Improved genetic algorithm for optimization design of a three-dimensional braided composite joint

Abstract In this study, we develop an algorithm for the optimization design of a three-dimensional (3D) braided composite joint. The design is a typical optimization problem with high computational expense in terms of objective evaluation and a highly dimensional design domain. The algorithm is established in the framework of a Genetic Algorithm (GA) that integrates the Steepest Descend Method (SDM), and it is improved according to the characteristics of the abovementioned problem. A Structural Strength Computational Model (SSCM) for the mechanical performance analysis of the joint is also developed to evaluate the optimization objective , and the model is experimentally verified. Results show that structural strength increases significantly after optimization. During optimization, the coupling relationship between stress distribution and material properties leads to the problem of nonlinearity and multimodality. The GA searches globally, which prevents the result from locally converging, and provides excellent initial solutions for the SDM. Then, the SDM searches locally with high-efficiency and increases the accuracy of the solutions. This method is precise and efficient for the abovementioned optimization problem.