基于EWT-KELM方法短期风电功率组合预测.docVIP

第44卷 第504期 电测与仪表 Vol.44 No.504 2017年 第12期 Electrical Measurement Instrumentation Dec.2017 PAGE 1 - 基于EWT-KELM方法的短期风电功率组合预测 卓泽赢1，曹茜2，李青3 （1. 国网新疆电力有限公司乌鲁木齐供电公司，乌鲁木齐 830001；2.国网新疆电力有限公司经济技术研究院，乌鲁木齐 830001；3. 国网新疆电力有限公司电力科学研究院，乌鲁木齐 830001） 摘要：针对短期风电功率预测，提出一种基于经验小波变换(empirical wavelet transform, EWT)预处理的核极限学习机(extreme learning machine with kernels, KELM)组合预测方法。首先采用EWT对风电场实测风速数据进行自适应分解并提取具有傅立叶紧支撑的模态信号分量，针对每个分量分别构建KELM预测模型，最后对各个预测模型的输出进行叠加得到风速预测值并根据风电场风功特性曲线可得对应风电功率预测值，为验证本文方法的有效性，将其应用于国内某风电场的短期风电功率预测中，在同等条件下，与KELM方法、极限学习机(extreme learning machine,ELM)方法、支持向量机(support vector machine,SVM)方法以及BP(back propagation neural network)方法对比，实验结果表明，本文所提方法具有较好的预测精度和应用潜力。 关键词：经验小波变换；核极限学习机；组合预测；风电功率；风速-功率特性曲线 中图分类号：TM641 文献标识码：A 文章编号：1001-1390（2019）00-0000-00 Wind power short-term combined forecasting based on EWT-KELM empirical wavelet transform and extreme learning machine with kernels method Zhuo Zeying1, Cao Qian2, Li Qing3 (1. Urumqi Power Supply Company of State Grid Xinjiang Electric Power Co., Ltd. Urumqi Power Supply Company, Urumqi 830001, China. 2. Economics and Technology Research Institute of State Grid Xinjiang Electric Power Co., Ltd. Economics and Technology Research Institute, Urumqi 830001, China. 3. Electric Power Research Institute of State Grid Xinjiang Electric Power Co., Ltd. Electric Power Research Institute, Urumqi 830001, China) Abstract: Aiming at short-term wind power forecasting, a kind of combining forecasting method for short-term wind power based on empirical wavelet transform (EWT) and extreme learning machine with kernels (KELM) is proposed in this paper. firstlyFirstly, the EWT method is used to decompose the wind speed data and extract the different modes which have a compact support Fourier spectrum. Secondly, different KELM forecasting models are constructed for the sub-sequences formed by the each m