【作者】 蔡绍宽； 宋洋； 钟炜； 钟登华；

【Author】 Cai Shaokuan Song Yang Zhong Wei Zhong Denghua (School of Civil Engineering Tianjin University,Tianjin 300072,China)

【机构】 天津大学建筑工程学院；

【摘要】 提出了一种基于小波神经网络(WNN)方法构建而成的水库日径流预测模型,该模型通过模糊聚类分析将水库历史径流数据进行分类,按不同类型并分别建立相应的小波神经网络预测模型,应用遗传算法对模型的参数进行优化,提高小波神经网络的预报精度,对某水库日平均采流进行分类预测,表明该方法具有较好的应用价值。更多还原

【Abstract】 This paper presents a daily runoff forecasting method for reservoir by means of wavelet neu- ral networks.The historical runoff data are divided using the fuzzy clustering analysis.The correspond- ing wavelet neural network(WNN)is built and genetic algorithm is used to train the network,so the precision of WNN forecast is enhanced.On the basis of fuzzy clustering analysis,the reservoir daily runoff are forecasted with the corresponding WNN.The forecasting results show that the proposed method will be of great value in application.更多还原