PSI - Issue 37

T. Oliveira et al. / Procedia Structural Integrity 37 (2022) 698–705

704

T. Oliveira et al. / Structural Integrity Procedia 00 (2022) 000–000

7

a

b

c

d

Fig. 8. Center damage scenario; DC1 (a) S FD (1) 1 − 10

; (b) S FD (2)

; (c) S FD (3)

; (d) S FD (4)

.

1 − 10

1 − 10

1 − 10

a

b

c

d

Fig. 9. Center + corner damage scenario; DC1 (a) S FD (1) 1 − 10

; (b) S FD (2)

; (c) S FD (3)

; (d) S FD (4)

.

1 − 10

1 − 10

1 − 10

Fig. 10. Damage severity according to damage depth.

a

b

c

Fig. 11. Damage detection with noise, S FD (2) 1 − 10

, DC2 (a) NL = 0, µ = 0 . 963; (b) NL = 3000, µ = 0 . 875; (c)NL = 5000, µ = 0 . 8348.

orders. Therefore, if one wants to detect damage in interior layers of a laminate, the quality of measurements are of outmost importance, since noisy data will easily mask the perturbations due to damage.

3.3. Influence of noise

This analysis is made in two complementary approaches: a qualitative analysis of the figures depicting the damage detection indices and a statistical analysis of the behavior of the quality index µ according to di ff erent noise levels. Figure 11 shows the borderline between detectable and undetectable damage, which is found to be at µ = 0 . 85 across all di ff erentiation orders and for multiple damage cases.