Hybrid multi-objective Harris Hawk optimization algorithm based on elite non-dominated sorting and grid index mechanism. (October 2022)

• Record Type: Journal Article
• Authors: Wang, Min; Wang, Jie-Sheng; Song, Hao-Ming; Zhang, Min; Zhang, Xing-Yue; Zheng, Yue; Zhu, Jun-Hua
• Is Part Of: Advances in engineering software. Volume 172(2022)
• Publication Date: 2022-10
• Subjects: Multi-objective optimization -- Pareto front -- HHO algorithm -- Elite non-dominant sorting -- Grid indexing mechanism; Computer-aided engineering -- Periodicals; Engineering -- Computer programs -- Periodicals; Engineering -- Software -- Periodicals; Periodicals; 620.0028553
• DOI: 10.1016/j.advengsoft.2022.103218
• Languages: English
• ISSNs: 0965-9978

Abstract

Highlights: A hybrid multi-objective Harris Hawk optimization algorithm was proposed. Elite non-dominated sorting and grid index mechanism combined to constitute the Pareto optimal solution. Hybrid structure to maintain population diversity and improve its effectiveness. Abstract: In order to find Pareto optimal solution set uniformly distributed along all objectives, a Hybrid Multi-Objective Harris Hawk Optimization Algorithm (H-MOHHO) was proposed based on elite non-dominated sorting and grid indexing mechanism. In order to maintain and improve the coverage of Pareto optimal solution, a method combining the two terms is adopted to obtain the optimal Pareto optimal solution set. Firstly, a non-dominated ranking mechanism based on elite was used to assign rank and sum to select the best solution set, and then the archived grid index mechanism with update mechanism was used to select the final solution set. This hybrid structure can not only obtain the optimal Pareto solution set but also keep the diversity of the population and improve the effectiveness of solving multi-objective optimization problems. In order to verify the performance of the proposed H-MOHHO algorithm, 22 test functions and 4 multi-objective engineering problems are used for simulation, and four performance indexes are compared with Multi-Objective Particle Swarm Optimization (MOPSO), Non-dominated Sorting Genetic Algorithm II (NSGA-II), Multi-Objective Ant Lion Optimizer (MOALO), Multi-Objective Salp Swarm Algorithm (MSSA) and Multi-Objective Dragonfly Algorithm (MODA). Experimental results show that the proposed H-MOHHO algorithm has better competitiveness and applicability.