Research On Optimization Of Wind Farm Energy Storage Capacity Based On Wind Power Forecast

The randomness,volatility,intermittency and peak-reversal characteristics that cannot be ignored in the operation of wind power have a serious impact on large-scale grid-connected consumption,and consequently lead to the phenomenon of“abandonment of wind”.And the increasingly mature energy storage technology can store wind power that cannot be consumed in time on the space-time scale,and suppress the fluctuation of wind power.Therefore,this paper takes wind power as the research object,from the perspective of wind power prediction and energy storage capacity optimization,to improve the shortcomings of large fluctuations in wind power,which is conducive to more reasonable planning and dispatching of wind power grid-connected,alleviating the current serious wind abandonment problem.This article adopts a research method that combines theoretical knowledge and case analysis.First,this article outlines the basic theory of wind power forecasting and energy storage capacity optimization.The DNN-SVM combined forecasting model is used to analyze the wind power data of a wind farm in January 2018.Carry out prediction training and testing,and compare the convolutional neural network and BP neural network algorithms to reflect the good prediction accuracy of the prediction model.Secondly,based on the research results of wind power prediction,for a153 MW wind farm,considering the operating characteristics of the all-vanadium flow battery,and then aiming at the best economic benefit and the minimum amount of wind abandonment,considering the constraints of real-time wind energy output,power quality and other constraints,calculate the energy storage system of the wind farm The capacity and the gray wolf optimization algorithm were used to solve the example,and it was concluded that for the 153 MW wind farm in this paper,the energy storage system with a capacity of 13.67 MWh can achieve a total revenue of24.504 million yuan and an abandoned wind volume of 0.Finally,the simulation model of the centralized energy storage combined system of the wind power system is established,and the wind power data of the wind power plant in June are used for simulation analysis to verify the energy storage's smoothing effect,and the integrated system power fluctuation rate after the energy storage is significantly smooth.This paper aims to provide new ideas for the smooth operation of wind power by predicting the output power of the wind power system and optimizing the energy storage capacity of the wind power plant in a certain area,so as to minimize the "windabandonment" phenomenon while ensuring maximum economic returns.