• Nuclear Engineering and Technology
• /
• v.55 no.12
• /
• pp.4607-4616
• /
• 2023

We implement machine learning regression models to predict peak pressures of primary and secondary systems, a major safety concern in Loss Of Condenser Vacuum (LOCV) accident. We selected the Multi-dimensional Analysis of Reactor Safety-KINS standard (MARS-KS) code to analyze the LOCV accident, and the reference plant is the Korean Optimized Power Reactor 1000MWe (OPR1000). eXtreme Gradient Boosting (XGBoost) is selected as a machine learning tool. The MARS-KS code is used to generate LOCV accident data and the data is applied to train the machine learning model. Hyperparameter optimization is performed using a simulated annealing. The randomly generated combination of initial conditions within the operating range is put into the input of the XGBoost model to predict the peak pressure. These initial conditions that cause peak pressure with MARS-KS generate the results. After such a process, the error between the predicted value and the code output is calculated. Uncertainty about the machine learning model is also calculated to verify the model accuracy. The machine learning model presented in this paper successfully identifies a combination of initial conditions that produce a more conservative peak pressure than the values calculated with existing methodologies.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1999.10a
• /
• pp.174.1-174
• /
• 1999

• Nuclear Engineering and Technology
• /
• v.54 no.5
• /
• pp.1698-1711
• /
• 2022

A combined safety approach, which uses a best-estimate computer code and adopts conservative assumptions for safety systems availability, is developed and applied to the safety margin evaluation for the Loss of Condenser Vacuum (LOCV) of the 1000 MWe Korean Nuclear Power Plant. The Multi-dimensional Analysis of Reactor Safety-KINS standard (MARS-KS) code is selected as a best-estimate code and the PAPIRUS program is used to obtain different initial operational conditions through random sampling of control variables. During an LOCV event, fuel integrity is not threatened by the increase in Departure from Nuclear Boiling Ratio (DNBR). However, the high pressure in the primary coolant system and the secondary system might affect the system integrity. Thus, the peak pressure becomes a major safety concern. Transient analyses are performed for 124 cases of different initial conditions and the most conservative case, which results in the highest system pressure is selected. It is found the suggested methodology gives similar peak pressures when compared to those predicted from existing methodologies. The proposed approach is expected to minimize the time and efforts required to identify the conservative plant conditions in the existing conservative safety methodologies.

• Journal of the Korean Society of Safety
• /
• v.14 no.1
• /
• pp.177-184
• /
• 1999

PSA(Probabilistic Safety Assessment) methodology is widely used on assessing the safety of Nuclear Power Plants(NPPs) quantitatively in the domestic nuclear field. Initiating event frequencies are absolutely needed to conduct PSA, and they considerably affect PSA results. There is no domestic database where domestic trip event cases are reflected, so they are used to assess the safety of NPPs that are from the foreign database. In this paper, operating experience data from the Korean NPPs was collected and analyzed for the trip event cases, which are necessary to determine the initiating events and their frequencies. Korean NPPs have experienced five of 16 initiating events, which we LOFW. LOCV, LOCCW, LOOP and GTRN as a result of analyzing the trip event cases. Initiating frequencies based on the domestic trip event cases are analyzed, and they are similar to that from the foreign database.