基于改进云推理模型的年径流预测  

Annual runoff forecasting based on cloud reasoning model

作　　者：景楠[1,2] 梁川[1,2] 龙贻东 刘述伊[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》2015年第1期1-4,16,共5页Journal of Engineering of Heilongjiang University

基　　金：国家自然科学基金资助项目(41271045)

摘　　要：逐年年径流量之间存在相关关系,且这种相关关系具有不确定性,然而云推理模型能有效将不确定关系定量表示。根据鸭绿江流域荒沟水文站1964-2001年实测年径流系列,利用改进的云推理模型模拟相邻年份年径流关系并进行径流预测,以预测值与实际值的相对误差评价预测效果。计算结果显示,当预见期为3a时,最大相对误差为4.21%,相对误差〈20%的达到100%;预见期为5a时,最大相对误差29.49%,相对误差〈20%的达到80%,表明云推理模型进行年径流预测具有较好的效果。There is a relationship between the year after year annual runoff,and the relationship is uncertain,however,cloud inference model can effectively and quantitatively express the uncertainty relation.According to the Yalu River basin Huanggou hydrological station＇s measured annual runoff series from 1964 to 2001,the improved cloud inference model was used to simulate the adjacent year annual runoff＇s relationship and to predicate the runoff,and evaluate the effect with relative error of predicted value and actual value.The results show,when the forecast period is three years,the maximum relative error is 4.21%,relative error less than 20% is 100%,and when foresee for a period of five years,the maximum relative error is 29.49%,relative error less than 20%is 80%,hence,the annual runoff forecast of cloud inference model has a good effect.

关 键 词：云推理模型 预测 年径流量 鸭绿江流域 

分 类 号：P338[天文地球—水文科学]