Research On Multi-model Predictive Control Of Wind Power Generation System Based On Gap Measurement

Due to the shortage of energy in the world and people’s attention to green energy,the new clean energy industry has developed rapidly,and wind power accounts for a large proportion.However,the wind power system is a nonlinear,multivariable complex system with instability and randomness.How to reduce the unbalanced load of wind power generator systems,prolong the service life of the unit,and reduce the adverse impact of power fluctuation on power grid is an urgent problem to be solved at present.Since the existing PID control technology is difficult to achieve the desired control effect,the control strategy and controller optimization have become the focus of research in the field of wind power system control.Aiming at the power control problems of wind power generation system,a multi-model predictive control strategy is studied in this paper.The main research contents are as follows:(1)The control strategies and control objectives of wind power generation systems are described,and the control strategies of wind power generation systems according to different operating conditions of wind turbines are analyzed.Meanwhile,the mathematical model of each subsystem is established according to the basic structure of the wind power generation systems,and a 5 MW wind turbine is simulated by the FAST software from NREL Laboratories of the United States and MATLAB/Simulink software.(2)Due to the incompleteness of model set and the redundancy of model set selection in the existing multi-model predictive control of wind power generation systems,a multi-model modeling method based on gap measurement for wind power generation systems are proposed.The gap measurement theory is used to estimate the neighborhood range of the linear model at the equilibrium point in the wind power generation systems,decompose the neighborhood of the whole working condition,and finally obtain the complete and non-redundant linear model set through model selection.In order to verify the accuracy and completeness of the model,a simulation is carried out by FAST and MATLAB/Simulink simulation software.The simulation results showthat the fitting degree of the model set meets the requirements for nonlinear conditions.(3)The problem of ensuring the switching stability between sub-predictive controllers is studied and a multi-model predictive control strategy with stable switching is discussed.In each subcase,the subcontroller is designed by using an infinite time-domain predictive control algorithm and the sub-controller switching rules are designed to ensure the stability by estimating the elliptical stability domain of the local predictive controller.In order to verify the proposed control strategy,a simulation study is carried out compared with the existing PID control,and the simulation results show that the control effect is obviously better than the existing PID control.