Wind speed is affected by many factors and has such features as randomness and nonstationarity,which makes it quite diffi-cult for wind power to be connected to the power grid.Accurate wind speed prediction has great research significance for wind power generation.A wind speed prediction model is proposed by combining the variational mode decomposition algorithm with the whale opti-mization algorithm to optimize the echo state network model.Firstly,the wind speed sequence is decomposed into multiple components by using the variational modal decomposition algorithm to reduce the coupling between the internal signals of wind speed and reduce the difficulty of modeling.Then the corresponding echo state network prediction models are established for these components.Targeting at the problem that the performance of the echo state network model is greatly affected by the parameters of the reservoir,the whale optimi-zation algorithm is used to optimize the parameters of the reservoir.The final predicted value of wind speed is obtained by adding the predicted values of each component after decomposition.Finally,taking the actual collected short-term wind speed data as the research object,the comparison and analysis with the other four prediction models show that the proposed wind speed prediction model has higher prediction accuracy and can better predict the wind speed change trend.
关键词
风速/预测/变分模态分解/回声状态网络/鲸鱼优化算法
Key words
wind speed/prediction/variational mode decomposition/echo state network/whale optimization algorithm
维普
万方数据