Modeling and visualizing zones within the spread of toxic clouds from fires and explosions during accidents at industrial facilities located near residential areas is of high practical value. This tool is critical for the rapid planning of population evacuation measures and emergency response. Of particular importance is the development of computer software that can quickly model the hazard zone of toxic cloud spread and superimpose it on a terrain map to determine the potential impact on residential areas. This software should be based on a mathematical model that can accurately predict the parameters of the hazard zone both near the industrial facility and beyond it, at a distance of more than 1 km. The objective of this study is to create algorithms for modeling the hazard zone during a fire or explosion at an industrial facility using a cellular automaton method and to develop a software tool for its visualization. The software must display the hazard zone for the population of a nearby residential area on a map in real time, which is necessary for assessing potential harm to residents’ health and in planning their rapid evacuation. To achieve this objective, this article presents a model for determining the boundaries and main parameters of a hazard zone based on the cellular automaton method (frontal and probabilistic). The proposed model takes into account both constants (properties of chemical substances, building parameters, population size, etc.) and variables (the mass of the substance at each explosion and fire, wind speed and direction, air temperature, etc.). The FireSoft III software, developed by the authors and based on the cellular automaton model, provides more rapid calculation of the parameters and delineation of the hazard zone boundaries compared to similar software, which was tested in cases of an ammonia tank explosion and a prolonged fire in a warehouse containing polyvinyl chloride at an enterprise. This makes FireSoft III promising for use in a fire and explosion response at enterprises.
Abu-Abed et al. (Thu,) studied this question.