PSI - Issue 52

Ilias N. Giannakeas et al. / Procedia Structural Integrity 52 (2024) 655–666 Ilias N. Giannakeas/ Structural Integrity Procedia 00 (2022) 000 – 000

662

8

Fig. 5: Effect of temperature difference on . Dashed lines indicated the 95% confidence interval. From Fig. 5 it is possible to extract the uncertainty for for different ∆ values. It is assumed that during the operation of the SHM system, this uncertainty will be added to the of each impact event. Therefore, for each ∆ , we can create samples from ̃ ~ ( , ) where = + , = while and denote the increase in the mean and standard deviation due to the temperature difference. In this section the location of the impact is set to its true value. The and values are computed for each temperature difference level. Using impact event FP1-1 as an example, the Monte Carlo methodology is followed to compute ̃ for different ∆ values. The obtained histograms are plotted in Fig. 6. As expected, because the temperature difference increases both and , the estimated ̃ value is also increased. The histogram of ̃ is shifted to larger damage area values while also a bigger spread is observed in the predictions.

Fig. 6: Effect of temperature difference on ̃ estimation for impact event FP1-1.

Fig. 7: Effect of temperature difference on A) ̃ and B) ̃ for each impact event.