Short-term wind speed combined prediction for wind farms

Abstract

Accurate prediction of short-term wind speed for wind farms will help to reduce the impact of wind power on the grid. In order to improve the prediction accuracy, a combined forecasting method is proposed. Firstly, the Ensemble Empirical Mode Decomposition (EEMD) of the original wind speed sequence is carried out to reduce the interaction between different feature scale sequences. Meanwhile, the Sample Entropy (SE) of each sub-sequence is calculated, and the sequences with similar complexity are merged to improve the prediction efficiency. Then the kernel width and regularization parameters of the Least Squares Support Vector Machine (LSSVM) are optimized by Particle Swarm Optimization (PSO) algorithm. Then the prediction model are used to predict the wind speed of the components, and the results of each component are superimposed, the final wind speed prediction result is obtained and compared with the results of other methods. The simulation results show that the proposed method can improve the prediction accuracy and have practical engineering application value.