Developing an enhanced thermal radiation model through a Semi-A priori approach

• Main Authors: Chong, Michael Vui San, Mohd Tohir, Mohd Zahirasri, Saadon, Syamimi, Abu Talib, Abd Rahim
• Format: Article
• Published: Elsevier Masson 2024
• Online Access: http://psasir.upm.edu.my/id/eprint/105855/; https://www.sciencedirect.com/science/article/pii/S1290072923006452
• Institution: Universiti Putra Malaysia

The accurate prediction of thermal radiation is crucial for effective fire safety assessment and the development of proactive prevention strategies. This article presents an innovative approach to thermal radiation modelling through the integration of a semi-a priori methodology. The proposed model combines the strengths of a priori modelling and data-driven approaches, utilising the well-established single point source model as a foundation. The proposed model undergoes refinement and development using experimental data involving propane gas fires, ethanol pool fires, and isopropyl alcohol pool fires. These datasets serve as fitting data, enabling the model to improve its accuracy and performance. The refined model's performance is further assessed through the validation process using experimental data and simulation data. Experimental data from different fire scenarios, i.e., acetone and methanol pool fires, are utilised for the validation. Additionally, the estimated heat flux values generated by the proposed model and the single point source model are also compared with simulation results obtained from Fire Dynamics Simulator (FDS), specifically for heptane and methane pool fires. The findings of this research highlighted the enhanced accuracy and reliability of the proposed model in predicting thermal radiation behaviour. This research offers insights into thermal radiation modelling, providing valuable information for fire safety practitioners and researchers in enhancing fire risk assessment and mitigation strategies. By combining physical principles with empirical data, the proposed approach offers an alternative method for thermal radiation prediction in diverse fire scenarios.