中长期负荷预测的多因素模型研究

Research of Multi-Factor Model for Medium and Long-Term Load Forecasting

为提高中长期负荷预测精度,提出了一种新的预测模型及参数计算方法。首先将负荷分解成基础负荷(Base Load,BL)、气象负荷(Temperature Load,TL)和不确定性负荷(Uncertain Load,UL),然后从气象、经济等因素分析了对电力需求的影响,特别是不确定因素使得负荷预测难度加大。基于回归分析,建立了BL之于地区生产总值、TL之于居民消费总额和极端气温持续时间以及UL之于大用户用电量的解析关系;继而将3类负荷相组合,提出多因素模型;采用最小二乘法原理提出参数的配置方法。实例表明,本文模型能够准确刻画负荷的生长趋势,具有较好的拟合性能,其预测精度能够满足工程应用的要求。

A novel model and parameter calculation method are proposed in order to improve the accuracy and fitting performance of the medium and long-term load forecasting.The load is decomposed to base load(BL),temperature load(TL)and uncertain load(UL).Several factors,such as meteorology and economy,affect the power demand.Especially,the uncertain factors make the load forecasting more difficult.Based on the regression analysis,BL is analyzed by GDP,TL by consumption and duration of the extreme temperature,and UL by electricity consumption of large user.According to the three kinds of load,the multi-factor model(MFM)is proposed.Then the way of those parameter configuration is presented by least square method.The example shows that the growth of load is described accurately by MFM,which have great fitting performance.And the prediction accuracy of the model meets some requirement at engineering application.