基于双向差分建模的优化GM(1,1)模型及其在中长期电力负荷预测中的应用  

An optimized GM(1,1) model based on bi-directional difference method and its application in long-term power demand forecasting

作　　者：李红伟[1] 毛文晋[2] 

机构地区：[1]西南石油大学电气信息学院,四川成都610500 [2]西南石油大学经济管理学院,四川成都610500

出　　处：《电力系统保护与控制》2011年第13期53-58,共6页Power System Protection and Control

基　　金：西南石油大学'中青年骨干教师培养计划'资金资助

摘　　要：针对电力负荷预测,论述了一种基于双向差分计算的GM(1,1)建模方法,采用向前差分与向后差分建模的拟合误差平方和最小的原则进行参数估计。为了进一步提高GM(1,1)模型的拟合及预测精度,利用拟合值和原始值误差平方和最小的原则对预测模型的初始值进行了优化,摆脱了原有模型的初始值必经过历史数据中的某一点的束缚。以太原地区电力负荷和重庆地区的农村用电量为例,采用了几种典型GM(1,1)模型进行了拟合和预测对比分析,结果表明所提出的优化GM(1,1)模型的拟合、预测精度更高,适应性更强。Aiming at the forecasting of power demand,this paper presents an optimized GM（1,1） model based on the bi-directional difference method,which adopts minimum sum of fitting error square of forward difference and backward difference modeling to estimate parameters.And in order to further improve the fitting and prediction precisions of GM（1,1） model,the initial value of GM（1,1） prediction model is optimized by using the method of minimum sum of error square of fitting value and initical value,therefore the initial value of the GM（1,1） model is free from necessarily passing a point of original data.To do fitting and forecasting comparative analysis,several typical GM（1,1） models have been built with the case of electrical load in Taiyuan and rural power demand in Chongqing.The results show that the new optimized GM（1,1） model has higher fitting and forecasting precision and stronger adaptability.

关 键 词：中长期电力负荷预测 GM(1 1)模型 双向差分 初始值优化 

分 类 号：TM715[电气工程—电力系统及自动化]