基于MATLAB的BP神经网络模型在漠阳江水质评价中的应用研究

为提高水质评价的准确性,基于MATLAB及人工神经网络理论,采用误差反向传播的BP算法建立漠阳江水质评价模型,充分利用神经网络的非线性映射特性,取7项常规地表水水质评价指标对漠阳江水质进行评价,并将BP神经网络评价结果与单因子评价法及综合指数法的评价结果进行比较,网络运行结果表明一致效果良好.同时较传统的水质评价方法,该网络具有较高的识别精度,提高了水质评价等级的准确性,使评价的结果更具有科学性.

Water Quality Evaluation for Moyang River Based on BP Neural Network with MATLAB

In order to improve the accuracy of water quality evaluation, this paper makes a water quality evaluation model for Moyang River using BP algorithm of error back propagation based on reviewing the history of MATLAB and artificial neural network. According to the nonlinear mapping characteristics of the BP neural network, water evaluation were carried out by 7 usual surface water quality evaluation indexes. In additon, it also analyzes the discrepancy between the BP neural network based on MTALAB and other traditional assessing methods, such as the single factor evaluation method and comprehensive index method. The result shows good optimizing effecting correspinding to the above two traditional methods. Meanwhile, the BP neural network based on MATLAB makes the evaluation result more scientific.