PSI - Issue 22

Abdelkader Guillal et al. / Procedia Structural Integrity 22 (2019) 201–210 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

207

7

B - Influence of parameters correlation of Paris law In order to investigate the effect of Paris law parameters correlation on the reliability estimation, the four mentioned cases above are deployed. Figure 4 plot the reliability index for various numbers of cycles. It’s clear from Figure 4 that all cases have the same trend of decreasing in the reliability index. The major difference is noticed during middle stages of growth ( 3*10 5 cycles) for case 3 and case 4 with relative difference equal to 9% , in other word between Cortie correlation and a negative correlation factor. In the same stage of growth, case four deviate from case two with a relative difference of 4.5%. This difference can be explained using sensitivity analysis. Figure 5 illustrate the variation of sensitivity to mean value crack growth rate in cases 2, 3 and 4.The sensitivity to mean value of crack growth rate in case 4 increases to reach a maximum value for 3*10 5 cycles. For the two other cases, the sensitivity have less importance. As a result, reliability index curve in the case 4 is deviated from the rest due to the influence of considered negative correlation between m and C .

Fig 4. Variation of reliability index for different cases of Paris law parameters

Fig 5. Variation of sensitivity to mean values of Paris law exponent in case 4

C – Comparison between LSFs of cracked pipeline surface The two limit state functions considered are presented in section (3.2). The variation in reliability index results considering two state functions, the first represent a brittle fracture While the second model are presented in figure 6.