Research On The Extended Model Predictive Control Based Optimal Load Frequency Control Issue Of Muti-area Intrtconnected Power System

An important feature of the modern power system operation is layered,divisional and interconnected.Therefore,load frequency control is of great significant for ensuring the stable operation of multi-area interconnected power system.With the large number of intermittent,stochastic distributed renewable energy,such as photovoltaic energy and wind energy,connected into electric power system,the problem of load frequency control issue for multi-area interconnected power system faces more challenges.In recent years,the application of model predictive control for fast process power electronic converter and electric power system has drawn more and more attentions.However,its application of load frequency control for complex multi-area interconnected power system with high penetration distributed renewable energy is still at its primary stage.In this thesis,it focuses on an improved model predictive control and extremal optimization algorithm for comprehensive study of load frequency control problem in traditional multi-area interconnected power system with and without renewable energy generations(e.g.,solar photovoltaic power generation and wind power generation).The main research work and innovative points of this thesis are as follows:(1)By introducing an extended state vector for the system state space model and a finite-horizon dynamic prediction model,it presents an extended model predictive control algorithm(EMPC)for two area interconnected power system with photovoltaic generation load frequency control.Compared with the existing research work,this thesis considers the nonlinear features such as speed governor dead band(DB)and generation rate constrain(GRC)in the thermal system.The simulation experiments for the cases under normal condition,dynamic load disturbance and parameter mismatch,demonstrate that EMPC has better transient,stable control performance and robustness than the existing firefly algorithm,genetic algorithm and population extremal optimization-based proportional integral control strategies.(2)Based on the research work(1),this thesis introduces the population extremal optimization algorithm into the rolling optimization strategy of EMPC.It proposes a population extremal optimization-based distributed extended model predictive control method(PEO-EDMPC)and applies it into the traditional load frequency control of multi-area interconnected power system.The simulation results on the two-area and three-area interconnected power systems present the PEO-EDMPC algorithm has better control performance and stronger robustness than classical distributed integral control,population extremal optimization based distributed PI and traditional model predictive control.(3)Based on the research work(2),this thesis extends the PEO-EDMPC algorithm into the load frequency control for the multi-area interconnected power system with wind turbines.By simulating the two-area and three-area interconnected power system considering wind turbine,it verifies the superiority of PEO-EDMPC to the traditional model predictive control in terms of transient,steady-state control performance and robust performance.