计及熵权指标及关联度排序的风电历史数据挖掘  

Study on mining in the historical data of wind power based on entropy-weight index and correlation sorting

作　　者：史坤鹏[1,2] 赵伟[1] 李婷 刘梦华[1] 王泽一[2] 

机构地区：[1]清华大学电机系,北京100084 [2]国网吉林省电力有限公司,长春130021 [3]吉林省加禧电力工程技术有限公司,长春130028

出　　处：《电测与仪表》2017年第4期1-5,99,共6页Electrical Measurement & Instrumentation

基　　金：国家科技支撑计划项目(2015BAA01B01)

摘　　要：预测风电功率对优化发电调度计划、促进风电消纳均具有重要意义。通过与供电负荷历史数据进行日间波动特性及其概率分布规律的对比研究,证明风电功率日间波动曲线在同比、环比方面均无明显规律可循,其幅频特性更适合以右偏态分布表征,但这会增大预测难度。为解决海量风电功率历史数据的有用信息挖掘问题,提出了一种基于熵权指标和关联度排序的亲密样本筛选方法,并将其应用于常见的几种短期风电功率预测模型。对北方某省实测数据的分析表明,所提出方法在提高风电功率预测准确度和计算效率方面均具显著效果。Wind power forecasting has important significance to optimize the power dispatching plan,and promote wind power acceptance. By comparing with the historical data of load concerning the daily fluctuation characteristics and the probability distribution,the volatility of the day data of wind power has no obvious law to follow on year-onyear and week-on-week basis,and its statistics law approximately meets Weibull distribution,which undoubtedly increases the difficulty of wind power prediction technology. In order to solve the problem of useful information mining from massive wind-power historical data,an identification method of the intimate sample based on entropy-weight index and correlation sorting has been proposed,and it is applied to several common models of short-term wind power forecasting. Through case study of measured data in northern province,it shows that the proposed method has significant effects on improving the wind-power prediction accuracy and computational efficiency.

关 键 词：风电预测 概率分布 熵权距离 关联度排序 亲密样本 

分 类 号：TM933[电气工程—电力电子与电力传动]