Limit Pressure Prediction for Steel Pipeline with Corrosion Defects Based on BP Neural Network

Based on the nonlinear finite element method, a numerical model for calculating the limit pressure of a pipeline with corrosion defects was established, and the accuracy of the method was verified using existing experimental data. According to orthogonal test combination, the effects of pipe diameter, wall thickness, depth, length and width of corrosion defects on the limit pressure of the pipeline were comprehensively considered. Based on this, the method for predicting limit pressure of pipeline with corrosion defects based on BP neural network was given. Finally, it was shown that the predicted results were highly accurate compared with the results from the actual operating parameters.