The objective of our project on design of risk-informed autonomous operation is to develop and demonstrate artificial reasoning systems for operator decision support, aided by autonomous control technology, for advanced nuclear reactors. One of its first tasks is the formulation of symptom-based conditional failure probabilities for selected structures, systems, and components (SSCs). This task falls under diagnostic modeling of the problem domain, since its primary goal is to aid plant personnel in deducing the probabilistic performance status of the monitored SSCs and in detecting impending faults/failure. As a bidirectional inference problem, the task of conditional failure probability estimation shall be logically tackled by the Bayesian network (BN) approach, which offers the capability of reasoning under uncertainty and graphical representation emulating the physical behavior of the target SSC. This report provides a comprehensive overview and evaluation of the BN technique and the software tools for handling implementation of BN models, along with the associated knowledge representation and reasoning paradigm that are adopted for use in this milestone. Both actual operational data and subjective expert knowledge can be readily incorporated into the knowledge base of a BN model. The challenges with data availability and collection are highlighted, and the general approach to target SSC identification is presented. Our focus is upon failure-prone and risk-important balance of plant assets—including rotating machinery and high-duty-cycle equipment—especially cases having strong operator involvement. While system analysis and risk assessment shall be further utilized to guide SSC selection when collaborating with utility partners, an example case study is conducted in this report on the failure of a general motor-driven centrifugal pump to demonstrate the usefulness and technical feasibility of the proposed artificial reasoning system using an expert system shell.
Advanced Nuclear Power Program