基于产能转移指标的中长期电力负荷组合预测方法  

Combining Method for Medium-Long Term Power Load Forecasting Based on Capacity Transfer Index

作　　者：赵阳[1] 蒋传文[1] 金尧[1] 谭胜敏 王璟[2] 

机构地区：[1]上海交通大学电子信息与电气工程学院,上海200240 [2]国家电网河南省电力公司经济技术研究院,河南郑州450000

出　　处：《电器与能效管理技术》2015年第20期55-60,共6页Electrical & Energy Management Technology

基　　金：国家电网公司科技项目(5217L0140009)

摘　　要：为提高产能转移产业的中长期电力负荷预测精度,提出产能转移表征指标体系,量化分析产能转移现象,建立了基于该指标体系的主成分回归模型,作为产能转移产业的负荷预测基础模型。最后,提出了加权百分方误差的概念,应用于电力负荷预测精度分析,并以其构建了电力系统中长期负荷组合预测模型。算例结果表明,该方法对于产能转移产业具有较强的实用性和较高的预测精度。In order to improve the accuracy of medium-long term load forecasting of capacity-transferred industry, a capacity transfer representation index system was presented, it can make quantitative analysis of the phenomenon of capacity transfer. Then, a principal component regression analysis based on the capacity transfer representation index system was built as the basic load forecasting model of capaclty-transferred industry. Finally, the concept of weighted squared percentage error （WSPE） was proposed in order to be applied to the power load forecasting accuracy analysis. And a combining model for medium-long term power load forecasting based on WSPE was built. An actual example proves the strong practicality and high accuracy of the method for capacity-transferred industry.

关 键 词：中长期负荷预测 产能转移指标 主成分回归 加权百分方误差 组合预测模型 

分 类 号：TM714[电气工程—电力系统及自动化]