Wind Power Prediction System Based On MEEMD-KELM And NCM

With the development of industry and the promotion of national sustainable development policy,wind energy as a clean energy has been developed rapidly.China's wind power installed capacity continues to grow,and wind power generation technology is becoming increasingly perfect.However,wind power has the characteristics of volatility,intermittence and non-stationarity.Large-scale wind turbine grid-connected brings challenges to the stable operation of power system.Improving the accuracy of the prediction system can improve the management of wind farms,reduce the phenomenon of "abandoning wind and limiting electricity",and contribute to the safe dispatch of power grids.Therefore,it is necessary to study wind power prediction methods.In this paper,the ultra-short-term prediction and short-term prediction of wind power are mainly studied.The specific contents are as follows:(1)Summarize the development status,market and existing problems of wind power generation at home and abroad,summarize the research status of wind power forecasting technology,and analyze and summarize the engineering significance and research significance of wind power forecasting methods.(2)Analyse the data of wind farm wind measurement and weather forecast,correct and fill the data of wind farm,and analyze the influence of data characteristics on wind power and mapping relationship.(3)In view of the shortcomings of the traditional mode decomposition method,MEEMD method is used to process wind power time series signals,and a new ultra-short-term prediction method based on MEEMD-KELM is proposed.First,MEEMD method is used to decompose the original power time series,and then KELM model is used to predict.The simulation results show that the prediction accuracy of this method is better.(4)A short-term combined forecasting method of wind power based on mid-intelligence clustering and IOWA operator is proposed.Firstly,the numerical weather information data are divided into several weather types by using the Chinese Intelligence Cluster.According to different weather types,GA-BP,GA-ELM and GA-WNN models are established respectively.The IOWA operator is used to assign weights to the three models,and the optimal combination forecasting model is obtained.The simulation results show that this method has higher accuracy than traditional methods.