Probabilistic fracture mechanics assessment of flaws in turbine disks including quality assurance procedures

A stochastic model for the failure of turbine disks including quality assurance procedures is established. The underlying reliability analysis is based on a fracture mechanics description using both a direct Monte Carlo simulation and a first-order reliability method. The failure probability and its sensitivity to input parameters are obtained together with confidence bounds with respect to uncertain input quantities. Assessment of the accuracy in probabilistic design is essential if only a limited amount of data is available. The results could be applied to extend the life in the respect that an inspection schedule can be derived from the calculated failure probabilities.