Development of a methodology for determining early SAMG entry condition considering operator action time in NPPs

Abstract

The importance of severe accident management is being re-examined because severe accident management failed during the Fukushima nuclear accident. Since the Fukushima nuclear accident, safety inspections have been carried out in many countries to check the safety of operating NPPs. Nonetheless, operators in NPPs may still be unable to properly mitigate accidents in severe situations. Furthermore, due to information overload and high workload in severe situations, the operators may experience confusion, which can lead to wrong decisions. Therefore, the operator must have enough time to properly mitigate the severe accident. The current severe accident management guideline (SAMG) entry condition is determined without considering operator action time during SAMG development. Therefore, it is necessary for the current SAMG entry condition to be optimized with the point of consideration of operator action time. Also, it is necessary to develop method that can quickly predict SAMG entry time in the initial phase after the accident because the operator rely on SAMGs during severe accident situations. The objective of this study is to develop a methodology for determining SAMG entry condition considering operator action time in NPPs. In this study, the margin of operator action time from SAMG entrance to reactor vessel (RV) failure was analyzed to compare the required operator action time and available operator action time based on a severe accident DB developed using MAAP 5.02 code. In order to analyze the margin of operator action time, dominant initiating accident are selected, such as small break loss of coolant accident (SBLOCA), medium break (MBLOCA), large break (LBLOCA), and station blackout (SBO) which account for about 80 percent of core damage frequency (CDF), and the worst scenario was selected in each initiating accident. In the analysis of the worst scenarios, it was confirmed that operator might be fail to prevent RV failure due to insufficient action time. Therefore, in this study, a method that can determine the SAMG entry condition in advance was developed by applying component state to function recovery procedure (FRP). To verify the developed method, a case study was performed based on the scenarios that feed & bleed (F&B) and aggressive cool down (ACD) operations are not successful within limited time after entering the FRP. As a result, the developed strategy flow chart can identify the SAMG entry condition in advance, and provide operator with sufficient action time to mitigate severe accidents. Also, having an accurate estimation of a severe accident is important for coping properly with unfavorable conditions. In this study, the extended group method of data handling (GMDH) with fuzzy concept is developed as means of predicting the major accident events that represent the SAMG entry time, including the time at which the RV water level decrease, the time when the core exit temperature (CET) reaches 450 degrees-centigrade, the time when the hydrogen concentration in the containment is over 4%, and the time when the reactor coolant system (RCS) pressure is over 2.86MPa under LOCAs. To train the extended GMDH model, it was necessary to acquire the data needed from a number of numerical simulations due to the lack actual LOCA data. The data was obtained by carrying out simulations using the MAAP5 code. To optimize the developed model, the optimal input selection processes were performed using the clustering analysis including self-organizing feature map (SOFM), hierarchical and non-hierarchical clustering methods. The prediction accuracy of the three types of initiating accident, SBLOCAs, MBLOCAs, and LBLOCAs was high enough to predict the SAMG entry time. When compared with other artificial intelligence (AI) methods, the extended GMDH was found to be superior to the original GMDH and support vector regression (SVR). Therefore, it is expected that this study can provide meaningful results for developing the integrated accident management procedures (IAMPs). Also, we expect that the developed method will help operators to predict the proper SAMG entry time during the initial phase after a reactor trip.