A new artificial ecosystem-based optimization integrated with Nelder-Mead method for PID controller design of buck converter
Over the last decade, there has been a constant development in control techniques for DC-DC power converters which can be classified as linear and nonlinear. Researchers focus on obtaining maximum efficiency using linear control techniques to avoid complexity although nonlinear control techniques ma...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Elsevier
2022
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/692389d4cae742b7802722d6d55ec873 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Over the last decade, there has been a constant development in control techniques for DC-DC power converters which can be classified as linear and nonlinear. Researchers focus on obtaining maximum efficiency using linear control techniques to avoid complexity although nonlinear control techniques may achieve full dynamic capabilities of the converter. This paper has a similar purpose in which a novel hybrid metaheuristic optimization algorithm (AEONM) is proposed to design an optimal PID controller for DC-DC buck converter’s output voltage regulation. The AEONM employs artificial ecosystem-based optimization (AEO) algorithm with Nelder-Mead (NM) simplex method to ensure optimal PID controller parameters are efficiently tuned to control output voltage of the buck converter. Initial evaluations are performed on benchmark functions. Then, the performance of AEONM-based PID is validated through comparative results of statistical boxplot, non-parametric test, transient response, frequency response, time-domain integral-error-performance indices, disturbance rejection and robustness using AEO, particle swarm optimization and differential evolution. A comparative performance analysis of transient and frequency responses is also performed against simulated annealing, whale optimization and genetic algorithms for further performance assessment. The comparisons have shown the proposed hybrid AEONM algorithm to be superior in terms of enhancing the buck converter’s transient and frequency responses. |
|---|