作　　者： ; ; ;

机构地区： 广东工业大学自动化学院

出　　处： 《太阳能学报》 2018年第11期3277-3285,共9页

摘　　要： 针对风电功率序列非线性、非平稳性等特点,提出一种基于变分模态分解(VMD).排列熵(PE)和混沌布谷鸟搜索算法(CCS)优化相关向量机的短期风电功率预测新方法。为降低风电功率序列非平稳性和减小计算规模,首先采用变分模态分解技术(VMD),将原始风电功率序列分解成一系列不同的子模态,利用排列熵(PE)分析其复杂度并重组得到子序列;然后采用CCS优化后的相关向量机(CCS.RVM)对各子序列进行提前24h预测;最后将预测结果叠加得到最终预测值,并利用某风电场实际采集数据进行仿真验证。结果表明,所提预测模型能有效提高风电功率预测的准确性。 According to the nonlinearity and non-stationarity of wind power series,a novel short-term wind power forecasting approach based on variational mode decomposition(VMD)-permutation entropy and relevance vector machine optimized by chaotic cuckoo search algorithm(CCS-RVM)is proposed in the paper.In order to reduce the influence of non-stationary effects of the wind power series and computational scale,firstly,the wind power series is decomposed into a series of different sub-modes by using variational mode decomposition(VMD).The permutation entropy(PE)value for each sub-mode is used to analyze its complexity.The sub-modes can be recombined to obtain a set of subsequences. Secondly,different prediction models based on relevance vector machine optimized by chaotic cuckoo search algorithm (CCS)are used to predict every subsequence24hours ahead.Finally,these forecasting results of subsequences are combined to obtain final forecasting result.Taking the actual collecting data of a certain wind farm as an example,the simulation results illustrate that the proposed prediction model can effectively improve the wind power forecasting accuracy.

关 键 词： 风电功率 预测模型 变分模态分解 相关向量机 排列熵 混沌布谷鸟搜索算法

领　　域： []