No abstract

This paper demonstrates the successful implementation of an artificial neural network to accurately predict the designated thermal radiation distance for jet fire, early pool fire, and late pool fire hazard consequence analysis. Specifically, integrated feedforward neural network models employing the backpropagation Levenberg–Marquardt algorithm were trained using data sets obtained through separate PHAST software simulations of 450 leak scenarios of 35 common flammable chemicals. For each fire model (jet, early, late pool), there are 11 input parameters spanning both chemical parameters and release conditions. Simulation data was randomly divided into 70% training, 15% validation, and 15% test sets to conduct cross-validation and provide an independent measure of predictive accuracy for the neural network models. Statistical values, namely, coefficient of determination (R 2) and mean-square error (MSE), are calculated to evaluate model regression performance. All three neural network predictive models achieved considerably accurate predictions of the logarithm format of the designated radiation effect distance. The models give an overall R 2 of 0.9930 and an MSE of 0.0022 for jet fire, an overall R 2 of 0.9909 and an MSE of 0.0016 for early pool fire, and an overall R 2 of 0.9899 and an MSE of 0.0015 for late pool fire.

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.