Fans Optimizer: A human-inspired optimizer for mechanical design problems optimization

Abstract

This paper proposes a human-inspired algorithm for optimizing various practical engineering problems, specifically, the Fans Optimization (FO) algorithm that is inspired by the Fans economy fused in the entertainment domain. Opposing current algorithms, the FO algorithm introduces a Multi-groups mechanism (Cooperation–Competition) and a Two-characteristic individual update mechanism to balance the exploration and exploitation (E&E). Additionally, the multi-phase optimization algorithm includes Roles information and Fans-community, Fans-community Switching, Resource Competition, Information Sharing, and K-Update. In the experiments, the FO algorithm is first compared with 12 meta-heuristic algorithms in four groups of benchmark functions (selected from the IEEE CEC and GECCO competitions), demonstrating that the FO algorithm has a superior convergence performance and E&E. Moreover, the Williams test prove the superiority of the FO algorithm over the competitor algorithms. Additionally, eight class practical engineering problems verify the FO’s practical engineering applicability. Finally, the FO algorithm solves the inverse kinematic solution problem of the 9-degree of freedom serial robot arm (9-DFSRA) , demonstrating superior results over the competitor schemes. Overall, the test results prove that the FO algorithm is superior to its competitor algorithms for complex engineering problems.