• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2017.11a
• /
• pp.339-340
• /
• 2017

화학공장의 누출사고는 초기에 적절히 대응하지 못할 경우 화재 폭발과 같은 2차 3차의 복합재난사고로 확산될 위험성이 매우 높다. 이러한 이유로 누출사고 발생 초기에 누출이 발생한 지점을 신속히 파악하여 현장안전요원에게 알림으로써, 보다 체계적이고 효율적인 초기대응을 가능하게 하여, 사고피해를 완화시킬 수 있는 통합적인 누출사고 대응시스템 구축은 매우 중요하다고 할 수 있다. 본 연구에서는, 통합적인 누출사고 대응시스템 구축을 위한 선행연구로, 딥러닝 기반의 누출원추적 모델 개발을 제안한다. 여수에 위치한 실제 화학공장을 대상으로 누출사고 시나리오에 대한 Computational Fluid Dynamics (CFD) 시뮬레이션을 진행한 뒤, 화학공장 경계면에 배치된 각 센서별 위치에서의 농도, 풍향 그리고 풍속데이터를 추출하고, 센서 좌표를 추가하여 인공신경망을 학습시켰다. 학습된 모델은 40개의 누출후보군에 대해 학습에 사용되지 않은 상황들에서도 75.43%의 정확도로 누출이 일어난 지점을 실시간 예측해냄을 확인하였다. 또한 누출지점 예측이 일치하지 않은 경우도, 예측된 지점이 실제 누출이 일어난 지점과 물리적으로 매우 인접함을 확인함으로써 제안된 모델을 실제 현장에 적용할시 기대되는 효과는 더 클 것으로 판단하였다.

• Journal of the Korean Institute of Gas
• /
• v.21 no.4
• /
• pp.1-5
• /
• 2017

Leaking of natural gas, which is mostly methane, in a confined living space creates flammable atmosphere and gives rise to explosion accident. The minimum amount of leaked methane for explosion is highly dependent on the degree of mixing in the confined space. This paper proposes a method for estimating minimum amount of flammable gas for explosion by using Gaussian distribution explosion model(GDEM) and experimental explosion data. The explosion pressure in the confined space can be estimated by assuming the Gaussian distribution of flammable gas along the height of an enclosure and estimating the maximum amount of gas within flammable limits, combustion of the estimated gas with constant volume and adiabatic or isothermal mixing in the confined space. The predicted minimum gas amount for an explosion is tied to explosion pressure that results in a given building damage level. The result shows that very small amount of methane leaking in the confined space may results in a serious gas explosion accident. This result could be applied not only to setting the leak criteria for developing a gas safety appliance but also to accident investigating of explosion.

• Journal of the Korean Institute of Gas
• /
• v.22 no.4
• /
• pp.55-62
• /
• 2018

There is growing interest in solar power generation due to global warming. As a result, demand for polysilicon, which is the core material for solar cells, is increasing day by day. As the market grows, large and small accidents occurred in the production process. In 2013, hydrochloric acid leaked from the polysilicon manufacturing plant in SangJu. In 2014, a fire occurred at a polysilicon manufacturing plant in Yeosu, and in 2015, STC(Silicon Tetrachloride) leaked at a polysilicon manufacturing plant in Gunsan City. Leakage of chemicals in the polysilicon manufacturing process can affect not only the workplace but also the surrounding area. Therefore, in this study, we identified the hazardous materials used in the polysilicon manufacturing process and quantitatively estimate the amount of leakage and extent of damage when the worst case scenario is applied. As a result, the damage distance by explosion was estimated to be 726 m, and the damage distance to toxicity was estimated to be 4,500 m. And, if TCS(Trichlorosilane), STC(Silicon Tetrachloride), DCS(Dichlorosilane) leaks into the air and reacts with water to generate HCl, the damage distance is predicted to 5.7 km.

• Journal of the Korean Institute of Gas
• /
• v.10 no.4 s.33
• /
• pp.52-62
• /
• 2006

In this study, we tried to propose a selection method of release scenarios and a method of consequence analysis at a gas explosion by pipe release. Thus, release rates, damage areas of the facilities, and fatality areas were estimated and analyzed at various release conditions(temperature, pressure, release material, etc). As a results, we could conclude that the rupture was the worst case of release scenarios, and at release rates and damage areas were better estimated by the weighted average method considering a generic failure frequency of the release hole than by an arbitrary selection of the release hole.

• Journal of the Korea Convergence Society
• /
• v.11 no.12
• /
• pp.23-30
• /
• 2020

In this study, the relationship between natural gas (NG) data and gas-related environmental elements was performed using machine learning algorithms to predict the level of gas leakage risk without directly measuring gas leakage data. The study was based on open data provided by the server using the IoT-based remote control Picarro gas sensor specification. The naturel gas leaks into the air, it is a big problem for air pollution, environment and the health. The proposed method is multivariate outlier removing method based Random Forest (RF) classification for predicting risk of NG leak. After, unsupervised k-means clustering, the experimental dataset has done imbalanced data. Therefore, we focusing our proposed models can predict medium and high risk so best. In this case, we compared the receiver operating characteristic (ROC) curve, accuracy, area under the ROC curve (AUC), and mean standard error (MSE) for each classification model. As a result of our experiments, the evaluation measurements include accuracy, area under the ROC curve (AUC), and MSE; 99.71%, 99.57%, and 0.0016 for MOL_RF respectively.

• Journal of the Korean Institute of Gas
• /
• v.20 no.3
• /
• pp.59-65
• /
• 2016

It is imperative to use suitable explosion proof equipments to prevent explosion in different gas facilities. There is no technical standard for the classification of hazardous areas though standard of explosion proof is regulated. In this study, we have adopted Industrial Standard KS to develop the methodology for the prediction of the explosion risk in the natural gas facility with low pressure using the important factors including hole size, hypothetical volume, validation of ventilation effectiveness. The applicability of the developed methodology was evaluated by the comparison with the data obtained from experiments of natural gas explosion.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.407-417
• /
• 2024

The objective of this study is to predict and reduce potential damage caused by chlorine gas leaks, a hazardous material, when vehicles transporting it overturn due to accidents or other incidents. The goal is to forecast the anticipated damages caused by chlorine toxicity levels (ppm) and to design effective response strategies for mitigating them. To predict potential damages, we conducted quantitative assessments using the ALOHA program to calculate the toxic effects (ppm) and damage distances resulting from chlorine leaks, taking into account potential negligence of drivers during transportation. The extent of damage from toxic gas leaks is influenced by various factors, including the amount of the leaked hazardous material and the meteorological conditions at the time of the leak. Therefore, a comprehensive analysis of damage distances was conducted by examining various scenarios that involved variations in the amount of leakage and weather conditions. Under intermediate conditions (leakage quantity: 5 tons, wind speed: 3 m/s, atmospheric stability: D), the estimated distance for exceeding the AEGL-2 level of 2 ppm was calculated to be 9 km. This concentration poses a high risk of respiratory disturbance and potential human casualties, comparable to the toxicity of hydrogen chloride. In particular, leaks in urban areas can lead to significant loss of life. In the event of a leakage incident, we proposed a plan to minimize damage by implementing appropriate response strategies based on the location and amount of the leak when an accident occurs.

• Journal of the Korea Convergence Society
• /
• v.10 no.10
• /
• pp.7-13
• /
• 2019

The natural gas (NG), mostly methane leaks into the air, it is a big problem for the climate. detected NG leaks under U.S. city streets and collected data. In this paper, we introduced a Deep Neural Network (DNN) classification of prediction for a level of NS leak. The proposed method is OrdinalEncoder(OE) based K-means clustering and Multilayer Perceptron(MLP) for predicting NG leak. The 15 features are the input neurons and the using backpropagation. In this paper, we propose the OE method for labeling target data using k-means clustering and compared normalization methods performance for NG leak prediction. There five normalization methods used. We have shown that our proposed OE based MLP method is accuracy 97.7%, F1-score 96.4%, which is relatively higher than the other methods. The system has implemented SPSS and Python, including its performance, is tested on real open data.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.33-41
• /
• 2022

In this article, we proposed to predict natural gas (NG) leakage levels through feature selection based on a factor analysis (FA) of the integrating the Korean Meteorological Agency data and natural gas leakage data for considering complex factors. The paper has been divided into three modules. First, we filled missing data based on the linear interpolation method on the integrated data set, and selected essential features using FA with OrdinalEncoder (OE)-based normalization. The dataset is labeled by K-means clustering. The final module uses four algorithms, K-nearest neighbors (KNN), decision tree (DT), random forest (RF), Naive Bayes (NB), to predict gas leakage levels. The proposed method is evaluated by the accuracy, area under the ROC curve (AUC), and mean standard error (MSE). The test results indicate that the OrdinalEncoder-Factor analysis (OE-F)-based classification method has improved successfully. Moreover, OE-F-based KNN (OE-F-KNN) showed the best performance by giving 95.20% accuracy, an AUC of 96.13%, and an MSE of 0.031.

• Applied Chemistry for Engineering
• /
• v.10 no.8
• /
• pp.1155-1160
• /
• 1999

The screening methodology modeling, dispersion modeling procedures for continuous and instantaneous releases of the gas phase flow from the storage tank and pressure relief valve were considered. This study was performed to develop the screening methodology for prediction and control of hazardous/toxic gas releases by estimating the 1-hr average maximum ground-level concentration of $Cl_2$ gas vs. downwind distance by incorporating source term model including the general/physical properties of released material and release mode of the $Cl_2$ storage tank of the chemical plant facilities, dispersion model, and meteorological/topographical data into the TSCREEN model. As the results of the study, it was found that dispersion modes of the dense gas were affected by the state of the released material, the released conditions, physical-chemical properties of released material, and the released modes (continuous and instantaneous releases), and especially largely affected by initial (depressurized) density of the released material and release emission rate as well as the wind velocity. Especially, this study was considered to release hazardous material as meteorological data. It was thought that this screening methodology can be useful as a preliminary guideline for application of the refined analysis model by developing the generic sliding scale methodology for various senarios selected.