【摘要】 将混沌理论和神经网络相结合,建立了径流预报的混沌神经网络模型.利用混沌理论的相空间重构技术计算饱和嵌入维数,将其作为神经网络的输入层神经元个数;根据模型预测步长确定输出层神经元个数.对黄河干流三门峡站的日流量时间序列进行了模拟和预报,取到了较好的预报效果,为河川径流的预报工作提供了新方法.更多还原

【Abstract】 The chaos theory and artificial neural networks(ANNs) are combined to establish a chaotic ANNs model for runoff forecast.The chaos theory′s phase space reconstruction is used to calculate saturated embedding dimensions which are served as the number of input layer neurons of ANNs.Then the number of output layer neurons is determined according to prediction step size of the model.The daily discharge time series of Sanmenxia station in the main stream of Yellow River is simulated and forecasted,from which the satisfactory results are obtained.The research provides a new way for rivers runoff forecast.更多还原