Performance Analysis of an Improved Particle Swarm Optimization and the Standard Particle Swarm Optimization
Patrick O. M. Ogutu1, Nicholas Oyie2, Winston Ojenge3
1Patrick O. M. Ogutu, Department of Electrical and Electronic Engineering, Murang’a University, Nairobi, Kenya, Kenya.
2Dr. Nicholas Oyie, Department of Electrical and Electronic Engineering, Murang’a University, Nairobi, Kenya, Kenya.
3Dr. Winston Ojenge, Department of Electrical and Electronic Engineering, Murang’a University, Nairobi, Kenya, Kenya.
Manuscript received on 06 September 2023 | Revised Manuscript received on 13 September 2023 | Manuscript Accepted on 15 October 2023 | Manuscript published on 30 October 2023 | PP: 37-42 | Volume-13 Issue-1, October 2023 | Retrieval Number: 100.1/ijeat.A42981013123 | DOI: 10.35940/ijeat.A4298.1013123
Open Access | Editorial and Publishing Policies | Cite | Zenodo | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Many industries employ different modes of control when it comes to PID parameter tuning. Previous authors have addressed the problem of tuning a control system for linear and nonlinear systems; however, the level of error reduction in system performance has not been achieved effectively. Hence, this study focuses on improving particle swarm optimisation using an improved Algorithm for PID parameter tuning. This paper addresses the optimisation of PID parameters using an improved PSO algorithm for nonlinear systems. Particle swarm optimisation is used to tune the PID parameters, ensuring an enhanced system response and operation. The PSO was deployed in a nonlinear system for application and validation of results achieved through PID tuning of the standard parameters on the MATLAB Simulink platform. The study ensured that the PID parameters were effectively tuned by applying an improved PSO Algorithm to the plant process. The research employed a standard nonlinear system to depict a real-life situation and an Improved Particle Swarm Optimisation Algorithm to analyse and compare the improved behaviour in the MATLAB/Simulink toolbox, as applied to the PID parameters. Finally, it was logically realised that the enhanced PSO Algorithm system response was significantly better compared to the non-PSO tuned system. The simulation was performed on the plant transfer function using the MATLAB and Simulink platforms, with various parameter choices and situations, and realisations were made from the obtained data. As the iteration was increased from 10, 50, and 100, there was a significant reduction in ITAE error from 0.054806 to a minimum of 0.01900, which is far better than the SPSO algorithm. SPSO reduces the error from 0.065143 to 0.020476. It was noted that the system behaviour was significantly improved in terms of settling time and peak overshoot for IPSO.
Keywords: MATLAB Simulink, PID parameters, Iteration, nonlinear system, Particle swarm optimization Algorithm
Scope of the Article: Swarm Intelligence