Crashworthiness design of 3D lattice-structure filled thin-walled tubes based on data mining

Lattice structures and thin-walled tubes are two types of energy-absorbers widely studied applied in engineering practice. In this study, a new type lattice-structure filled tube (LFT) was proposed by combining these energy absorbers designed via explainable data mining method. LFT, BCC-Z (Body- centred cubic unit cell with vertical struts) lattice structure into square tube. Then, the LFT parameterized five design variables, that is, number layers longitudinal direction (n), cells transverse (m), rod diameter (d), thickness (t) height difference between (h). Using Latin hypercubic sampling algorithm, 150 cases were generated. Numerical models then developed to simulate their crush behavior, simulation dataset used for mining. The results showed (1) Filling BBC-Z can significantly improve absorption (EA) capacity structure. (2) SEA (Specific Energy Absorption) increased increasing d, m n. (3) decision trees generated process indicated d is key variable has most significant impact on EA, followed (4) rules build LFTs high EA efficiency (SEA ≥ 16 kJ/kg CFE (Crush Force Efficiency)≥45%), total 6 kJ) lightweight Mass ≤ 0.45 kg) obtained from trees. ideal configurations corresponding three objectives are: > 2 mm, n 3 efficiency; EA; 2, 4 t 1.7 mm lightweight.