Multi-Area Power System Data-Driven Load Frequency Control

Load frequency control(LFC)is one of the fundamental tasks and key issues used in multi-area power systems to maintain real-time power balance between generation and demand,which tracks the load variation of the power system by regulating the active output of generating units to maintain the system frequency at the rated value.Load frequency control is of great significance to ensure power quality,improve economic efficiency and prevent large area blackouts.As the scale and complexity of modern power systems gradually increase,load frequency control also faces more challenges and demands,such as the effects of nonlinearity,time-varying,uncertainty,communication delays,and other factors,as well as new energy access,market operation,and intelligent development,etc.Therefore,load frequency control needs to be studied and designed using advanced theories and methods.For the sake of research convenience,the current research on the control of power systems mainly uses linear models,due to the generator generation rate constraint,governor deadband effect,the actual power system is a complex nonlinear system,it is difficult to obtain accurate power system model information,so it is important to reduce the dependence of the controller on the power system model.In this paper,data-driven load frequency control of power systems is studied in the following aspects:(1)A fuzzy model-free adaptive control strategy for multi-area power systems based on Takagi-Sugeno(T-S)fuzzy model is proposed for the existence of nonlinear links such as generator generation rate constraint and governor dead zone in multi-area power systems.Firstly,a T-S fuzzy model of the power system is established,a fuzzy model-free adaptive controller is designed based on the T-S fuzzy model,and the effectiveness of the scheme is verified by mathematical analysis and numerical examples.(2)Discusses the event-triggered model-free adaptive predictive load frequency control(LFC)problem for a multi-area interconnected power system.First,assuming that the power system model is unknown,the input and output data of the powerline system are used to convert the nonlinear power system into an equivalent linear data model.Combined with the idea of predictive control,a model-free adaptive predictive controller is designed.An event-triggered scheme is also developed in the design of the LFC to save communication and computational burden.Finally,the effectiveness of the scheme is verified by a rigorous mathematical analysis and a numerical example of a three-area power system.(3)A resilient distributed model-free adaptive predictive control scheme is proposed for a multi-region interconnected power system under Denial of Service(DoS)attack.First,considering the uncertainty and high-dimensional nonlinearity,the power system is modelfree adaptive predictive control(MFAPC)model,in which only input and output(I/O)data are used.Second,a prediction compensation algorithm is designed to mitigate the effects of disturbance attacks so that the frequency tracking errors of the multi-area interconnected power system converge consistently under the prediction compensation algorithm.Simulation results demonstrate the effectiveness of the method.