Obtenção do modelo matemático e mapa de calor do consumo de combustível de um caminhão na mineração à céu aberto, utilizando algoritmo genético e regressão linear múltipla

• Ano de defesa: 2019
• Autor(a) principal: Oliveira, Jean Carlos de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Algoritmo Genético; Genetic Algorithm; Regressão Linear Múltipla; Multiple Linear Regression; Consumo de Combustível; Fuel Consumption; Redução de Custos; Cost Reduction; Rota Econômica; Economic Route; Engenharia Elétrica; Electrical Engineering; Análise de Regressão; Analysis of Regression; Redes Elétricas - Redução de Custos e Tempo; Electrical Networks - Reduction of Costs and Time; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA::ELETRONICA INDUSTRIAL, SISTEMAS E CONTROLES ELETRONICOS::AUTOMACAO ELETRONICA DE PROCESSOS ELETRICOS E INDUSTRIAIS
• Link de acesso: https://repositorio.ufu.br/handle/123456789/25515; http://dx.doi.org/10.14393/ufu.di.2019.39

The global business environment forces organizations to improve its processes and services as a means to success and survival. Seeking for optimization has become essential for better use of resources, reducing costs and maximizing results. This work proposes, with the help of both tools, Multiple Linear Regression (MLR) and Genetic Algorithm (GA), to obtain mathematical models and a heat map of fuel consumption, of a mining truck in operation based on its routes physical characteristics. In addition to, the developed system is responsible for data processing, the genetic algorithm parameterization and obtained models validation by comparison among them and reference values. Through the mathematical model and the heat map, it is possible to create routines, provide information for truck dispatch systems and achieve consequent mine-operating costs reduction The developed system plays an important during the definition and creation of new routes, helping to indicate the economic one. Moreover, it can be useful when reviewing existing routes, supporting changes in their topography. The research uses real and current data collected from a telemetry system of an open pit mine. The Multiple Linear Regression was performed in MS Excel® environment, while the Genetic Algorithm was implemented at Matlab® software. Studies which intend to reduce fuel consumption, provide significant information for open-pit mining companies once this consumption represents a large part of operating costs. It is also worth emphasizing the benefit of reducing greenhouse gases emissions, which interest and concern are general.