Implementação do algoritmo inspirado no comportamento das colônias de formigas para a problemática da condução de trens de carga
The behavior-based algorithm of ant colonies is a metaheuristic that had its trigger in the 1990 by Marco Dorigo. The main idea is based on the behavior of real ants and their ability to find the best path between their nest and food. This skill is based on exploring pheromone trails, which are chem...
| Autor principal: | Bernardo, Wesley |
|---|---|
| Formato: | Trabalho de Conclusão de Curso (Graduação) |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2020
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/16006 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
The behavior-based algorithm of ant colonies is a metaheuristic that had its trigger in the 1990 by Marco Dorigo. The main idea is based on the behavior of real ants and their ability to find the best path between their nest and food. This skill is based on exploring pheromone trails, which are chemicals left on the way to the nest each time food is found. Due to this cooperative and efficient search behavior, they build better path alternatives to find the food. In the scenario of freight train driving, the route of a trip was subdivided into fragments, for a fractional representation of the entire trip, named as the point of measurement. Each measurement point represents the moment of application of an acceleration point, where an acceleration point is represented as an artificial ant. Therefore, Ant Colony Optimization was developed, which was integrated in a computational environment that simulates the real driving of a freight train. In this sense, it was possible to construct a set of acceleration points and apply them during a trip, without causing damage to the track and the train, thus showing the applicability of a computational algorithm in a problem with real characteristics. The application of the set of acceleration points developed with the optimization resulted in a similarity index of 87% compared to the driving performed by the driver. Above all, optimization also resulted in lower fuel consumption and time spent on the entire journey compared to driving a driver. |
|---|