Prediction of external corrosion rate of buried pipeline based on CS-SVM model

In order to predict the corrosion rate of buried pipeline, this paper firstly analyzes the main influencing factors of the corrosion rate of buried pipeline. On this basis, combined with 74 groups of measured data of external pipeline corrosion in a certain area in western China, 62 groups of them were used to establish the CS-SVM model, and the remaining 12 groups of data were predicted. Meanwhile, the prediction accuracy and calculation time of CS-SVM, PSO-SVM, GA-SVM, LS-SVM four models were compared. The results show that CS-SVM has the highest prediction accuracy, the shortest calculation time and strong applicability, which can be popularized and applied.