PSI - Issue 42

C. Boursier Niutta et al. / Procedia Structural Integrity 42 (2022) 1449–1457 Auth r name / Structur l Integrity Procedia 00 (2019) 000 – 0 0

1456 8

(a) (b) Fig. 7. Influence of the diameter variation: a) SEA variation for different percentages of defectivity; b) confidence interval variation for different percentages of defectivity. As expected, by increasing the number of defective beams, the SEA of the structure consistently reduces. About 45% of reduction of the SEA is obtained when 50% of the beams have a defect. The confidence interval plot reveals that the location of the defect within the unit-cell is also an important factor. For a small defect population, the pristine material is able to compensate the presence of the defects and their location does not significantly affect the SEA, i.e., all the simulations return similar SEA values. Similarly, in the case of large defectivity, the defects are prevalent over the pristine material and SEA does not significantly vary with their location. On the contrary, in the middle range of defectivity, the location of the defect influences the crushing performance, resulting in an increasing scatter of the data, i.e., increasing confidence interval. Similar considerations can be done for the lack-of-fusion defect. Results of the SEA variation and the 95% confidence interval variation are reported in Fig. 8.

(a) (b) Fig. 8. Influence of the lack-of-fusion defect: a) SEA variation for different percentages of defectivity; b) confidence interval variation for different percentages of defectivity. According to Fig. 8, lack-of-fusion defects are more detrimental for the SEA of the lattice structure. When the defect affects 50% of cell beams, 88% of SEA reduction is obtained. This can be appreciated also through the confidence interval plot. The scatter of data is yet reduced at 25% of defectivity, that is the pristine material cannot compensate the presence of the defects yet at this percentage (Cf. Fig. 7b and 8b) and their location is not significant for the SEA performance of the structure. In conclusion, the analysis shows that the defects consistently affect the SEA of the lattice structure. Curves of variation of SEA with the defectivity, as those shown in Fig. 7a and 8a, can be adopted both in the design of components made of lattice structures for energy absorbers and in the quality control stage by a proper analysis of the defectivity.