基于ANFIS的水库年径流预报  

Reservoir annual runoff forecast based on ANFIS

作　　者：马细霞[1] 胡铁成[1] 

机构地区：[1]郑州大学环境与水利学院,郑州450001

出　　处：《水力发电学报》2008年第5期33-36,22,共5页Journal of Hydroelectric Engineering

基　　金：河南省自然科学基金资助项目(0411050800);河南省杰出青年科学基金资助项目(512002500)

摘　　要：水库年径流预报是充分利用水资源、真正实现水库优化运行、发挥电站经济效益的有力手段和重要环节。本文在分析以往年径流预报方法优缺点的基础上,采用自适应神经模糊推理系统(adaptive neural fuzzy inference system,ANFIS)建立了年径流预报模型,并将其应用到某水库年径流预报中。实例结果表明,与回归分析法、BP神经网络法、改进的Elman神经网络相比较,该方法计算速度快、泛化能力强、预报精度高,能够有效地应用于水库年径流预报。In order to fully utilize the water resources, really realize the reservoir optimized operation, improve the economic benefit of hydropower stations, it is important to forecast reservoir annual runoff. Based on the analysis of advantages and disadvantages of the traditional methods of annual runoff forecast, a new annual runoff forecast model is established by using adaptive neural fuzzy inference system（ANFIS）. As an example, it is used for reservoir annual runoff forecast. The research result indicates that this method compared with regression analysis BP network and modified Elman neural network has the advantage of high accuracy, strong generalization ability and fast calculation. It can effectively be used in reservoir annual runoff forecast.

关 键 词：水资源 径流预报 ANFIS 人工神经网络 

分 类 号：TV121.4[水利工程—水文学及水资源]