Research On Voltage Fluctuation Suppression Strategy Optimization Of Wind Farms Based On Power Prediction

As a clean renewable energy,wind power can effectively reduce the human society's demand for non-renewable energy,avoid environmental pollution,and ensure the stable and sustainable development of the social economy.Therefore,it is significance to vigorously develop and utilize wind energy resources.Improving the utilization rate of wind energy is the key to the development of wind power.So,it is worthy of people's attention how to effectively solve the problem of wind power grid connection and improve the grid power quality.The background of this study is to suppress the fluctuation of grid-connected voltage in wind farms.STATCOM optimal control strategy based on fan output power prediction is proposed to suppress voltage fluctuation based on reactive power compensation.The main contents of this thesis are as follows:Firstly,by analyzing the electric line structure and grid-connection characteristics of gridconnection system in wind farms,it is determined that the grid-connection voltage fluctuation is mainly caused by the varying reactive power.So,based on the purpose of suppressing voltage fluctuation,a scheme is suggested that STATCOM device is used for reactive power compensation of the system.Meanwhile,the traditional STATCOM control strategy is studied and analyzed.It is concluded that the traditional control strategy has some limitations and needs to be improved in the application of wind farm under special conditions.In order to improve the traditional control strategy,it is necessary to study the variation law of wind power output.An ensemble learning method based on Bagging Algorithm is proposed by studying and comparing the traditional single neural network predicting mode.This method has better performance in the prediction of fan output power and can be applied to the optimization of traditional STATCOM control strategy.Based on the prediction of power change,the control strategy is suggested of PI parameter setting for the inner current loop and outer voltage loop according to the predicted power level,by analyzing the characteristics of the double closed-loop structure of feedforward decoupling of STATCOM.In this strategy,the current inner ring adopts the parameter setting scheme of variable damping coefficient,while the voltage outer ring adopts the parameter setting scheme based on Wu elimination method constrained by dynamic performance.Simulation results show that the dynamic response time of STATCOM,under the PI parameters after setting,satisfies the requirements of the project under different working conditions.Finally,the experimental platform of grid connection system of wind farm was built to verify the feasibility of the theoretical method in practical application.The application of STATCOM is realized in voltage fluctuation suppression under the parameter optimization control strategy based on power prediction.The experimental results show that the response time of voltage fluctuation suppression is less than 10 ms under different power levels,which meets the requirements of the grid-connected system for the response speed of voltage fluctuation suppression.