基于分解去噪和LSSVM的短期风速预测

Short-term wind speed prediction based on multi-objective optimization and error correction

随着风能在电力系统运行中的重要性不断加强,准确可靠的风速预测可以有效提高电网运行的稳定性,提高电网经济效益。提出了一种分解去噪、智能算法优化LSSVM的短期风速混合预测模型,首先对初始风速数据进行变分模态分解(VMD),然后利用样本熵(SE)评估各子序列的复杂程度,采用奇异谱分析(SSA)对最无序子序列进行降噪处理;一种改进的灰狼优化算法(IGWO)优化LSSVM的关键参数,提升了预测精度;最后将所有子序列的预测值叠加得到最终预测结果,以华中某风电场实际运行数据进行算例分析,结果表明该模型性能优于其他比较模型,在风速预测的实际应用中具有很大潜力。

With the increasing importance of wind energy in power system operation,accurate and reliable wind speed prediction can effectively improve the stability of power grid operation and improve the economic benefits of power grid.In this paper,a short-term wind speed hybrid prediction model based on decomposition、denoising and intelligent algo-rithm optimization LSSVM is proposed.First,the initial wind speed data were decomposed by variational mode(VMD),then the complexity of each subsequence was evaluated by using sample entropy(SE),and the most disor-dered subsequence was de-noised by singularity spectrum analysis(SSA).An improved grey wolf optimization algo-rithm(IGWO)optimizes the key parameters of the LSSVM and improves the prediction accuracy.Finally,the predic-ted values of all the sub-sequences are superimposed to get the final predicted results.Based on the actual operation data of a wind farm in central China,the results show that the performance of this model is better than other compara-tive models,and it has great potential in the practical application of wind speed prediction.