Intelligent algorithm-based optimization of lattice structures with asymmetric unit cells for minimizing dynamic compliance

Abstract

Lattice structures with asymmetric unit cells have attracted significant attention due to their expanded design space introduced by the variable geometry parameters. However, existing optimization methods are predominantly based on gradient algorithms, which pose great challenges for sensitivity analysis, especially in time-domain problems. Therefore, the aim of this study is to propose a sensitivity-free intelligent algorithm-based optimization framework to minimize the dynamic compliance of lattice structures with asymmetric unit cells under dynamic loads. Firstly, the geometric model of the asymmetric unit cell is constructed using the parametric level-set method, and its effective mechanical properties are obtained through numerical homogenization method. Second, the Newmark-𝛽 method is employed to calculate the transient dynamic responses of the lattice structure. Third, intelligent algorithms directly update design variables, thereby bypassing sensitivity analysis. Finally, several 2D and 3D numerical examples are provided to demonstrate the effectiveness and advantages of the proposed method. The results show the optimized lattice structures with asymmetric unit cells significantly outperform the conventional symmetric designs in both dynamic compliance and vibration amplitude. Specifically, under asymmetric boundary conditions, they are reduced by approximately 20%-30%.