Research on Prediction Model of External Corrosion Rate for Buried Pipelines Based on Response Surface Methodology

Abstract

The factors influencing the external corrosion rate of buried oil pipelines are extremely complex, and the interactions between these factors make accurate prediction of the pipeline corrosion rate challenging.Response surface methodology can analyze the interactions between various influencing factors.Therefore, this study considered four affecting factors: pH value of the soil surrounding buried oil pipelines, Clconcentration, NO3－ concentration and SO42－ concentration.Experiments were designed using response surface methodology to establish a prediction model for pipeline corrosion rates in soil environments.Regression and variance analyses were performed to evaluate the model’s effectiveness and accuracy.The established response surface methodology prediction model was compared with linear regression and PSO-BP prediction models.Results showed that the difference of lack-of-fit terms in the prediction model based on response surface methodology was not significant, and the model itself passed the significance test, demonstrating its reliability.Compared to the other two models, the response surface methodology model had the highest coefficient of determination (R2) of 0.941 8, and the smallest mean absolute error, mean relative error and root mean square error,which were 0.0205,4.53%and 0.027 7,respectively.These results verified the accuracy of the response surface methodology prediction model, showing that it could accurately assess the external corrosion rate of pipelines and provide a reliable basis for developing protection measures and maintenance plans for pipelines.

Keywords

• buried pipelines； corrosion rate； prediction model； influencing factors； response surface methodology