Open AccessHybrid Computational Model to Assist in the Location of Victims Buried in the Tragedy of Brumadinho
Author(s) -
Rafael Fernandes Pinheiro,
Harold Ivan Angulo Bustos,
Bruno Bestle Turrin,
Francisco Monteiro,
Diego Colón,
Mirelly Ferreira Silveira,
Dário José Aloise,
Alysson Mendes de Oliveira
Publication year - 2021
Publication title -
journal of environmental analysis and progress
Language(s) - English
Resource type - Journals
ISSN - 2525-815X
DOI - 10.24221/jeap.6.3.2021.4103.183-193
Subject(s) - computer science , tragedy (event) , support vector machine , work (physics) , regression , regression analysis , extreme learning machine , dynamics (music) , machine learning , artificial intelligence , mathematics , artificial neural network , engineering , mechanical engineering , statistics , social science , physics , sociology , acoustics
This paper presents a hybrid computational model based on regression techniques, machine learning and physicomathematical algorithms developed for assistance in locating victims in the Brumadinho tragedy in 2019. The physicomathematical model, which provided results to help search teams, is based on integral and vector calculus, and fluid mechanics concepts. In addition, from data provided by the physicomathematical algorithm, two hybrid model were developed. One of them uses regression statistical and the other one uses support vector regression which is a type of machine learning. With good prospects of the advances in research, it is expected in future work, a more accurate model that can be used in other possible situations of dam-break. Moreover the model can be applied to situations involving computational fluid dynamics in general