IDENTIFICATION OF KEY SAFETY DESIGN PARAMETERS IN THE FMR DESIGN USING MELCOR

Abstract

The Gas-Cooled Fast Reactor (GFR) is a generation IV reactor type that was first proposed in the 1960s with the primary objective of being a breeder reactor. This was a promising concept of GFRs as uranium deposits were thought to be limited at the time and the concept of breeder reactors were highly sought after. However, after the discovery of more fuel deposits around the world the development of GFRs were halted. Currently, the GFR concept is again gaining traction for the capability of being able to reach a higher operating temperature which can yield higher efficiencies in the power conversion cycle as well as being utilized in other industrial applications. The GFR operates using a fast neutron spectrum to burn its fuel, a key distinction from conventional Light Water Reactors (LWRs), resulting in significantly higher fuel burn-up levels. This unique characteristic, combined with the use of gas as a coolant, has led researchers to identify forced convective cooling of the gas coolant as a critical feature of GFRs. This cooling mechanism enables effective heat dissipation under both normal operation and accidents. One of the most severe accident scenarios for a GFR is a Depressurized Loss of Forced Coolant (DLOFC), where the loss of pressurized flow through the core eliminates a primary heat transfer pathway. In such a scenario, the core must rely solely on radiative and conduction heat transfer to dissipate decay heat as temperatures rise. This thesis examines the response of a GFR-type reactor to a DLOFC scenario by conducting a sensitivity analysis using MELCOR, providing insights into the key parameters that influence heat dissipation within the core. The modeling of the core, specifically, the modeling of the core geometry and the thermal properties of the materials was determined to be the most important in the study of heat dissipation of the core. This was a challenging task as there are limited research done on the highly heat resistive core components the make up the core which introduces uncertainties. To address this uncertainty a safety analysis was necessary to validate the results of MELCOR. The radiative heat transfer path within the core was modeled by using a percentage-based value that indicates how much radiative heat is transferred from one core cell to another called the radiative view factor. The view factors were calculated using the default MELCOR model provided in its reference manual. This calculation yielded higher core temperatures compared to using the thermal properties of the core provided by GA-EMS and developing an effective thermal conductivity. This resulted in prompting two separate base case when conducting the sensitivity analysis (conservative and best-estimate base case). Additionally, the view factor that was used in the heat structures that make up the RPV wall, the core barrel, and the RVCS panels were more straightforward and were separately calculated using the Monte Carlo method. The GFR type Fast Modular Reactor (FMR), which is a reactor that is currently being developed by General Atomics Electromagnetic Systems (GA-EMS) is a GFR, the main focus of this thesis where a sensitivity analysis is performed to test and confirm temperature limits of the PCT of the reactor under a DLOFC accident. This was done by modeling the radiative heat transfer within the core and modeling the heat dissipation through the RVCS using MELCOR. The thesis tackles the

heat dissipation of the core of the FMR and the RVCS separately by decoupling these two complicated engineering components into two separate MELCOR models. It was hypothesized that the radiative heat transfer through the core after the accident will dominate shortly after the accident rendering the relatively lower temperature of the RVCS a nonfactor in the peaking of PCT. Replacing the RVCS with simple heat structures with a constant bounding water temperature, helps with the complexity of the overall model of the FMR as well as save time in our simulations. The results of the sensitivity analysis of the FMR confirms the hypothesis that was made validating the decision of decoupling the RVCS from the main reactor. The PCT analysis was done by determining parameters to change in the model and comparing the PCT of these different cases to the conservative base case. Findings of the PCT analysis indicates that the most important parameters that affect the PCT are the active core radiative view factors. Two cases were considered one where the view factors were doubled and the other where it was halved. Both cases yielded a difference of two orders of magnitude. Surprisingly, the emissivity cases, where the emissivity values of both the cladding (SiC) and the stainless steel were changed, had negligible effects on the PCT. Other parameters that are worth mentioning are the difference in pressure across the power conversion unit (PCU) and the RVCS water temperature, which had second order effects on the PCT. The RVCS analysis consisted of determining a base case for the MELCOR model by matching the flow characteristics of the model to the experimental results of the UW-Madison RVCS facility. This was accomplished by changing certain parameters of the model to best match the power input into the water through the risers, flow rate in both single and two phase, pressure in the tank, and the inlet and outlet temperatures through the risers. MELCOR was able to predict the results of the experiment with one exception: oscillations in the two-phase region of the flow rate. MELCOR was predicting much higher oscillation compared to the experiment. This was determined to be the case due to predictions that MELCOR was making for the void fraction through the flow system. Additionally, MELCOR predicted flashing occurring much earlier in the flow system compared to the experiment. Based on these results, a sensitivity analysis was conducted based on which parameters in the RVCS design had the most effect on the void behavior of the flow system. The parameters chosen for the sensitivity analysis was determined with the focus on decreasing the oscillation amplitudes down. These include: the loss coefficients through the lower and upper chimney, the tank pressure and water elevation, the input power, the heat transfer through the pipes, the friction factor multiplier, and the height of the RVCS model. Findings of the sensitivity analysis concluded that the overall parameter that had the most effect in decreasing the fluctuations of the oscillations was the loss coefficient. Other parameters had effects on the flow rate but not as pronounced as the loss coefficient parameter. Parameters worth mentioning are the RVCS height where a gradual increase in oscillation peaks were observed, the friction factor multiplier which introduced chaotic behaviors in the flow rate, and the heat transfer where the oscillation frequency decreased as the heat transfer coefficient (HTC) increased