门限自回归模型及其在水文随机模拟中的应用  

Application of Threshold Auto-regressive Model in Hydrologic Stochastic Simulation

作　　者：王文圣[1] 袁鹏[1] 丁晶[1] 邓育仁[1] 

机构地区：[1]四川大学水电学院,四川成都610065

出　　处：《四川水力发电》2001年第B07期47-50,共4页Sichuan Hydropower

基　　金：国家自然科学基金 ( 4 98710 18);高速水力学国家重点实验室开放基金资助项目 (编号 2 0 0 8)

摘　　要：为了客观描述日流量变化的非线性特性 ,将一种非线性时序模型——门限自回归模型引入日流量随机模拟。根据我国金沙江流域屏山站观测资料建立了日流量随机模拟的门限自回归模型。实用性检验结果表明 ,该模型用于模拟日流量过程是可行的 ,成果实用。这种尝试为日流量随机模拟提供了一种考虑日流量非线性变化特性的新模型。In order to characterize the nonlinear behavior of daily flow,a new nonlinear time series model,i.e.threshold auto regressive model has been introduced firstly into daily flow stochastic simulation in this paper.Based on the observed data at Pingshan Station at Jinsha River in China,the threshold auto regressive model has been set up to conduct stochastic simulation of daily flow.The results from model verification have shown that the suggested model is functional and applicable for daily flow simulation.

关 键 词：门限自回归模型 日流量 水文随机模拟 

分 类 号：P333[天文地球—水文科学]