Abstract The microbiologically influenced corrosion (MIC) is a serious issue that should be considered for effective risk-based integrity management of offshore systems under MIC. This paper presents a proper methodology by using a hybrid Bayesian network (BN) and Markov process to predict the MIC rate, failure probability, and critical failure year of an internally corroded subsea pipeline. The BN model is developed to probabilistically obtain the MIC rate, considering the dynamic non-linearity and interdependency among vital input factors. The effects of the nonlinear interactions of various prominent factors are evaluated, and their degree of influence is explored. The Markov process is employed to predict the failure probability, critical failure year, and the time evolution MIC pit depth distribution using the predicted MIC rate as a transition intensity. The developed model is adaptive and captures the evolving impact of MIC. The proposed integrated methodology is tested on a case study, and the most critical parameters that influence the MIC rate and system failure are identified. The proposed approach would provide an early warning guide for a timely intervention to prevent total failure of corroded subsea pipelines and associated consequences.

中文翻译：

---

微生物腐蚀下海上部件的综合动态失效评估模型

摘要 微生物影响腐蚀（MIC）是MIC下海上系统有效的基于风险的完整性管理应考虑的严重问题。本文提出了一种通过使用混合贝叶斯网络 (BN) 和马尔可夫过程来预测内部腐蚀海底管道的 MIC 率、故障概率和临界故障年份的适当方法。考虑到重要输入因素之间的动态非线性和相互依赖性，开发了 BN 模型以概率地获得 MIC 率。评估了各种突出因素的非线性相互作用的影响，并探讨了它们的影响程度。马尔可夫过程用于预测失效概率、关键失效年份、以及使用预测的 MIC 速率作为过渡强度的时间演化 MIC 凹坑深度分布。开发的模型是自适应的，可以捕捉 MIC 不断变化的影响。建议的集成方法在案例研究中进行了测试，并确定了影响 MIC 率和系统故障的最关键参数。提议的方法将为及时干预提供早期预警指南，以防止腐蚀的海底管道完全失效和相关后果。