B－P网络用于环境质量分类研究

用全国２５个湖泊的水质指标资料作为训练样本，应用人工神经网络Ｂ－Ｐ算法，分别建立了３参数和４参数的水质分类模型。并用所建立的模型对６个湖泊水质进行分类检验效果。结果表明Ｂ－Ｐ模型应用于环境质量分类是可行的。且由于Ｂ－Ｐ网络有自学习、自适应的能力，因而与其它的环境质量分类法相比具有优越性。

Study on Environmental Quality Classification Based on B- P Neural Network

Based on the back-propagation (B-P) algorithm of neural network, the classification models of water quality with 3 and 4 parameters were developed by taking the data on water quality indexes of 25 lakes as training samples. The verification of models for water quality classification of 6 lakes indecated that B-P neural network possessed not only good practicability, but also superiority over other classification methods of environmental quality because of its self-learning and self-adaptabity.