Hybridised differential evolution and equilibrium optimiser with learning parameters for mechanical and aircraft wing design

Abstract

Metaheuristics (MHs) have been widely used for aeroelastic optimisation of aircraft wings and other types of aircraft structures. Using such methods offers some advantages e.g. flexibility for coding, robustness, global optimisation capability, and a derivative-free feature. Moreover, unconventional design problems can be posed when using metaheuristics. This paper proposes a new hybrid algorithm, named HDEEO-LP, with a learning control parameter for aeroelastic optimisation. The new optimiser is obtained from hybridising differential evolution and the recently invented equilibrium optimisation, while a learning scheme for control parameter tuning is integrated. The new method is tested against a number of established and recently invented MHs, such as a grey wolf optimiser (GWO), a salp swarm algorithm (SSA), an equilibrium optimiser (EO), an artificial bee colony (ABC), teaching-learning based optimisation (TLBO), water cycle algorithm (WCA), self-adaptive spherical search algorithm (SASS) using the CEC-RW-2020 test suite and the Goland wing aeroelastic optimisation. The results reveal that the proposed hybrid algorithm is among the top performers.