An Improved Hybrid Multi-Objective Particle Swarm Optimization to Enhance Convergence and Diversity
An Improved Hybrid Multi-Objective Particle Swarm Optimization to Enhance Convergence and Diversity
复制标题
DOI：
10.1145/3583133.3596365
复制
发表时间：
2023-07
期刊：
影响因子：
--
通讯作者：
Nazrul Islam;J. Oyekan
中科院分区：
文献类型：
--
作者：
Nazrul Islam;J. Oyekan
粒子群优化算法(PSO)以其快速的收敛能力受到越来越多研究者的关注。提出了一种新的基于Pareto优势的多目标粒子群算法的改进算法。本研究旨在利用有限的迭代过程开发出一种收敛速度更快的高效勘探开发算法。研究人员已经证明，流行的EA算法可以更快地收敛，并提供有效的探索。然而，在保持种群多样性和最佳解之后，在较小的迭代结果中建立成功是有一些缺点的。因此，我们提出的方法将混合进化计算过程与粒子群算法相结合来解决多目标问题。结果表明，该方法具有很强的竞争性，能够逼近其他算法难以达到的Pareto前沿。
Particle Swarm Optimization (PSO) has received increasing attention from researchers due to its fast convergence ability. This paper proposes a new improvement in Multi-objective Particle Swarm Optimization based on Pareto dominance. This research aims to develop an efficient exploration and exploitation algorithm with faster convergence using a limited iterative process. Researchers have already proved that popular EA algorithms can converge faster as well as provide efficient exploration. However, there are some shortcomings in establishing success in smaller iterative outcomes after maintaining both population diversity as well as best optimal solutions. Therefore, our proposed approach incorporates the hybrid evolutionary computation process with PSO to address multi-objective problems. The results indicate that the proposed method is highly competitive and is able to approximate the Pareto Front where other algorithms struggle.