This study focuses on a first-stage stator vane of a gas turbine that developed fatigue cracks during service. A finite element–based fracture mechanics simulation was performed under thermo-mechanical coupling conditions with the aim of providing a practical approach for fatigue crack growth (FCG) analysis and FCG life prediction of the vane.
First, the potential crack growth directions were identified through a crack growth likelihood analysis. Then, initial cracks were introduced based on actual inspection data from the turbine blade, and the crack growth process was analyzed. Finally, the randomness of FCG life was investigated assuming that the Paris model parameters (C, m) follow a bivariate normal distribution with a correlation coefficient of −0.9, with a focus on the distribution characteristics of FCG life.
The crack growth is more likely to occur toward the leading edge of the blade and along the rib direction. The crack growth rate throughout the process exhibits non-monotonic behavior with turning points. The FCG life can be approximately described by a Weibull distribution, and both the mean and dispersion of the FCG life increase as the coefficients of variation for C and m increase.
This research provides the FCG characteristics and the probabilistic distribution of FCG life for the turbine vane, offering a useful reference for probabilistic damage tolerance assessment of turbine blades under thermo-mechanical loading.
