A novel epsilon-dominance Harris Hawks optimizer for multi-objective optimization in engineering design problems

摘要

In this article, A Multi-Leaders Guided Harris Hawks optimizer using Epsilon-Dominance relation is developed for solving multi-objective optimization problems. For this reason, the standard HHO algorithm is equipped with a fixed-size external archive to ensure the elitism concept. On the other hand, both crowding distance computation and epsilon dominance relation are adopted when updating the archive in the hope of improving the diversity of solutions. Moreover, an efficient leader selection procedure is proposed to guarantee convergence towards less-crowded Pareto regions. Our algorithm's performance is validated on 18 test functions in all, 5 with two objectives and 13 with three objectives, and it is compared with four well-regarded algorithms, namely: Multi-objective Evolutionary Algorithm Based on Decomposition (MOEA/D), Multi-Objective Grey Wolf Optimizer (MOGWO), Multi-Objective Particle Swarm Optimization (MOPSO), and Multi-objective Salp Swarm Algorithm (MSSA). Also, it is applied to solve four engineering real-world problems, namely: Four bar truss, Speed reducer, Disk brake design, and Welded beam design problems. Inverted Generational Distance (IGD) metric and Hypervolume (HV) metric were used to quantify the behaviors of multi-objective algorithms. The obtained results show the performance of the proposed algorithm in terms of convergence and diversity for the benchmark functions and the engineering real-world problems.