Bio-inspired optimization algorithms applied to the GAPID control of a Buck converter
The objective of this work is to realize a comparative study of which bio-inspired metaheuristic strategies are the best to be used when finding the optimal gains for a Gaussian Adaptive PID control system applied to a Buck Converter. This comes from two necessities, the first one is that to surpass...
| Autor principal: | Itaborahy Filho, Marco Antonio |
|---|---|
| Formato: | Dissertação |
| Idioma: | Inglês |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2021
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/25060 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
The objective of this work is to realize a comparative study of which bio-inspired metaheuristic strategies are the best to be used when finding the optimal gains for a Gaussian Adaptive PID control system applied to a Buck Converter. This comes from two necessities, the first one is that to surpass the performance of linear control systems a more advanced strategy such as adaptive control is required; the second is that since there are no deterministic optimization methods for such strategies, the use of metaheuristics is deemed necessary, and according to the No Free Lunch theorem such comparative studies ought to be made to understand what optimization techniques ares better suitable for such a problem. The chosen techniques were the Genetic Algorithm, the Particle Swarm Optimization and the Differential Evolution. The Gaussian function is employed as the adaptive rule because it is a smooth curve with smooth derivatives that avoids the potential of chattering and abrupt control changes. After evaluating the responses found from simulating all variations of each of the optimization strategies we were able to create an analytical and statistical comparison of each technique and show how an adaptive control system can create a faster and more robust response when compared to the linear PID. |
|---|