximum. In addition, the UPC-mode DG will return to UPC mode if its

output get back to the reference power and the front UPC-mode DG is in

the FFC mode.


63

Fig. 4.3. Algorithm of the control-mode change of UPC-mode DGs

4.2.2. Control Strategy for Multi-Feeder Microgrid

A microgrid with m feeders is depicted in Fig. 4.4. As mentioned earlier in

this chapter, the first DG of each feeder is in the FFC mode. The feeder

flow references of the feeders and control strategy of single feeder case

presented in 4.2.1 can be applied to the multi-feeder case because, in grid-

connected mode with the first DG in the FFC mode, the feeders operate

independently with each other. Therefore, the operation of each feeder is

same to a single feeder microgrid. The feeder flow reference of each feeder,

P1 ≥ P1max

Yes

Mode = UPC

Front UPC-DGis in FFC?

Front UPC-DG is in FFC?

Mode = FFC Mode = UPC

No

Yes Yes

PDG ≤ PDG0

Yes

No

Yes

CHAPTER IV

64

FLiref, is determined by equations (4.6) and thus the feeder flow at PCC is

equal to the sum of all feeders’ power flow. Each feeder flow is regulated

unchanged therefore the power from the utility side does not change and

equal to1

m

ii

FL=∑ ; m: number of parallel feeders. Each feeder flow is

minimized and hence FLPCC is minimized. During transition mode, the

change of frequency is determined as in equation (3.11) and depends on

FLPCC0. Therefore frequency change in transition mode is also minimized.

DG11

LD11 LDn1

DGn1DG21

LD21 LD31

…

MainGrid

FLPCC

…

FL1

DG12

LD12 LDn2

DGn2DG22

LD22 LD32

…

FL2

DG1m

LD1m LDnm

DGnmDG2m

LD2m LD3m

…

FLm

Fig. 4.4 Microgrid with m feeders


65

4.3. Control Strategy for Microgrid in the Islanded Mode

In the conventional control strategy, the loads are shared by the DGs

according to the droop characteristic as following:

‐ The change in load ith will be first compensated by the FFC-mode

DGs between the load and the PCC, DGj, j = 1...i-1, and frequency

does not change.

‐ When all DGj (j< i) reach to their maximum, the load change will be

picked up by all remaining DGs, and the frequency will be changed.

The frequency change is depicted in Fig. 4.5.

Using the control strategy presented in the grid-connected mode for

islanded mode, the UPC-mode DGs will change the control mode

according to the control algorithm shown in Fig. 4.3. By means of the

proposed control algorithm, the microgrid’s frequency does not change as

long as the DGs’ output reach maximum.

For multi-feeder microgrids, the operation of DGs in each feeder is

same to single-feeder case. However, when all DGs in a feeder increase to

Fig. 4.5 Power sharing for an islanded microgrid

f 1 f 0

FL211

f

P221P22

0

f

FL210 P22

2 FL231 FL23

0 FL232

FL212

f 2

f

CHAPTER IV

66

limits the load will be shared by DGs of the remaining feeders according to

droop characteristic and thus the frequency changes. The power sharing

during load variation of multi-feeder microgrid in the islanded mode can be

summarized as follows:

f 1 f 0

FL111

f

P121P12

0

f

FL110 P12

2 FL1n1 FL1n

0 FL1n2FL11

2

f 2

f 1 f 0

FL211

f

P221P22

0

f

FL210 P22

2 FL231 FL23

0 FL232FL21

2

f 2

f

f

f 1 f 0

FLm11

f

Pm21Pm2

0

f

FLm10 Pm2

2 FLmn1 FLmn

0 FLmn2FLm1

2

f 2

f

Fig. 4.6 Power sharing for a multi-feeder microgrid in islanded mode


67

‐ When a variation in load, the extra power is compensated by other

FFC-mode DGs in the same feeder, from the nearest to the furthest

DG (the first DG - DGi1). The frequency remains unchanged until

DGi1 output reaches to its limit.

‐ When DGi1 output is the limit, other UPC-mode DGs in the same

feeder will change the control mode to FFC. The manner to change

the control mode of UPC-mode DGs is same to the single-feeder case

(presented in the above).

‐ When all UPC-mode DGs in front of the load change, e.g. LDk, the

load variation is compensated by all other DGs in the microgrid,

including other feeders’ DGs. Therefore, the frequency of the

microgrid is changed. Due to the change of frequency, the load

sharing in other feeder is also changed.

‐ The power sharing is depicted in Fig. 4.6.

4.4. Simulation Studies and Results

4.4.1. Test System and Simulation Scenarios

In order to verify viability of the proposed power sharing method, a grid-

connected microgrid with three DGs and three-phase loads was simulated

by using PSCAD. The system configuration is shown in Fig. 4.7, and the

system parameters are listed in Table 4.1.

The control modes of DG1 through DG3 were FFC, UPC, and UPC,

respectively. The feeder flow reference of DG1 was determined to 5 kW,

based on equation (4.6) and the parameters shown in Table 4.1, and this

value was used in all simulation cases, including conventional method and

CHAPTER IV

68

proposed method. LD1 and LD2 were not changed during the simulation,

while the LD3 value was varied as follows. LD3 was initially 10 kW and

changed to 20 kW, 27 kW, 20 kW, and 10 kW at 2 s, 4 s, 6 s, and 8 s,

respectively (see Fig. 4.8 (b)).

Fig. 4.7. The Grid-connected Microgrid with Three DGs

TABLE 4.1 THE SYSTEM PARAMETERS

Parameter Value Unit PDG1

max 17 kW PDG2

max 15 kW PDG3

max 9 kW PDG2

0 5 kW PDG3

0 5 kW FLPCC

ref 5 kW LD1

max 5 kW LD2

max 11 kW LD3

max 30 kW

Main Grid

LD1 FLPCC0

PDG

LD2 LD3

FFC UPC UPC

P1max P3

max P2max

FLPCCref P2

0

LD1max LD2

max LD3

max

P30

DG1 DG2 DG3


69

4.4.2. Simulation Results in the Grid-Connected Operation Mode

This section describes the simulation results during the grid-connected

operation mode according to different power sharing method (conventional

and proposed). Fig. 4.8(a) shows the simulation results for the

conventional power sharing method. During the low load (from 1 s to 2 s),

the power from the main grid was regulated to the reference value (5 kW).

At 2 s, DG1 increased its output to compensate for the increase of LD3.

However, since DG1 reached its maximum (17 kW), the remainder load

was picked up by the main grid, and thus the FLPCC was changed to 8 kW.

Since the power output of DG2 and DG3 were fixed (control mode was not

changed), between 2s and 8 s, the microgrid was no longer a constant load

from the main grid viewpoint, as was discussed in Section III-A.

Fig. 4.9 shows the simulation results for the proposed power sharing

method. The DGs power and/or flow references and the load demand were

identical to those of the above case.

However, in the proposed method, the control mode of the UPC-

mode DGs were controlled according to the algorithm shown in Fig. 4.3. It

can be seen from Fig. 4.9 (a) that, by using the proposed method, the

feeder flow at PCC (FLPCC) remained unchanged, i.e. although the load

demand LD3 increaseed between 2 s and 8 s, FLPCC was regulated to

reference value (5kW). To accomplish this, the control modes of DG2 and

DG3 were changed from UPC to FFC at 2 s and 4 s, respectively, as shown

in Figs. 4.9 (b) and 4.9 (c).

CHAPTER IV

70

(a)

(b)

Fig. 4.8. Simulation results for the conventional method (Control mode of

DGs are not changed). (a) Active power generation (FLPCC changes during

heavy load conditions). (b) Load demand (LD1,LD2 remain unchanged,

LD3 changes).

LD3

LD1+LD2

FLPCC

FLPCC > FLPCCref

PDG1 PDG2+PDG3


71

(a)

(b)

FFC mode

UPC mode

PDG2 PDG1

FLPCC = FLPCCref

PDG3

CHAPTER IV

72

(c)

Fig. 4.9. The simulation results for the proposed method. (a) Active power

generation (FLPCC remains unchanged). (b) DG2 control-mode change. (c)

DG3 control-mode change.

When LD3 increased to 19 kW at 2 s, DG1 firstly increased its output,

but the variation could not be compensated by DG1 within its limit.

Therefore, DG2 changed its control mode to FFC, and increased output

until the load demands were matched. When LD3 increased further at 4 s,

DG2 also reached its maximum limit (15 kW). Accordingly, DG3 started to

increase its output by changing the control mode to FFC. When LD3

decreased at 6 s, DG3, which is the nearest DG upstream from LD3, firstly

decreased its output. Since the DG3 output returned to its initial power

reference (5 kW), the control mode was changed to UPC, and DG2

decreased its output. As LD3 decreased further at 8 s, the control mode of

FFC mode

UPC mode


73

DG2 was also returned to UPC, and the DG1 output was decreased.

4.4.3. Simulation Results in the Transition Mode

(a)

(b)

Δf=0.5Hz

Δf=1.05Hz

CHAPTER IV

74

(c)

Fig. 4.10. The frequency change in the transition mode. (a) Conventional

method, disconnection at 5s. (b) Conventional method, disconnection at 7s.

(c) Proposed method, disconnection at 7s.

In this section, we investigated the advantage of the proposed method

during the transition mode. To accomplish this, we simulated the

intentional islanding, and observed the microgrid frequency during the

transition mode. In order to investigate the effect of load level on the

microgrid frequency, we simulated the islanding at 5 s and 7 s, respectively,

for each power sharing method.

Fig. 4.10 summarizes the simulation results. It can be seen that the

change of frequency in the proposed method was always the same since the

feeder flow at PCC point was unchanged as depicted in Fig. 4.10 (c). On

the other hand, in the conventional method, FLPCC depended on the load

Δf=0.3Hz


75

condition and it was no longer a constant if the first DG output reached its

maximum as shown in Fig. 4.8 (a). Therefore, the change of frequency due

to a disconnection from the main grid was not constant, but depended on

the feeder flow power at which the disconnection occurred. Figs. 4.10 (a)

and 4.10 (b) show the changes of frequency during the transition mode at

5s and 7s, respectively. The frequency change in case of disconnection at

5s (1.05Hz) was larger than that in case of disconnection at 7s (0.5Hz),

since the FLPCC at 5s was larger than FLPCC at 7s.

Additionally, if the transition mode occurred between 2s and 8s,

FLPCC was larger than the feeder flow reference (5kW), and hence the

change of frequency in the conventional method was always larger than the

change in the proposed method. As depicted in Figs. 4.10 (a) and 4.10 (b),

the changes of frequency in conventional method were 1.05Hz and 0.5Hz

respectively, whereas that was 0.3Hz for the proposed method (see Fig.

4.10 (c)).

From the simulation results it is seen that with proposed method the

change of frequency during the transition mode (grid-connected

mode/islanded mode) is minimized, the feeder flow at PCC is remained

unchanged during grid-connected mode, and frequency of islanding

operation mode can be regulated unchanged as long as the load is shared

by its own feeder. According to the proposed method, the DG output is

mobilized during heavy load condition thus reduce the burden to the utility

grid in the peak time and the microgrid operates more stably. Otherwise,

the UPC-mode DGs do not change their control mode, and thus they can

operate at high efficiency (e.g. economical condition, high performance

CHAPTER IV

76

band…). The control strategy is also applied to the multiple feeder

microgrid and simulation results will be presented in Chapter VI.

Chapter V:

CONTROL STRATEGIES FOR HYBRID SOURCE IN

MICROGRID

5.1. Overview

Renewable energy is currently widely used. One such resource is solar

energy. The PV array normally uses maximum power point tracking

(MPPT) techniques to continuously deliver the highest power to the load

when there are variations in irradiation and temperature. The disadvantage

of PV energy is that the PV output power depends on weather conditions

and cell-temperature making it become an uncontrollable source [49].

Furthermore, it is not available during night time. In order to overcome

these inherent drawbacks, alternative source, such as the proton exchange

membrane fuel cell (PEMFC) that will likely play a major role in

distributed generation and microgrids in near future [27, 50], should be

installed in the hybrid system. By changing FC output power, the hybrid

source output becomes controllable. However, PEMFC, in its turn, works

only at a high efficiency within a specific power range ( low upFC FCP P÷ ) [51, 52].

The hybrid source like PV-FC is connected to feeders of microgrid

and work as a distributed generator. Therefore the hybrid source as a DG

also has two control mode, UPC mode and FFC mode. In the UPC mode,

variations of load demand are compensated by the main grid because the

hybrid source output is regulated to a reference power. Therefore, the

CHAPTER V

78

reference value of the hybrid source output PMSref must be determined. In

the FFC mode, the feeder flow is regulated to a constant, the extra load

demand is picked up by the hybrid source and hence the feeder reference

power Pfeederref must be known.

The purposes of hybrid source operation are to operate PV in MPPT

mode to mobilize the free of charge energy, and to operate FC in a high

efficiency band. Therefore, the control mode of hybrid source must be

determined so that it can meet the above assumptions.

The proposed operating strategy is to coordinate the two control

modes and determine the reference values of the UPC mode and FFC mode

so that all constraints are satisfied, the operation of such a complicated

system should be simplified, and the performance of microgrid operation is

improved.

5.2. Hybrid System Description

5.2.1 Structure of Grid Connected Hybrid System

The system consists of a PV-FC hybrid source with the main grid

connecting to loads at the PCC as shown in Fig. 5.1. The Photovoltaic [53,

54] and the PEMFC [55, 56] are modeled as nonlinear voltage sources.

These sources are connected to DC/DC converters which are coupled at the

DC side of a DC/AC inverter. The DC/DC connected to the PV array

works as an MPPT controller. Many MPPT algorithms have been proposed

in the literature such as Incremental Conductance (INC), Constant Voltage

(CV), and Perturbation and Observation (P&O). The P&O method has

been widely used because of its simple feedback structure and fewer

CONTROL STRATEGY FOR HYBRID SOURCE IN MICROGRID

79

measured parameters. The P&O algorithm with power feedback control

[57-59] is shown in Fig. 5.2. As PV voltage and current are determined, the

power is calculated. At the maximum power point, the derivative (dP/dV)

is equal to zero. The maximum power point can be achieved by changing

the reference voltage by the amount of ΔVref.

Fig. 5.1. Grid Connected PV-FC Hybrid System

5.2.2 PV Array Model

The mathematical model [53, 54, 60-62] can be expressed as below

AC DC

DC DC

MPPT

PV

PWM1

Vref

D1

V I

DC DC PWM2

VDCref

D2

PCC

FC

DC Bus

PFeeder

PMS

PLoad

Load

VDC

CHAPTER V

80

( ) ( )5.1ph sat sqI I I exp V IR 1

AKT⎧ ⎫⎡ ⎤= − + −⎨ ⎬⎢ ⎥⎣ ⎦⎩ ⎭

Where:

:Electronic charge:A dimensionless factor:Boltzmann constant

qAk

Equation (1) shows that the output characteristic of a solar cell is non-

linear and vitally affected by solar radiation, temperature and load

condition.

Photo-current Iph is directly proportional to solar radiation Ga

( ) ( )5.2aph a sc

as

GI G IG

=

The short-circuit current of solar cell Isc depends linearly on cell

temperature:

( ) ( ) ( )5.3sc scs sc sI T I 1 ΔI T T⎡ ⎤= + −⎣ ⎦

Thus, Iph depends on solar irradiance and cell-temperature:

( ) ( ) ( )5.4aph a scs sc s

as

GI G ,T I 1 ΔI T TG

⎡ ⎤= + −⎣ ⎦

The saturation current Isat also depends on solar irradiation and cell-


81

temperature and can be mathematically expressed as follows:

( ) ( )( )( )

( )5.5oc

t

ph asat a V T

V T

I G ,TI G ,T

e 1⎛ ⎞⎜ ⎟⎜ ⎟⎝ ⎠

=

−

Where

Gas: Standard irradiation (1000 W/m ).

Isc: Short-circuit current.

Iscs: Short-circuit current at standard condition (Gas and Ts)

Ts: Standard temperature (298 K).

Voc: Open-circuit voltage

Vt: Thermal voltage

5.2.3 PEMFC Model

The PEMFC steady-state feature of a PEMFC source is assessed by means

of a polarization curve, which shows the non-linear relationship between

the voltage and current density. The PEMFC output voltage is as follows

[52, 56, 63-65]:

( )5.6out Nerst act ohm concV E V V V= − − −

Where, ENerst is the “thermodynamic potential” of Nerst, which represents

the reversible (or open-circuit) voltage of the fuel-cell. Activation voltage

drop Vact is given in Tafel equation as below:

CHAPTER V

82

( ) ( )5.7actV T a bln I⎡ ⎤= +⎣ ⎦

a, b: Constant terms in Tafel equation (in volts per Kelvin)

The overall ohmic voltage drop Vohm can be expressed as:

( )5.8ohm ohmV IR=

The ohmic resistance Rohm of PEMFC consists of the resistance of the

polymer membrane and electrodes, and the resistances of the electrodes.

Concentration voltage drop Vconc is expressed as:

( )5.9conclimit

RT IV ln 1zF I

⎛ ⎞= − −⎜ ⎟

⎝ ⎠

5.2.4 Maximum Power Point Tracking Control

Many MPPT algorithms have been proposed in the literature such as INC

[66], CV [67], and P&O [57, 59]. The two algorithms often used to achieve

maximum power point tracking are the P&O and INC methods. The INC

method offers good performance under rapidly changing atmospheric

conditions. However, four sensors are required to perform the

computations. If the sensors require more conversion time, then the MPPT

process will take a longer time to track the maximum power point. During

tracking time, PV output is less than its maximum power. This means that

the longer the conversion time the larger amount of power loss. On the

contrary, if the execution speed of the P&O method increases, then the


83

system loss will decrease. Moreover, this method only requires two sensors,

which results in a reduction of hardware requirements and cost [68].

Therefore, the P&O method is used to control MPPT process.

Fig. 5.2. P&O MPPT Algorithm.

In order to achieve maximum power, two different applied control

methods often chosen are voltage-feedback control and power-feedback

control [58, 59]. Voltage-feedback control uses the solar array terminal

voltage to control and keep the array operating near its maximum power

point by regulating the array’s voltage and matching the voltage of the

array to a desired voltage. The drawback of the voltage-feedback control is

CHAPTER V

84

its neglect of the effect of irradiation and cell-temperature. Therefore, the

power-feedback control is used to achieve maximum power. The P&O

MPPT algorithm with a power-feedback control [57, 59] is shown in Fig.

5.2. As PV voltage and current are determined, the power is calculated. At

the maximum power point, the derivative (dP/dV) is equal to zero. The

maximum power point can be achieved by changing the reference voltage

by the amount of ΔVref.

In order to implement the MPPT algorithm, a Buck-Boost DC/DC

converter is used as depicted in Fig. 5.3.

a)

b)

Fig. 5.3. Buck-Boost converter and control: a) Buck-Boost topology,

b) PWM circuit


85

The Buck-Boost converter consists of one switching device (GTO)

that enables it to turn on and off depending on the applied gate signal D.

The gate signal for the GTO can be obtained by comparing the saw-tooth

waveform with the control voltage [58]. The change of the reference

voltage ΔVref obtained by MPPT algorithm shown in Fig. 5.2 becomes the

input of the Pulse-Width-Modulation (PWM) circuit depicted in Fig. 5.3b.

The PWM circuit generates a gate signal to control the Buck-Boost

converter and thus maximum power is tracked and delivered to the AC side

via a DC/AC inverter.

5.3. Control Algorithm of the Hybrid System

As mentioned above, the purpose of the operating algorithm is to

determine the control mode of the hybrid source and the reference value

for each control mode so that the PV is able to work at maximum output

power and the constraints are fulfilled. Once the constraints (PFClow, PFC

up,

and PFmax) are known, the control mode of the hybrid source (UPC mode

and FFC mode) depends on load variations and PV output. The algorithm

to choose control mode of hybrid source will be presented in section 5.3.2.

In the UPC mode, the reference output power of the hybrid source PMSref

depends on PV output and the constraints of FC output. The algorithm

determining PMSref is presented in 5.3.1 and is depicted in Fig. 5.4.

5.3.1 Control Strategy for the Hybrid System in the UPC mode

In this subsection, the presented algorithm determines the hybrid source

works in the UPC mode. Such an algorithm allows the PV to work at its

maximum power point, and the FC to work within its high efficiency band.

CHAPTER V

86

In the UPC mode, the hybrid source regulates the output to the reference

value. Then

refPV FC MSP P P+ = (5.11)

Equation (5.11) shows that the variations of PV output will be

compensated for by the FC power and thus the total power will be

Fig. 5.4 Operation Strategy of Hybrid Source in the UPC Mode.

PMSref PPV

PMS3ref = PMS2

ref + ΔPMS

PMSiref = PMSi-1

ref + ΔPMS

: : PPV3

PPV2

PPV1

PPVi

Area 2

Area 1

PMSnref = PMSn-1

ref + ΔPMS PPVn

: :

ΔPMS

ΔPMS PMS2ref = PMS1

ref + ΔPMS

PMS1ref = PFC

max

PFClow

0 0

PPV0 = 0


87

regulated to the reference value. However, FC output must satisfy its

constraints and hence PMSref must set at an appropriate value. Fig. 5.4

shows the operation strategy of the hybrid source in UPC mode to

determine PMSref . The algorithm includes two areas, Area 1 and Area 2.

In Area 1, PPV is less than PPV1, and then the reference power PMS1ref

is set at PFCup. Where:

( )1 – 5.12up lowPV FC FCP P P=

( )1 5.13ref upMS FCP P=

If PV output is zero, then equation (5.11) deduces PFC to be equal to PFCup.

If PV output increases to PPV1, then, from equations (5.11) and (5.12), we

get PFC equal to PFClow. In other words, when PV output varies from zero to

PPV1, the FC output will change from PFCup to PFC

low. As a result, the

constraints for FC output are always reached in Area 1. It is noted that the

reference power of the hybrid source during UPC mode is fixed at a

constant PFCup.

Area 2 is for the case in which PV output power is greater than PPV1.

As examined earlier, when PV output increases to PPV1, the FC output will

decrease to its lower limit, PFClow. If PV output keeps increasing, the FC

output will decrease below its limit, PFClow. In this case, to operate the PV

at its maximum power point and the FC within its limit, the reference

power must be increased. As depicted in Fig. 5.4, if PV output is larger

than PPV1, the reference power will be increased by the amount of ΔPMS,

CHAPTER V

88

and we get

( )2 1 5.14ref refMS MS MSP P P= + Δ

Similarly if PPV is greater than PPV2, the FC output becomes less than

its lower limit and the reference power will be thus increased by the

amount of ΔPMS. In other words, the reference power remains unchanged

and equal to PMS2ref if PPV is less than PPV2 and greater than PPV1. Where

( )2 1 5.15PV PV MSP P P= + Δ

It is noted that ΔPMS is limited so that with the new reference power, the

FC output must be less than its upper limit, PFCup. Then we have

( )– 5.16up lowMS FC FCP P PΔ ≤

In general, if PV output is between PPVi and PPVi-1 (i=2, 3, 4…), then we

have

( )1 5.17ref refMSi MSi MSP P P−= + Δ

( )1 5.18PVi PVi MSP P P−= + Δ

Equations (5.17) and (5.18) show the method of finding the reference


89

power when PV output is in Area 2. The relationship between PMSiref and

PPVi is obtained by using equations (5.12), (5.13), and (5.18) in (5.17), and

then

( ), 2, 3, 4 5.19ref lowMSi PVi FCP P P i= + = …

It can be generalized for the determination of PMSref in both Area 1 and

Area 2 by starting the index i from 1. Therefore, if PV output is

1 , 1, 2, 3PVi PV PViP P P i− ≤ ≤ = …

then we have

( )min , 1, 2, 3 5.20refMSi PVi FCP P P i= + = …

( )1 , 2, 3, 4 5.21PVi PVi MSP P P i−= +Δ = …

It is noted that, when i=1, PPV1 is given in equation (5.12), and

( )1 0 0 5.22PVi PVP P− = =

In brief, the reference power of the hybrid source is determined

according to PV output power. If PV output is in Area 1, the reference

power will always be constant and set at PFCup. Otherwise, the reference

value will be changed by the amount of ΔPMS, according to the change of

PV power. The reference power of the hybrid source PMSref in both Area 1

and Area 2 is determined by equations (5.20) and (5.21). PPV0, PPV1, and

CHAPTER V

90

ΔPMS are shown in equations (5.22), (5.12), and (5.16) respectively.

Fig. 5.5 shows the control algorithm diagram determining the

reference power automatically. The constant C must satisfy equation (5.16).

If C increases, the number of change of PMSref will decrease and thus the

performance of system operation will be improved. However, C should be

small enough so that the frequency does not change over its limits (±5%).

PMSref= PFC

max, ∆PMS = 0

PMSref(new) = PMS

ref(old) + ∆PMS

PPV > PPV1

PFC ≤ PFCmin

PFC ≥ PFCmax

∆PMS = –C ∆PMS = +C ∆PMS = 0

No

Yes

No

Yes

Yes

No

Fig. 5.5. Control Algorithm Diagram in UPC mode (PMSref

Automatically Changing)


91

In order to improve the performance of the algorithm, a hysteresis is

included in the simulation model. The hysteresis is used to prevent

oscillation of the setting value of hybrid system reference power PMSref. At

the boundary of change in PMSref, the reference value will be changed

continuously due to the oscillations in PV maximum power tracking. To

avoid the oscillations around the boundary, a hysteresis is included and its

control scheme to control PMSref is depicted in Fig.5.6.

C

PMSref = PMS

ref –C

PPV

PPV2

PPV

PPVn

PMSref = PMS

ref

+C

PMSref = PMS

ref –C

PMSref = PMS

ref –C

PMSref = PMS

ref

+C

PMSref = PMS

ref

+C

I

II

: :

+ C– C0

Fig. 5.6. Hysteresis Control Scheme for PMSref Control

CHAPTER V

92

5.3.2 Overall Control Strategy for the Hybrid System

It is well known that, in the microgrid, each DG as well as the hybrid

source has two control modes, the UPC mode and FFC mode. In the above

subsection, a method to determine PMSref in the UPC mode is proposed. In

this subsection, an operating strategy is presented to coordinate the two

control modes. The purpose of the algorithm is to decide when each control

mode is applied and to determine the reference value of the feeder flow

when the FFC mode is used. Such an operating strategy must enable the

PV to work at its maximum power point, FC output and feeder flow to

satisfy their constraints.

If the hybrid source works in the UPC mode, the hybrid output is

regulated to a reference value and the variations in load are matched by

feeder power. With the reference power PMSref proposed in subsection A,

the constraints of FC and PV are always satisfied. Therefore, only the

constraint of feeder flow is considered. On the other hand, when the hybrid

works in the FFC mode, the feeder flow is controlled to a reference value

Pfeederref and thus the hybrid source will compensate for the load variations.

In this case, all constraints must be considered in the operating algorithm.

Based on those analyses, the operating strategy of the system is proposed

as demonstrated in Fig. 5.7.

The operation algorithm in Fig. 5.7 involves two areas (Area-I and

Area-II) and the control mode depends on the load power. If load is in

Area-I, the UPC mode is selected. Otherwise, the FFC mode is applied

with respect to Area-II.


93

In the UPC area, the hybrid source output is PMS

ref. If load is lower

than PMSref, the redundant power will be transmitted to the main grid.

Otherwise, the main grid will send power to the load side to match load

demand. When load increases, the feeder flow will increase

correspondingly. If feeder flow increases to its maximum Pfeedermax, then

the feeder flow cannot meet load demand if the load keeps increasing. In

order to compensate for the load demand, the control mode must be

MS

refP

2loadP

P FC

1loadP

P Fee

derm

ax

FC

PV

Feeder

FC

Load Shedding

P FC

up–

P FC

Area-I(UPC)

Area-II(FFC)

P FC

max

– P F

C

PLoad maxloadP

P PV

0

Fig. 5.7. Overall Operating Strategy for the Grid Connected Hybrid System

CHAPTER V

94

changed to FFC with respect to Area-II. Thus the boundary between Area-I

and Area-II Pload1 is

( )max1 5.23ref

Load Feeder MSP P P= +

When the mode changes to FFC the feeder flow reference must be

determined. In order for the system operation to be seamless, the feeder

flow should be unchanged during control mode transition. Accordingly,

when the feeder flow reference is set at Pfeedermax then we have

max ref

Feeder FeederP P= (5.24)

In the FFC area, the variation in load is matched by the hybrid source.

In other words, the changes in load and PV output are compensated for by

PEMFC power. If FC output increases to its upper limit and load is higher

than total generating power, then load shedding will occur. The limit that

load shedding will be reached is

( )max2 5.25up

Load FC Feeder PVP P P P= + +

Equation (5.25) shows that Pload2 is minimal when PV output is at 0 kW.

Then

( )min max2 5.26up

Load FC FeederP P P= +


95

Equation (5.26) means that, if load demand is less than Pload2min, load

shedding will never happen.

From the beginning, FC has always worked in high efficiency band

and FC output has been less than PFCup. If load is less than Pload2

min, load

shedding is ensured not to occur. However, in severe conditions, FC should

mobilize its availability, PFCmax, to supply the load. Thus, the load can be

higher and the largest load is

( )max max max 5.27Load FC FeederP P P= +

If FC power and load demand satisfy equation (5.27), load shedding will

never happen. Accordingly, based on load forecast, the installed power of

FC can be determined following equation (5.27) to avoid load shedding.

Corresponding to the FC installed power the width of Area-II is calculated

as follows:

( )max – 5.28upArea II FC FCP P P− =

In order for the system to work more stably, the number of mode

changes should be decreased. As seen in Fig. 5.7, the limit changing the

mode from UPC to FFC is PLoad1, which is calculated in equation (5.23).

Equation (5.23) shows that PLoad1 depends on PFeedermax and PMS

ref.

PFeedermax is a constant, thus PLoad1 depends on PMS

ref. Fig. 5.4 shows that in

Area 2 PMSref depends on ΔPMS. Therefore, to decrease the number of mode

changes, PMSref changes must be reduced. Thus ΔPMS must be increased.

CHAPTER V

96

However, ΔPMS must satisfy condition (5.16) and thus the minimized

number of mode change is reached when ΔPMS is maximized

( )max – 5.29up lowMS FC FCP P PΔ =

In summary, in a light load condition the hybrid source works in UPC

mode, the hybrid source regulates output power to the reference value

PMSref, and the main grid compensates for load variations. PMS

ref is

determined by the algorithm shown in Fig. 5.4, and thus the PV always

works at its maximum power point and the PEMFC always works within

the high efficiency band (PFClow...PFC

up). In heavy load conditions, the

control mode changes to FFC, and the variation of load will be matched by

the hybrid source. In this mode, PV still works with the MPPT control, and

PEMFC operates within its efficiency band until load increases to a very

high point. Hence FC only works outside the high efficiency band

(PFCup...PFC

max) in severe conditions. With an installed power of FC and

load demand satisfying equation (5.27), load shedding will not occur.

Besides, to reduce the number of mode changes, ΔPMS must be increased

and hence the number of mode changes is minimized when ΔPMS is

maximized, as shown in equation (5.29). Additionally, in order for system

operation to be seamless, the reference value of feeder flow must be set at

PFeedermax.


97

5.4. Simulation Studies and Result

5.4.1 Simulation Results in the Case without Hysteresis

A simulation was carried out using the system model shown in Fig. 5.2 to

verify the operating strategies. The system parameters are shown in Table

5.1.

In order to verify the operating strategy, the load demand and PV

output were time varied in terms of step. According to the load demand

and the change of PV output, PFC, PMSref, PFeeder

ref, and operating mode are

determined by the proposed operating algorithm. Fig. 5.8 shows the

simulation results of the system operating strategy. The changes of PPV and

PLoad are shown in Fig. 5.8-a (Δ line) and Fig. 5.8-b (ο line) respectively.

TABLE 5.1 THE HYBRID SYSTEM PARAMETERS

Parameter Value Unit PFC

low 0.01 MW PFC

up 0.07 MWPFeeder

max 0.01 MW ∆PMS 0.03 MW

Based on PPV and the constraints of PFC shown in Table 5.1, the

reference value of the hybrid source output PMSref is determined as dipicted

in Fig. 5.8-a (ο line). From 0s to 10s, PV operates at standard test

conditions to generate a constant power, and thus PMSref is constant. From

10s to 20s, PPV changes step by step and thus PMSref is defined as the

algorithm shown in Fig. 5.4 or Fig. 5.5. PEMFC ouput PFC, as shown in

Fig. 5.8-a (• line), changes according to the change of PPV and PMS. Figure

CHAPTER V

98

5.8(c) shows the system operating mode. The UPC mode and FFC mode

correspond to values 0 and 1, respectively. From 4s to 6s, the system works

in FFC mode and thus PFeedermax becomes the feeder reference value.

During FFC mode, the hybrid source output power changes with respect to

the change of load demand, as in Fig. 5.8(b). On the contrary, in UPC

mode, PMS changes following PMSref, as shown in Fig. 5.8(a).

It can be seen from Fig. 5.8 that, the system only works in FFC mode

when the load is heavy. The UPC mode is the major operating mode of the

system, and hence the system works more stably.

It can also be seen from Fig. 5.8(a) that, at 12s and 17s, PMSref

changes continuously. This is caused by variations of PPV in the maximum

power point tracking process. As a result, PMS and PFC oscillate and are

unstable. In order to overcome these drawbacks, a hysteresis was used to

control the change of PMSref, as shown in Fig. 5.6. The simulation results of

the system, including the hysteresis, are depicted in Fig. 5.9.

Fig. 5.8. (a)


99

Fig. 5.8. (b)

Fig. 5.8. (c)

Fig. 5.8. Simulation result without hysteresis: (a) The operating strategy of

the hybrid source; (b) The operating strategy of the whole system; (c) The

change of operating modes.

5.4.2 Improving Operation Performance by Using Hysteresis

Fig. 5.9 shows the simulation results when a hysteresis was included with

the control scheme shown in Fig. 5.6. From 12s to 13s and from 17s to 18s,

the variations of PMSref (Fig. 5.9a, ο line), FC output (Fig. 5.9a, • line), and

feeder flow (Fig. 5.9b, Δ line) are eliminated, and thus the system works

Feeder control

Unit power control mode

CHAPTER V

100

more stably compared to a case without hysteresis (Fig. 5.8). Fig. 5.9d

shows the frequency variations when load changes or when the hybrid

source reference power PMSref changes (at 12s and 18s). The parameter C

was chosen at 0.03MW, and thus the frequency variations did not reach

over its limit (±5%*60 = ±0.3Hz).

5.4.3 Discussion

It can be seen from Fig. 5.9(b) that, during the UPC mode, the feeder flow

(Δ line) changes due to the change of load ( line) and hybrid source

output (• line). This is because in the UPC mode the feeder flow must

change to match the load demand.

However, in a real-world situation, the microgrid should be a

constant load from the utility view point [9]. In reality, the microgrid

includes some DGs connected in parallel to the feeder [16, 47]. Therefore,

in the UPC mode, the changes of load will be compensated for by other

FFC mode DGs and the power from the main grid will be controlled to

remain constant.

In the case in which there is only one hybrid source connected to the

feeder, the hybrid source must work in the FFC mode to maintain the

feeder flow at constant. Based on the proposed method, this can be done by

setting the maximum value of the feeder flow (PFeedermax) to a very low

value, and thus the hybrid source is forced to work in the FFC mode.

Accordingly, the FC output power must be high enough to meet the load

demand when load is heavy and/or at night time without solar power.


101

Fig. 5.9(a)

Fig. 5.9(b)

Fig. 5.9(c)

Feeder control mode

Unit power control mode

C=0.03MW

CHAPTER V

102

Fig. 5.9(d)

Fig. 5.9. Improving operation performance by using hysteresis: (a) The

operating strategy of the hybrid source; (b) The operating strategy of the

whole system; (c) The change of operating modes; (d) Frequency

variations occur in the system.

From the above discussions, it can be said that the proposed

operating strategy is more applicable and meaningful to a real-world

microgrid with multi DGs.

In summary, the main operating strategy shown in Fig. 5.7 is to

specify the control mode and feeder flow reference; the algorithm shown in

Fig. 5.4 is to determine PMSref in the UPC mode. With the operating

algorithm, PV always operates at maximum output power, PEMFC

operates within the high efficiency range (PFClow ÷ PFC

up), and feeder

power flow is always limited to its maximum value (PFeedermax). The change

of the operating mode depends on the current load demand, the PV output

and the constraints of PEMFC and feeder power.


103

With the proposed operating algorithm, the system works flexibly,

exploiting maximum solar energy; PEMFC works within a high efficiency

band and hence improves the performance of the system’s operation. The

system can maximize the generated power when load is heavy and

minimize the load shedding area. When load is light, the UPC mode is

selected and thus the hybrid source works more stably.

The changes of operating mode only happen when the load demand is

at the boundary of mode change (PLoad1); otherwise, the operating mode is

either UPC mode or FFC mode. Besides, the variation of hybrid source

reference power PMSref is overcome by means of hysteresis. Additionally,

the number of mode changes is reduced. As a consequence, the system

works more stably due to the minimization of mode changes and reference

value variation.

CHAPTER V

104

Chapter VI:

CASE STUDY

Previous chapters have presented the control strategies for a microgrid in

general cases and the operation algorithm for a hybrid source connected to

the microgrid. The simulation was implemented to verify the algorithms

separately. In this chapter, overall control algorithms are together applied

to a tested microgrid with controllable DGs and PV-FC hybrid source.

6.1. Introduction to Case Study

Fig. 6.1 System Configuration

D

FL1 1

FFC LD

DUPC F

Main Grid

P

FL2 1 FL1 2 FL2 2 FL1 HB FL2 HB

D

FL1 3

FFC LD

FL2 3

PHB P2P1

SS

CB2

CB1

CHAPTER VI

106

6.1.1. System Configuration

The tested system configuration shown in Fig. 6.1 has two feeders. The

first feeder has two controllable DGs and a PV-FC hybrid source supplying

to load demand LD1. The second feeder has one controllable DG supplying

power to load LD2. The two feeders connect to the utility grid at PCC

through a static switch (SS). The circuit breakers CB1 and CB2 are to

isolate the feeder separately if necessary. The system is simulated in

PSCAD/EMTDC as shown in Appendix A.

6.1.2. System Parameters

General parameters:

‐ Microgrid nominal voltage: 380V

‐ Main grid nominal voltage: 22.9kV

‐ Nominal frequency: 60Hz

‐ Number of feeders: 2

DG1:

‐ P1max = 50kW,

‐ FFC mode: FLref = 0kW

DG2:

‐ P2max = 50kW

‐ UPC mode: P2ref = 20kW

Hybrid source:

‐ FC: Pfcmax = 50kW, Pfcmin = 10kW

‐ Pmsref = 40kW ( = Ppv1 = Pfcmax – Pfcmin)

‐ FL31max = 100kW ( = P1max + P2max + FL0)

CASE STUDY

107

‐ HB source change mode when: LD1 ≥ FL31max + Pmsref =

140kW.

6.1.3. Simulation Scenarios

TABLE 6.1

CASES WITHOUT HYBRID SOURCE

Operation mode

Control method

Grid connected

mode

Islanded mode

Without proposed control algorithm Case 1-1 Case 2-1

With proposed control algorithm Case 1-2 Case 2-2

TABLE 6.2 CASES WITH HYBRID SOURCE

Operation mode

Control method

Grid connected

mode

Islanded mode

Without proposed control algorithm Case 3-1 Case 4-1

With proposed control algorithm Case 3-2 Case 4-2

TABLE 6.3 CASES WITH MULTI-FEEDER MICROGRID

Operation mode Cases Grid Connected mode Case 5-1

Islanded mode Case 5-2

CHAPTER VI

108

TABLE 6.4 FREQUENCY RESTORATION CONTROL METHOD VS. PROPOSED METHOD

Operation mode Control method

Islanded mode

Frequency restoration control scheme Case 6-1 Proposed control scheme Case 6-2

6.2. Results and Discussion

6.2.1. Simulation Results in case without Hybrid Source

Case 1-1: Grid-connected mode, without control strategy

Fig. 6.2 Active power response as LD1 changes

CASE STUDY

109

(a)

(b)

Fig. 6.3 Voltage and frequency response in grid-connected mode

CHAPTER VI

110

Fig. 6.4 Frequency change due to islanding

Islanding at 5 seconds, Δf = 1.5Hz

Case 1-2: Grid-connected mode, with control strategy

Fig. 6.5 Active power response as LD1 changes

CASE STUDY

111

(a)

(b)

Fig. 6.6 Voltage and frequency response in grid-connected mode

Frequency remains unchanged in grid-connected mode and equals to

nominal value, 60Hz.

CHAPTER VI

112

Fig. 6.2 shows that, without control strategy the feeder flow is

change d when load is heavy, between 2s and 8s, and feeder flow increases

over the reference value FLref, 0kW. In other words, the microgrid

becomes uncontrollable load from the utility point of view. With control

strategy, Fig. 6.5, DG2 changes its mode according to the control

algorithm and thus the feeder flow remains unchanged all the time.

Frequency remains unchanged in grid-connected mode and equals to

nominal value 60Hz as depicted in Fig. 6.3(b) and Fig. 6.6(b).

Fig. 6.7 Frequency change due to islanding

Frequency does not change as islanding occurs.

The simulation results of case1-1 and case 1-2 show that without

control strategy, as load LD1 increases the feeder flow increases whereas it

remains unchanged in case with the control strategy as shown in Fig. 6.2

and Fig. 6.5, respectively.

CASE STUDY

113

In the transition mode the frequency drop in conventional method is

larger than in case with proposed control scheme as shown in Fig. 6.4 and

Fig. 6.7, respectively.

Case 2-1: Islanded mode, without control algorithm.

Fig. 6.8 Active power response as LD1 changes during island mode

Fig. 6.9 System voltage

CHAPTER VI

114

Fig. 6.10 System frequency

Frequency decreases when DG2 output increases

Fig. 6.8 shows simulation results of conventional method in islanded

mode. The DG2 control mode does not change all the time and it is in the

UPC mode. As DG1 reaches maximum, DG2 increases output according to

P-f droop characteristic to meet the load demand, 50kW. The frequency,

therefore, decreases as DG2 output increases as shown in Fig. 6.10.

CASE STUDY

115

Case 2-2: Islanded mode, DG2 does change its mode according to the

control algorithm.

Fig. 6.11 Active power response as LD1 changes during island mode

Fig. 6.12 DG2 control mode

CHAPTER VI

116

(a)

(b)

Fig. 6.13 System Voltage and frequency

Using the control algorithm, f is regulated to nominal value, 60Hz, all the

time

CASE STUDY

117

Fig. 6.11 shows that as DG1 reaches maximum 50kW, from 3s to 9s,

DG2 increases output to meet the load demand. However, with the

proposed control strategy the DG2’s control changes according to the

algorithm as shown in Fig. 6.12, and the frequency, therefore, remains

unchanged as shown in Fig. 6.13(b).

The simulation results of case 2-1 and case 2-2 show that with the

proposed control strategy the frequency remains unchanged all the time

whereas it changes according to P-f droop characteristic in the

conventional control method.

6.2.2. Simulation Results in case with the PV-FC Hybrid Source

The PV-FC hybrid source is controlled according to the algorithm

presented in Chapter V.

Case 3-1: Grid-connected mode, without control strategy

Fig. 6.14(a) shows that the feeder flow in front of HB source,

Pf1_HB, is limited to Pf1_HBmax, 100kW. Between the period of 5s and

7s Pf1_HB reaches maximum, the hybrid source’s control mode changes to

the FFC mode as shown in Fig. 6.14(b).

From Fig. 6.15 it is seen that when hybrid source’s control mode

changes to the FFC mode, from 5s to 7s, its output increases to track the

load demand and the feeder flow Pf1_HB is regulated unchanged.

Therefore, the presence of the hybrid source can limit the power shared by

other DGs and it actively participates in sharing the load with other DGs.

Otherwise, the hybrid source works in UPC mode and controls its output to

the reference power, 40kW.

CHAPTER VI

118

(a)

(b)

Fig. 6.14 (a) Load demand and feeder flow in front of HB source,

(b) Control mode of the hybrid source

CASE STUDY

119

Fig. 6.15 Power response in case without control strategy

Without proposed control method the feeder flow at PCC is changed as

P_DG1 reaches maximum (50kW).

The frequency and voltage are regulated to the nominal values and

do not change during the grid-connected mode as shown in Fig. 6.16.

CHAPTER VI

120

(a)

(b)

Fig. 6.16 System voltage and frequency

CASE STUDY

121

Case 3-2: Grid-connected mode, with control strategy

Fig. 6.17 Load demand and feeder flow in front of HB source

These parameters are same to Case 3-1.

Fig. 6.18 Power response in case with control strategy

CHAPTER VI

122

(a)

(b)

Fig. 6.19 The control mode of DG2 and hybrid source: (a) DG2 control

mode, (b) HB source control mode

CASE STUDY

123

(a)

(b)


CHAPTER VI

124

The operation of the hybrid source does not change the operation of

other DGs in proposed control method, the feeder flow at PCC (Pgrid)

remains unchanged as DG2 changes it control mode to increase the output

(P_DG2) as depicted in Fig. 6.18 by changing the DG2 control mode to

FFC mode between 3s and 9s as load is heavy, Fig. 6.19(a). The frequency

and voltage in proposed method are also regulated to the nominal values

and do not change during the grid-connected mode.

The simulation results of case 3-1 and case 3-2 show that, in grid-

connected mode, the participation of the hybrid source with its control

strategy does not change the microgrid operation performance. The feeder

flow is regulated unchanged with the proposed control strategy. In addition,

the operation of hybrid source allows limiting the feeder flow in front of its

connection point, Pf1_HB, therefore the operation of DGs between hybrid

source and PCC are easier and reduce the burden to theses DGs. The

simulation results show that the control strategy of the hybrid source

actively coordinate with the control strategy for controllable DGs

presented in Chapter IV.

Again the feeder flow does not change in case with proposed control

method, otherwise it changes according to the load condition and hybrid

source capacity.

CASE STUDY

125

Case 4-1: Islanded mode, without control strategy.


Feeder flow Pf1_HB is limited to 100kW

Fig. 6.22 Power response in case without control strategy

CHAPTER VI

126

Fig. 6.23 The hybrid source control mode

CASE STUDY

127

(a)

(b)


The frequency decreases according to the P-f droop characteristic of DG2

CHAPTER VI

128

Similar to grid-connected mode, the feeder flow at hybrid source is

limited to 100kW. The hybrid source’s control mode changes to the FFC,

from 5s to 7s, as its feeder flow reaches 100kW as depicted in Fig. 6.23

and, therefore, its output increases to meet the load demand, Fig. 6.22.

The voltage is regulated constant. However the frequency is changed

according to the droop characteristic as shown in Fig. 6.24 as the DG2’s

output increases between the duration of 3s to 9s.

Case 4-2: With HB source, islanded mode, DG2 changes its control mode.


CASE STUDY

129

Fig. 6.26 Power response in case with control strategy

CHAPTER VI

130

(a)

(b)

Fig. 6.27 The control mode of DG2 and hybrid source: (a) DG2 control

mode, (b) HB source control mode

CASE STUDY

131

(a)

(b)


The microgrid frequency remains unchanged as DG2 output increases

according to the FL-f droop characteristic.

CHAPTER VI

132

The parameters used in this case are same to those in case 4-1. When

DG1 output reaches maximum, DG2 increases its output to match the load

demand, as shown in Fig. 6.26. The hybrid source control mode changes to

the FFC when its feeder flow reaches limit, Fig. 6.27(a). The changes in

power output of DG1, DG2 and hybrid source are same to those in case 4.1.

However, in this case the DG2’s control mode changes to the FFC

whenever its output power increases as shown in Fig. 6.27(b) and,

therefore, the frequency does not change as depicted in Fig. 6.28(b).

Simulation results of case 4-1 and case 4-2 show that in the islanded

mdoe the hybrid source coordinates well with other DGs in the microgrid

and similarly to grid-connected mode. When the hybrid source increases its

output the control mode changes to the FFC, from 5s to 7s - Fig. 6.27b,

therefore the system frequency does not change. Additionally, with the

proposed control strategy the frequency does not change when the DGs

increase the outputs, Fig. 6.28, whereas it changes adequately according to

load in case without control strategy, Fig. 6.24.

6.2.3. Simulation Results with Multiple-Feeder Microgrid

In the grid-connected mode each feeder in the microgrid operates

independently and the operation of each feeder is similar to the single

feeder microgrid considered earlier in this section.

CASE STUDY

133

Case 5-1: Grid-connected mode, multi-feeder

Fig. 6.29 Active power responses of feeder 1

Fig. 6.30 Active power responses of feeder 2

CHAPTER VI

134

Fig. 6.31 Power from the main grid

Grid power increases to meet the load demand in feeder 2, LD2.

DGs in feeder 1 have sufficient capacities to meet the local load and

the feeder flow Pf1_DG1 is unchanged in grid-connected mode as shown

in Fig. 6.29. DG3 increases output to meet load demand LD2 as long as its

output is less than the maximum (50kW) and the feeder flow increases due

to the short of generation, Fig. 6.30.

The simulation results show that the control strategy applied to multi-

feeder microgrid in grid-connected mode is same to the single feeder case.

CASE STUDY

135

Case 5-2: Multi-feeder, islanded mode

Fig. 6.32 Active power responses of feeder 2 in island mode

Fig. 6.33 Active power responses of feeder 1 in island mode

CHAPTER VI

136

Fig. 6.34 Feeder flow in front of DG2, Pf1_DG2

Pf1_DG2 decreases due to the reversed power flow.

Fig. 6.35 Microgrid frequency

Frequency change according to droop characteristic

CASE STUDY

137

Fig. 6.32 shows that from 2s to 8s, the DG1 reaches its maximum

(50kW) thus the load LD2 is compensated by DGs on feeder-1 side and

feeder flow FL2 increases. Fig. 6.33 shows that both DG1 and DG2 in the

feeder 1 increase the outputs to compensate the load LD2 in the feeder 2.

When the load in feeder 2 is shared by DGs in the other feeder, feeder 1,

the frequency therefore changes according to the droop characteristic as in

Fig. 6.35.

From the simulation results it can be seen that the load shedding can

be achieved easily according to the change of frequency.

If DGs of each feeder can supply the load in its own feeder, the

power sharing among DGs in each feeder is same to the single-feeder case.

Therefore, the frequency does not change if the proposed control strategy

is applied as presented the earlier cases. Otherwise, the load in this feeder,

e.g. feeder 2, is compensated by the other feeder, e.g. feeder 1. The feeder

flow vs. frequency droop is applied and the frequency will be decreased in

case of high load. In such a manner load is shed according to the deviation

of frequency from the nominal value.

6.2.4. Proposed control method versus frequency restoration control

scheme in island mode:

Without control scheme, the frequency can be changed according to the

droop characteristic as microgrid is in the islanding mode due to the load

variations. With the proposed control scheme, the frequency can be kept

unchanged by changing the DGs’ operation mode. However, the frequency

can also be regulated to nominal value by means of frequency restoration

controller [69, 70].

CHAPTER VI

138

The following simulation results of single feeder microgrid show the

differences between the two control methods.

Case 6-1: Using frequency restoration control scheme

Fig. 6.36 Active power responses in case of frequency restoration method

Fig. 6.37 Microgrid frequency in case of frequency restoration method

CASE STUDY

139

Case 6-2: Proposed control scheme

Fig. 6.38 Active power responses in case of proposed method

Fig. 6.39 Microgrid frequency in case of proposed method

CHAPTER VI

140

The change in output power of DGs in case 6-1 and case 6-2 are

same as shown in Fig. 6.36 and Fig. 6.38.

From Fig. 6.37 and Fig. 6.39 it is seen that the frequency is

controlled unchanged in both two method, frequency restoration method

and proposed method. However, the frequency deviation due to the load

change in proposed method is much smaller than the one in frequency

restoration method. The frequency transient in conventional method

depends on the changes of load. The large load changes the higher

frequency deviates as shown in Fig. 6.37 because in the load is primarily

shared according to the droop characteristic. On the contrary, in the

proposed method the control mode of DG changes to the FFC therefore the

frequency does not change according to the droop characteristic. The

frequency deviation is caused by the mode change and thus it is much

smaller as shown in Fig. 6.39. As the result, the power quality of the

microgrid is improved in island mode. In addition, this characteristic

facilitates the load shedding as the generation is insufficient based on the

change of frequency from the nominal value.

Chapter VII:

CONCLUSIONS AND FUTURE EXTENSIONS

7.1. Conclusions

This dissertation has presented a control algorithm for the DGs and the

operation algorithm for the PV-FC hybrid source connected in a microgrid.

In order to achieve the control strategy, the frequency and active power

responses according to the droop characteristic have been investigated. The

analysis based on the three conditions of operations and four possible

configurations of the microgrid. According to the analysis, the proposed

control algorithm is to take the advantages and overcome the drawbacks of

each configuration. In addition, the operation of hybrid source is a

particular case and a control algorithm has proposed to operate in the

microgrid with a high efficiency and maximizing the solar energy.

Analysis of frequency and active power responses:

The analysis shows that: 1) in both parallel and series configurations, the

FFC mode has more advantages over the UPC mode in terms of frequency

change and reserved active power; 2) in islanded operation mode, the

configuration of series–FFC has more reserve to regulate frequency

unchanged and therefore it has more advantage over other configurations;

3) Otherwise, in grid-connected mode and transition mode, the parallel–

FFC configuration is more advantageous.

CHAPTER VII

142

Control strategies for controllable DGs in microgrid

With proposed method the change of frequency during the transition mode

(grid-connected mode/islanded mode) is minimized, the feeder flow at

PCC is remained unchanged during grid-connected mode and therefore the

microgrid becomes controllable load from the utility point of view. The

frequency of islanding operation mode can be regulated unchanged as long

as the load is shared by DGs in its own feeder. According to the proposed

method, the DG output is mobilized during heavy load condition thus

reduce the burden to the utility grid in the peak time and the microgrid

operates more stably. Otherwise, the UPC-mode DGs do not change their

control mode, and thus they can operate at high efficiency such as

economic condition, high performance band etc.

Control strategies for hybrid source

With the proposed operating algorithm, the system works flexibly,

exploiting maximum solar energy; PEMFC works within a high efficiency

band and hence improves the performance of the system operation. The

system can maximize the generated power when load is heavy and

minimize the load shedding area. When load is light, the UPC mode is

selected and thus the hybrid source works more stably.

The changes of operating mode only happen when the load demand is

at the boundary of mode change (PLoad1); otherwise, the operating mode is

either UPC mode or FFC mode. Besides, the variation of hybrid source

reference power PMSref is overcome by means of hysteresis. Additionally,

the number of mode changes is reduced. As a consequence, the system

CONCLUSIONS AND FUTURE EXTENSIONS

143

works more stably due to the minimization of mode changes and reference

value variation.

Microgrid operation with presence of hybrid source

The presence of the hybrid source together with other DGs in the microgrid

does not change the microgrid operation performance in both grid-

connected and islanded mode. It can also reduce the burden and facilitate

the operation of other DGs by limiting the feeder flow at its connection

point.

In the islanded mode, the frequency variation of proposed control

strategy is smaller than the one of the method with frequency restoration

scheme. Therefore, the power quality in islanded mode is improved. In

addition, the control strategy also facilitates the load shedding scheme in

the islanded operation mode by the frequency deviation.

7.2. Future Extensions

Control strategy for microgrid

The control strategy for microgrid did not mention the characteristic of

DGs. In a real microgrid various types of DGs are used such as PV, Wind,

FC, battery energy storage system (BESS), super capacitor (SC), micro-gas

turbine (MGT) etc. Each type of DGs has different operation characteristic,

for instance, MGT response time is slow, SC is a fast dynamic storage

(from seconds to minutes), BESS is long-term storage (from minutes to

hours) [71-75], etc. Among the available energy storage technologies,

batteries, flywheel and super-capacitors are more applicable for microgrid

CHAPTER VII

144

[76-78]. Therefore, the characteristic of DGs, especially the one of energy

storage technologies, should be taken into account to build up a control

strategy for whole system. For further extension, the microgrid control

strategy taken the DGs’ characteristics will be studied.

On the other hand, the control strategy has been proposed under the

technical view point. In order to, however, minimize the global energy cost

in the microgrid the control strategy should be put together with the

Economic Dispatch and the Unit Commitment problems. In other word, the

proposed control algorithm could be considered under circumstance of

energy management system.

Control strategy for hybrid source

The Fuelcell responses to the change of power slowly therefore the use of

fast response DGs such as BESS or SC can improve the operating

efficiency. The operating algorithm taking operation of BESS and SC into

account to enhance operation performance of the system will be considered.

BIBLIOGRAPHY

[1] P. Biczel, et al., "Power Electronic Devices in Modern Power

Systems," in EUROCON, 2007. The International Conference on

Computer as a Tool, 2007, pp. 1586-1586.

[2] L. Fresis and D. Infield, Renewable Energy in Power Systems: A

John Wiley & Sons, 2008.

[3] "European Technology Smartgrid Platform, "Smartgrids: Vision and

strategy for European electricity networks of the future"," 2006.

[4] R. H. Lasseter, "MicroGrids," in Power Engineering Society Winter

Meeting, 2002. IEEE, 2002, pp. 305-308 vol.1.

[5] R. Lasseter and P. Piagi, "Providing premium power through

distributed resources," in System Sciences, 2000. Proceedings of the

33rd Annual Hawaii International Conference on, 2000, p. 9 pp.

[6] P. Piagi and R. H. Lasseter, "Autonomous control of microgrids," in

Power Engineering Society General Meeting, 2006. IEEE, 2006, p. 8

pp.

[7] J. A. P. Lopes, et al., "Defining control strategies for MicroGrids

islanded operation," Power Systems, IEEE Transactions on, vol. 21,

pp. 916-924, 2006.

[8] J. A. P. Lopes, et al., "Management of Microgrids," JIEEC2003,

Bilbao, 2003.

[9] H. Jiayi, et al., "A review on distributed energy resources and

MicroGrid," Renewable and Sustainable Energy Reviews, vol. 12, pp.

2472-2483, 2008.

BIBLIOGRAPHY

146

[10] A. G. Tsikalakis and N. D. Hatziargyriou, "Centralized Control for

Optimizing Microgrids Operation," Energy Conversion, IEEE

Transactions on, vol. 23, pp. 241-248, 2008.

[11] F. Katiraei, et al., "Micro-grid autonomous operation during and

subsequent to islanding process," Power Delivery, IEEE


[12] F. Katiraei and M. R. Iravani, "Power Management Strategies for a

Microgrid With Multiple Distributed Generation Units," Power

Systems, IEEE Transactions on, vol. 21, pp. 1821-1831, 2006.

[13] N. L. Soultanis, et al., "A Stability Algorithm for the Dynamic

Analysis of Inverter Dominated Unbalanced LV Microgrids," Power

Systems, IEEE Transactions on, vol. 22, pp. 294-304, 2007.

[14] A. P. S. Meliopoulos, "Challenges in simulation and design of

μGrids," in Power Engineering Society Winter Meeting, 2002.

IEEE, 2002, pp. 309-314 vol.1.

[15] A. A. Lasseter R, Marnay C, Stephens J, Dagle J, Guttromson R, et

al., "CERTS microgrid concept. White Paper on Integration of

distributed energy resources, prepared for Transmission Reliability

Program," Office of Power Technologies, U.S. Department of

EnergyApril 2002.

[16] R. H. Lasseter and P. Piagi, "Control and Design of Microgrid

Components," Jan. 2006.

[17] I. A. Hiskens and E. M. Fleming, "Control of inverter-connected

sources in autonomous microgrids," in American Control

Conference, 2008, 2008, pp. 586-590.

BIBLIOGRAPHY

147

[18] H. N, et al., "Centralized and decentralized control of microgrids,"

International Journal of Distributed Energy Resource vol. 1, pp.

197–212, 2005.

[19] P. a. L. JA, et al., "Control strategies for microgrids black start and

islanded operation," Int J Distr Energy Resour, vol. 2, pp. 211–31,

2006.

[20] H. N., "Active distribution network. The Effect of Distributed and

Renewable Generation on Power Systems Security," Available:

http://www.microgrids.eu/micro2000/presentations/21.pdf., 2005.

[21] M. Barnes, et al., "Microgrid laboratory facilities," in Future Power

Systems, 2005 International Conference on, 2005, pp. 6 pp.-6.

[22] T. Logenthiran, et al., "Multi-agent coordination for DER in

MicroGrid," in Sustainable Energy Technologies, 2008. ICSET 2008.

IEEE International Conference on, 2008, pp. 77-82.

[23] K. De Brabandere, et al., "Control of Microgrids," in Power

Engineering Society General Meeting, 2007. IEEE, 2007, pp. 1-7.

[24] G. Celli, et al., "Optimal participation of a microgrid to the energy

market with an intelligent EMS," in Power Engineering Conference,

2005. IPEC 2005. The 7th International, 2005, pp. 663-668 Vol. 2.

[25] J. D. Kueck, et al., "Microgrid Energy Management System,"

CERTSJanuary 29, 2003.

[26] L. K. Siow, et al., "Wi-Fi based server in microgrid energy

management system," in TENCON 2009 - 2009 IEEE Region 10

Conference, 2009, pp. 1-5.

[27] "Program on technology innovation: The Galvin Path to Perfect

BIBLIOGRAPHY

148

Power-A Technical Assessment," project report, EPRI, Mar. 2007.

[28] C. Gyu-Yeong, et al., "Comparative study of power sharing

algorithm for fuel cell and photovoltaic hybrid generation system," in

Power Electronics Conference (IPEC), 2010 International, 2010, pp.

2615-2620.

[29] A. M. Sharaf and A. A. A. El-Gammal, "A novel PSO-based hybrid

PV-FC-Diesel-Battery electric PID-controller drive system for

electric vehicle traction," in Electric Power and Energy Conference

(EPEC), 2010 IEEE, 2010, pp. 1-6.

[30] W. Caisheng and M. H. Nehrir, "Power Management of a Stand-

Alone Wind/Photovoltaic/Fuel Cell Energy System," Energy

Conversion, IEEE Transactions on, vol. 23, pp. 957-967, 2008.

[31] K. Loc Nguyen, et al., "Improvement of a PV-FC hybrid source

operation in a microgrid," in Power Electronics Conference (IPEC),

2010 International, 2010, pp. 717-720.

[32] K. Loc Nguyen, et al., "Power-Management Strategies for a Grid-

Connected PV-FC Hybrid System," Power Delivery, IEEE


[33] T. K. Abdel-Galil, et al., "Project report, "Protection Coordination

Planning with Distributed Generation", Clean Energy Technologies

Techniques, CANMET Energy Technology Centre, June 2007."

[34] J. Balakrishnan, "Distributed generator interconnection protection,"

in Industrial and Information Systems, 2007. ICIIS 2007.

International Conference on, 2007, pp. 367-372.

[35] IEEE, "IEEE P-1547 Update on the Current Status of DG

BIBLIOGRAPHY

149

Interconnection Protection," ed.

[36] W. Hartmann and B. Electric. (Nov. 2002) Winding Arrangements

for Distributed Generation. On-Peak Performance.

[37] J. M. Guerrero, et al., "Decentralized Control for Parallel Operation

of Distributed Generation Inverters Using Resistive Output

Impedance," Industrial Electronics, IEEE Transactions on, vol. 54,

pp. 994-1004, 2007.

[38] N. Pogaku, et al., "Modeling, Analysis and Testing of Autonomous

Operation of an Inverter-Based Microgrid," Power Electronics, IEEE


[39] C. L. Moreira, et al., "Using Low Voltage MicroGrids for Service

Restoration," Power Systems, IEEE Transactions on, vol. 22, pp.

395-403, 2007.

[40] K. De Brabandere, et al., "A voltage and frequency droop control

method for parallel inverters," in Power Electronics Specialists

Conference, 2004. PESC 04. 2004 IEEE 35th Annual, 2004, pp.

2501-2507 Vol.4.

[41] B. M. Weedy and B. J. Cory, Electric Power Systems, 4 ed.: New

York: Wiley, 1998.

[42] H. Nikkhajoei and R. H. Lasseter, "Distributed Generation Interface

to the CERTS Microgrid," Power Delivery, IEEE Transactions on,

vol. 24, pp. 1598-1608, 2009.

[43] A. Engler and N. Soultanis, "Droop control in LV-grids," in Future

Power Systems, 2005 International Conference on, 2005, pp. 6 pp.-6.

[44] H. Laaksonen, et al., "Voltage and frequency control of inverter

BIBLIOGRAPHY

150

based weak LV network microgrid," in Future Power Systems, 2005

International Conference on, 2005, pp. 6 pp.-6.

[45] A. Seon-Ju, et al., "Power-Sharing Method of Multiple Distributed

Generators Considering Control Modes and Configurations of a

Microgrid," Power Delivery, IEEE Transactions on, vol. 25, pp.

2007-2016, 2010.

[46] N. K. Loc, et al., "Analysis of Active Power and Frequency Response

in Microgrid," in IEEE Trondheim PowerTech 2011, to be published.

[47] M. Barnes, et al., "Real-World MicroGrids-An Overview," in System

of Systems Engineering, 2007. SoSE '07. IEEE International

Conference on, 2007, pp. 1-8.

[48] I. S. P. D. 10, "Draft Application Guide for IEEE Standard 1547,

Interconnecting Distributed Resources with Electric Power Systems,"

ed, Mar. 2008.

[49] N. K. Loc, et al., "Analysis of the Effects of Irradiation and Cell-

Temperature on the Dynamic Responses of PV System with MPPT,"

in KIEE Sumer Conference, July 2008.

[50] "Large scale integration of micro-generation to low voltage grids,"

project report, 2005, available online at http://www.microgrids.eu.

[51] T. Bocklisch, et al., "Predictive and Optimizing Energy Management

of Photovoltaic Fuel Cell Hybrid Systems with Short-time Energy

Storage," in 4th European Conference PV-Hybrid and Mini-Grid,

2008, pp. 8-15.

[52] J. Larmine and A. Dicks, Fuel Cell Systems Explained: New York,

Wiley, 2003.

BIBLIOGRAPHY

151

[53] X. Weidong, et al., "A novel modeling method for photovoltaic

cells," in Power Electronics Specialists Conference, 2004. PESC 04.

2004 IEEE 35th Annual, 2004, pp. 1950-1956 Vol.3.

[54] D. Sera, et al., "PV panel model based on datasheet values," in

Industrial Electronics, 2007. ISIE 2007. IEEE International

Symposium on, 2007, pp. 2392-2396.

[55] H. Chihchiang and S. Chihming, "Comparative study of peak power

tracking techniques for solar storage system," in Applied Power

Electronics Conference and Exposition, 1998. APEC '98. Conference

Proceedings 1998., Thirteenth Annual, 1998, pp. 679-685 vol.2.

[56] W. Caisheng, et al., "Dynamic models and model validation for PEM

fuel cells using electrical circuits," Energy Conversion, IEEE


[57] E. Koutroulis, et al., "Development of a microcontroller-based,

photovoltaic maximum power point tracking control system," Power

Electronics, IEEE Transactions on, vol. 16, pp. 46-54, 2001.

[58] H. Chihchiang and L. Jong Rong, "DSP-based controller application

in battery storage of photovoltaic system," in Industrial Electronics,

Control, and Instrumentation, 1996., Proceedings of the 1996 IEEE

IECON 22nd International Conference on, 1996, pp. 1705-1710

vol.3.

[59] H. Chihchiang, et al., "Implementation of a DSP-controlled

photovoltaic system with peak power tracking," Industrial

Electronics, IEEE Transactions on, vol. 45, pp. 99-107, 1998.

[60] S. Weixiang, et al., "Mathematical model of a solar module for

BIBLIOGRAPHY

152

energy yield simulation in photovoltaic systems," in Power

Electronics and Drive Systems, 2009. PEDS 2009. International

Conference on, 2009, pp. 336-341.

[61] J. Leuchter, et al., "Mathematical modeling of photovoltaic systems,"

in Power Electronics and Motion Control Conference (EPE/PEMC),

2010 14th International, 2010, pp. S4-1-S4-4.

[62] R. F. Coelho, et al., "A proposed photovoltaic module and array

mathematical modeling destined to simulation," in Industrial

Electronics, 2009. ISIE 2009. IEEE International Symposium on,

2009, pp. 1624-1629.

[63] C. Wang, et al., "Control of PEM fuel cell distributed generation

systems," Energy Conversion, IEEE Transactions on, vol. 21, pp.

586-595, 2006.

[64] M. Tanrioven and M. S. Alam, "Modeling, Control, and Power

Quality Evaluation of a PEM Fuel Cell-Based Power Supply System

for Residential Use," Industry Applications, IEEE Transactions on,

vol. 42, pp. 1582-1589, 2006.

[65] Y. Zhu and K. Tomsovic, "Development of models for analyzing the

load-following performance of microturbines and fuel cells," Electric

Power Systems Research, vol. 62, pp. 1-11, 2002.

[66] L. Fangrui, et al., "A Variable Step Size INC MPPT Method for PV

Systems," Industrial Electronics, IEEE Transactions on, vol. 55, pp.

2622-2628, 2008.

[67] Z. Ye and X. Wu, "Compensation Loop Design of a Photovoltaic

System Based on Constant Voltage MPPT," in Power and Energy

BIBLIOGRAPHY

153

Engineering Conference, 2009. APPEEC 2009. Asia-Pacific, 2009,

pp. 1-4.

[68] T. Esram and P. L. Chapman, "Comparison of Photovoltaic Array

Maximum Power Point Tracking Techniques," Energy Conversion,

IEEE Transactions on, vol. 22, pp. 439-449, 2007.

[69] J. A. P. Lopes, et al., "Control strategies for microgrids emergency

operation," in Future Power Systems, 2005 International Conference

on, 2005, pp. 6 pp.-6.

[70] A. Madureira, et al., "Secondary Load-Frequency Control for

MicroGrids in Islanded Operation," in International Conference of

Renewable Energy Power Quality, Spain, 2005.

[71] L. Di, et al., "Design of a power management system for an active

PV station including various storage technologies," in Power

Electronics and Motion Control Conference, 2008. EPE-PEMC

2008. 13th, 2008, pp. 2142-2149.

[72] J. Li Jun, "Modeling and Simulation of Micro Gas turbine Generation

System for Grid Connected Operation," in Power and Energy

Engineering Conference (APPEEC), 2010 Asia-Pacific, 2010, pp. 1-

4.

[73] L. Xiao, et al., "Modeling and control of aggregated Super-Capacitor

Energy Storage system for wind power generation," in Industrial

Electronics, 2008. IECON 2008. 34th Annual Conference of IEEE,

2008, pp. 3370-3375.

[74] W. Juanhua, et al., "Study on a Super Capacitor Energy Storage

system for improving the operating stability of Distributed

BIBLIOGRAPHY

154

Generation system," in Electric Utility Deregulation and

Restructuring and Power Technologies, 2008. DRPT 2008. Third

International Conference on, 2008, pp. 2702-2706.

[75] B. Wu, et al., "Super-capacitors energy storage system applied in the

microgrid," in Industrial Electronics and Applications (ICIEA), 2010

the 5th IEEE Conference on, 2010, pp. 1002-1005.

[76] T. Sels, et al., "Electrical Energy Storage Systems: Existing Systems

versus Newest Systems-an Overview," in Power Generation and

Sustainable Development International Conference, 2001, pp. 215-

220.

[77] "Review of Electrical Energy Storage Technologies and Systems and

of their Potential for the UK," EA Technology, 2004.

[78] N. W. A. Lidula and A. D. Rajapakse, "Microgrids research: A

review of experimental microgrids and test systems," Renewable and

Sustainable Energy Reviews, vol. 15, pp. 186-202, 2011.

Appendix A:

FREQUENCY AND ACTIVE POWER RESPONSES IN THE ISLANDED MODE

A.1 PSCAD Model of Three-DG Microgrid

Fig. A.1 Three-DG microgrid, series configuration

APPENDIX A

156

Fig. A.2 Three-DG microgrid, parallel configuration

APPENDIX A

157

A.2 Simulation results

Configuration 1: Series-FFC

Fig. A.3 Power sharing of three DGs in series-FFC configuration

Fig. A.4 Frequency response in series-FFC configuration

APPENDIX A

158

Configuration 2: Series UPC

Fig. A.5 Power sharing of three DGs series-UPC configuration

Fig. A.6 Frequency response in series-UPC configuration

APPENDIX A

159

Configuration 3: Parallel-FFC

Fig. A.7 Power sharing of three DGs in parallel-FFC configuration

Fig. A.8 Frequency response in parallel-FFC configuration

APPENDIX A

160

Configuration 4: Parallel-UPC

Fig. A.9 Power sharing of three DGs in parallel-UPC configuration

Fig. A10 Frequency response in parallel-UPC configuration

Appendix B:

CASE STUDY

B.1 PSCAD Model of Whole System

APPENDIX B

162

B.2 PSCAD Model of Photovoltaic

Short circuit current of solar cell:

Photon current:

The saturation current:

APPENDIX B

163

PV voltage and current:

V-I characteristic:

B.3 PSCAD Model of MPPT Algorithm of PV

APPENDIX B

164

APPENDIX B

165

B.4 PSCAD Model of PEMFC

Enerst:

APPENDIX B

166

Activation loss, Vact:

APPENDIX B

167

Concentration loss, Vconc:

APPENDIX B

168

Ohmic loss, Vohm:

FC output voltage and current:

APPENDIX B

169

B.5 PSCAD Model of Buck-Boost DC/DCs and Controllers

PV’s DC/DC controller (MPPT controller):

FC’s DC/DC controller (DC voltage controller):

APPENDIX B

170

B.6 PSCAD Model Inverter Controller

APPENDIX B

171

Submitted to the Faculty of INHA UNIVERSITY In Partial ... · CONTROL STRATEGIES FOR DGS TO IMPROVE...

Documents

Transcript of Submitted to the Faculty of INHA UNIVERSITY In Partial ... · CONTROL STRATEGIES FOR DGS TO IMPROVE...