Chaos-based Swarm Intelligence Algorithms for Optimal Design of Truss Structures

Abstract

The incorporation of chaos functions into metaheuristic algorithms leads to significant progress in the results of optimal design of truss structures. Chaos functions, by forming chaotic mutations, create the necessary conditions to create a balance between exploration and exploitation. With this balance, the algorithm is saved from premature convergence and, by forming chaotic series, a jump from local optima to global optima is achieved. In this research, chaos functions are formed in the basic steps of three meta-heuristic swarm intelligence algorithms and three new chaos algorithms. These algorithms include the Chaotic Grey Wolf Optimizer (CGWO), the Chaotic Crow Search Algorithm (CCSA), and the Chaotic Cyclical Parthenogenesis Algorithm (CCPA). To improve the optimization results, three different scenarios are examined and the chaotic results are compared with the standard case. In these scenarios, chaos series replace the exploration, exploitation, or both stages simultaneously.