PPG.EGPSA/ITEGAM

URI permanente desta comunidadehttps://rigalileo.itegam.org.br/handle/123456789/1

A comunidade dispõe da produção técnica e científica do Programa de Pós-graduação em Engenharia, Gestão de Processos, Sistema e Ambiental (PPG.EGPSA) do Instituto de Tecnologia e Educação Galileo da Amazônia (ITEGAM), fruto da atividade de pesquisa e desenvolvimento (P&D). É possível acessar os trabalhos de conclusão do programa de pós-graduação, artigos e livros vinculados a pesquisa, desenvolvimento, inovação e extensão.

Navegar

Resultados da Pesquisa

Agora exibindo 1 - 4 de 4
  • Imagem de Miniatura
    Item
    Modelo Híbrido com Redes Neurais Artificiais e Algoritmos Evolucionários para Otimização do Consumo de Combustível em Embarcações que Utilizam Motor de Combustão Interna a Diesel
    (Instituto Nacional da Propriedade Industrial (INPI), 2021-07-27) SIQUEIRA JUNIOR, Paulo Oliveira; NASCIMENTO, Manoel Henrique Reis; http://lattes.cnpq.br/0850846128967798
    This document describes a computer program registered under number BR512023001328-5, which presents a hybrid model combining artificial neural networks and evolutionary algorithms. The objective of the program is to optimize fuel consumption on vessels that use diesel internal combustion engines. The hybrid approach allows to improve energy efficiency and reduce operational costs, being especially relevant for the naval sector. The program uses advanced artificial intelligence and optimization techniques, providing a powerful tool for marine engineers and operators seeking sustainable and efficient solutions. Program registration is valid for 50 years from January 1, 2022.
  • Imagem de Miniatura
    Item
    JP- NSGA-III
    (Instituto de Tecnologia e Educação Galileo da Amazônia, 2023-05-16) SIQUEIRA JUNIOR, Paulo Oliveira; LEITE, Jandecy Cabral
    Registration of a computer program registered with the National Institute of Industrial Property (INPI), work derived from the dissertation "Hybrid model with artificial neural networks and evolutionary algorithms for optimizing fuel consumption on vessels using diesel internal combustion engines".
  • Imagem de Miniatura
    Item
    Computational meta-heuristics based on Machine Learning to optimize fuel consumption of vessels using diesel engines
    (Instituto de Tecnologia e Educação Galileo da Amazônia, 2021-05-01) SIQUEIRA JUNIOR, Paulo Oliveira; NASCIMENTO, Manoel Henrique Reis; SILVA, Ítalo Rodrigo Soares; PARENTE, Ricardo Silva; FONSECA JUNIOR, Milton; LEITE, Jandecy Cabral;
    With the expansion of means of river transportation, especially in the caseof small and medium-sized vessels that make routes of greater distances, the cost of fuel, if not taken as an analysis criterion for a larger profit margin, is considered to be a primary factor , considering that the value of fuel specifically diesel to power internal combustion machines is high. Therefore, the use of tools that assist in decision making becomes necessary, as is the case of the present research, which aims to contribute with a computational model of prediction and optimization of the best speed to decrease the fuel cost considering the characteristics of the SCANIA 315 machine. propulsion model, of a vessel from the river port of Manaus that carries out river transportation to several municipalities in Amazonas. According to the results of the simulations, the best training algorithm of the Artificial Neural Network (ANN) was the BFGS Quasi-Newton considering the characteristics of the engine for optimization with Genetic Algorithm (AG).
  • Imagem de Miniatura
    Item
    Modelo híbrido com redes neurais artificiais e algoritmos evolucionários para otimização do consumo de combustível em embarcações que utilizam motor de combustão interna a diesel
    (Instituto de Tecnologia e Educação Galileo da Amazônia, 2021-07-27) SIQUEIRA JUNIOR, Paulo Oliveira; NASCIMENTO, Manoel Henrique Reis
    The waterway transport is the main means of locomotion in the North Region, feeding cities through boats, speedboats, ships and ferries with the transport of goods and/or passengers. However, one of the problems of this modality of transport is the cost of supply, considering that the lack of technologies/resources that allow or facilitate a strategic vision of the business is a reality. The flow of passenger transport concentrates an average turnover of 9 million people, while the cargo transport with approximately 3 million, both distributed throughout the northern region. This fact characterizes a considerable demand in the water transportation sector, bringing to light the perspective of this research to study methods of analysis and support for decision making on the basis of fuel consumption. This dissertation aims to present the results about the development of an optimization model of fuel consumption considering the optimal speed for small vessels that operate on a regular basis in the Manaus River Port. In view of this, the research meets the objectives of mapping the variables related to the specifications of the vessel and engine to be analyzed, present the methods and results about the development of the computational hybrid model for optimization of fuel consumption when considering as a parameter of regulation the optimal speed for minimizing the predictor variable and the distance of the projected path for 3 scenarios: Manaus to Itacoatiara, Manaus to Barcelos and Manaus to Parintins, determine by statistical error analysis the best model of Artificial Neural Network (ANN) when considering number of neurons, hidden layers, activation functions (hyperbolic Tangent, Sigmoid and Linear) and training algorithm being the latter 12 possibilities each with different objectives and convergence strategies, To test the hybrid model analyzing the performance of 3 optimization algorithms (Particle Swarm, Genetic Algorithm and Simulated Annealing) as a function of computational cost and error rate in each generation of elites, and finally, to present the simulation results of the 3 scenarios mentioned above when using the hybrid model with the winning algorithm as a function of the requirements mentioned above. To analyze the results provided by the simulations, scenarios and tests for the acquisition of thebest models, the statistical techniques of Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Normalized Root Mean Squared Error (NRMSE) and Mean Absolute Percentage Error (MAPE) were used, the data regarding the vessel were acquired through meetings and interviews with experts (vessel owner) to formalize a specific database for the study delimited in this dissertation, 12 training algorithms were used to choose the best ANN, according to the results the Levenberg-Marquardt presented 100% correlation between the output variables and the Particle Swarm obtained the lowest computational cost in relation to the others proving the effectiveness of the computational hybrid model.