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

Abstract

For effectively predicting the tank failure time and analyzing the key elements influencing the reliability of corroded bottom plates, this article presents a model for calculating the reliability of corroded tank bottom plates based on a hierarchical Bayesian corrosion growth model. Firstly, the growth of corrosion defect depth is expressed by the gamma process, and the hierarchical Bayesian model is used to calculate the corrosion depth growth. After that, the reliability calculation model of the corroded tank base plate is established by combining the results of the hierarchical Bayesian model with the stress-strength interference theory, and the three uncertain factors of the base plate thickness, radius, and yield strength are considered in the model. Finally, the reliability assessment and sensitivity analysis of corroded bottom plate are carried out. The results show that the proposed reliability calculation model can provide more accurate failure state prediction results than the reliability calculation model which only considers the influence of corrosion depth, and can provide reference for reducing the failure rate of tank floor and reasonably formulating the maintenance plan of tank floor.