12th International Symposium on Symbiotic Nuclear Power ( ISSNP-2026)

Passive residual heat removal (PRHR) systems rely on buoyancy-driven circulation to transfer decay heat without active pumps or external power. In the configuration considered in this study, the cooling water inventory is divided between a PRHR heat exchanger room and a return water tank, which are connected by a riser and a downcomer. Compared with an integral tank, this arrangement modifies the hydrostatic driving head, hydraulic resistance, thermal storage, and temperature distribution of the system, making a system-level thermal-hydraulic assessment necessary.

A RELAP5/MOD4.0 model is developed for the split-tank PRHR system. The nodalization represents the main natural circulation loop, the two cooling-water volumes, the PRHR heat exchanger and its heat structures, steam-side boundary volumes, the makeup line, and the major hydraulic junctions. The model is assessed against previously published steady-state experimental data obtained from the corresponding facility. Comparisons focus on the circulation flow rate, pressure distribution, fluid temperatures at representative elevations, differential pressure across the heat exchanger, and heat-removal rate. Both mean values and temporal trends over the established steady-state interval are considered to evaluate the consistency of the calculated hydrostatic head, circulation behavior, and heat-transfer response.

Sensitivity calculations are performed by varying local hydraulic-resistance coefficients, elevation differences among major components, and selected nodalization parameters. The heat-transfer model is kept unchanged so that the effects of hydraulic layout and model design can be evaluated separately. The analysis examines how these parameters influence the predicted circulation flow rate, pressure redistribution, temperature distribution, and heat-removal performance. The study provides a traceable RELAP5 modeling framework for the split-tank PRHR system and establishes a practical basis for model refinement and design-oriented sensitivity analysis using available steady-state data.