Short Term Wind Speed Prediction Based on Linear Combination and Error Correction

Abstract

In order to improve the accuracy of wind speed prediction, a model based on linear combination and error correction is proposed. Firstly, sustainability model, grey verhulst model and weibull model are modified to obtain three predictions; secondly, the three predictions are matrix empowering analyzed based on the proximity to the ideal value to gain weights and linearly combined based on weights to gain the combination result; finally, the error between the actual value and the combination value is predicted by ARMA model, to correct the prediction wind speed to improve accuracy. The wind speed prediction results in the future for a wind farm in china are evaluated by RMSE (Root Mean Square Error) and MAPE (Mean Absolute Percentage Error), demonstrating that the proposed model is reasonable and effective.