论文部分内容阅读
利用前馈神经网络模型和自适应调整学习率的反向传播算法,分析处理了嘉陵江流域北碚水文观测站30多年的实测数据,对未来几年含沙量变化的趋势进行了非线性预测,并讨论了隐层神经元个数以及迭代的误差标准等参数的最优选择问题.
Using the feedforward neural network model and back propagation algorithm adaptively adjusting the learning rate, the measured data of Beibei Hydrological Observatory over the past 30 years in Jialing River Basin are analyzed and treated, and the trend of sediment concentration in the next few years is predicted nonlinearly The optimal selection of parameters such as the number of hidden neurons and the error standard of iteration is discussed.