A Conservative Preliminary Analysis for Multi-Module PSA of SMRs Considering Dependencies of Shared Systems
Authors
PrimarySEJIN BAEK— Korea Hydro & Nuclear Power Co., Ltd · white.sejin@khnp.co.kr
Co-authorjeonhojun@khnp.co.kr— jeonhojun@khnp.co.kr Edit Profile Co-authorswhwang1221@khnp.co.kr— swhwang1221@khnp.co.kr Edit Profile Some small modular reactor(SMR) designs currently under development worldwide adopt a multi-module configuration in which several reactor modules are integrated within a single plant unit. In such configurations, each module is designed to ensure independent safety performance based on passive safety systems. However, certain plant level systems, such as electrical systems, cooling water systems, and the control room, exist. These shared systems can introduce inter-module dependencies, and therefore a multi-module risk assessment is required.
In the current domestic regulatory framework, multi-unit or multi-module probabilistic safety assessment(PSA) is not included. in addition, considering the design characteristics of passive safety systems, the risk of multi-module SMRs is expected to be very low. However, for innovative SMRs under development in Korea, where the standard design approval(SDA) review is at an early stage, a preliminary quantitative estimation of multi-module risk can provide meaningful insights from both regulatory and technical perspectives. From this perspective, this study propose a conservative approach for evaluating multi-module risk without detailed modeling.
The analysis conditions applied in this study are as follows. (1) Initiating events are selected as loss of offsite power(LOOP) and loss of component cooling water(LOCCW) based on previous multi-unit PSA studies. (2) The internal event model of a single-module innovative SMR is used. (3) Seismic events are excluded from the analysis since the site has not yet been specified and a seismic margin assessment(SMA) would be required. (4) Common cause failures (CCFs) associated with shared systems are conservatively assumed to occur with complete dependency either between two modules or across all modules. Under these conditions, multi-module risk is conservatively estimated using single-module PSA results.
Although uncertainties exist due to the underlying assumptions, the proposed approach can serve as a preliminary evaluation to efficiently identify trends and insights related to multi-module risk while reducing time and economic costs.
✅Status: The abstract has been accepted! This abstract is indicated as Abstract + Presentation only, so no paper is required.
← Check another abstract