.png){kind=icon}
.png){kind=icon}
Return
Opposition-Based Chaotic Tunicate Swarm Algorithms for Global Optimization
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2024.3359587.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Tunicate Swarm Algorithm (TSA) is a novel swarm intelligence algorithm developed in 2020. Though it has shown superior performance in numerical benchmark function optimization and six engineering design problems over its competitive algorithms, it still needs further improvements. This article proposes two improved TSA algorithms using chaos theory, opposition-based learning (OBL) and Cauchy mutation. The proposed algorithms are termed OCSTA and COCSTA. The static and dynamic OBL are used respectively in the initialization and generation jumping phase of OCTSA, whereas centroid opposition-based computing is used, in the same phases, in COCTSA. The proposed algorithms are tested on 30 IEEE CEC2017 benchmark optimization problems consists of unimodal, multimodal, hybrid, and composite functions with 30, 50, and 100 dimensions. The experimental results are compared with the classical TSA, TSA with the local escaping operator (TSA-LEO), Sine Cosine Algorithm (SCA), Giza-Pyramid Construction Algorithm (GPC), Covariance Matrix Adaptation Evolution Strategy (CMAES), Archimedes Optimization Algorithm (AOA), Opposition-Based Arithmetic Optimization Algorithm (OBLAOA), and Opposition-Based Chimp Optimization Algorithm (ChOAOBL). The statistical analysis of experimental results using the Wilcoxon Signed Rank Test establishes that the proposed algorithms outperform TSA and other algorithms for most of the problems. Moreover, high dimensions are used to validate the scalability of OCTSA and COCTSA, and the results show that the modified TSA algorithms are least impacted by larger dimensions. The experimental results with statistical analysis demonstrate the effectiveness of the proposed algorithms in solving global optimization problems.
Keywords:
Tunicate swarm algorithm
swarm intelligence
metaheuristic
opposition-based learning
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W
View more organizations