Adaptive Chaotic Marine Predators Hill Climbing Algorithm for Large-scale Design Optimisations

Meta-heuristic algorithms have been effectively employed to tackle a wide range of optimisation issues, including structural engineering challenges. The the shape and size large-scale truss structures is difficult due nonlinear interplay between cross-sectional nodal coordinate pressures structures. Recently, it was demonstrated that newly proposed Marine Predator Algorithm (MPA) performs very well on mathematical MPA meta-heuristic simulates essential hunting habits natural marine predators. However, this algorithm has some disadvantages, such as becoming locked in locally optimal solutions not exhibiting high level exploratory behaviour. This paper proposes two hybrid predator algorithms, Nonlinear (HNMPA) Nonlinear-Chaotic (HNCMPA), improved variations paired with hill-climbing (HC) technique for form size. major advantage these techniques are they seek overcome MPA’s disadvantages by using values prolonging exploration phase chaotic values; also, HC used avoid optimum solutions. In terms performance, compared fourteen well-known meta-heuristics, Dragonfly (DA), Henry Gas Solubility (HGSO), Arithmetic (AOA), Generalized Normal Distribution Optimisation (GNDO), Salp Swarm (SSA), Predators (MPA), Neural Network (NNA), Water Cycle (WCA), Artificial Gorilla Troops Optimiser (GTO), Gray Wolf (GWO), Moth Flame (MFO), Multi-Verse (MVO), Equilibrium (EO), Cheetah (CO). Furthermore, seven were chosen test HNCMPA performance benchmark sets, MPA, MVO, PSO, MFO, SSA, GWO, WOA. results experiment demonstrate put forth surpass previously established meta-heuristics field optimisation, encompassing both traditional CEC problems, margin over 95% attaining superior ultimate solution. Additionally, regards solving difficulties real-world challenge, outcomes indicate boost 65% obtaining significantly better problems involving 260-bar 314-bar; conversely, case 340-bar improvement rate slightly lower at almost 25%.