WANN模型的改进及日径流预测中的应用

介绍基于小波分析建立的人工神经网络模型的方法原理，并给出构造模型的一般步骤及关键算法。针对一般BP算法收敛速度慢，易陷入局部极小值的缺陷，受Fletcher—Reeves线性搜索方法的启发．提出基于改进共轭梯度法的BP算法。利用此优化模型对日径流进行模拟与预测，实验表明，基于小波分析的人工神经网络模型在日径流模拟过程中具有很好的仿真能力．训练后的模型用于预测具有较高的精度。

Improvement of WANN Model and Its Application to Daily Runoff Forecasting

The principle of artificial neural network model based on wavelet analysis was introduced, and the general steps and key algorithm of the model were expounded. Dealing with the defects of the steepest descent and slowly converging, easy to immerging in partial minimum frequently, after analyzing the linear hunting method developed by Fletcher and Reeves, the improvement was made of conjugating gradient with BP algorithm to solve the problem. After training, when the model used to simulate and forecast daily runoff, it was shown by numerical experiments that the model has better capability of simulation for the process of daily runoff with higher accuracy.