An estimation of an operator's action time by using the MARS code in a small break LOCA without a HPSI for a PWR

Abstract

To estimate the success criteria of an operator's action time for a probabilistic safety/risk assessment (PSA/PRA) of a nuclear power plant, the information from a safety analysis report (SAR) and/or that by using a simplified simulation code such as the MAAP code has been used in a conventional PSA. However, the information from these is often too conservative to perform a realistic PSA for a risk-informed application. To reduce the undue conservatism, the use of a best-estimate thermal hydraulic code has become an essential issue in the latest PSA and it is now recognized as a suitable tool. In the same context, the ‘ASME PRA standard’ also recommends the use of a best-estimate code to improve the quality of a PSA. In Korea, a platform to use a best-estimate thermal hydraulic code called the MARS code has been developed for the PSA of the Korea standard nuclear power plant (KSNP). This study has proposed an estimation method for an operator's action time by using the MARS platform. The typical example case is a small break loss of coolant accident without the high pressure safety injection system, which is one of the most important accident sequences in the PSA of the KSNP. Under the given accident sequence, the operator has to perform a recovery action known as a fast cooldown operation. This study focuses on two aspects regarding an operator's action; one is how they can operate it under some restrictions; the other is how much time is available to mitigate this accident sequence. To assess these aspects, this study considered: (1) the operator's action model and (2) the starting time of the operation. To show an effect due to an operator's action, three kinds of control models (the best-fitting, the conservative, and the proportional-integral) have been assessed. This study shows that the developed method and the platform are useful tools for this type of problem and they can provide a valuable insight related to an operator's actions.