Improving teaching-learning-based optimization algorithm with golden-sine and multi-population for global optimization

Aosheng Xing, Yong Chen, Jinyi Suo and Jie Zhang

Mathematics and Computers in Simulation (MATCOM), 2024, vol. 221, issue C, 94-134

Abstract: Teaching-learning-based optimization (TLBO) is an optimization algorithm that has become very popular in recent years and has shown excellent performance in solving many scientific development issues. However, several recent studies have revealed that TLBO struggles with handling complicated issues and has a significant propensity to move back to the original. This research suggests a unique golden-sine and multi-population teaching-learning-based algorithm (GMTLBO) to address these issues. The main innovations of the algorithm are: Firstly, Instead of using the conventional random approach to find the initial individuals, the good point set strategy is utilized, which results in a more uniform initialization of the number of individuals in the search space and enhances the level of accuracy of the initial solution. Second, in the teacher phase, the origin offset issue is resolved by using the golden-sine search model to maintain the appropriate balance between worldwide exploration and regional exploitation. Finally, we incorporated the multi-population learner phase following the learner phase. Based on individual fitness values, the population is segmented into three identically sized subpopulations. Distinct mechanisms are then employed to allocate movement strategies to each sub-population for additional diversification and to prevent the algorithm from converging on regionally optimal solutions. Several validation tests were carried out on 23 traditional standard functions and the CEC2017 and CEC2019 test sets for assessing the effectiveness of the GMTLBO algorithm for resolving global optimization issues. The results demonstrate that GMTLBO converges faster and solves with higher accuracy compared to other algorithms. Additionally, GMTLBO was tested on four engineering design problems to assess its capability to solve restricted optimization issues. The suggested algorithm exhibits outstanding efficacy and competitiveness, according to experimental data.

Keywords: Teaching-learning algorithm; Global optimization; Golden-sine search model; Multi-population optimization (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0378475424000570
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:matcom:v:221:y:2024:i:c:p:94-134

DOI: 10.1016/j.matcom.2024.02.008

Access Statistics for this article

Mathematics and Computers in Simulation (MATCOM) is currently edited by Robert Beauwens

More articles in Mathematics and Computers in Simulation (MATCOM) from Elsevier
Bibliographic data for series maintained by Catherine Liu ().