Data-driven design and crashworthiness optimization of novel hybrid TPMS lattice structures
摘要
Design and development of novel hybrid TPMS structures were achieved by integrating the level-set functions of different base Triply Periodic Minimal Surfaces (TPMS) using a weighted calculation method. Firstly, the mechanical response of the base-TPMS structures was investigated through uniaxial compression experiments, which validated the high simulation accuracy of the established numerical model. Subsequently, a systematic numerical simulation study was conducted to explore the influence of design parameters on the crashworthiness of the hybrid TPMS structures. Results indicated that within the constructed hybrid TPMS structures, the Primitive + Gyroid (PG) and Gyroid + Diamond (GD) types exhibited a significantly enhanced energy absorption capacity. Specifically, the GD structure achieved a 73.3% and 73.8% increase in specific energy absorption (SEA) over the base Gyroid (G) and Diamond (D) structures, respectively. Further analysis revealed that the topological configuration of the TPMS is the dominant factor influencing its crashworthiness. A comparative study with traditional lattice structures demonstrated a substantial advantage of the hybrid TPMS lattice, with the GD configuration exhibiting 6.9% higher energy absorption than the strongest performing traditional lattice (SC) at equivalent mass. Finally, this study integrated the Crowned Porcupine Optimized Backpropagation (CPOBP) algorithm with the Non-dominated Sorting Whale Optimization Algorithm (NSWOA) multi-objective optimization framework to achieve the collaborative optimization of key design parameters for the hybrid TPMS structures, providing a novel solution and a theoretical foundation for their engineering applications in energy-absorbing components.