考虑城市化因素的中长期负荷预测模型  

Mid / long-term load forecasting model considering urbanization characteristics

作　　者：李亦言 严正[1] 冯冬涵[1] 

机构地区：[1]上海交通大学电气工程系电力传输与功率变换控制教育部重点实验室,上海200240

出　　处：《电力自动化设备》2016年第4期54-61,共8页Electric Power Automation Equipment

基　　金：国家自然科学基金资助项目(51377103);国家电网公司科技项目(SGHB0000KXJS1400044)~~

摘　　要：随着城市化进程的推进,部分大城市的负荷增长开始缓和,波动性逐渐凸显,加大了中长期负荷预测工作的难度。提出一种基于城市化特性的中长期负荷预测模型。首先分析城市化在人口、经济、地域三方面的特征,提取城市化影响负荷的主要构成要素;然后,利用层次分析法结合专家经验,实现各种城市化要素对电力负荷影响的主观赋权;最后,采用模糊聚类分析法预测多因素影响下的城市负荷。所提模型能够较好地解决负荷的波动性给中长期负荷预测带来的困难,适用于城市化背景下的中长期负荷预测。算例分析表明,相较于时序外推以及弹性系数等常规预测方法,所提方法具有较好的预测精度。Along with the enhancement of urbanization level,the power load of some big cities begins to present a certain trend of saturation and fluctuation,which makes the mid / long-term load forecasting much harder. A mid / long-term load forecasting model based on the urbanization characteristics is proposed. The urbanization characteristics are analyzed in three aspects：population,economic and area,and the main influencing factors of urbanization on power load are extracted. Combined with the experiences of experts,the analytic hierarchy process is applied to set the subjective weighs for the main influencing factors and the fuzzy clustering analysis is adopted to forecast the power load affected by multiple factors. The proposed model can fairly well deal with the fluctuation of power load,suitable for the mid / long-term load forecasting during the urbanization. Case analysis shows that,compared with the traditional methods,such as the time series extrapolation and the elastic coefficient,the proposed method has higher forecasting accuracy.

关 键 词：中长期预测 负荷预测 城市化特性 波动性 层次分析法 模糊聚类分析 

分 类 号：TM715[电气工程—电力系统及自动化]