Abstract: In order to accurately predict the maximum corrosion depth of the pipe, the influencing factors of corrosion and the maximum corrosion depth of pipe was simulated by using support vector regression (SVR).A new algorithm, Multi-Objective Slime Mould Algorithm (MOSMA)was used to simplify the hyperparameter tuning of SVR, and an efficient intelligent hybrid model MOSMA-SVR was built to predict the maximum pitting corrosion depth of oil and gas pipes.The prediction results were compared with WOA-SVR, GWO-SVR and AOA-SVR hybrid algorithm models.The results show that the MSE of MOSMA-SVR is only 0.002 in the test set, the MAPE is 3.659%, and the R² of the model is 0.999.All of the evaluation values are better than other hybrid algorithm models, which means that the MOSMA-SVR has higher precision and better generalization, which can provide reference for pipeline corrosion protection decision.