Abordagem evolutiva para o ajuste automático dos parâmetros de um modelo de propagação de incêndios baseado em autômatos celulares

Abstract

Wildfires are a growing global concern, impacting both the environment and public health. To mitigate these effects, it is crucial to understand the dynamics of fire spread and to develop effective firefighting strategies. This study presents the construction of a genetic algorithm for the automatic adjustment of parameters in a wildfire propagation model based on cellular automata, with the goal of replicating patterns observed in historical wildfire data. The research builds on a wildfire propagation model from the literature and implements a genetic algorithm capable of optimizing simulation parameters, an otherwise highly complex task to perform manually. The results validated the effectiveness of this evolutionary approach, demonstrating that, even when optimal values were not fully achieved, the simulations produced a faithful representation of fire spread. The analysis of the algorithm's performance indicates that this method can be applied to a range of real-world scenarios, contributing to the design of more effective firefighting strategies. This work outlines clear directions for future research on wildfire simulation, including proposed enhancements such as refining cell type definitions and recovery parameters to achieve even greater accuracy and efficacy in modeling.