Assessment of the EM Plus Advanced BE Code Platform LOCA Analysis Approach With Application in a Chinese 300MW NPP

Abstract

In order to develop licensing LOCA analysis program platform for nuclear power plant (NPP), a relatively realistic technique have been adopted in this paper, which is to satisfy the requirements of conservative evaluation models (EM) in related regulation (10CFR50, Appendix K) by modifying related models or correlations of the best estimate (BE) program RELAP5/MOD3, in an effort to form a licensing LOCA analysis tool. By using this “advanced program platform plus conservative EM” technique, LOCA analysis on the Chinese 300MW NPP, of which LOCA used to be conservatively evaluated, is attempted so that further potential of margins should be achieved. By comparing with Appendix K, 10 models or correlations in RELAP5/MOD3 code are identified, which need to be modified, invoked, evaluated or verified. Modifications or control have been separately introduced on such models or correlations as fission product decay model, correlations for critical heat flux (CHF), post-CHF heat transfer correlations, logic of preventing from returning nucleate boiling and transition boiling heat transfer prior to reflood, discharge model, metal-water reaction model, emergency core cooling (ECC) bypass model and related models for pressurized water reactor (PWR) refill and reflood. With the accomplishment of modification or invoking the above-mentioned models, conservatism verifications by utilizing proper standard curve or separate test data have been carried out. The standard curve or separate effect tests data includes: (1) 1971 fission product decay standard curve; (2) related ORNL THTF test data; (3) Marviken test-22 data; (4) Cathcart oxidation test data; and (5) Westinghouse FLECHT - SEASET test data, etc. Further, for the integral effect test LOFT L2–5, calculations with both original RELAP5/MOD3 program and the program of which some related models modified in compliance with the requirement of 10CFR50 Appendix K have been made. The results indicate that, on the one hand, ability of the original BE program simulating the accident progress is identified; and on the other hand, the modified models are confirmed to be conservative. Moreover, both individual contributions and combined effects of the model modification to conservatism of the integral test analysis are observed and analyzed. To assess effects of the “advanced program platform plus conservative EM” approach on LOCA analysis, finally, calculations with the modified code for both hypothetical small break and large break LOCA of the Chinese 300MWe NPP are conducted. Compared with the calculated results of original RELAP5/MOD3 program and that delivered in Final Safety Analysis Report (FSAR) of the NPP, mutual and integrated effects with different model modifications are investigated. And finally, safety margin achieved by the present calculation is preliminarily discussed.