Subset simulation based simplified approach for pipeline with multiple irregular corrosion defects in time-dependent reliability analysis

Abstract

Pipeline system plays an important role in the natural gas and petroleum transportation, and it is widely employed in the engineering application. However, the pipeline system is subjected to corrosion given the industry environment condition, or soil condition when it is buried. In this case, it is important to assess the remaining life of corroded pipeline. Consequently, it is important to predict the failure probability considering the corrosion growth over time and operating pressure. Nevertheless, the prediction of failure probability in corroded pipeline in not an easy task, due to the fact that a realistic corrosion usually has an irregular geometry, especially, when multiple irregular corrosion is involved in the analysis. To simplify the problem, this work presents a simplified procedure for time-dependent reliability analysis to predict the failure probability in pipeline with multiple irregular corrosion defects, considering two failure modes, the burst and leak mode. The approach is based on Subset Simulation and Weighted Depth Difference method, where the multiple irregular corrosion is treated by a discretization procedure and a weighting coefficient is evaluated in every discretization points. Then, this weighting coefficient is introduced into burst pressure assessment, which is employed by burst failure mode limit state function. In this work, the corrosion growth is modelled by power function corrosion model, and initial corrosion depth is generated randomly. The Subset Simulation is employed to evaluate the failure probability, where the Markov Chain Monte Carlo is adopted to evaluate the conditional probability and Metropolis-Hasting algorithm is employed to solve the problem. Finally, several scenarios with single and multiple irregular corrosion defects are analyzed to demonstrate the effectiveness of presented procedure.