Cover: Energy Storage for Power System Planning and Operation by Zechun Hu

Energy Storage for Power System Planning and Operation

Zechun Hu

Department of Electrical Engineering
Tsinghua University
China








No alt text required.

Preface

The installed capacity of renewable energy generation (REG), represented by wind power and photovoltaic power generation, has been growing rapidly, changing the generation mix of traditional power systems. REG can be connected to the transmission network in a centralized manner, or can be connected to the distribution network in the form of distributed generation, thereby changing the operating rules and power flow direction of the power networks. Unlike traditional thermal power or hydropower units, wind and photovoltaic power outputs are subject to weather conditions, which are random, intermittent, and difficult to predict precisely. In order to cope with the challenges brought by the large‐scale REG integration to the planning and operation of power systems, the deployment of energy storage system (ESS) has become an important and even essential solution.

At present, pumped hydroelectric storage (PHS) is the largest and most mature energy storage type applied in power systems. The optimal planning and operation methods for PHS power plants are quite mature. However, the PHS power plant has a long construction period and a large investment scale, and its development is constrained by multiple factors such as geographical conditions, land occupation, and environmental impacts. With the advancement of new energy storage technologies, e.g. chemical batteries and flywheels, in recent years, they have been applied in power systems and their total installed capacity is increasing very fast. The large‐scale development of REG and the application of new ESSs in power system are the two backgrounds of this book.

In Chapter 1, energy storage technologies and their applications in power systems are briefly introduced. In Chapter 2, based on the operating principles of three types of energy storage technologies, i.e. PHS, compressed air energy storage and battery energy storage, the mathematical models for optimal planning and scheduling of them are explained. Then, a generic steady state model of ESS is derived.

Chapters 3 and 4 of this book mainly discuss the joint scheduling and bidding of ESS with REG. It is assumed that ESS and REG form a union to maximize their total revenue from electricity markets. The key challenge for the union is the randomness of the REG power output. Chapter 3 describes day‐ahead scheduling models and bidding strategies of a REG‐ESS union in the energy market, with a focus on the methods to deal with the random power output of REG. In Chapter 4, the scheduling and bidding strategies of a REG‐ESS union are further discussed. A combined day‐ahead bidding and real‐time operating strategy based on linear decision rules is presented, which considers the revenues and costs of the REG‐ESS union both at day‐ahead and real‐time stages. Forecast error of wind and solar power decreases distinctly with the reducing span of time horizon. The intraday energy markets have been built in the electricity markets of some countries/regions. In Sections 4.4 and 4.5, a rolling optimization strategy for the REG‐ESS union, which considers the day‐ahead and intraday biddings and real‐time operational optimization decisions, is explained.

In Chapters 5–7, this book discusses three areas of ESS participating in optimal dispatch and control of power systems from different time scales, namely unit commitment (UC), optimal power flow (OPF) and automatic generation control (AGC). Energy balance is the key issue for dispatching or controlling an ESS for the optimal operation of power system, which means that the energy coupling constraints of an ESS should be carefully taken into account. For the UC problem, it is quite simple to consider the energy coupling constraints of an ESS. While for the OPF problem, the operating conditions under multiple successive time intervals, rather than a single time period, should be considered and optimized. Chapter 5 first introduces a deterministic UC formulation, which considers the optimal scheduling of ESS. Then, a scenario‐based and a robust security constrained UC formulations are derived. In Chapter 6, two types of solutions for the multi‐period OPF problem are explained, namely an interior point method and a semi‐definite programming method. Interior point methods are popular for solving the single‐period OPF problem, and methods based on semi‐definite programming are research hotspots in recent years. The semi‐definite programming method introduced in Chapter 6 is relatively basic, and readers can learn more comprehensive research results and in‐depth discussions following the references. In Chapter 7, this book introduces the methods of frequency regulation simulation, control and capacity quantification considering the participation of ESS in secondary frequency regulation (i.e., AGC). Compared with the thermal generating units, a battery or flywheel energy storage system can respond to frequency regulation commands sent from power system control center with shorter delay and faster ramping speed. Therefore, technical problems on how to optimally make use of ESS for AGC are discussed, including the optimal allocation of frequency regulation commands to ESS and the required capacities for AGC contributed from both ESS and thermal generating units.

Chapters 8 and 9 of this book discuss the optimal planning of ESS connected to transmission network and distribution network, respectively. For the transmission expansion planning problem considering ESS deployment, the decision variables include the locations, power and energy capacities of ESSs, which greatly increase the difficulties of modeling and solving the problem. A joint planning model for transmission network expansion and ESS deployment is derived in Section 8.3, which considers the active power loss costs. In Section 8.4, the mathematical formulation for transmission expansion and ESS deployment planning considering the daily operating conditions is explained in order to evaluate the energy capacity of ESS more precisely. Chapter 9 introduces the optimal planning models and solution methods for deploying ESSs in distribution network with distributed generations (DGs). By transforming the nonlinear power flow constraints of a distribution network into second‐order cone constraints, the ESS planning problem is formulated as a mixed integer second‐order cone programming (SOCP) problem. In order to consider the uncertain power outputs of DGs and load variations, multiple typical days are selected to represent the changing operating states within a whole year. For large‐scale distribution networks, the established mixed‐integer SOCP problem is difficult to solve directly. In Section 9.4, a solution method based on generalized Benders decomposition is derived. Section 9.5 briefly discusses the optimization model and solution method for the joint planning of distribution network and ESS. This book does not specifically describe the optimal operation problem for distribution network with ESS. However, the sub‐problem on the daily operation optimization of ESS in Section 9.4 can be used as a basic model for this problem.

In summary, this book focuses on the joint operation of REG and ESS, optimal operation of power system with ESS, and optimal planning of ESSs for the power networks. This book can be used as a reference book for graduate students and researchers who are interested in operation and planning of power systems. It should also be useful for technicians in power network planning, power system dispatch, and energy storage investment/operation companies. The ESS technologies and their application in power systems is developing rapidly, and more advanced research results are being and will be published. This book can be served as a basis for understanding the relevant technologies and tracking the latest research achievements.

Acknowledgements

The content of this book is mainly from the research work I carried out with the postgraduates of our research lab in Tsinghua University. The research work was supported in part by National Natural Science Foundation of China (51107060 and 51477082) and also in part by National Key Research and Development Program (2016YFB0900500).

I would like to specially thank Professor Yonghua Song for his support. I must also thank Huajie Ding, Fang Zhang, Shu Zhang, Haocheng Luo, Ge Gao and Zhe Lin for their valuable contributions. Thanks to Professor Pierre Pinson for the contributions on optimization methods.

To my family for their selfless and constant support to me.

Abbreviation List

AC
Alternating current
ACE
Area control error
ADMM
Alternating direction method of multipliers
AGC
Automatic generation control
BESS
Battery energy storage system
CAES
Compressed air energy storage
CG
Conventional generator
CSP
Concentrating solar power
CVaR
Conditional value at risk
DC
Direct current
DESS
Distributed energy storage system
DG
Distributed generation
DNO
Distribution network operator
DNP
Distribution network planning
DoD
Depth of discharge
EMS
Energy management system
ES
Energy storage
ESS
Energy storage system
FES
Flywheel energy storage
GBD
Generalized Benders decomposition
IPM
Interior point method
LHS
Latent heat storage
MILP
Mixed‐integer linear programming
OPF
Optimal power flow
PHS
Pumped hydroelectric storage
PHSP
Pumped hydroelectric storage plant
PV
Photovoltaic
REG
Renewable energy generation
SCADA
Supervisory control and data acquisition
SCES
Supercapacitor energy storage
SCUC
Security‐constrained unit commitment
SDP
Semidefinite programming
SFC
Secondary frequency control
SHS
Sensible heat storage
SMES
Superconducting magnetic energy storage
SO
System operator
SoC
State of charge
SOCP
Second‐order cone programming
TEP
Transmission expansion planning
TES
Thermal energy storage
UC
Unit commitment
UPS
Uninterruptible power supply
V2G
Vehicle‐to‐grid
VaR
Value at risk
VRB
Vanadium redox battery
WF
Wind farm