基于改进ARMA模型的时用水量预测  

Analysis of Prediction for Hourly Water Consumption Based on Improved ARMA Model

作　　者：吉乔伟[1] 毛根海[1] 郑冠军[2] 程伟平[1] 

机构地区：[1]浙江大学建筑工程学院,浙江杭州310027 [2]杭州市自来水总公司,浙江杭州310016

出　　处：《江南大学学报（自然科学版）》2008年第2期216-220,共5页Joural of Jiangnan University (Natural Science Edition)　

基　　金：浙江省科技计划项目(2005C23069)

摘　　要：针对时用水量存在周期性的特点,引入季节因子,采用时变ARMA季节模型对非线性非平稳系统的时用水量进行模拟.以杭州市实测用水量为例,分别采用时不变ARMA季节模型和时变ARMA季节模型进行时用水量预测,结果表明时变ARMA季节模型预测精度相对较高,且预测误差相对稳定,较时不变模型更具有实用性.To solve the. problem where the hourly water consumption contains seasonality, seasonal parameter is proposed. The time -vary ARMA model is used to simulate the hourly water consumption since the water consumption is the nonstationary and nonlinear process. The model is applied to the prediction of water consumption in Hangzhou. The result shows that the prediction based on time-vary ARMA model has the advantage of high precision. Besides, the error of the prediction based on the model is more placidity than that of original model. The time-vary ARMA model proposed in this paper is a feasible and effective model of forecasting hourly water consumption.

关 键 词：时用水量 自回归滑动平均模型 预测 时变模型 

分 类 号：TU991.9[建筑科学—市政工程]