基于粒计算的电力系统中长期负荷动态聚类预测模型  

Dynamic Clustering Model for Medium- and Long-Term Power Load Forecasting Based on Granular Computing

作　　者：顾洁[1] 杨熠娟 施伟国 

机构地区：[1]上海交通大学电子信息与电气工程学院,上海市闵行区200240 [2]上海市南供电公司,上海市徐汇区201100

出　　处：《电网技术》2009年第20期120-124,共5页Power System Technology

摘　　要：结合聚类分析与时间序列数据挖掘技术,提出了基于粒计算的动态聚类预测模型。该模型有助于消除聚类结果与先验知识之间的主观不协调性,使聚类结果与客观实际相符。基于该模型得到的预测结果是区间值,这降低了预测风险。某地区需电量的预测结果表明,该模型能显著提高预测精度,适用于电力系统中长期负荷预测。In order to improve the accuracy of load forecasting in the context of uncertain environment, a granular computing based dynamic clustering model （DCM）, in which the clustering analysis is integrated with time series data mining, is proposed. The proposed DCM is conductive to eliminate subjective incompatibility between the clustering results and the priori knowledge. The forecasting result by the proposed model is an interval, so the risk of forecasting is reduced. Forecasting results of electricity quantity demand of a certain region show that using the proposed model, the forecasting accuracy can be improved. The proposed model is suitable for medium- and long-term load forecasting.

关 键 词：电力系统 中长期负荷预测 粒计算 粗糙集 动态聚类 时间序列数据挖掘技术 

分 类 号：TM615[电气工程—电力系统及自动化] TM744