Method for predicting stray current corrosion based on PSO-BP algorithm

The invention belongs to the technical field of stray current prediction, and relates to a method for predicting stray current corrosion based on a PSO-BP algorithm, compared with a prediction method based on a BP neural network, the calculation amount is relatively large, and if the calculation capability is sufficient, the calculation amount is relatively large. By using the technical scheme of the invention, the problems that the BP neural network is weak in global search capability, poor in robustness, easy to fall into local minimum and the like when facing partial data can be solved. Moreover, only the individual extremum data at the first position pbest and the global extremum data at the second position gbest are used in the operation, the pbest can retain the possibility that each particle explores a better global extremum point, and the gbest can retain the found global extremum point, thereby ensuring that the final algorithm can converge. 本发明属于杂散电流预测技术领域，涉及一种基于PSO-BP算法预测杂散电流腐蚀预测的方法，相比于基于BP神经网络的预测方