Optimizing System-Level Reliability of Fail-Safe Isolation Valves in SMR Using Fault Tree Analysis

Fail-safe valves are critical components in Small Modular Reactors (SMRs), specifically NuScale type SMR, ensuring safety of module under both normal and abnormal conditions. However, they face two competing risks: Spurious Operation (SO) during normal operation and Demand Failure (DF) during accident scenarios. This study introduces a quantitative framework to optimize system-level reliability of fail-safe valve based passive safety systems, demonstrated through the Containment Vessel Isolation System (CVIS). Using Fault Tree Analysis (FTA) aligned with NuScale's reference design. We evaluate 24 representative configurations that combine key optimization factors, including valve arrangement, actuator diversity, and the presence of an Inadvertent Actuation Block (IAB). Results indicate that with an IAB, series valve arrangements improve DF unreliability when two valves per train are used, while heterogeneous actuators provide limited reduction of SO unreliability. Without an IAB, actuator diversity significantly reduces SO unreliability, although there is a slight penalty of DF unreliability. The proposed framework offers practical guidance for balancing risks of SO and DF unreliability, supporting more resilient and efficient SMR safety system designs.