• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.2
• /
• pp.75-93
• /
• 2022

Machine learning (ML) data-driven meta-model is proposed as a surrogate model to reduce the excessive computational cost of the physics-based model and facilitate the real-time prediction of a nuclear power plant's transient response. To forecast the transient response three machine learning (ML) meta-models based on recurrent neural networks (RNNs); specifically, Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), and a sequence combination of Convolutional Neural Network (CNN) and LSTM are developed. The chosen accident scenario is a control element assembly withdrawal at power concurrent with the Loss Of Offsite Power (LOOP). The transient response was obtained using the best estimate thermal hydraulics code, MARS-KS, and cross-validated against the Design and control document (DCD). DAKOTA software is loosely coupled with MARS-KS code via a python interface to perform the Best Estimate Plus Uncertainty Quantification (BEPU) analysis and generate a time series database of the system response to train, test and validate the ML meta-models. Key uncertain parameters identified as required by the CASU methodology were propagated using the non-parametric Monte-Carlo (MC) random propagation and Latin Hypercube Sampling technique until a statistically significant database (181 samples) as required by Wilk's fifth order is achieved with 95% probability and 95% confidence level. The three ML RNN models were built and optimized with the help of the Talos tool and demonstrated excellent performance in forecasting the most probable NPP transient response. This research was guided by the Systems Engineering (SE) approach for the systematic and efficient planning and execution of the research.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1996.05b
• /
• pp.622-627
• /
• 1996

Detailed analysis of Feedwater Line Break (FLB) event for the fuel failure point of view are lack because the event was characterized as the increase in reactor coolant system (RCS) pressure. Up to now, the potential of the rapid system heatup case has been emphasized and comprehensively studied. The cooldown effects of FLB event is considered to be bounded by the Steam Line Break (SLB) event since the cooldown effect of SLB event is larger than that of the FLB event. This analysis provides a new possible path which can cause the fuel failure. The new path means that the fuel failure can occur under the heatup scenario because the Pressurizer Safety Valves (PSVs) open before the reactor trips. The 1000 MWe typical C-E plant FLB event assuming Loss of Offsite Power (LOOP) at the turbine trip has been analyzed as an example and the results show less than 1% of the fuel failure. The result is well within the acceptance criteria. In addition to that, a study was accomplished to prevent the fuel failure for the heatup scenario case as an example. It is found that giving the proper pressure gap between High Pressurizer Pressure Trip (HPPT) analysis setpoint and the minimum PSV opening pressure could prevent the fuel failure.

• Proceedings of the Korean Radioactive Waste Society Conference
• /
• 2004.06a
• /
• pp.361-367
• /
• 2004

For assessment of tritium and radiocarbon ingestion dose to off site individuals, water, hydrogen, and carbon content of main farm produce of Korea were investigated to replace the existing data in K-DOSE60, the Offsite Dose Calculation Manual (ODCM) of Korea Hydro ＆ Nuclear Power Co. Ltd, (KHNP). Main items and weighting factors of farm produce were determined with the nationwide food intake data in 2001, 2002. Main farm produce were sampled around Kori, Wolsong, Ulchin, Younggwang nuclear power sites, Content of each produce was multiplied by weighting factor and summed up to make the weighted mean group value For grains, water, hydrogen, and carbon content was not much different from the existing data currently used in K-DOSE60, but root vegetables had 3.5 times more hydrogen, and leafy vegetables and fruits had 0.7∼1.3 times more or less water, hydrogen, and carbon contents than K-DOSE60.

• Journal of Radiation Protection and Research
• /
• v.28 no.3
• /
• pp.189-197
• /
• 2003

The Level 3 PSA being termed accident consequence analysis is defined to assess effects on health and environment caused by radioisotopes released from severe accidents of nuclear power plants. In this study consequence analysis on health effects depending on release characteristics of radioisotopes has been peformed using the 3 MACCS code in severe accidents. The results of this study may contribute to identifying the relative importance of various parameters occurred in consequence analysis as well as to assessing risk reduction accident management strategies. Especially three parameters for the purpose of consequence analysis, such as the release height, the heat content, and the duration time, are used to analyze the variation of early fatalities and latent cancer fatalities. Also, in this study risk assessment using the concept, 'products of uncertainty and consequences', has been performed using consequence of MACCS and frequency on source term category 19 scenarios from IPE (Individual Plant Examination) analysis.

• Nuclear Engineering and Technology
• /
• v.27 no.4
• /
• pp.483-490
• /
• 1995

The objective of the PKL III A 4.4 experiment is to examine that the plant could be controlled by manually operative actions "after Steam Generator Tube Rupture under Offsite Power Available without Safety Injection". In order to verify the limitation and ability of the system code NLOOP in the expeiment simulation, the behaviors of the PKL III facility obtained in the experiment are compared with the results of NLOOP code. NLOOP code, which is originally developed to simulate the transients of the Westinghouse type PWRs by KAERI/SIEMENS, modified properly to simulate the PKL III facility. Particular attention is given to the RCS mass How rate of the natural circulation in loops and the termination behavior of the natural circulation in the isolated loop. The comparisons between the experimental and calculational results show the simulation ability and problems of the code. the code.

• Journal of the Korea Safety Management & Science
• /
• v.18 no.3
• /
• pp.63-70
• /
• 2016

The main point of this study is to find out duplicates and differences among various regulations from different organizations. Also, it focuses on creating a reasonably unified regulation system to standardize safety & environment management. In this study, I analyzed the commonalities and the differences of two systems which are typical korean Process Safety Management System and off-site Consequence Analysis. It is confirmed that there are 25 species of overlapped material of those two systems and assessment like handling material information, facilities lists, hazardous substances and list of machine power. Process safety report focuses on onsite workers and facility protect. On the other hand, off-site Consequence Analysis focuses on design, arrangement and management of handling facility from off-site influence. I found difference two system of Enforcement purposes and way. Contradiction of Harmful information of Chemicals Control Act and occupation safety and health acts from same material. To be specific, There are no unit rule of occupation safety and health acts. so it permit inch, psi etc. But Chemicals Control Act provides that m, Mpa units. Therefore, Each regulatory duplication of items for chemicals management, standardization is writing so that you can coordinate overlapping items in the measures the need to be presented.

• Journal of the Korean Society of Systems Engineering
• /
• v.16 no.2
• /
• pp.97-109
• /
• 2020

On March 11, 2011, an earthquake followed by a tsunami caused an extended station blackout (SBO) at the Fukushima Dai-ichi NPP Units. The accident was initiated by a total loss of both onsite and offsite electrical power resulting in the loss of the ultimate heat sink for several days, and a consequent core melt in some units where proper mitigation strategies could not be implemented in a timely fashion. To enhance the plant's coping capability, the Diverse and Flexible Strategies (FLEX) were proposed to append the Emergency Operation Procedures (EOPs) by relying on portable equipment as an additional line of defense. To assess the success window of FLEX strategies, all sources of uncertainties need to be considered, using a physics-based model or system code. This necessitates conducting a large number of simulations to reflect all potential variations in initial, boundary, and design conditions as well as thermophysical properties, empirical models, and scenario uncertainties. Alternatively, data-driven models may provide a fast tool to predict the success window of FLEX strategies given the underlying uncertainties. This paper explores the applicability of Artificial Intelligence (AI) to identify the success window of FLEX strategy for extended SBO. The developed model can be trained and validated using data produced by the lumped parameter thermal-hydraulic code, MARS-KS, as best estimate system code loosely coupled with Dakota for uncertainty quantification. A Systems Engineering (SE) approach is used to plan and manage the process of using AI to predict the success window of FLEX strategies under extended SBO conditions.