Development and Improvement of Artificial Neural Network Corrosion Model for Low Carbon Steels

In order to develop the corrosion model of low carbon steels in acid rain,corrosion data of low carbon steels Q235 and 35 in simulated acid rain were obtained by electrochemical work station with the brand of Auto-lab.Based on the data,a three-layer BP neural network model was built and used to forecast the corrosion rate.The cross-verification method was used to analyze and improve the model.The results indicated that the model prediction precision was higher than before,and the model could be used to simulate the corrosion experiment in acid rain.