Back to search

University of British Columbia

Self-adapting control parameters in particle swarm optimization

Abstract

dc:description

This study focuses on the development of a scheme for self-adapting the Particle Swarm Optimization (PSO) method to solve constrained optimization problems. PSO is a powerful nature-inspired metaheuristic optimization method. Compared to other methods, PSO has the ability to determine the optimal solution in fewer evaluations and in general performs in a more efficient and effective manner. However, researches show that the PSO method suffers from premature convergence and a dependence on the initial control settings. Due to these flaws, the application of PSO could lead to a failure in obtaining the global optimal solution. An extensive parametric sensitivity analysis was conducted to understand the impact of the individual control parameters and their respective influence on the performance of PSO. Results of the sensitivity analysis revealed that PSO was most sensitive to the inertia weight, cognitive component and social component. Modifications were performed on the original PSO algorithm to adapt the control parameters with respect to the circumstances of the particles at a specific moment. The modified PSO variant is called the Unique Adaptive Particle Swarm Optimization (UAPSO). Unique control parameters were established for each particle through using a novel term known as the evolutionary state. In the developed approach, constraints were handled by forcing the particles to learn from their personal feasible solutions only. Therefore, in the proposed method, the constraint handling technique worked in accord with the adapting scheme to ensure that the particles were adapting to the environment by directing itself to the feasible regions. Furthermore, particles were reinitialized whenever they stagnated in the design space. Verification of the performance of the proposed method was done by means of a comparative study with other well-known algorithms. The comparative study demonstrated that UAPSO proved to be effective and efficient in solving the considered problems and especially in terms of the speed of convergence. Furthermore, design of a three-bar truss was investigated through the application of UAPSO along with multiple variants of PSO. The numerical results showed the superiority of UAPSO compared to the other variants, its ability in avoiding premature convergence and its consistency and efficiency.

Degree

thesis:*
Name thesis:degree_name
Master of Applied Science - MASc
Level thesis:degree_level
master's
Discipline thesis:degree_discipline
Mechanical Engineering
Grantor dc:publisher
University of British Columbia
Year dc:date
2019

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Isiet, Mewael Daniel

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/68364

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Isiet, Mewael Daniel. Self-adapting control parameters in particle swarm optimization. master's thesis, University of British Columbia, 2019. http://hdl.handle.net/2429/68364