Issue 37

V. Shlyannikov et alii, Frattura ed Integrità Strutturale, 37 (2016) 193-199; DOI: 10.3221/IGF-ESIS.37.25 198 Figure 8: Crack growth rate as a function of plastic SIF. Figure 9 : Lifetime prediction based on elastic and elastic-plastic solutions. Fig. 9 represents the comparison between the predicted change in crack length a on the free surface of disc and the crack depth b along the slot of key as a function of fatigue load cycles for both type of solutions. The elastic-plastic solution on based on the plastic SIF's show that the crack would grow on the free surface from 1.6 mm to 10 mm and in the depth direction from 1.6 mm to 20 mm in about 1,300 cycles. At the same time the elastic solution using the elastic SIF's gives overestimate the lifetime in about 1,700 cycles. As the predicted crack growth rate according to nonlinear fracture mechanics approach is much faster that elastic modeling, this indicates that the plastic material properties have a significant effect on the damage accumulation and growth in an critical zone of turbine disc. It should be pointed out that the elastic solution is not accounted for the stress- strain state redistributions at the plastic zone close to the crack tip. The implications due to this limitation may give non- conservative predictions of crack growth rate for this case as the actual stress and strain may be higher than predicted by elastic solution. As the purpose for analyzing the edge crack in the turbine disc considered at the operation was to estimate the crack growth rate under extreme situation, the residual fatigue life should be determined based on the elastic- plastic solution. S UMMARY study on residual life assessment of disc with initial flaws in a power plant steam turbine was carried using full- size 3D analyses and fatigue crack growth predictions. The predictions of the rate of crack growth and residual lifetime of steam turbine disk are compared for elastic and elastic-plastic solutions. It is shown that the previously proposed elastic crack growth models provide overestimate the lifetime with respect to the present one. An advantage to use the plastic stress intensity factor to characterize the fracture resistance as the self-dependent unified parameter for a variety of turbine disk configurations rather than the magnitude of the elastic stress intensity factors alone is shown. The methodology of the lifetime assessment of steam turbine components presented in this study has considerable practical importance. The approach described in the present work is implemented in practical engineering tools, and its usefulness was illustrated with an example of its practical application to a steam turbine rotor. A