Load Frequency Control of Power Systems With Electric Vehicles and Diverse Transmission Links Using Distributed Functional Observers

What is it about?

This paper presents a load frequency control scheme using electric vehicles (EVs) to help thermal turbine units to provide the stability fluctuated by load demands. First, a general framework for deriving a state-space model for general power system topologies is given. Then, a detailed model of a four-area power system incorporating a smart and renewable discharged EVs system is presented. The areas within the system are interconnected via a combination of alternating current/high voltage direct current links and thyristor controlled phase shifters. Based on some recent development on functional observers, novel distributed functional observers are designed, one at each local area, to implement any given global state feedback controller. The designed observers are of reduced order and dynamically decoupled from others in contrast to conventional centralized observer (CO)-based controllers. The proposed scheme can cope better against accidental failures than those CO-based controllers.

Featured Image

Why is it important?

In this paper, we present a novel LFC scheme that incorporates EVs into the stabilization of load fluctuations. To demonstrate the feasibility of our scheme, we consider a fourarea power system where each area is interconnected to others via a combination of ac/HVDC links and TCPS. A detailed mathematical model of a four-area power system including the dynamics of EVs is derived in this paper. Accordingly, with the availability of a detailed mathematical model, a state feedback control law can be easily designed to optimally determine the charging/discharging behaviors of EVs and the generating units’ power output. However, any feasible optimal state feedback control law inevitably requires the information of all the state variables in order to generate a control input signal. Hence, it becomes unrealistic if some state variables are not available for feedback control. To overcome this problem, conventional CO-based controllers where state observers were used to reconstruct the unmeasured states have been proposed. However, CO-based control schemes require complex hardware and a central facility for processing very large amount of information in real-time and on-line. Therefore, to be able to implement any optimally designed global state feedback control law, it is important to develop some reduced-order distributed observer-based schemes where the processing of information and the control task are shared among the local controllers. For security and economical considerations, it is desirable that these controllers share as little information among themselves as possible