基于最优交集相似日选取的短期母线负荷综合预测  

Short-term Bus Load Integrated Forecasting Based on Selecting Optimal Intersection Similar Days

作　　者：孙谦[1] 姚建刚[1] 赵俊[2] 金敏[1] 毛李帆 毛田[1] 

机构地区：[1]湖南大学电气与信息工程学院,湖南省长沙市410082 [2]湖南省电力公司益阳调度管理所,湖南省益阳市413000 [3]海南省电力公司,海南省海口市571000

出　　处：《中国电机工程学报》2013年第4期126-134,17,共9页Proceedings of the CSEE

基　　金：国家高技术研究发展计划项目(863计划)(2008AA042802);国家重点产业振兴和技术改造项目([2010]2272)~~

摘　　要：准确的短期母线负荷预测是实现节能降耗与调度精细化管理的基础,提出一种基于最优相似日选取的综合预测方法。利用改进的聚类分析算法,得到历史标幺曲线的形状相似集与特征曲线。通过构造反映数据点性质的横向及纵向特征向量矩阵,辨识出坏数据并进行调整。计算日特征相关因素对负荷水平的影响,并将各因素的重要程度加权于模糊目标函数,得到目标日的负荷水平相似集。建立各类形状相似集的判别函数,并将目标日归类。对待预测日的负荷水平与曲线形状相似集,取两者的交集作为相似日选择结果。以该交集中与目标日日期差最小的样本为虚拟预测对象,计算综合预测中各算法的权重。实例分析表明,所提方法可有效改善原始数据的质量,提高母线负荷预测精度。An accurate short-term bus load forecasting is the basis of realizing energy saving,consumption reducing and meticulous management of dispatching.The integrated forecasting method presented was based on the selection of optimal similar days,using improved clustering analysis algorithm to obtain characteristic curves and sets of historical per-unit curves with similar shape.Identifying bad data and making adjustments by constructing vertical and horizontal eigenvectors reflecting the properties of the sample points.In order to obtain a set of curves whose load level is similar to the target date,the effects of daily characteristics factors on the load level was calculated,and fuzzy object function is weighted over each factor importance degree.Establishing discriminant function of all sets with similar shape,and target date was classified.Taking the intersection of load level set and curve shape set as the result of similar day selection.Sample with minimum date interval to target date was selected as virtual predict object,and then the weight of each algorithm in integrated forecasting was calculated.Case study shows that the proposed method can be effective in improving the quality of raw data,as well as the bus load forecasting accuracy.

关 键 词：短期母线负荷预测 坏数据处理 最优交集 相似日选取 综合预测 

分 类 号：TM715[电气工程—电力系统及自动化]