Influencing Factors of Crude Oil Corrosion Based on Artificial Neural Network

For the complexity of crude oil corrosion to transportation equipment, influencing factors of crude oil corrosion were cut down from 18 to 9 by means of a screening principle of input nodes in Artificial Neural Networks (ANN). It was found that the network model with 9 nodes in input layer had more prediction accuracy, than that with 18 nodes. The influence law of each selected factor on corrosion rate was obtained by sensitivity analysis.