Determining the best practice – Optimal designs of composite helical structures using Genetic Algorithms

Abstract

Abstract Composite helical structures (CHSs) can store and release strain energy through elastic deformation, which have been used in automobile and aerospace structures. The compressive stiffness to weight ratio is the core in the design of these structures, requiring an optimal geometric configuration. Seven state-of-the-art Genetic Algorithms were employed and benchmarked to optimise two conflicting objectives, maximising the compressive stiffness while minimising the weight. All design variables that having effects on the compressive stiffness and weight of CHSs had been considered, which are the helix angle, the number of active coils, the helix diameter, the outer and inner diameter of cross-section, and the ply angle. A quantitative analysis method, mimicked inverted generational distance (mIGD), was used to determine the best practice of Genetic Algorithms. This study shows the selection of the Genetic Algorithm is crucial and multi-objective evolutionary algorithm based on decomposition (MOEA/D) is the best solver on searching the designs of the maximum compressive stiffness and the minimum weight.