雅鲁藏布江中下游径流预测方法的对比分析  

Comparative analysis of runoff prediction methods in the middle and lower reaches of Yarlung Tsangpo River

作　　者：关静[1,2] 梁川[1,2] 王欣[1,2] 杨瑞祥[1,2] 

机构地区：[1]四川大学水利水电学院,成都610065 [2]水力学与山区河流开发保护国家重点实验室,成都610065

出　　处：《黑龙江大学工程学报》2017年第2期7-13,共7页Journal of Engineering of Heilongjiang University

基　　金：水文学及水资源博士点基金资助项目(20130181110045)

摘　　要：径流预测是水资源研究领域中的一项重要内容,对于区域社会经济规划具有重大的意义。引用雅鲁藏布江中下游羊村站和奴下站1980—2010年的径流资料,采用自回归模型、门限回归模型和人工神经网络模型进行年径流预测,采用自回归模型、季节性自回归模型和季节性人工神经网络模型进行月径流预测。结果表明:年径流预测中,几种预测模型的预测精度均不高,相比较而言,一阶自回归模型和人工神经网络模型的预测结果更优;月径流预测中,季节性自回归模型和季节性人工神经网络模型在枯水期的预测精度均较高,但在丰水期径流预测效果欠佳。Runoff prediction is an important content in the water resources research, and it is of great significance for the regional social-economic planning. Based on the runoff data of two stations of the middle and lower reaches of Yarlung Tsangpo River of 1980-2010, annual runoff prediction and lunar runoff prediction using autoregressive model, artificial neural network model were studied. The results showed that several prediction models in the annual runoff prediction all had not high precision. By contrast, autoregressive model and artificial neural network model were better. Seasonal autoregressive model and seasonal artificial neural network model in the lunar runoff prediction had high precision in low water season, but the satisfied. prediction effect in high water season was not very

关 键 词：径流预测 自回归模型 人工神经网络模型 雅鲁藏布江 

分 类 号：P333[天文地球—水文科学]