PSI - Issue 22

S.C. Wu et al. / Procedia Structural Integrity 22 (2019) 211–218 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

217

7

Fig. 4. The difference methods for fitting P-S-N curves of specimens and axle: (a) compared with TCM, (b) compared with BSIM

But Considering the difference of STD, the shape of the P-S-N curve of the ISIA is a downward opening bell shape and more actual. According to Eq. (6), Fig. 2 and Fig. 4 , the fatigue life of full-scale axle for TCM is 2.49 × 10 16 km and for the BSIM is 9.26×10 11 km, but the life for the ISIA is 4.7×10 10 km, which is about 17% of the BSIM and 0.1% of TCM. Obviously, the prediction result of the ISIA is more reliable. The results show that the new method is more reliable in predicting axle life and more suitable for testing axle design life in engineering. From the forgoing, since the design life of axle is 20-30 years according to EN 13104, e.i. 8 × 10 6 -1.2 × 10 7 km ， the service life of axle in high-speed train predicted by the ISIA meets the design acquirement. Remarkably, the actual operating environment for axle is worse than the variable amplitude (VA) loading (as shown in Fig. 2 ). For example, the high speed train axle will be subjected to foreign object damage, extreme cold, high temperature, corrosion and so on. Therefore, there are still great challenges in the accurate assessment of axle strength and residual life for the critical safety components of high-speed train. 4. Welded structures of high-speed train It is well-known that the various defects were introduced into the welded components during the welding process, such as the porosity, the inclusion, the undercut, the looseness and so on. As a result, the dispersion of fatigue life is great for these components. In order to verify the applicability for high-speed rail welded structures (such as pantograph, car body and frame), the ISIA is used to process the high-cycle fatigue data of two welded materials and the results are shown in Table 4. In this study, the fatigue properties of welds joints were studied by comparing the results with the International Institute of Welding (IIW) recommendations (Hobbacher, 2016) , as illustrated in Fig. 5 Fig. 5.When p =97.5%, the decrease extent of the conditioned fatigue strength is different for various base metals compared with median SN curve. In other words, the dispersion of fatigue life is markedly different for the two base metals as reported in Table 4. For example, the lowered degree for 2024-T4 with the FSW is 33%, and the value for 2024-T3 is 25.9%. In effect, the data of 2024-T4 with the FSW has much distinct dispersion. On the other hand, the characteristic fatigue strength calculated by the ISIA is obviously smaller than that of the TCM for various welded materials as shown in Table 4, Especially the decrease percentage for 2024-T4 FSW joints is 16.7%. Therefore, the fatigue limit predicted by ISIA is more reliable than that predicted by the TCM. Meanwhile, the P-S-N curve of the ISIA ensure the validity by considering the STD distributed in gradient.

Table 4. The axial fatigue properties of welded joints

P =97.5%

Welding methods

∆ σ D with P 50%

Materials

Author

R

∆ σ D (ISIA)

∆ σ D (TCM)

Relative error

Friction stir weld (FSW) Friction stir weld (FSW)

2024-T4

Di et al. (2006)

0.1

80

60

70

16.7%

Zhou et al. (2006)

0.1

73

58

66

13.8%

2024-T3