Dissertações

URI permanente para esta coleçãohttps://rigalileo.itegam.org.br/handle/123456789/174

Dissertações produzidas oriundas de convênios

Navegar

Resultados da Pesquisa

Agora exibindo 1 - 2 de 2
  • Imagem de Miniatura
    Item
    Sistema Automático de Balanceamento de Unidades Consumidoras Monofásicas Conectadas na Rede de Distribuição de Baixa Tensão​
    (Instituto de Tecnologia, 2018) Alex Sander Leocádio Dias; Manoel Henrique Reis Nascimento
    The dissertation presents an automatic system for load balancing in single-phase consumer units connected to the low-voltage distribution network. The objective is to minimize imbalances between phases, reducing technical losses and improving the efficiency of the electrical system. To this end, an approach based on fuzzy logic is used, which allows real-time decision-making to redistribute loads between phases. The study includes simulations and tests in a controlled environment, demonstrating the effectiveness of the proposed method in optimizing load balancing. The results indicate that the system can significantly contribute to improving the quality of the electrical energy supplied, reducing voltage variations and increasing the useful life of the network equipment.
  • Imagem de Miniatura
    Item
    Sintonia de Controladores PID Aplicada a Robôs Uniciclos Utilizando o Algoritmo do Lobo Cinzento (GWO)
    (Universidade Federal do Pará, 2018) MEDEIROS, Antonio Benjamin Leão de; MARANHÃO, Geraldo Neves de Albuquerque
    Industrial AGVs have a wide range of applications in manufacturing processes, providing flexibility and safety for operators. PID controllers are commonly used for controlling these robots. This study proposes optimizing a PID controller for a unicycle-type AGV using the Gray Wolf Optimization (GWO) algorithm, a heuristic technique inspired by the hunting behavior of gray wolves. The PID controller’s performance is compared with a conventional kinematic controller, considering different trajectories and running simulations to evaluate standard and mean errors in robot responses. The study also examines the impact of iteration numbers and search agents on algorithm performance and suggests future improvements.