Short-term Wind Speed Prediction Method For Wind Farms Based On Cluster Analysis

The increasing demand for energy under the current world situation has brought profound changes to the development of our country’s energy sector.To ensure the smooth realization of the ambitious goal of carbon neutrality by 2060,the key is to vigorously develop renewable energy.As an important part of renewable energy,wind energy has received extensive attention because of its clean and pollution-free characteristics.However,when wind energy is connected to the power grid,the stability of power grid will be seriously affected due to the intermittent and fluctuating characteristics of itself.One of the keys to solving the above problems is to accurately predict the wind speed of the wind turbines in the wind farm.However,in the existing methods for predicting wind turbines,the prediction result obtained by selecting a wind turbine as a representative wind turbine for prediction is too large to be directly used as the prediction result of the entire wind farm;the process of modeling and forecasting for wind turbines is too complicated and consumes many unnecessary calculation examples.How to predict the wind speed in wind farms efficiently and accurately has become an urgent problem.In order to solve the above problems,this paper proposes a short-term wind speed prediction method for wind farms based on cluster analysis.The method is based on analytic hierarchy process,entropy method,genetic algorithm,K-means clustering,density peaks clustering and long short-term memory.The method firstly preprocesses the data with optimal comprehensive weighting and dimensionality reduction,which improves the problem that the traditional clustering algorithm does not take into account the unequal importance of the various indicators of the data in the clustering process;secondly,the clustering algorithm is used to classify the numerous and scattered wind turbines in the wind farm,and the wind turbines in each category are regarded as a whole;finally,by considering the simultaneous influence of historical data and current data,a deep learning neural network based on the long short-term memory model is constructed,and it is used to iteratively predict the wind speed in the section for a period of time in the future.This paper takes the measured data of a wind farm in northern China as an example to comprehensively compare and evaluate the prediction performance of the proposed model according to the two error indicators: average absolute percentage error and root mean square error.