基于云推理的年降水预测模型  

Prediction of Annual Precipitation Based on the Cloud Model

作　　者：高文荣[1] 段春青[2] 陈晓楠[3] 邱林[2] 

机构地区：[1]河海大学水环学院,南京210098 [2]华北水利水电学院水利系,郑州450011 [3]西安理工大学水利水电学院,西安710048

出　　处：《中国农村水利水电》2009年第3期32-35,共4页China Rural Water and Hydropower

基　　金：河南省高校创新人才培养工程项目(HNCX2003-17)

摘　　要：通过分析年降水量时间序列的特性,提出了基于当前趋势以及相邻年份降水量的年降水预测规则。利用云模型描述每条规则,实现不确定性推理。云模型结合了随机不确定性和模糊不确定性,通过云推理实现未来年份年降水预测。通过与传统预测方法相比较,发现云推理模型很擅长从大量数据中挖掘出不确定性知识,有较高的预测精度和推广应用价值。According to the features of precipitation time-series, prediction rules based on the current tend and the neighbor year are suggested. Each rule is described by the cloud model that combines stochastic uncertainty with fuzzy uncertainty and precipitation prediction is realized with uncertain cloud reasoning. Results indicate that annual precipitation prediction with the cloud model is good at excavatinguncertain knowledge and can find more information than traditional methods. In fact, this is a new data mining model for time-series. Any similar problem can be solved better by the method, such as runoff prediction, water demand of city prediction and so on. It is significant to popularize this method.

关 键 词：云理论 云推理 年降水量 

分 类 号：S274.1[农业科学—农业水土工程]