Impact of Mixed-Precision: A Way to Accelerate Data-Driven Forest Fire Spread Systems

Producció científica: Capítol de llibreCapítolRecercaAvaluat per experts

Resum

Every year, forest fires burn thousands of hectares of forest around the world and cause significant damage to the economy and people from the affected zone. For that reason, computational fire spread models arise as useful tools to minimize the impact of wildfires. It is well known that part of the forest fire forecast error comes from the uncertainty in the input data required by the models. Different strategies have been developed to reduce the impact of this input-data uncertainty during the last few years. One of these strategies consists of introducing a data-driven calibration stage where the input parameters are adjusted according to the actual evolution of the fire using an evolutionary scheme. In particular, the approach described in this work consists of a Genetic Algorithm (GA). This calibration strategy is computationally intensive and time-consuming. In the case of natural hazards, it is necessary to maintain a balance between accuracy and time needed to calibrate the input parameters. Most scientific codes have over-engineered the numerical precision required to obtain reliable results. In this paper, we propose to use a mixed-precision methodology to accelerate the calibration of the input parameters involved in forest fire spread prediction without sacrificing the accuracy of the forecast. The proposed scheme has been tested on a real fire. The results have led us to conclude that using the mixed-precision approach does not compromise the quality of the result provided by the forest fire spread simulator and can also speed up the whole evolutionary prediction system
Idioma originalEnglish
Títol de la publicacióComputational Science – ICCS 2023 - 23rd International Conference, Proceedings
EditorsJiří Mikyška, Clélia de Mulatier, Valeria V. Krzhizhanovskaya, Peter M.A. Sloot, Maciej Paszynski, Jack J. Dongarra
Pàgines62-76
Nombre de pàgines15
Volum14074
DOIs
Estat de la publicacióPublicada - 2023

Sèrie de publicacions

NomLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volum14074 LNCS
ISSN (imprès)0302-9743
ISSN (electrònic)1611-3349

Fingerprint

Navegar pels temes de recerca de 'Impact of Mixed-Precision: A Way to Accelerate Data-Driven Forest Fire Spread Systems'. Junts formen un fingerprint únic.

Com citar-ho