城市用水量的混沌特性与预测  

Chaotic Characters and Forecasting of Urban Water Consumption

作　　者：赵鹏[1] 张宏伟[1] 

机构地区：[1]天津大学环境科学与工程学院,天津300072

出　　处：《中国给水排水》2008年第5期90-93,97,共5页China Water & Wastewater

基　　金：国家自然科学基金资助项目(50578108);天津市科技发展计划项目(033113811)

摘　　要：采用混沌理论对城市用水量时间序列的混沌特性进行了判定。基于最大Lyapunov指数法,提出了城市用水量短期预测的模型以及多步预测的步骤。实例研究表明,基于最大Lya-punov指数法的预测结果明显好于BP神经网络法,在计算得出的最大可预测时间尺度内预测精度较高,而在最大可预测时间尺度外的预测精度大为下降。The chaotic characters of time series of urban water consumption were judged using the chaos theory.A short-term forecasting model on urban water consumption and a multi-step forecasting approach were put forward based on the largest Lyapunov exponent.The example research shows that the forecasted result is better than that based on the artificial neural network.The forecasting precision is higher within maximum predictable time scale than that beyond of it.

关 键 词：最大LYAPUNOV指数 混沌理论 城市用水量 时间序列 短期预测 

分 类 号：TU991.31[建筑科学—市政工程]