.png){kind=icon}
.png){kind=icon}
Return
Opposition-based moth swarm algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2021.115481.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Nowadays, resource-optimizing techniques are required in many engineering areas to obtain the most appropriate solutions for complex problems. For this reason, there is a trend among researchers to improve existing swarm-based algorithms through different evolutionary techniques and to create new population-based methods that can accurately explore the feature space. The recently proposed Moth swarm algorithm (MSA) inspired by the orientation of moths towards moonlight is an associative learning mechanism with immediate memory that uses Le ' vy mutation to cross-population diversity and spiral movement. The MSA is a population-based method used for tackling complex optimization problems. It presents an adequate capacity for exploration and exploitation trends; however, due to its nature of operators, this type of method is prone to get stuck in sub-optimal locations, which affects the speed of convergence and the computational effort to reach better solutions. To mitigate these shortcomings, this paper proposes an improved MSA that combines opposition-based learning (OBL) as a mechanism to enhance the exploration drifts of the basic version and increase the speed of convergence to obtain more accurate solutions. The proposed approach is called OBMSA. It has been tested for solving three classic engineering design problems (welded beam, tension/compression spring, and pressure vessel designs) with constraints, 19 benchmark functions comprising 7 unimodal, 6 multimodal, and 6 composite functions. Experimental results and comparisons provide evidence that the performance and accuracy of the proposed method are superior to the original MSA. We hope the community utilizes the proposed MSA-based approach for solving other complex problems.
Keywords:
Moth swarm algorithm
Opposition-based learning
Optimization techniques
Metaheuristics
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W
Organization
View more organizations