Self-adjusting a Genetic Algorithm Using Fuzzy Logic Techniques
One of the most important tasks in approximately solving an optimisation problem is to adjust the parameters of the metaheuristic used as a solution method. As the metaheuristics are usually general in purpose, it is necessary to make adjustments to them for each optimisation problem to which they a...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2017 |
| País: | México |
| Institución: | Tecnológico Nacional de México |
| Repositorio: | Redalyc-TNM |
| OAI Identifier: | oai:redalyc.org:265253895002 |
| Acceso en línea: | https://www.redalyc.org/articulo.oa?id=265253895002 |
| Access Level: | acceso abierto |
| Palabra clave: | Computación Fuzzy logic parameter tuning genetic algorithm |
| Sumario: | One of the most important tasks in approximately solving an optimisation problem is to adjust the parameters of the metaheuristic used as a solution method. As the metaheuristics are usually general in purpose, it is necessary to make adjustments to them for each optimisation problem to which they are applied to get high-quality solutions. In this paper, we propose the use of a Type 1 Fuzzy Inference System and a Type 2 Fuzzy Logic Inference System to select the operators of a Genetic Algorithm during execution time to solve a set of ten test functions from the literature. The results of computational experiments show that the fuzzy selection of operators improves the performance of the original GA on nine of the ten test functions with practically the same execution time. |
|---|