Toward risk-informed infrastructure engineering: development of risk-informed in-service inspection methodology

风险知情基础设施工程是一门新兴的工程学科，它试图为“摇摇欲坠的基础设施”提供一个整体和生命周期的解决方案。拟议的研究计划旨在为核电厂（NPP）开发一种先进的风险知情在役检查（RI-ISI）方法，核电厂是一个关键的物理基础设施系统，安全是一个非常重要的问题。先进的RI-ISI方法将概率风险分析中的风险洞察与服务经验、操作条件和其他确定性信息相结合，以优化检查决策。本文将重点研究核电厂组件老化管理的随机退化建模、寿命预测与可靠性分析、检查与维修优化三个关键且相互依存的问题。随机退化建模为随时间变化的可靠性分析和寿命预测提供了可靠的退化模型，从而指导了最优检测位置和维修策略的选择。同时，精心设计的检测程序将为退化建模提供高质量的退化数据。本研究将运用随机过程理论和现代统计方法，为核电厂的RI-ISI计划提供坚实的理论基础。该研究的新颖之处在于，它对与退化相关的所有不确定性进行了彻底的处理，并将其整合到可靠性分析和检验决策中。拟议的RI-ISI方法将使加拿大公用事业公司能够开发更有效和经济的ISI计划，以提高核电站的可靠性和运行安全性。主要发现也可以很容易地转移到其他基础设施系统的生命周期管理，如道路、铁路、桥梁、公共交通、水和废水系统。

Risk-informed infrastructure engineering is a newly emerging engineering discipline that attempts to provide a holistic and life-cycle solution to the "crumbling infrastructure". The proposed research program aims at developing an advanced risk-informed in-service inspection (RI-ISI) methodology for a nuclear power plant (NPP), a key physical infrastructure system that safety is so important an issue. The advanced RI-ISI methodology integrates risk insights from probabilistic risk analysis with service experience, operating conditions, and other deterministic information to optimize inspection decisions. The proposed research will focus on the following three key and interdependent issues: stochastic degradation modeling, life prediction and reliability analysis, and inspection and maintenance optimization for aging management of NPP components. The stochastic degradation modeling provides reliable degradation model for time-dependent reliability analysis and life prediction, which in turn guides the selection of optimal inspection locations and maintenance strategy. Meanwhile, the well-designed inspection program will provide high-quality degradation data for the degradation modeling. Using stochastic process theory and modern statistical methods, the proposed research will provide a solid theoretical basis for RI-ISI programs of NPP. The novelty of the research is that it provides a thorough treatment of all uncertainties associated with the degradation and integrates it into the reliability analysis and the inspection decision-making. The proposed RI-ISI methodology will enable Canadian utilities to develop more efficient and economical ISI programs that improve the reliability and operational safety of the nuclear power plants. The major findings can also readily be transferred to life-cycle management of other infrastructure systems such as roads, rails, bridges, public transits, and water and wastewater systems.