基于Census X12-SARIMA模型的中长期负荷预测  

Medium and Long-term Load Forecasting Based on Census X12-SARIMA Model

作　　者：乔占俊[1] 

机构地区：[1]华北科技学院机电工程学院,北京101601

出　　处：《电力系统及其自动化学报》2014年第1期34-38,共5页Proceedings of the CSU-EPSA

基　　金：中央高校基本科研业务费资助项目(3142013065)

摘　　要：中长期区域负荷时间序列具有明显的循环性和季节周期性等非平稳特点,预测难度较大。尝试应用SARIMA模型处理具有季节周期性的非平稳负荷时间序列,同时应用Census X12季节调整方法将呈明显趋势循环性、季节周期性的区域负荷时间序列分解成具有实际经济含义的趋势循环要素、季节要素、不规则要素并进行中长期区域负荷的分析与预测。通过在苏州地区115个月的负荷实证检验,结果表明Census X12-SARIMA季节调整模型及方法在中长期区域负荷的预测中有效。Mid-long term regional power load series manifests the obviously trend of circulation and seasonal cycle.Handling time series with seasonal periodic feature,SARIMA model have unique advantage.In addition,the time series can be decomposed into trend circulation element,seasonal element and irregular element based on the SARIMA model and Census X12 season adjustment method.This paper attempts to establish a Census X12-SARIMA season adjustment model for medium and mid-long term regional power load analysis and prediction.Through empirical test for 115 months＇ load of Suzhou area,12 months＇ load from August 2012 to July 2013 is predicted.The results confirm that Census X12-SARIMA model is effective in mid-long term regional power load analysis and prediction.

关 键 词：负荷预测 区域负荷 时间序列 季节调整 模型 

分 类 号：TM715[电气工程—电力系统及自动化] TM743