基于小波包和支持向量回归的风速预测.pdfVIP

第35 卷 第5 期 电 网 技 术 Vol. 35 No. 5 2011 年5 月 Power System Technology May 2011 文章编号：1000-3673 （2011 ）05-0177-06 中图分类号：TM 712 文献标志码：A 学科代码：470·4051 基于小波包和支持向量回归的风速预测 1 1 1 2 1 陈盼 ，陈皓勇 ，叶荣 ，陈天恩 ，李丹 （1．华南理工大学 电力学院，广东省 广州市 510640 ；2．西北电网公司交易中心，陕西省 西安市 710049 ） Wind Speed Forecasting Based on Combination of Wavelet Packet Analysis With Support Vector Regression 1 1 1 2 1 CHEN Pan , CHEN Haoyong , YE Rong , CHEN Tian’en , LI Dan (1. School of Electric Power, South China University of Technology, Guangzhou 510640, Guangdong Province, China; 2. Northwest China Grid Company Trader Center, Xi’an 710049, Shaanxi Province, China) ABSTRACT: By use of the combination of wavelet package 以控制，迫切需要准确度较高的风电功率预测机制 transform with support vector regression (SVR), the one hour 指导电网调度运行。 to six hour-ahead wind speed forecasting of each ten minute is 风速预测是风电功率预测的基础，其中提前 researched. Firstly, in view of the non-stationary and nonlinear 1~6 h 、间隔为10 min 的预测在电网实时调度中发 features of wind speed, the original wind speed sequence is 挥着重要作用[2-3] 。目前关于风速及风电功率预测的 decomposed into a series of sub-sequences; then these sub-sequences are respectively forecasted by SVR; finally, 研究已在国内外展开。持续法是最早的方法，它以 respective outputs are superposed to obtain final forecasted 简单快捷取胜，但时效较长时预测精度有限，无法 wind speed. Taking two sets of measured data possessing 满足实际需要，常作为评价标准用于与新提出的算 different characteristics in a certain wind farm as application 法比较。文献[4]将时间序列法用