基于ARIMA模型的广州市年用电量预测  

Prediction of Guangzhou′s Annual Electricity Consumption Based on ARIMA Model

作　　者：贾朝勇 潘玉荣 夏福全 JIA Chao-yong;PAN Yu-rong;XIA Fu-quan(School of Science,Bengbu University,Bengbu,233030,Anhui)

机构地区：[1]蚌埠学院理学院

出　　处：《蚌埠学院学报》2019年第5期72-75,共4页Journal of Bengbu University

基　　金：安徽省高等学校自然科学研究重点项目(KJ2017A568);安徽省高等学校自然科学一般项目(113052015KJ06);安徽省质量工程项目(2017zhkt323);蚌埠学院科研项目(2018CXY045);蚌埠学院教研项目(2018JYXML8)

摘　　要：将ARIMA模型运用到1990年至2016年广州市年用电量序列中,利用单位根检验确定模型中的差分次数d为1,并通过时间序列相关图对模型中的参数p,q进行初步识别,最后根据赤池信息准则等方法确定最优的年用电量预测模型为ARIMA(1,1,2)。应用ARIMA(1,1,2)对未来几年广州市年用电量进行预测,研究表明预测值与实际值的误差较小,说明该模型预测效果较好。It applied the ARIMA model to the annual electricity consumption time series of Guangzhou from1990 to2016 in this paper,and used the unit root test to determine the difference order d of the model is1,identified preliminarily the parameters p,q in the model by time series correlation graph.Finally based on Akaike Information Criterion and other methods,it built the optimal annual electricity consumption forecasting model which is the ARIMA(1,1,2). ARIMA(1,1,2) was used to forecast the annual electricity consumption in Guangzhou in the next few years. The study showed that the error between the predicted value and the actual value is small,which stated that the model has a good prediction effect.

关 键 词：年用电量 EVIEWS软件 ARIMA模型 预测 

分 类 号：TM715[电气工程—电力系统及自动化]