Smart Power Distribution Systems: Control, Communication, and Optimization

By Qiang Yang and Ting Yang

Smart Power Distribution Systems: Control, Communication, and Optimization explains how diverse technologies work to build and maintain smart grids around the globe. Yang, Yang and Li present the most recent advances in the control, communication and optimization of smart grids and provide unique insight into power system control, sensing and communication, and optimization technologies. The book covers control challenges for renewable energy and smart grids, communication in smart power systems, and optimization challenges in smart power system operations. Each area discussed focuses on the scientific innovations relating to the approaches, methods and algorithmic solutions presented.

Readers will develop sound knowledge and gain insights into the integration of renewable energy generation in smart power distribution systems.

• Presents the latest technological advances in electric power distribution networks, with a particular focus on methodologies, approaches and algorithms
• Provides insights into the most recent research and developments from expert contributors from across the world
• Presents a clear and methodical structure that guides the reader through discussion and analysis, providing unique insights and sound knowledge along the way

• Language: English
• Publisher: Academic Press
• Release date: Oct 17, 2018
• ISBN: 9780128123256

Book preview

Smart Power Distribution Systems

Control, Communication, and Optimization

First Edition

Qiang Yang

Ting Yang

Wei Li

Table of Contents

Cover image

Title page

Copyright

List of contributors

About the editors

Preface

Acknowledgments

Organization of this book

Part One: Modeling and control of smart power distribution network

Part Two: ICT technologies for smart power distribution networks

Part Three: Optimization models and methods in smart power distribution networks

Part One: Modeling and Control of Smart Power Distribution Network (Control Aspect)

1: An overview of codes and control strategies for frequency regulation in wind power generation

Abstract

Acknowledgment

1.1 Introduction

1.2 Grid codes on frequency regulation

1.3 Frequency regulation framework

1.4 System-level control

1.5 Plant/farm-level coordinated control

1.6 WTG-level control strategy

1.7 Discussion

1.8 Conclusion

2: A two-stage reserve scheduling considering wind turbine generator's de-loading control

Abstract

Acknowledgment

2.1 Introduction

2.2 WTG-integrated dispatch mode and DFIG de-loading operation

2.3 A bi-level optimization model for the two-stage reserve scheduling problem

2.4 Case studies

2.5 Conclusion

Appendix. The single-level model formulation

3: Dynamic energy management and control of a grid-interactive DC microgrid system

Abstract

3.1 Introduction

3.2 System description

3.3 Dynamic energy management and control

3.4 Results and discussion

3.5 Conclusion

4: Modeling, control, and energy management for DC microgrid

Abstract

Acknowledgments

4.1 Introduction

4.2 DC MG structure and modeling

4.3 DC MG experimental set-up

4.4 Optimal control and energy management for DC MG

4.5 Results and discussions

4.6 Conclusions

5: Hybrid AC/DC distribution network voltage control

Abstract

Acknowledgment

5.1 Introduction

5.2 VSC-based hybrid AC/DC MG (lower layer)

5.3 Proposed voltage control scheme (upper layer)

5.4 Case study

5.5 Conclusions

6: Controlling the distributed energy resources under fading channel

Abstract

6.1 Introduction

6.2 Microgrid state-space model

6.3 LQG controller under fading channel

6.4 Simulation results and discussions

6.5 Conclusions and future work

7: Cooperative energy dispatch for multiple autonomous microgrids with distributed renewable sources and storages

Abstract

7.1 Introduction

7.2 Autonomous microgrid optimization model

7.3 Cooperative operation control strategy (w/o storages)

7.4 Numerical experiments and result

7.5 Remarks

7.6 Cooperative scheduling strategies (with storages)

7.7 Case study with the IEEE 33-bus network scenario

7.8 Simulation experiment and numerical result

7.9 Remarks

7.10 Conclusions

Part Two: ICT Technologies for Smart Power Distribution Networks

8: Privacy of energy consumption data of a household in a smart grid

Abstract

8.1 Introduction

8.2 Smart grid and its many benefits

8.3 Security vulnerabilities of smart grid and its impact

8.4 Security objectives of smart grid

8.5 Privacy preserving techniques in smart grids

8.6 Conclusions

9: Microgrid communication system and its application in hierarchical control

Abstract

Acknowledgments

9.1 Introduction

9.2 Communication construction based on hierarchical control

9.3 Consensus algorithm based on microgrid communication system

9.4 Case studies

9.5 Conclusions

10: ICT technologies standards and protocols for active distribution network

Abstract

10.1 Introduction to the concept of information and communication technology (ICT)

10.2 Introduction to active distribution network

10.3 ICT technologies in the active distribution networks

10.4 Conclusion

11: Virtual power plant communication system architecture

Abstract

11.1 Introduction

11.2 VPPs in the smart grid concept

11.3 Communication system architecture

11.4 Communication protocols

11.5 Communication system performance analysis

11.6 Conclusions

12: Inertia emulation from HVDC links for LFC in the presence of smart V2G networks

Abstract

12.1 Introduction

12.2 Inertia emulation from SPC-based HVDC systems for LFC

12.3 Introduction to V2G network

12.4 Simulation studies

12.5 Conclusions

13: Internet of things application in smart grid: A brief overview of challenges, opportunities, and future trends

Abstract

13.1 Introduction

13.2 Demand response opportunities in smart distribution systems

13.3 IOT cyber physical security in smart grid

13.4 Modeling and simulation challenges of IoT in smart grid

13.5 Conclusions

Glossary

14: H-infinity-based microgrid state estimations using the IoT sensors

Abstract

14.1 Introduction

14.2 Observation model

14.3 H-Infinity for microgrid state estimation

14.4 Microgrid modeling and simulation results

14.5 Conclusion and future work

Part Three: Optimization Models/Methods in Smart Distribution Networks (Optimization Aspects)

15: Management of renewable energy source and battery bank for power losses optimization

Abstract

15.1 Introduction

15.2 Energy management system for DC microgrid

15.3 Results and discussions

15.4 Conclusion

16: Scenario-based methods for robust electricity network planning considering uncertainties

Abstract

16.1 Introduction

16.2 The mathematical model of network planning problem

16.3 Scenario generation methods

16.4 The solving process of the robust network planning

16.5 Case studies

16.6 Conclusion

Appendix

17: Scenarios/probabilistic optimization approaches for network operation considering uncertainties

Abstract

17.1 Introduction scenario

17.2 Basic problems of power system optimization with large-scale wind power integration

17.3 Research status of power system optimization with large-scale wind power integration

17.4 p-Efficient point theory

17.5 Moment matching theory

17.6 Conclusion

18: The optimal planning of wind power capacity and energy storage capacity based on the bilinear interpolation theory

Abstract

18.1 Introduction

18.2 Research status of wind power accommodation

18.3 Adequacy indices with wind power integration

18.4 Estimation of wind power accommodation

18.5 The optimal allocation of the wind power capacity and ESS capacity based on bilinear interpolation

18.6 Case study

18.7 Conclusions

19: Optimal energy dispatch in residential community with renewable DGs and storage in the presence of real-time pricing

Abstract

Acknowledgment

19.1 Introduction

19.2 System model and problem formulation

19.3 Optimal energy dispatch approach

19.4 Simulation experiment and numerical result

19.5 Conclusions and future work

20: Evaluation on the short-term power supply capacity of an active distribution system based on multiple scenarios considering uncertainties

Abstract

20.1 Introduction

20.2 Analysis of uncertainty factors in evaluating PSC

20.3 Definition of PSC evaluation index

20.4 Short-term PSC evaluation algorithm based on multiscene technology

20.5 Case study

20.6 Conclusions

21: Multi-time-scale energy management of distributed energy resources in active distribution grids

Abstract

21.1 Introduction

21.2 System modeling

21.3 Hierarchical multi-time-scale energy management system

21.4 Simulation configuration and implementations

21.5 Results and discussion

21.6 Conclusion

22: Distribution Network planning considering the impact of Electric Vehicle charging station load

Abstract

22.1 Introduction

22.2 Different operating parameters of distribution network

22.3 Impact of EV charging load on different operating parameters of distribution network

22.4 Optimal placement of charging stations in distribution network

22.5 Case study

22.6 Conclusions

23: Distribution systems hosting capacity assessment: Relaxation and linearization

Abstract

Acknowledgment

23.1 Introduction

23.2 HC mathematical modeling

23.3 Linear model of HC

23.4 Simulations

23.5 Conclusions

Index

Copyright

Academic Press is an imprint of Elsevier

125 London Wall, London EC2Y 5AS, United Kingdom

525 B Street, Suite 1650, San Diego, CA 92101, United States

50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom

Copyright © 2019 Elsevier Inc. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher's permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices

Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

ISBN: 978-0-12-812154-2

For information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Joe Hayton

Acquisition Editor: Maria Convey

Editorial Project Manager: Mariana L. Kuhl

Production Project Manager: R.Vijay Bharath

Cover Designer: Christian J. Bilbow

Typeset by SPi Global, India

List of contributors

Sandhya Armoogum       Department of Industrial Systems Engineering, School of Innovative Technologies and Engineering, University of Technology Mauritius, La Tour Koenig, Pointe-aux-Sables, Mauritius

Birgitte Bak-Jensen       Department of Energy Technology, Aalborg University, Aalborg, Denmark

Vandana Bassoo       Department of Electrical and Electronic Engineering, Faculty of Engineering, University of Mauritius, Reduit, Mauritius

Bishnu P. Bhattarai       Department of Power and Energy System, Idaho National Laboratory, Idaho Falls, ID, United States

Rajeev Kumar Chauhan       School of Computing and Electrical Engineering, Indian Institute of Technology, Mandi, India

Kalpana Chauhan       Department of Electrical and Electronics Engineering, Galgotias College of Engineering and Technology, Greater Noida, India

Weirong Chen       School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China

Bo Chen       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

B. Chitti Babu       University of Nottingham Malaysia Campus, Semenyih, Malaysia

S.S. Choi       School of Electrical, Electronic and Computer Engineering, University of Western Australia, Perth, WA, Australia

Sanchari Deb       Centre for Energy, Indian Institute of Technology, Guwahati, India

Sanjoy Debbarma       Electrical Engineering Department, NIT Meghalaya, Shillong, India

Ali Ehsan       College of Electrical Engineering, Zhejiang University, Hangzhou, People's Republic of China

Xinli Fang

PowerChina Huadong Engineering Corporation Limited

Hangzhou Huachen Electric Power Control Co. LTD., Hangzhou, People's Republic of China

Deqiang Gan       College of Electrical Engineering, Zhejiang University, Hangzhou, China

Yajing Gao       School of Electrical and Electronic Engineering, North China Electric Power University, Baoding, China

Ying Han       School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China

Xiaoqing Han       Shanxi Key Laboratory of Power System Operation and Control, Taiyuan University of Technology, Taiyuan, China

Yujin Huang       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

Le Jiang       College of Electrical Engineering, Zhejiang University, Hangzhou, People's Republic of China

Karuna Kalita       Department of Mechanical Engineering, Indian Institute of Technology, Guwahati, India

Mitja Kolenc       ELES, d.o.o., Ljubljana, Slovenia

Qi Li       School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China

Jinghua Li       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

Bo Lu       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

Jin Ma       School of Electrical and Information Engineering, The University of Sydney, Sydney, NSW, Australia

Pinakeswar Mahanta       Department of Mechanical Engineering, Indian Institute of Technology, Guwahati, India

Yuhong Mo       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

Qitian Mu       School of Electrical and Electronic Engineering, North China Electric Power University, Baoding, China

Kurt S. Myers       Department of Power and Energy System, Idaho National Laboratory, Idaho Falls, ID, United States

Bonu Ramesh Naidu       Department of Electrical Engineering, Indian Institute of Technology Kharagpur, Kharagpur, India

Gayadhar Panda       Department of Electrical Engineering, National Institute of Technology Meghalaya, Shillong, India

Feng Qiao       School of Electrical and Information Engineering, The University of Sydney, Sydney, NSW, Australia

M.M. Rana       College of Science and Engineering, James Cook University, Townsville, QLD, Australia

Mohammad Seydali Seyf Abad       School of Electrical and Information Engineering, The University of Sydney, Sydney, NSW, Australia

Rituraj Shrivastwa       Department of Electrical Power Engineering (IEE), Grenoble Institute of Technology, Grenoble, France

Nermin Suljanović       Milan Vidmar Electric Power Research Institute, Ljubljana, Slovenia

Robert J. Turk       Department of Power and Energy System, Idaho National Laboratory, Idaho Falls, ID, United States

Zhen Wang       College of Electrical Engineering, Zhejiang University, Hangzhou, China

Eric Wang       College of Science and Engineering, James Cook University, Townsville, QLD, Australia

Zhibang Wang       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

Shanyang Wei       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

Kit Po Wong       School of Electrical, Electronic and Computer Engineering, University of Western Australia, Perth, WA, Australia

Wei Xiang       College of Science and Engineering, James Cook University, Townsville, QLD, Australia

Zhengqing Yang       College of Electrical Engineering, Zhejiang University, Hangzhou, China

Qiang Yang       College of Electrical Engineering, Zhejiang University, Hangzhou, People's Republic of China

Hanqing Yang       School of Electrical Engineering, Southwest Jiaotong University, Chengdu, China

Ting Yang       Key Laboratory of Smart Grid of Ministry of Education, School of Electrical and Information Engineering, Tianjin University, Tianjin, China

Matej Zajc       Faculty of Electrical Engineering, University of Ljubljana, Ljubljana, Slovenia

Jiasheng Zhou       Department of Electrical Engineering, Guangxi University, Nanning, People's Republic of China

Xiaojie Zhou       School of Electrical and Electronic Engineering, North China Electric Power University, Baoding, China

Jing Zhu       Maintenance Branch Company of State Grid Fujian Electric Power Co., Ltd, Xiamen, China

Yi Zong       Center for Electric Power and Energy (CEE), Department of Electrical Engineering, Technical University of Denmark, Roskilde, Denmark

About the editors

Dr. Qiang Yang holds a BS degree (first class honors) in electrical engineering and received an MSc (with distinction) and a PhD degree both in electronic engineering and computer science from Queen Mary College, University of London, London, UK, in 2003 and 2007, respectively. He has worked as a postdoctoral research associate at the Department of Electrical and Electronic Engineering, Imperial College London, UK, from 2007 to 2010 and involved in a number of high-profile UK EPSRC and European IST research projects. He had visited University of British Columbia and University of Victoria Canada as a visiting scholar in 2015 and 2016. Currently, he is an associate professor at College of Electrical Engineering, Zhejiang University, China, and has published more than 150 technical papers, applied for more than 50 national patents, and coauthored 2 books and 10 book chapters. He has received more than 20 research grants in the past 5 years, including the National Key Research and Development Program of China, the National Natural Science Foundation of China, and the National High-Tech Research and Development Program of China (863 Program). His research interests over the years include communication networks, smart energy systems, and large-scale complex network modeling, control, and optimization. He is the senior member of IEEE, member of IET, and IEICE as well as the senior member of China Computer Federation (CCF).

Prof. Ting Yang is currently a chair professor of Electrical Theory and Advanced Technology, at the School of Electrical Engineering and Automation, Tianjin University, China. He was the cooperative research staff of Imperial College London (2008); visiting professor of University of Sydney, Australia (2015). Prof. Yang is the winner of the New Century Excellent Talents in University Award from Chinese Ministry of Education. He is the leader of tens of research grant projects, including the International S&T Cooperation Program of China, the National High-Tech Research and Development Program of China (863 Program), the National Natural Science Foundation of China, and so on. Prof. Yang is the chairman of two workshops of IEEE International Conference, and the editor in chief of one of the special issues of the International Journal of Distributed Sensor Network (DSN). He has authored/coauthored four books, more than 100 publications in internationally refereed journals and conferences. Prof. Yang is a senior member of the Chinese Institute of Electronic, the fellow of Circuit and System Committee, the fellow of Theory and Advanced Technology of Electrical Engineering, and the member of International Society for Industry and Applied Mathematics. His research fields include smart energy systems, artificial intelligence, and internet of things.

Dr. Wei Li has received his PhD from School of Information Technologies at The University of Sydney. He is currently a research fellow of Centre for Distributed and High Performance Computing, and School of Information Technologies at The University of Sydney. His research is supported by Early Career Researcher (ECR) funding scheme and Clean Energy and Intelligent Networks Cluster funding scheme, The University of Sydney. He is the recipient of four IEEE or ACM conference best paper awards. His research interests include Internet of things, edge computing, energy efficient, task scheduling, and optimization. He is a senior member of IEEE and a member of ACM.

Preface

Qiang Yang, College of Electrical Engineering, Zhejiang University, Hangzhou, China

In recent decades, the electric power grid is experiencing a fundamental transformation. Smart grids are considered as one of the biggest technological revolutions since the advent of Internet, as they have the potential to reduce carbon dioxide emissions, increase the reliability of electricity supply, and increase the efficiency of our energy infrastructures. In particular, the electrical network is integrated with the information and communication network in order to improve the stability, efficiency, and robustness. The control, communication, and optimization technologies explain how diverse technologies play hand in hand in building and maintaining smart grids around the globe. This book aims to highlight the recent advances in the control, communication, and optimization aspects of smart distribution grids, provides incredible insight into power system control, sensing and communication, and optimization technologies, and points out the potentials for new technologies and markets.

First, the current resurgence of interest in the use of renewable energy is driven by the need to reduce the high environmental impact of fossil-based energy systems. Smart distribution grids promise to facilitate the integration of distributed renewable sources and provide other benefits as well. Advanced control technology is a key enabling technology for the deployment of renewable energy systems including solar, wind, and other small-scale renewable sources in a reliable and friendly fashion. Second, the underlying communication network of smart distribution grids is a critical enabler of new functions, such as demand response, dynamic pricing, robust distributed generation system, and so forth. It is expected to support the functionalities of enhancing energy savings, cost reductions, and increased reliability and security. In fact, the smart grid sets a novel context for addressing communication problems and devising innovative solutions. Finally, the optimization-based techniques have received a lot of attention in overcoming the outstanding challenges of optimal planning and operation of the smart power distribution systems.

This book is an excellent reference for researchers and postgraduate students working in the area of smart power distribution networks. It also targets professionals interested in gaining deeper knowledge and technical challenges of the smart distribution grids. This book is mainly for readers who have a good knowledge of electrical engineering. We also tried to include sufficient details and provide the necessary background information in each chapter to help the readers to easily understand the content. We hope the readers will enjoy reading this book.

Acknowledgments

Qiang Yang; Ting Yang; Wei Li

We would like to express our gratitude to everyone who participated in this project and made this book a reality. In particular, we would like to acknowledge the hard work of the authors and their patience during the revisions of their chapters.

We would also like to acknowledge the outstanding comments of the reviewers, which enabled us to select these chapters out of the many we received and improved their quality. Some of the authors also served as referees and hence their double task is highly appreciated.

We also thank our family members for realizing the importance of this project and their consistent support throughout the project. Special thanks to Angelia for her continuous love, support, and encouragement over the years.

The editors hope to acknowledge the following grants for supporting their research work: the National Key Research and Development Program of China (Basic Research Class 2017YFB0903000)-Basic Theories and Methods of Analysis and Control of the Cyber Physical Systems for Power Grid, the Natural Science Foundation of China (51777183, 61571324, 51407146, 61473238, 51607068, 51377027), the Natural Science Foundation of Zhejiang Province (LZ15E070001) and the Natural Science Foundation of Tianjin (16JCZDJC30900).

Lastly, we are very grateful to the editorial team at Elsevier Press for their support through the stages of this project. We enjoyed working with Mariana Kühl Leme, who was involved in all the phases, from the time this project was just an idea through the writing and editing of the chapters, and then during the production process. We would also like to thank R.Vijay bharath, from Elsevier, for managing the production process of this book.

Organization of this book

Qiang Yang, College of Electrical Engineering, Zhejiang University, Hangzhou, China

There have been quite a few books out related to the smart grid. However, most of the books attempt to cover the concepts and technologies in the scope of very broad area: smart grid, covering the power generation, transmission, and distribution networks. This book aims to restrict the view to the control, communication, and optimization of smart active power distribution networks with integration of renewable distributed generations. Also, some of them lack comprehensive analysis of the control and optimization-related models and algorithms, besides the communication and data traffic models and analysis are not discussed in details.

This book aims to present the latest technology advances in electric power distribution networks, with a particular focus on the scientific innovations of the methodologies, approaches, and algorithms in enabling efficient and secure operation of smart distribution networks.

This book is divided into three parts, each of which is devoted to a distinctive area.

Part One: Modeling and control of smart power distribution network

Part I of this book aims to focus on theoretical, experimental, and proof-of-concept results of the modeling and control methodologies in the context of active distribution networks.

Chapter 1 presents an overview of grid-integration codes on wind turbine generator (WTG's) frequency regulation in several typical countries. The typical WTG frequency control strategies, such as inertial emulation, de-loading control, overproduction, and droop control are further investigated in detail to reflect the development of related WTG technologies.

Chapter 2 investigates one of the active power control technologies, WTG de-loading control, which is briefly reviewed first. A bi-level optimization model of two-stage reserve scheduling problem considering WTG integration is then proposed to evaluate the benefits of WTG de-loading control on system reserve scheduling.

Chapter 3 presents the control and energy management in a grid-connected DC microgrid comprising solar PV array, battery, and supercapacitor banks in order to maximize the synchronized grid integration of the renewable energy along with the objective of stabilizing the DC microgrid.

Chapter 4 studies the system structure and the math model of each DC MG subsystem. Also, a DC microgrid experiment platform is set up to verify its control ability and energy management performance.

Chapter 5 incorporates the VSC-based MG into the voltage control scheme of the distribution network by implementing a two-layer control structure. With the distribution network central controller at the upper layer, the tap position of on-load tap changer (OLTC), the position of shunt capacitors (SCs), and reactive power injection from MGs can be efficiently coordinated.

Chapter 6 proposes a discrete-time linear-quadratic Gaussian (LQG) controller to stabilize the microgrid states under fading channel condition. Numerical studies have confirmed the effectiveness of using smaller number of step sizes and fading parameters to stabilize the distributed energy resource (DER) states.

Chapter 7 investigates the optimal coordinated operation of multiple autonomous MGs and shows the potential technical benefits. The proposed solution identifies the optimal network topologies and allocates the critical loads (CLs) to appropriate DGs based on the minimum spanning tree (MST) algorithm with power loss and reliability considerations.

Part Two: ICT technologies for smart power distribution networks

Part II of this book focuses on the advanced information and communication technologies, high-performance computing, and cyber security issues in the smart distribution grids.

Chapter 8 identifies different security vulnerabilities in smart grids. The impact of consumer data privacy and confidentiality breach are discussed and existing techniques as proposed in the literature to protect the privacy of customer information in a smart grid are presented.

Chapter 9 studies the structure and categories of microgrid communication system, and overviews the application of microgrid communication system integrating consensus algorithm. In addition, a distributed hierarchical control method is established based on the communication system to balance the battery state of charge in the decentralized battery energy storage system.

Chapter 10 begins with a brief introduction to the basic concepts of smart grids and active distribution networks, followed by an introduction to the areas of active grid that require information and communication technology (ICT) technologies, the power system communication standard, and the supervisory control and data acquisition (SCADA) systems. In addition, the cyber-security issues are also discussed.

Chapter 11 investigates the communication system architecture of VPPs, giving an overview of current communication technologies and communication protocols. The study focused on the downstream communication between the virtual power plant and distributed energy resources, and upstream communication between the virtual power plant, transmission system operator, distribution system operator, electricity market, and retailers.

Chapter 12 presents virtual synchronous power concept (SPC)-based high-voltage DC transmission (HVDC) systems for emulating the virtual inertia in order to regulate grid frequency. This work also considers mobile electric vehicles (EVs) network as energy storage with smart V2G algorithm taking into consideration future driving demand of EV owners.

Chapter 13 discusses the current challenges and opportunities of internet of things (IoT)-enabled smart energy systems from a number of aspects. Existing approaches and recent solutions with respect to domestic demand response, IoT cyber security, and modeling and simulation challenges faced by current smart grid are provided.

Chapter 14 proposes an H-infinity-based microgrid state estimation algorithm. First of all, the renewable microgrid is represented by the state-space framework. The IoT-based smart sensors are used to obtain the system measurements. The energy management system adopts the H-infinity-based state estimation algorithm where it is not required to know the exact noise statistics.

Part Three: Optimization models and methods in smart power distribution networks

Part III of this book aims to focus on the optimization models, methodologies, and techniques of smart distribution network planning and operation management.

Chapter 15 presents an intelligent control strategy for photovoltaic (PV) and multi-battery bank for a DC microgrid in grid connected as well as isolated mode.

Chapter 16 introduces scenario-based methods for tackling the uncertainties of renewable generation in the electricity distribution network planning problem and two scenario-generation methods are adopted.

Chapter 17 proposes two approaches for network operation in consideration of renewable generation uncertainties using probabilistic-based and scenarios-based approaches.

Chapter 18 proposes a method to access the wind power accommodation by considering adequacy indexes aiming at the statistical characteristics of wind power. The optimal planning of wind power capacity and energy storage capacity is addressed based on the bilinear interpolation theory.

Chapter 19 explores the optimal energy dispatch problem in the scope of residential community with penetration of renewable DGs and energy storage in the presence of real-time pricing. An efficient algorithmic solution is presented and implemented at two levels: optimal control within individual households and energy trading among neighboring households.

Chapter 20 carries out the evaluation on the short-term power supply capacity of active distribution system based on multiple scenarios in order to fully consider the operational uncertainties.

Chapter 21 proposes an integrated multi-time-scale energy management approach for active distribution networks. The suggested solution can not only maximize the deployment of flexibility from spatially distributed resources, but also enable single flexible resource to provide multiple grid support functionalities.

Chapter 22 investigates the impact of EV charging station load on different operational parameters of the distribution network. The method of optimal placement of the EV charging stations in the distribution network is also studied considering different operating network parameters.

Chapter 23 studies the relaxation and linearization methods for the hosting capacity assessment in power distribution systems.

Part One

Modeling and Control of Smart Power Distribution Network (Control Aspect)

1

An overview of codes and control strategies for frequency regulation in wind power generation

Zhen Wang⁎; Kit Po Wong†; S.S. Choi†; Deqiang Gan⁎; Yi Zong‡    ⁎ College of Electrical Engineering, Zhejiang University, Hangzhou, China

† School of Electrical, Electronic and Computer Engineering, University of Western Australia, Perth, WA, Australia

‡ Center for Electric Power and Energy (CEE), Department of Electrical Engineering, Technical University of Denmark, Roskilde, Denmark

Abstract

With rapid growth of installed wind power capacity worldwide and significant development of wind turbine generator (WTG) technologies, WTGs gradually have the capability to participate in system frequency regulation. To this regard, an overview of grid-integration codes on WTG's frequency regulation in several typical countries is first carried out. Then, typical WTG frequency control strategies, such as inertial emulation, de-loading control, overproduction, and droop control, are investigated in detail to reflect the development of related WTG technologies. In particular, existing research works about frequency coordination control at three levels, WTG level, wind-farm level, and system level, are discussed in depth and summarized. Future trends are discussed in the end.

Keywords

Grid code; Active power control; Frequency response and regulation; Wind power integration

Acknowledgment

The National Natural Science Foundation of China (No. 51677165) and the Danish Agency for Science, Technology and Innovation (No. 4070-00023B) are to be acknowledged for their funding support.

1.1 Introduction

In recent years, significant increase of wind power generation has emerged for the purpose of releasing the pressure of energy shortage and low-carbon power supply. However, with a large amount of wind power integrated, the daily operation of conventional power systems can be inevitably affected (Ummels et al., 2007; Banakar et al., 2008). For example, more synchronous reserve capacities are desired for frequency regulation because of the intermittent, volatile, and antipeaking characteristics of wind power (Ernst et al., 2007).

Those variable-speed wind turbine generators (VSWTGs) such as double-fed induction generators (DFIGs) and permanent magnet synchronous generators (PMSGs), have gained high market share during the past few years due to their high efficiency and easy-controllability performance over a wide range of wind speeds. Generally, typical VSWTG is connected with a grid through power electronic converters. Therefore, its rotational speed is decoupled with system frequency and it cannot provide any rotational inertia to the main power grid (Mullane and O'Malley, 2005; Holdsworth et al., 2004). As a result, the system equivalent inertia will greatly decrease when large numbers of VSWTGs are introduced to replace synchronous generators (SGs). On the one hand, a low inertia system becomes more prone to large frequency oscillation when any serious disturbance occurs (Lalor et al., 2005). On the other hand, as most VSWTGs are designed to operate at the maximum power point tracking (MPPT) mode and they usually have no frequency regulation ability, in consequence the frequency regulation burden of conventional SGs will be aggravated; even the risk of system frequency collapse is thus exposed (Lalor et al., 2005; Conroy and Watson, 2008). Therefore, when large-scale wind power is integrated into a power grid, more challenges regarding frequency regulation and system reserve schedule and dispatch will emerge for the independent system operator (ISO).

Compared with conventional SG, WTG has its own characteristics regarding frequency control: (1) WTG usually has a faster frequency response than conventional units, but its response capability is greatly limited by real wind condition; (2) WTG designed as other control scheme may exhibit quite different characteristics. Therefore, it's very important to investigate WTGs’ frequency regulation ability according to their control schemes.

Only years ago, most WTGs were still regarded as unscheduled units due to their intermitted power output characteristics and were exempted from system frequency regulation. With the development of control technologies and increasing pressure of frequency regulation, WTGs now can participate in different levels of frequency control events. In fact, in many European countries such as Denmark and Germany, there exist corresponding operation codes (Energinet, 2004; Tennet TSO GmbH, 2012). As a result, the potentials of frequency support function by WTGs are gradually attracting attention from the power industry as well as academia (Sun et al., 2010).

The purpose of this chapter is to investigate in depth and summarize the development of frequency regulation issues in wind power generation. The remainder of the chapter is organized as follows: in Section 1.2, existing grid codes of frequency regulation for WTG in several countries are investigated and compared. A three-level hierarchical dispatch framework is introduced in Section 1.3. System-level and plant-level control schemes are presented in Sections 1.4 and 1.5, respectively. Section 1.6 introduces four types of control schemes for WTGs to provide frequency support and their control mechanisms and characteristics are elaborated. In addition, future trends on a three-level frequency regulation framework are discussed in Section 1.7. Section 1.8 gives the conclusions.

1.2 Grid codes on frequency regulation

After dozens of years’ development, several representative countries have established frequency regulation rules on wind farms in various grid codes, which are summarized in Fig. 1.1, including representative countries such as Germany (Tennet TSO GmbH, 2012), South Africa (Eskom System Operation and Planning, 2012), England and Scotland (UK National Grid, 2009), Sweden (Nordel, 2007), Ireland (Irish EirGrid, 2015), Denmark (Energinet, 2004), USA ERCOT (USA ERCOT, 2013) and Canada (Hydro-Québec TransÉnergie, 2009), China (GB Institute, 2011). It can be seen that in most grid codes the regulated system frequency can be divided into five bands (based on a 50 Hz rate frequency):

•serious supersynchronous band (SSUP, > 52.0 Hz)

•medium supersynchronous band (MSUP, [51.0,52.0]Hz)

•synchronous band (SYN, [49.0,51.0]Hz)

•medium subsynchronous band (MSUB, [47.0,49.0]Hz)

•severe subsynchronous band (SSUB, < 47.0 Hz)

Fig. 1.1 A comparison of grid codes according to the frequency regulation requirement.

It is very clear that in the SYN range, in most countries above WTGs will maintain MPPT operation (full production). When the system frequency increases and enters into the MSUP range, WTGs need to actively reduce their output in order to prevent the system from overfrequency. On the other hand, when the system frequency is in the MSUB range, power reduction is strictly controlled to prevent the underfrequency situation. And if system frequency is in the SSUP or SSUB range, WTGs are mostly allowed to be tripped off entirely.

The representative five-band frequency regulation can be clearly seen in the Denmark P-f curve of Fig. 1.2A, which can be outlined by the five P-f points A-B-C-D-E. In Fig. 1.2, there particularly exist two separate operation codes: (1) under MPPT operation or unreduced production, WTG farms can only make downward regulation; and (2) under active power control (APC) or reduced production, WTG farms will operate at 50% MPPT output when the system frequency is in the 49.5–50.5 Hz range, in which the opportunity cost incurred would be compensated accordingly; as a result the WTG can make an up- or down-regulation. The corresponding P-f curves are given in Fig. 1.2B, which can reflect the relationship between the system frequency and WTG power output. In these curves, there usually exists a dead band for WTG control to prevent frequency action near synchronous frequency.

Fig. 1.2 Denmark and Ireland P - f curves. (A) Ireland P - f curve ( Irish EirGrid, 2015) (B) Denmark P-f curve ( Energinet, 2004).

According to different thresholds and regulation requirements in Fig. 1.1, European countries such as Germany and Sweden have relatively high requirements regarding ramp rate and lasting time than Canada and China. As a matter of fact, the grid codes continuously evolved as WTG's control technologies advance or administration manner changes. For example, from April 1, 2005 the UK Grid Code has taken effect in England, Scotland, and Wales, replacing the Scottish Grid Code (Scottish TSO, 2010). In China, WTGs had no duty to participate in frequency regulation before 2011, but the new national standards drafted in 2011 clearly raise a similar frequency regulation requirement for European countries. In addition, there is evidence that the grid code on a wind farm's power ramp rate very specifically caters for the frequency regulation requirement (GB Institute, 2011):

•If PR < 30 MW, then R10 ≤ 10 MW/10 min, and R1 ≤ 3 MW/10 min

•If 30 MW ≤ PR ≤ 150 MW, then R10 ≤ PR/3, and R1 ≤ PR/10 within 10 min

•If PR ≥ 150 MW, then R10 ≤ 50 MW/10 min, and R1 ≤ 15 MW/10 min

where PR, R10, and R1 denote the installed capacity and the 10- and 1-min power ramp rate, respectively.

Despite nonspecific requirement regarding WTGs’ frequency regulation in the FERC's grid code (USA FERC, 2005), some regional power systems in United States such as Texas ERCOT also issue rules that wind farms/plants should provide a proper frequency response to severe system disturbances (USA ERCOT, 2013; Ackermann, 2012). In addition, the frequency ancillary service can be captured by those mature frequency regulation markets in United States, for example, PJM and ERCOT.

It should be noted that the P-f curve described in Fig. 1.2 actually is the WTG frequency response resulting from APC technologies, which have enabled WTG to adjust its actual operating point according to the operator's command or system grid codes.

1.3 Frequency regulation framework

As wind power's penetration increases in a power system, the coordination of WTGs, wind farms, and conventional power system equipment would attract considerable attention for the purpose of secure and economic operation of the power grid. When there are wind farms participating in system frequency regulation, a hierarchical three-level dispatch framework can be utilized as there exist different time-scale characteristics among all participants involved (Wang et al., 2015a): the WTG level, the plant level, and the system level as illustrated in Fig. 1.3. On the system level, ISO will concentrate on co-dispatching wind farms with conventional generators or energy storage system for frequency regulation; on the plant level, wind farms are committed to organize and dispatch the WTGs, and determine which control mode the WTG will rest on, MPPT or APC; on the WTG level, some network-friendly and closed-loop control strategies, such as pitch control and de-loading control, can be adopted for WTG to participate in the system ancillary service.

Fig. 1.3 Three-level frequency regulation framework considering a wind farm's active power control ability.

In the three-level dispatch framework, a bi-directional data and information exchange is established among WTGs, wind farms, ISO, and other conventional plants, including: (a) WTG operating condition, including the instant wind speed vk and the WTG's state such as rotor speed ωk and pitch angle βk related to the kth WTG in a wind farm; and (b) the active power adjustment of other fossil plants PC/ΔPC and hydropower plants, PE/ΔPE;. In this framework, there exist two implementation schemes: passive frequency regulation (Wang et al., 2015a) and active frequency regulation (Kroposki et al., 2017).

1.3.1 Passive frequency regulation

Normally, all WTGs in a wind farm will operate at MPPT in the SSF band (| Δf | ≤ 0.1 Hz). All aggregated WTG information is sent to the ISO and a specific frequency regulation scheme is therefore formulated according to the grid-code requirement (Energinet, 2004). When there is frequency deviation beyond the SSF band, the ISO will send a frequency regulation command in the form of power adjustment to the ith wind farm, either the upward ΔPi+ (MSF or SSBF) or the downward ΔPi−, in which the WTGs will receive the command of power output adjustment from the wind farm-level supervisory control (WFSC).

1.3.2 Active frequency regulation

The main difference compared with the above scheme is that wind farms will actively reschedule their operating curves when they participate in competitive ancillary services markets and provide a frequency regulation service (Singarao and Rao, 2016), which will also prespecify the wind farm's operation mode such as MPPT or APC (Badihi et al., 2015). In this way, some WTGs will actively keep reserve capacities for frequency regulation. Therefore, the active regulation scheme can be more credible. However, the establishment of a prescheduling mechanism in the active regulation scheme is quite complicated and will be discussed below.

1.4 System-level control

On the system level, the required power regulation, ΔPtotal, can be captured by the ISO in two ways: (1) the ISO can purchase the required capacity via some optimal bidding in a joint energy and regulation market (He et al., 2017); and (2) there is preallocation among conventional thermal/hydro/gas power plants as well as wind farms (if available) according to some negotiated rate and allocation coefficients (Wang et al., 2017), similar to conventional AGC procurement. For example, the corresponding adjustment on the power reference for the ith power plant, ΔPi, can be proportionally allocated:

   (1.1)

where the allocation coefficient pfi meets ∑ pfi = 1 and can be specified according to the power plant's available reserve capacity, ramp ratio, rated capacity, etc. (Zhang et al., 2014; Li et al., 2015).

To fulfil the system control purpose, some two-level hierarchical control schemes can be an alternative to achieve satisfactory performance of dynamic frequency and damping response (Leon et al., 2012) without taking WTG's APC ability into account. The upper level includes a centralized controller based on synchronized wide-area signals, and the decentralized controller in the lower level can coordinate all conventional SGs, wind farm converters, energy storage system, FACTS, etc. (He et al., 2017; Leon et al., 2012; Kothari, 2005).

On the system level, the ISO needs to consider an overall dispatch among wind farms and conventional power plants to maintain system reserve capacity adequacy.

1.5 Plant/farm-level coordinated control

The purpose of plant/farm-level control is to cooperate all aggregated WTGs in a wind farm. Usually, the wind farm will arrange the WTG sequence to participate in frequency regulation according to the WTG status (on or off service) in the WFSC system and thus the wind farm can effectively respond to the ISO's command to adjust its power outputs. The operating points of the active and reactive power for these WTGs can be scheduled according to some dynamic distribution factor (Chang-Chien et al., 2008), or systematically determined by some advanced optimization algorithms, for example, in (de Almeida et al., 2006) a primal-dual predictor-corrector interior point method is developed to carry out ISO requests.

A general optimal model related to this reserve scheduling problem can be formulated as follows:

   (1.2)

   (1.3)

The object is to pursue a minimum regulation cost, where ak and ΔPi,k are the cost coefficiency and power adjustment of the kth WTG in the ith wind farm, respectively. The related constraints in Eq. (1.3) include: (1) the total frequency regulation capacity (ΔRi+/ΔRi−) required should be respected; and (2) each WTG has its power adjustment bounds (Rk−,max/Rk+,max) that are dependent on its operating conditions. In addition, there exist other plant-level control targets, such as minimum power reduction of WTG for maintaining a frequency regulation reserve (Diaz et al., 2013) and minimum number of acted WTGs, for example, a WTG with more reserve has priority to respond to the dispatch command.

The challenge on this level lies in the fact that the operating condition for each WTG in a farm actually differs due to environmental and spatial effects, and thus a predictive evaluation of the WTG's adjustment bounds requires those advanced real-time measurement technologies.

1.6 WTG-level control strategy

WTG-level control is usually implemented by an auxiliary closed-loop control such as inertial emulation (Gautam et al., 2011), overproduction control with step input (Keung et al., 2009), de-loading control (Chang-Chien et al., 2011), and droop control (Vidyanandan and Senroy, 2013), which will be elaborated below.

1.6.1 Inertial emulation control

By inertial emulation control, WTG can emulate the inertial response and frequency regulation similar to that of SG, which is mainly implemented by adding some PI loop and introducing the frequency signal into the torque or power control (Anaya-Lara et al., 2006). Typical control strategies are given in Fig. 1.4A. In mathematics, the control law in Fig. 1.4A can be formulated in Eq. (1.4):

   (1.4)

where f is the system frequency; Tgen and Tm are the electromagnetic torque and mechanical torque, respectively; and KD is the derivative time constant. Further, Eq. (1.4) can be rewritten as Eq. (1.5) by introducing the rotor speed ω:

   (1.5)

Fig. 1.4 (A) Inertial emulation control schemes; (B) overproduction power reference signal.

On the other hand, the torque-rotor speed relationship of any SG has similar expression in Eq. (1.6), where H denotes the inertial constant and other variables are SG counterparts:

   (1.6)

By comparing Eq. (1.5) with Eq. (1.6) above, it is observed that WTG can emulate a conventional generator with an inertia time constant KD/4π. In short, WTGs with inertia emulation control are capable of increasing system rotational inertia (Kayikci and Milanovic, 2009). The emulated inertia will be released or absorbed when the system frequency varies. The corresponding power injection or absorption is completely uncontrollable and only depends on df/dt or final Δf once the PI parameters are set. That means, WTGs with inertia emulation control only cannot maintain long-term APC ability.

Despite the underlying common philosophy, the inertial emulation control can be classified into several categories (Wu et al., 2017): (1) natural inertial control, which can emulate the inertial response of a conventional SG by introducing a df/dt loop control (Hwang et al., 2016; Van de Vyver et al., 2016; Zhao et al., 2016); and (2) virtual synchronous control, which enables a DFIG to deliver the inertial response to enhance the frequency stability without the traditional PLL loop (Wang et al., 2015b,c; Huang et al., 2017) and is particularly helpful when WTG is integrated into a weak AC power grid with low short-circuit ratio.

1.6.2 Overproduction

By adding a step input signal, WTG with overproduction control could increase the power output when the system frequency drops. A typical step input signal of overproduction is shown in Fig. 1.4B, where Pe0 represents the WTG steady power. When the system frequency drops, there will be a power reference increment ΔPe(d) and this process will last for td. Kinetic energy stored in rotational mass is extracted to increase electrical power output and the rotor decelerates during the deceleration period. In order to avoid stalling, there exists an acceleration process (ΔPe(a), ta) (Rawn et al., 2010).

Another overproduction scheme can be implemented by overloading a converter for a short time when the wind speed is above a rated value (Rawn and Lehn, 2008). It should be noted that, the over-loading capability of a wind generator is determined by the maximal excess power that the drive train, the generator, and the converter can withstand without adding damaging fatigue loads on wind turbine structure.

Overproduction is similar to inertia emulation in making use of kinetic energy stored in rotational mass to serve extra output. The characteristic of these kinetic energy releasing-based control strategies (inertia emulation and overproduction) is that there exists an acceleration process to recover rotor speed for preventing rotor stalling. In the acceleration process, the power output will be less than in the normal state. Therefore, there exists a secondary frequency drop event (Wang et al., 2015a). The difference is that the output power profile of the overproduction scheme can be set manually. Hence, overproduction has APC capability and can be used as an active frequency regulation strategy.

1.6.3 De-loading operation

The de-loading control can make WTG maintain a higher rotate speed and less power output compared with MPPT that has the potential to keep reserve capacity for frequency regulation (de Almeida and Lopes, 2007). The de-loading operation principle is illustrated in Fig. 1.5, in which a group of bell curves reflects the relationship between mechanical power and rotor speed; the MPPT curve and de-loading curve reflect the relationship between electrical power output and rotor speed. In Fig. 1.5, Pb represents de-loading power while Pa represents MPPT power. There are two ways to implement the de-loading operation of WTG: (1) the rotational speed control, which means the P-ω curve changes from the MPPT curve to the de-loading curve (Fig. 1.5A); and (2) the pitch angle control, which means the Pm curve changes from a solid to a dashed bell curve (Fig. 1.5B). Under rotational speed control (Fig. 1.5A), the power reserve for frequency regulation can be calculated by Eq. (1.7):

   (1.7)

where (Pa − Pb) means power deviation between the MPPT mode and the de-loading mode; H(ωb² − ωa²) means kinetic energy stored in rotational masses, which will release when the operation point moves from point b to point a; H is the inertia constant and T is the conversion time from point b to point a. Under pitch angle control (Fig. 1.5B), the power reserve of WTG can be calculated as follows:

   (1.8)

Fig. 1.5 Principle illustration of two de-loading schemes: (A) rotational speed control and (B) pitch angle control.

Differing from formula (1.7), there is no additional item caused by kinetic energy transformation, because acceleration energy for a rotor to move from point b to point a is extracted from wind energy. Learning from Eqs. (1.7), (1.8), rotational speed control usually has more reserve capability than the pitch angle control under the same wind condition. But limited by the maximum rotor speed, rotational speed control can only be applied under middle wind speed.

A typical rotational speed control scheme is proposed in Fig. 1.6A, in which the WTG is de-loaded in advance and there will not be an additional power signal △P until the system frequency drops. And, a typical pitch angle control scheme is proposed in Fig. 1.6B.

Fig. 1.6 Two types of de-loading control strategies: (A) rotational speed control and (B) pitch angle control.

Under de-loading control, WTG's output power is not fully rated in advance and the power reserve can be released faster and last for a more long-term timescale than the inertial emulation control. What's more, the pitch angle control usually acts more slowly than rotational speed control, because the pitch angle control relies more on mechanical action while rotational speed control basically relies on a converter's fast response (Attya and Hartkopf, 2014).

1.6.4 Droop control

A typical droop control scheme is given in Fig. 1.7, where the additional power signal ΔP is calculated through Eq. (1.9):

   (1.9)

Fig. 1.7 Droop control scheme.

The droop control is to make a WTG turbine emulate SG's droop characteristic in frequency regulation, which is usually combined with other strategies such as the de-loading control to ensure that the output of WTG increases with the drop of system frequency gradually. However, droop control also has its own disadvantage: the droop control with fixed droop coefficient may lead to WTG instability. A variable droop control strategy can not only solve this problem but also improve the efficiency of frequency regulation (Vidyanandan and Senroy, 2013).

1.6.5 Hybrid control schemes

To make full use of the above WTG-level control strategies, several of the above control strategies can be combined. For example, a joint inertia emulation and droop control can achieve remarkable frequency dynamics (Hwang et al., 2016; Lee et al., 2016); a hybrid strategy scheme with the droop control and the de-loading control combined has long-term frequency regulation performance similar to SG (Huang et al., 2017; Lu et al., 2014).

Another hybrid scheme is to switch over among control strategies according to the real wind condition (Attya and Hartkopf, 2014). As listed in Table 1.1, the wind speed can be classified into four continuous intervals: (1) the cut-in speed vin; (2) the cut-out speed vout; (3) the low speed vlow and (4) the high speed vhigh. Here, vlow and vhigh are the wind speed at which the WTG power is equal to 0.2 and 1.0 p.u., respectively, as denoted at points A and B of Fig. 1.5A, respectively.

Table 1.1

Normally, WTG operates in the MPPT mode. When there is any frequency support requirement ordered from the ISO, WTG can be mutually switched to frequency control modes, for example, when there is vlow ≤ v ≤ vhigh, WTG will be switched to the de-loading mode to hold a frequency regulation reserve.

1.6.6 Performance comparison

A performance comparison of the various control strategies above is given in Table 1.2, in which C1-C4 represent the implementation methods these control schemes will adopt; F1-F2 are two typical types of frequency controls with different timescales. It is obvious that only in the middle or high wind speed band can WTGs have large frequency regulation capacity. And there is no single control strategy that can provide steady frequency support under full-scope wind conditions.

Table 1.2

C1, step reference command; C2, converter overloading; C3, rotational speed control; C4, pitch angle control; F1, the primary control; F2, the secondary control.

1.7 Discussion

In this section, future work under the three-level frequency regulation framework is discussed.

1.7.1 WTG level

1.7.1.1 Advanced control strategies

During the process of inertial emulation control, a secondary frequency drop phenomenon is likely to occur during the rotor speed recovery due to the reduced output power; thus, it is desirable to design advanced control schemes to provide inertial response support while preventing the rotor speed from overdeceleration and mitigating the impact of rotor speed recovery on the overall frequency performance (Wu et al., 2017).

1.7.1.2 WTG regulation margin assessment

At some given wind speed, the regulation margin, that is, the maximum de-loading ratio (Lu et al., 2014) or the additional power reference signal ΔPe(d) in Fig. 1.4B, needs to be accurately assessed for each type of WTG in order to establish the link between the wind speed and the frequency regulation capacity and fully exploit frequency regulation potential.

1.7.2 Wind plant/farm level

1.7.2.1 WTG inner coordination

Due to the impact of local-scale terrain, WTGs’ site-specific information should be collected for decision-making, so that the regulation reference can be determined to achieve a smooth frequency regulation (Piwko et al., 2012).

1.7.2.2 Frequency regulation capacity assessment

As previously mentioned, diverse control strategies will be triggered as the wind speed changes, as reported by Lei and Infield (2013). As terrain wind difference exists, assessment of the aggregated frequency regulation capacity considering site-specific information is desired for system planning and scheduling (Yan and Saha, 2015).

1.7.3 System level

1.7.3.1 Dynamic allocation scheduling

An ordinary allocation mechanism among various aggregated devices has been introduced in Section 1.4, in which there is a fixed allocation coefficient. Actually, the allocation scheme will be affected by the system operation status and wind condition. That means the allocation coefficient pfi in Eq. (1.1) should be dynamic and time varying. The solution process of this problem can be built on the actual system scheduling scheme (Wang et al., 2015d).

1.7.3.2 Economics of frequency service

As active power output and generation benefits will decrease when wind power participates in system frequency regulation, a market mechanism needs to be introduced to price the ancillary service provided by related wind farms (Saiz-Marin et al., 2012).

1.8