Desenvolvimento de método de inteligência artificial baseado no comportamento de enxames do gafanhoto-do-deserto

• Ano de defesa: 2017
• Autor(a) principal: RIBEIRO, Tiago Martins
• Orientador(a): PAUCAR, Vicente Leonardo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Maranhão
• Programa de Pós-Graduação: PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA DE ELETRICIDADE/CCET
• Departamento: DEPARTAMENTO DE ENGENHARIA DA ELETRICIDADE/CCET
• País: Brasil
• Palavras-chave em Português: Inteligência artificial; inteligência de enxames; enxame de gafanhotosdo-deserto;serotonina;algoritmos bioinspirados
• Palavras-chave em Inglês: Artificial intelligence; swarm intelligence; desert locust swarm; serotonin; bioinspired algorithms
• Área do conhecimento CNPq: Análise de Algoritmos e Complexidade de Computação
• Link de acesso: http://tedebc.ufma.br:8080/jspui/handle/tede/1294

Complex optimization problems have been studied over the years by researchers seeking better solutions, these studies have encouraged the development of several algorithms of artificial intelligence, and a part of them are bio-inspired methods, based on the behavior of populations. These algorithms target to develop techniques based on nature in search of solutions to these problems. In this work, was introduced as a purpose, an algorithm based on the behavior of locust swarms, the Locust Swarm Optimizer (LSO). The behavior of the desert locust is introduced highlighting the formation of clouds of attacks caused by a synthesized neurotransmitter monoamine, present on the insect, known as serotonin. Observing this behavior, the LSO was developed. It was compared to other known artificial intelligence techniques through 23 benchmark functions and also tested on an power system economical dispatch problem. From the point of view of the results and the ease of implementation, it can be concluded that the LSO algorithm is very competitive as compared to existing methods