Experimental Machine Learning Study on CO<sub>2</sub> Gas Dispersion

Gwak, Kiuk; Rho, Young J.

doi:10.1109/iscaie.2019.8743776

Cited by 2 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is possible that DLP models of more than 3 hidden layers show lower performance [3]. In this research, the DLP model of 2 hidden layers showed better performance than the MLP model of one hidden layer [4]. PPM values are predicted by the both models of MLP and DLP for ensemble.…”

Section: B Deep Multi-layer Perceptronmentioning

confidence: 65%

Gas Diffusion Simulation Based on Ensemble Approach

Gwak¹,

Rho²

2020

IJMO

View full text Add to dashboard Cite

The 4th 1 industrial revolution is promoting manufacturing industry to be vitalized again. The manufacturing industry requires many industrial materials. Among them, different gases are also used in many fields. While they are useful, industrial gases can be also hazardous at the same time. In order to control those bad features of gases, their dynamic characteristics are required to be understood. In this paper we tried to understand the characteristics by applying several machine learning methods such as MLP, DLP and LSTM. Two ensemble methods are applied to compensate the lack of raw data. Simulation outputs are compared each other to know which method is proper for this case.

show abstract

Section: B Deep Multi-layer Perceptronmentioning

confidence: 65%

Gas Diffusion Simulation Based on Ensemble Approach

Gwak¹,

Rho²

2020

IJMO

View full text Add to dashboard Cite

show abstract

“…21 Gwak and Rho compared three different machine learning techniques in predicting CO 2 dispersion in a laboratory environment. 22 However, these works examined the techniques using only one or two chemicals (chlorine, carbon dioxide, or sulfur dioxide) under limited leaking conditions, which is not sufficient to construct a widely applicable model for real-world applications.…”

Section: Introductionmentioning

confidence: 99%

“…Wang et al compared two different machine learning algorithms for use in gas dispersion using field experiments, which showed the effectiveness of machine learning for predicting hazardous gas dispersion . Gwak and Rho compared three different machine learning techniques in predicting CO 2 dispersion in a laboratory environment . However, these works examined the techniques using only one or two chemicals (chlorine, carbon dioxide, or sulfur dioxide) under limited leaking conditions, which is not sufficient to construct a widely applicable model for real-world applications.…”

Section: Introductionmentioning

confidence: 99%

Development of Flammable Dispersion Quantitative Property–Consequence Relationship Models Using Extreme Gradient Boosting

Jiao

Sun

Hong

et al. 2020

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Uncontrolled release of flammable gases and liquids can lead to the formation of flammable vapor clouds. When their concentrations are above the lower flammable limit (LFL), or 1/2 LFL for conservative evaluation, fires and explosions can happen in the presence of an ignition source. The objective of this work is to develop highly efficient consequence models to precisely predict the downwind maximum distance, minimum distance, and maximum vapor cloud width within the flammable limit. In this work, the novel methodology named quantitative property–consequence relationship (QPCR) is proposed and constructed to precisely predict flammable dispersion consequences in a machine learning and data-driven manner. A flammable dispersion database consisting of 450 leak scenarios of 41 flammable chemicals was constructed using PHAST simulations. A state-of-art machine learning regression method, the extreme gradient boosting algorithm, was implemented to develop models. The coefficient of determination (R 2) and root-mean-square error (RMSE) were calculated for statistical assessment, and the developed QPCR models achieved satisfactory predictive capabilities. All developed models had high precision, with the overall RMSE of three models being 0.0811, 0.0741, and 0.0964, respectively. The developed QPCR models can be used to obtain instant flammable dispersion estimations for other flammable chemicals and mixtures at much lower computational costs.

show abstract

Experimental Machine Learning Study on CO₂ Gas Dispersion

Cited by 2 publications

References 0 publications

Gas Diffusion Simulation Based on Ensemble Approach

Gas Diffusion Simulation Based on Ensemble Approach

Development of Flammable Dispersion Quantitative Property–Consequence Relationship Models Using Extreme Gradient Boosting

Contact Info

Product

Resources

About

Experimental Machine Learning Study on CO2 Gas Dispersion

Cited by 2 publications

References 0 publications

Gas Diffusion Simulation Based on Ensemble Approach

Gas Diffusion Simulation Based on Ensemble Approach

Development of Flammable Dispersion Quantitative Property–Consequence Relationship Models Using Extreme Gradient Boosting

Contact Info

Product

Resources

About

Experimental Machine Learning Study on CO₂ Gas Dispersion