三重周期极小曲面多孔结构的可恢复性机理与优化设计方法研究

电动汽车轻量化的发展趋势下，如何同时保证安全性要求成为了工程设计的难题，尤其是保证在连环相撞恶劣工况下的安全性。传统吸能结构在碰撞过程中通过发生塑性变形吸收能量，耐撞性能随着碰撞次数的增加迅速下降，不能满足多次碰撞的要求。本项目发现了一种性能新奇，具有可恢复性的轻质三重周期极小曲面（TPMS）多孔结构，旨在应用于汽车吸能结构中，提高多次碰撞工况下的吸能效果。本项目通过分析TPMS结构在不同材料、不同构型和不同循环加载工况下的可恢复性，探究其可恢复性机理，用于指导优化设计。本项目提出了一种基于机器学习和纳米压痕试验的多尺度模型建立方法，并通过在贝叶斯优化中引入基于汉明距离（Hamming distance）的协方差矩阵，解决了同时考虑离散变量和连续变量问题。本课题研究对于提高汽车吸能结构的可恢复性和吸能效率、推动我国电动汽车轻量化和安全性发展具有重要意义。

With the lightweight development of electric vehicles, it is a challenge for engineering design to simultaneously satisfy their safety requirements，especially under the severe traffic accident: vehicle pileup. The traditional energy-absorbing structure absorbs energy through plastic deformation during a collision, and its crashworthiness decreases rapidly with the increase of the number of collisions, which cannot meet the requirements of multiple collisions. This project has discovered a novel and recoverable lightweight Triply Periodic Minimal Surface (TPMS) cellular structure, which is designed to be used in automobile energy-absorbing structures to improve the energy-absorbing properties under multiple collision conditions. This project analyzes the recoverability of the TPMS structure with different materials, different topological configurations, and different cyclic loading conditions, and explores its recoverability mechanism to guide the optimization design. This project proposes a Machine Learning-based multi-scale modeling method based on nanoindentation test and introduces a covariance matrix based on Hamming distance in Bayesian optimization, which solves the problem of considering both discrete variables and continuous variables. The research of this subject is of great significance for improving the recoverability and energy absorption efficiency of the energy-absorbing structure of automobiles and promoting the lightweight and safety development of electric vehicles in China.