摘要
风速预测在风能开发和利用中起着关键作用,然而风速序列往往存在强波动性和非平稳性的特征。为了提高风速预测的精度,文章提出变分模态分解(variational mode decomposition,VMD)和神经网络相结合的风速组合预测模型。首先采用变分模态分解将风速序列分解为若干不同频率的子序列;其次计算各子序列的样本熵(sample entropy,SE)以量化复杂程度,引入熵值法建立神经网络组合预测模型,对复杂度较高的分量采用神经网络组合预测模型,其余分量采用支持向量机(support vector machine,SVM)模型进行预测;最后将各分量预测结果运用BP神经网络拟合得到最终预测值。针对北京测风塔实测样本进行建模预测,验证所提出预测模型的可行性,并与6种不同风速预测组合模型开展对比分析,证明所提出的预测模型具有更好的鲁棒性和预测精度。
Wind speed prediction plays a key role in the development and utilization of wind energy.Wind speed series is usually characterized by strong fluctuation and non-stationarity.In order to improve the accuracy of wind speed prediction,a combination prediction model based on variational mode decomposition(VMD)and neural network is proposed.The wind speed series is decomposed into several intrinsic mode functions(IMFs)based on VMD.Then,the sample entropy(SE)of each IMF is calculated to investigate the complexity of the signal.The neural network combination prediction model is proposed based on the SE.The IMF components with higher complexity are predicted using the neural network combination prediction model.The signals with lower complexity are obtained by the support vector machine(SVM)model.The final predicted wind speed series is fitted by the analysis results of each component using back propagation(BP)neural network.The feasibility of the proposed wind speed prediction model is verified on the basis of wind speed data measured from the Beijing meteorological tower.Comparisons are made between the proposed method and six different wind speed combination prediction models.The results show that the proposed method has better robustness and higher prediction accuracy.
作者
郅伦海
訾勇
徐凯
ZHI Lunhai;ZI Yong;XU Kai(School of Civil and Hydraulic Engineering,Hefei University of Technology,Hefei 230009,China)
出处
《合肥工业大学学报（自然科学版）》
CAS
北大核心
2022年第11期1505-1510,1584,共7页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金资助项目(51978230)
安徽省自然科学基金杰出青年科学基金资助项目(2108085J29)
中央高校基本科研业务费专项资金资助项目(PA2019GDZC0094)。
关键词
变分模态分解(VMD)
支持向量机(SVM)
样本熵(SE)
BP神经网络
组合预测模型
风速预测
variational mode decomposition(VMD)
support vector machine(SVM)
sample entropy(SE)
back propagation(BP)neural network
combination prediction model
wind speed prediction
