Evaluation of passively safe boiling water reactor design with integral tests and codes

Jun Yang, Purdue University

Abstract

This thesis study presents the research work aimed to investigate the safety features of a next generation Boiling Water Reactor (BWR) design by conducting the experiments on a scaled down facility and simulating the reactor transients with thermal-hydraulic system analysis code (RELAP5). The Economic Simplified Boiling Water Reactor (ESBWR), designed by General Electric (GE), is a nuclear reactor containing passively safety characteristics. A series of integral tests simulating different types of Loss-Of-Coolant Accidents (LOCAs) were conducted using the Purdue University Multi-Dimensional Integral Test Assembly for ESBWR applications (PUMA-E) facility to support the licensing process to this reactor design. This research work was sponsored by the U.S. Nuclear Regulatory Commission (NRC). Several code models have been developed based on the prototypic power plant design, test facility configuration and experiment conditions. Test transients were simulated on the computer and the simulation results were presented as the counterpart of experiment data. Various kinds of Design Basis Accident (DBA) and Beyond Design Basis Accident (BDBA) were simulated in the PUMA-E facility. Depending on the location and size of the break, four types of LOCA tests were performed, namely, Bottom Drain Line Break (MSLB), Gravity Drain Line Break (GDLB), Feed Water Line Break (FWLB) and Main Steam Line Break (MSLB). In addition to the line break, partly failure of safety system, like ICS, PCCS or GDCS was applied on some tests to address the safety margin. This thesis focused on BDLB and MSLB tests particularly since they represented significantly different break size, location and flow type. Generally, the experiment results showed that passively safety system performance was satisfied under those types of LOCAs. Specifically, Automatic Depressurization System (ADS) and Gravity-Driven Cooling System (GDCS) collaborated to fulfill the function of Emergency Core Cooling System (ECCS), providing adequate supply of coolant to keep the Reactor Pressure Vessel (RPV) coolant level above the Top of the Active Fuel (TAF). The Isolation Condenser System (ICS) and Passive Containment Cooling System (PCCS) kept the containment system, composed of Drywell (DW) and Wetwell (WW), below the design pressure limit. No serious challenge to fuel intact and containment integrity had been encountered even for BDBA tests with partly failure of safety system. Several issues like GDCS oscillation and potential low coolant level in the RPV found during the tests have been addressed and summarized for further evaluation. RELAP5 was claimed to be a "best estimate" system code suitable for the analysis of all transients and postulated accidents in Light Water Reactor (LWR) systems, including both large and small break LOCAs as well as the full range of operational transients. The version of code used in this research is RELAP5/Mod3.3. Generally the code models gave reasonably accurate predictions of the system thermal hydraulic behaviors compared with the experiment data. The agreement between the code simulation results and experimental data demonstrated that the RELAP5 had a good capacity to analyze the LOCA transients, thus still had some limitation to reproduce some particular local phenomena. The similarity between the code results of the plant model and test facility model demonstrated the scale up capacity of RELAP5 and scalability of integral test. The code sensitivity study has been performed to investigate the strength of safety system and code stability upon the perturbation of break size, initial conditions and PCCS capacity. Several comments and suggestion has been made regarding the potential improvement of certain calculation models in the code.

Degree

Ph.D.

Advisors

Ishii, Purdue University.

Subject Area

Nuclear engineering

COinS