• 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.

• Nuclear Engineering and Technology
• /
• v.30 no.3
• /
• pp.235-244
• /
• 1998

The parametric model method determines the accident source term which is Presented by a set of source term parameters. In this method, the cumulative distribution of each source term parameter should be derived for its uncertainty analysis. This paper introduces a method of generating the parameters in the form of cumulative distribution using MAAP version 4.0. In MAAP, there are model parameters which could incorporate uncertain physical and/or chemical phenomena. In general, the model parameters do not have a point value but a range. In this paper, considering that, the input values of model parameters influencing each parameter are sampled using LHS. Then, the computation results are shown in cumulative distribution form. For a case study, the CDFs of FCOR and WES of Kori Unit 1 are derived. The target scenarios for the computation are the ones whose initial events are large LOCA, small LOCA and transient, respectively. It is found that the computed CDF's in this study are consistent to those of NUREG-1150 and the use of MAAP is proven to be adequate in assessing the parameters of the severe accident source term.

• Nuclear Engineering and Technology
• /
• v.41 no.7
• /
• pp.945-952
• /
• 2009

The Three Mile Island Unit 2 (TMI-2) accident provides unique full scale data, thus providing opportunities to check the capability of codes to model overall plant behavior and to perform a spectrum of sensitivity and uncertainty calculations. As part of the TMI-2 analysis benchmark exercise sponsored by the Organization for Economic Cooperation and Development Nuclear Energy Agency (OECD NEA), several member countries are continuing to improve their system analysis codes using the TMI-2 data. The Republic of Korea joined this benchmark exercise in November 2005. Seoul National University has analyzed the TMI-2 accident as well as the currently proposed alternative scenario along with a sensitivity study using the Modular Accident Analysis Program Version 4.03 (MAAP4.03) code in collaboration with the Korea Hydro and Nuclear Power Company. Two input files are required to simulate the TMI-2 accident with MAAP4: the parameter file and an input deck. The user inputs various parameters, such as volumes or masses, for each component. The parameter file contains the information on TMI-2 relevant to the plant geometry, system performance, controls, and initial conditions used to perform these benchmark calculations. The input deck defines the operator actions and boundary conditions during the course of the accident. The TMI-2 accident analysis provided good estimates of the accident output data compared with the OECD TMI-2 standard reference. The alternative scenario has proposed the initial event as a loss of main feed water and a small break on the hot leg. Analysis is in progress along with a sensitivity study concerning the break size and elevation.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.20 no.6
• /
• pp.63-68
• /
• 2006

Recent statistical data regarding electric shock accidents have been analyzed to examine the electric shock accidents occurred at the voltage of 22,900[V], In order to demonstrate the mechanism of the 22,900[V] electric shock accident, a number of experiments to simulate electric shock accidents have been performed based on the analysis results. In the experiment, the current flowing through a human body model was measured to quantitatively analyze the hazards of the simulated electric shock accidents in various situations. As a result of the experiment, it was shown that once an electric shock accident occurred the accident proved fatal to the human body, regardless of electric shock situation.

• The Korean Journal of Applied Statistics
• /
• v.27 no.1
• /
• pp.147-157
• /
• 2014

Women's roles have changed substantially in economically developed countries; subsequently, the ratio of female drivers has also increased. In such countries, there has been considerable interest in assessing gender differences in vehicle accident risks and reasons to explain the gender differences. This study investigates the gender differences in vehicle accident risk based on 500,000 drivers randomly selected from a population sample. A Heckman model is used for accident damage and a negative binomial model is used for the accident frequency. Empirical results show that male drivers are 8.3% riskier than female drivers in terms of accident damage; however, female drivers are 113% risker than male drivers in term of accident frequency. We can implement more practical policies to reduce vehicle accidents if we can understand the reasons for the gender differences.

• Proceedings of KOSOMES biannual meeting
• /
• 2004.05b
• /
• pp.29-34
• /
• 2004

Ship is bring operated under a highly dynamic environments and many factors are related whit marine accident and those factors are interacting. An analysis on the marine accident is very important to prepare countermeasures which will ensure the safe navigation This paper aims to build a model of the rouses and improved policy for marine accident using SD(System Dynamics) approach and to measure a effect which is risk control countermeasures of marine accident The methodology of this paper is to perform the causes and improved policy for marine accident using Brainstorming method, and was to changed by quantitutive, qualitative factors and their feedback loops in casual map. This model was performed over 23 years($1997\~2020$) in a standard simulation model and 4 policy simulation models.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.4 no.2
• /
• pp.1-11
• /
• 1984

This aims of this study is to analyse the correlationship between traffic accident s and traffic characteristic variables in at-grade intersections of urban area, to build up an accident forecasting model and to propose an evaluation method of hazardous at-grade intersections. The accident forecasting model is formulated by the use of residual indexes that is selected by principal component analysis and its statistical significance is tested by step-wise regression analysis. Effective countermeasures for safety can be established on the basis of identifying high accident intersections, because the validity of this model was examined and found to coincide with real world situations.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2018.11a
• /
• pp.122-123
• /
• 2018

Although the disaster rate of the industry as a whole is on a downward trend, the disaster rate of the construction industry is on an ongoing trend. Therefore, in this study, we analyzed safety accident statistical data of the construction site over the past three years. As a result of the analysis, the incidence of disasters at small construction sites was very high. And the proportion of disaster occurred for workers who worked in less than 6 months even roughly 92.6%. In addition, as a result of analyzing the form of disaster occurrence, the crash was 34.1% and the fall was 15.1%. The analysis results of these construction safety accidents are to provide as a basic material for developing a policy that can prevent safety accidents and a safety accident prediction model.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.23 no.3
• /
• pp.254-262
• /
• 2015

This paper presents 3D collision deformation modelling methodologies using photogrammetry for reconstruction of vehicle accidents. A vehicle's deformation shape in collision provides important information on how the vehicle collided. So effective measurement(scanning) and construction of a corresponding appropriate model are essential in the analysis of collision deformation shape for obtaining much information related to collision accident. Two measurement methods were used in this study: Indirect-photogrammetry which requires relatively small amount of photos or videos, and direct-photogrammetry which requires large amount of photos directly taken for the purpose of 3D modelling. When the indirect-photogrammetry method, which was mainly used in this study, lacked enough photographic information, already secured 2D numerical deformation data was used as a compensation. This made 3D collision deformation modelling for accident reconstruction analysis possible.

• Journal of the Society of Disaster Information
• /
• v.17 no.3
• /
• pp.427-441
• /
• 2021

Purpose: This study proposes a method for developing a model that predicts the probability of traffic accidents in advance to prevent the most frequent traffic accidents in the military. Method: For this purpose, CRISP-DM (Cross Industry Standard Process for Data Mining) was applied in this study. The CRISP-DM process consists of 6 stages, and each stage is not unidirectional like the Waterfall Model, but improves the level of completeness through feedback between stages. Results: As a result of modeling the same data set as the previously constructed accident investigation data for the entire group, when the classification criterion was 0.5, Significant results were derived from the accuracy, specificity, sensitivity, and AUC of the model for predicting traffic accidents. Conclusion: In the process of designing the prediction model, it was confirmed that it was difficult to obtain a meaningful prediction value due to the lack of data. The methodology for designing a predictive model using the data set was proposed by reorganizing and expanding a data set capable of rational inference to solve the data shortage.