Improvement and application of particle swarm optimization algorithm
Particle Swarm Optimization (PSO) remains straightforward and has many scientific and engineering applications. Most real-world optimization problems are nonlinear and discrete with local constraints. The PSO algorithm encounters issues such as inefficient solutions and early convergence. It works b...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Published: |
SAGE Publications
2025
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/120545/ |
| _version_ | 1848868201208741888 |
|---|---|
| author | Deevi, Durga Praveen Kodadi, Sharadha Allur, Naga Sushma Dondapati, Koteswararao Chetlapalli, Himabindu Perumal, Thinagaran |
| author_facet | Deevi, Durga Praveen Kodadi, Sharadha Allur, Naga Sushma Dondapati, Koteswararao Chetlapalli, Himabindu Perumal, Thinagaran |
| author_sort | Deevi, Durga Praveen |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | Particle Swarm Optimization (PSO) remains straightforward and has many scientific and engineering applications. Most real-world optimization problems are nonlinear and discrete with local constraints. The PSO algorithm encounters issues such as inefficient solutions and early convergence. It works best with well-tuned attribute weights, improving case retrieval accuracy. Using case-based reasoning to optimize pressure vessel models improves PSO performance, resulting in predictions closer to true values and fulfilling real-world engineering requirements. When developed for a group of Wheeled Mobile Robots (WMR), a Fault Tolerant Formation Control (FTFC) technique is designed to protect against serious actuator defects. At the outset of the study, the WMRs are arranged very orderly. When severe actuator faults impede certain robots, functioning wheeled mobile robots (WMRs) adjust their formation to reduce the consequences of the malfunction. An ideal assignment technique assigns new duties to each functioning robot, followed by evolutionary algorithms and Particle Swarm Optimization (PSO) to design pathways to the reconfigured positions. The CPTD approach uses a piecewise linear approximation to overcome obstacles in optimization problems with continuous switch inputs. This method combines CPTD with the Genetic Algorithm and PSO (GAPSO), resulting in an effective strategy for dynamic formation reconfiguration and path optimization. This holistic method reduces the time required to achieve the configuration while considering the physical restrictions of WMRs and avoiding collisions. Finally, real-world tests are performed to verify the proposed Algorithm's efficacy compared to existing optimization methods. The proposed GAPSO algorithm will achieve an average relative error reduction of 2%, accuracy will improve by 96%, the maximum performance will be achieved by 95%, the F1 score will develop by 95%, and the training error cure rate will improve by 94%. |
| first_indexed | 2025-11-15T14:48:37Z |
| format | Article |
| id | upm-120545 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-15T14:48:37Z |
| publishDate | 2025 |
| publisher | SAGE Publications |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1205452025-10-06T01:57:51Z http://psasir.upm.edu.my/id/eprint/120545/ Improvement and application of particle swarm optimization algorithm Deevi, Durga Praveen Kodadi, Sharadha Allur, Naga Sushma Dondapati, Koteswararao Chetlapalli, Himabindu Perumal, Thinagaran Particle Swarm Optimization (PSO) remains straightforward and has many scientific and engineering applications. Most real-world optimization problems are nonlinear and discrete with local constraints. The PSO algorithm encounters issues such as inefficient solutions and early convergence. It works best with well-tuned attribute weights, improving case retrieval accuracy. Using case-based reasoning to optimize pressure vessel models improves PSO performance, resulting in predictions closer to true values and fulfilling real-world engineering requirements. When developed for a group of Wheeled Mobile Robots (WMR), a Fault Tolerant Formation Control (FTFC) technique is designed to protect against serious actuator defects. At the outset of the study, the WMRs are arranged very orderly. When severe actuator faults impede certain robots, functioning wheeled mobile robots (WMRs) adjust their formation to reduce the consequences of the malfunction. An ideal assignment technique assigns new duties to each functioning robot, followed by evolutionary algorithms and Particle Swarm Optimization (PSO) to design pathways to the reconfigured positions. The CPTD approach uses a piecewise linear approximation to overcome obstacles in optimization problems with continuous switch inputs. This method combines CPTD with the Genetic Algorithm and PSO (GAPSO), resulting in an effective strategy for dynamic formation reconfiguration and path optimization. This holistic method reduces the time required to achieve the configuration while considering the physical restrictions of WMRs and avoiding collisions. Finally, real-world tests are performed to verify the proposed Algorithm's efficacy compared to existing optimization methods. The proposed GAPSO algorithm will achieve an average relative error reduction of 2%, accuracy will improve by 96%, the maximum performance will be achieved by 95%, the F1 score will develop by 95%, and the training error cure rate will improve by 94%. SAGE Publications 2025 Article PeerReviewed Deevi, Durga Praveen and Kodadi, Sharadha and Allur, Naga Sushma and Dondapati, Koteswararao and Chetlapalli, Himabindu and Perumal, Thinagaran (2025) Improvement and application of particle swarm optimization algorithm. Intelligent Decision Technologies, 19 (4). pp. 2347-2367. ISSN 1872-4981; eISSN: 1875-8843 https://journals.sagepub.com/doi/10.1177/18724981251331781 10.1177/18724981251331781 |
| spellingShingle | Deevi, Durga Praveen Kodadi, Sharadha Allur, Naga Sushma Dondapati, Koteswararao Chetlapalli, Himabindu Perumal, Thinagaran Improvement and application of particle swarm optimization algorithm |
| title | Improvement and application of particle swarm optimization algorithm |
| title_full | Improvement and application of particle swarm optimization algorithm |
| title_fullStr | Improvement and application of particle swarm optimization algorithm |
| title_full_unstemmed | Improvement and application of particle swarm optimization algorithm |
| title_short | Improvement and application of particle swarm optimization algorithm |
| title_sort | improvement and application of particle swarm optimization algorithm |
| url | http://psasir.upm.edu.my/id/eprint/120545/ http://psasir.upm.edu.my/id/eprint/120545/ http://psasir.upm.edu.my/id/eprint/120545/ |