PID control optimisation using an enhanced hybrid PSOGA algorithm with reinforcement-driven modifications

Journal article

Authors/Editors

Strategic Research Themes

Publication Details

Author list: K.; Assawinchaichote, W.; Shi, M.; Pattaramalai, S.; Zhang, H.

Publisher: Taylor and Francis Group

Publication year: 2025

Journal: International Journal of Systems Science (0020-7721)

ISSN: 0020-7721

eISSN: 1464-5319

URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-105026374055&doi=10.1080%2F00207721.2025.2603559&partnerID=40&md5=5b6e8775f13f2224d65ad11e451aed7a

Languages: English-Great Britain (EN-GB)

View on publisher site

Abstract

Optimising PID controllers in complex environments requires algorithms that are efficient, adaptive and capable of learning. This research presents a Reinforcement-Driven Modified PSOGA (MPSOGA) for PID controller optimisation. The algorithm embeds Q-learning into the PSO-GA framework, enabling real-time adaptation of search strategies and improving the balance between exploration and exploitation. This reinforcement-driven mechanism enhances convergence reliability and overall solution quality. Simulation studies on benchmark PID control tasks demonstrate that MPSOGA achieves faster convergence, reduced overshoot and improved transient stability compared with PSO, GA, GWO, PSOGA and mJS. Additional evaluations on standard optimisation benchmarks indicate robustness across diverse problem settings. The results highlight the potential of MPSOGA as a practical and adaptable approach to control optimisation, contributing to ongoing research on reinforcement-driven metaheuristics. © 2025 Informa UK Limited, trading as Taylor & Francis Group.

Keywords

No matching items found.