Design and optimization of additively manufactured sinusoidally-corrugated multi-cell energy absorbers

Abstract This study presents the design and optimization of additively manufactured sinusoidally-corrugated multi-cell (SCMC) energy absorbers produced from PLA+ using fused deposition modeling. First, an experimental investigation was performed to evaluate the influence of geometric design on the crashworthiness performances of a simple tube without corrugation (STWC) and a baseline SCMC. The results showed that the baseline SCMC displayed a progressive and stable folding mode resulting in improvements in specific energy absorption (SEA) and crush force efficiency (CFE) of 13 % and 77 %, respectively, compared with STWC. Next, the crashworthiness performance of SCMC was optimized by using machine learning (ML) models. For that purpose, two different support vector regression (SVR) models that employed linear (SVR-L) and Gaussian (SVR-G) kernels were developed. Finally, these SVR models were integrated into a non-dominated sorting genetic algorithm II (NSGA-II) for optimization. The optimized configuration, obtained from the SVR-L model, achieved an SEA of 21.61 kJ kg −1 and a CFE of 0.732, corresponding to 37 % and 2 % improvements over the baseline SCMC and 55 % and 82 % over STWC, respectively. Overall, the results showed that combining additive manufacturing with an ML-driven optimization approach provides an efficient and reliable approach for developing high-performance energy absorbers with enhanced crashworthiness.