Research On The Wind Speed And Wind Power Forecasting Bansed On The Actual Measured Data

With the massive consumption of fossil energy and the aggravation of the worldwide energy crisis,as a new type of clean energy utilization,wind power has been attracted more and more attention from all over the world.Due to the characteristics of randomness and volatility of wind speed,leading it to be a kind of intermittent energy source.Therefore,it will bring new challenges to the safe and stable operation of electric power system in the process of integration of large-scale wind turbines.Simultaneously,it will restrict the further development of wind power.But,the accurate prediction of wind power provides an effective way to solve the above problems.While wind speed is easily affected by meteorological factors and its dimension is high.So it is difficult to predict and analyze directly.Consequently,it is necessary to screen out the optimal attribute samples.In this paper,the maximum correlation minimum redundancy algorithm is applied to select the attributes set which has the greatest correlation with wind speed as input variables of forecasting model.The degree of relevance between different characteristics of wind speed is different.If treat them equally,it can't be able to highlight the role of high correlation attributes.Then the Pearson correlation coefficient method is used to weigh the wind speed attributes.In order to verify the accuracy of the prediction model,the mean average percentage error and the root mean square error are used as the evaluation index of the prediction model.In order to further improve the accuracy of prediction model,taking the characteristics of wind speed as clustering index,optimal Fuzzy C mean clustering algorithm and K-means clustering algorithm were used to cluster the training samples.According to the clustering results,the optimized ELM networks are used to predict the wind speed for each category training samples.And the combined wind speed prediction model is formed.Experimental simulations are carried out based on the measured data of American Wind Energy Technology Center to verify the feasibility and accuracy of the new method.Owing to the existence of certain errors in the standard wind speed-power characteristic curve provided by manufacturer,in order to further improve the accuracy of the wind power model,the wind speed-power characteristic curve based on the actual measured data is built.On this basis,the FCM clustering algorithm optimized by Genetic simulated annealing algorithm(GA+SA)is used to cluster and divide wind turbines,and the wind power equivalent model is established.Finally,taking a wind farm as an example to verify the simulation results based on the MATLAB 2015 a platform.