短期洪水预报的模糊联想记忆网络模型

A Fuzzy Assocative Memory Model for Short-term Flood Forecasting

径流是水文系统的输出之一，径流的形成是一个复杂的过程，难以用精确的数学模型进行描述，而应采用从定性到定量的综合集成的方法，将专家群体、数据和各种信息与计算机结合起来，把理论与人的经验结合起来对其进行研究。人们对经流形成的认识常常以一种结构性语言知识的形式来表达，模糊联想记忆网络模型具有模拟人类联想记忆的功能，为应用结构性语言知识进行综合推理提供了一条有效的途径，使得在径流预报中模拟应用人的经验成为可能。本文以紫坪埔洪水预报为实例。

People's knowledge of the formulatton of flood is always expressed in a structural linguistic form. Fuzzy associative memory model provide an efficient route to apply the structural linguistic knowleedge to the integrating inference because it can imitate some functios of human brain. So it becomes possibe to apply human experience to the flood forecasting. In this paper, a fuzzy associative memory model is discussed and the practical efficiency is showed.