APPLICATION OF BP ARTIFICAL NURAL NETWORK FOR PREDICTION OF ATMOSPHERE CORROSION OF CARBONSTEEL AND LOW ALLOY STEEL
BP神经网络在碳钢及低合金钢大气腐蚀预测中的应用

According to environment data and material corrosion data collected by atmospheric corrosion sites of China and using artificial neural network method,models for prediction of atmospheric corrosion behavior for carbon steel and low alloy steel were proposed to describe the relationship between corrosion rate and weather,corrosive environment factors.The choice of input factors and transform functions of output layer were also discussed.Results showed that the models could predict the corrosion rates of steels in atmosphere.