A Nonparametric Bayesian Network Model for Predicting Corrosion Depth on Buried Pipelines

Abstract

The present study develops a nonparametric Bayesian network (NPBN) model to predict the corrosion depth on buried pipelines using the pipeline age and local soil properties. The dependence structure and parameters of the NPBN model are extracted from Velázquez’s dataset, which consists of 250 samples of corrosion depths, pipeline age, and such local soil properties as the water content, redox potential, and pH value. The NPBN models the joint distribution of the corrosion depth, pipeline age, and local soil parameters by a Gaussian copula. The five-fold cross-validation is used to examine the predictive capability of the developed NPBN model. The results indicate that the predicted mean values of corrosion depths in general agree well with the corresponding field measurements, and more than 95% of the field-measured depths are within the 5 to 95 percentile range of the predicted distribution for the corrosion depth. The NPBN and the associated model mining method provide an effective data-driven approach to develop predictive models of corrosion depths using soil parameters as predictors. The developed NPBN will benefit the corrosion management of pipelines for which direct inspections are infeasible.

Data Augmentation of a Corrosion Dataset for Defect Growth Prediction of Pipelines Using Conditional Tabular Generative Adversarial Networks

“…The Velázquez dataset was obtained from 259 underground pipelines in operation in southern Mexico over the course of three years. The Velázquez’s dataset has extensive information on various kinds of coating and cathodic protection, in contrast to the NIST dataset, which only includes data from uncoated pipes without cathodic protection [ 14 ]. Thus, Velázquez’s dataset is regarded as a more accurate representation of actual pipeline corrosion and is highly favored by scholars.…”

Reliability assessment method for tank bottom plates based on hierarchical Bayesian corrosion growth model

“…For example, Maes et al 15 proposed a hierarchical Bayesian framework for corrosion defect expansion considering the uncertainties in the pipeline deterioration process, and combined hierarchical Bayesian analysis with likelihood estimation methods for remaining life estimation. Zhou et al [16][17][18][19][20] used power law process, gamma process, nonparametric Bayesian network model, and dynamic Bayesian network model to simulate the corrosion defect growth, respectively, and the results showed that these models can accurately and effectively predict the corrosion depth growth of defects. Aulia et al 21,22 developed a dynamic Bayesian network model to forecast the future corrosion state for the corrosion of submarine pipelines, which provides an effective alternative for the residual lifetime prediction of corroded pipelines.…”

Influence of the Corrosion Anomaly Class on Predictive Accuracy of Burst Capacity Models for Corroded Pipelines

“…Corrosion on buried pipelines is largely influenced by properties of the surrounding soils such as the pH value, soil resistivity, water content and dissolved chloride. Extensive research has been reported in the literature to predict the severity of corrosion on pipeline using soil parameters as predictors [2][3][4]. In practice, pipeline engineers carry out the fitness-for-service (FFS) assessment to evaluate the structural integrity of corroded pipelines and then determine necessary, if any, mitigation actions.…”