Research Of The Methods Of Medium And Long-term Wind Power Forecasting

In recent years,the development of wind power is rapid,but because of the wind power's characteristics,such as intermittent,randomness and uncertainty,which brings a huge impact to the stability of the power system,it is necessary to predict wind power to maintain the stability of power system operation.But in the long term wind power prediction,it is crucial to the future development such as accuracy of wind farm location,reserve capacity and storage capacity selection,wind turbine maintenance and wind power bidding.Therefore,it is necessary to do the medium and long term forecast for wind power generation.Due to the lately start of wind power generation in China,there is not a large number of historical data,so it is impossible to establish long-term forecasting model.In consideration of the wind power generation mainly depends on wind speed,it is possible to obtain the historical year of wind speed data and wind turbine power forecasting model can be used to simulate the historical year of the power generation data.Firstly,this paper analyzes the time series of wind power generation in Fujin wind farm 11,discussing the influence of the dimension of the grey GM(1,1)model on the prediction accuracy.On the basis of selecting the best modeling dimension,the model is optimized from the point of view of the model itself and the original data.The result shows that the improved GM(1,1)model,which is based on the background value and the initial value of the model itself,and the incremental processing of the original data,can effectively improve the prediction accuracy of GM(1,1)model.Secondly,this paper uses the principal component analysis(PCA)to extract the principal component of wind speed information,modeling the neural network of wind speed forecasting model based on PCA.Meanwhile,we used the measured data of Fujin wind farm in Heilongjiang in the years of history.In order to verify the necessity of the principal component extraction and the applicability of the wind speed prediction model,we used all the weather factors as the input of neural network prediction model.The simulation results show that the wind speed forecasting model based on principal component analysis can improve the prediction accuracy.Finally,this paper studied the combination forecasting model.We used linear weighting method to make a combination forecasting analysis between prediction model based on Grey Theory and neural network prediction model based on PCA.Compared with other prediction models,the results show that the prediction model based on Grey Theory and the combination model based on PCA have the highest precision.Finally,the this paper studied the combination forecasting method.The prediction model based on Grey Theory and neural network prediction model based on principal component analysis were analyzed by using linear weighted method.Compared with the model,the results showed that the combination model based on the two prediction models has the highest precision.Principal component analysis neural network forecasting method and combination forecasting method to solve China's wind power in the long term prediction of the problem of low accuracy in a certain extent based on providing a solid theoretical foundation and basis for research work for China's future long-term wind power forecasting.This method can solve the problem of low accuracy of wind power forecasting in a certain extent,which provides a solid theoretical basis and basis for the follow-up of the long-term wind power forecasting.