A probabilistic model for forging flaw crack nucleation processes for heavy duty gas turbine rotor operations

We present a probabilistic model for quantifying the number of load cycles for nucleation of forging flaws -for a 3.5NiCrMoV high strength low alloy rotor steel - into a crack under gas turbine operating conditions. The model correlates low cycle fatigue data, ultrasonic testing indication data, flaw morphology, and type with the nucleation process. This paper is the third of a series of publications presenting this modeling approach progressively. It focuses on the effect of temperature variation on the nucleation life of forging flaws. We quantified the number of cycles to crack nucleation was for specimens that included forging flaws at elevated temperatures. Flaws of different sizes and shapes are effectively described at respective temperature and stress levels by either an ellipsoidal finite element model or an analytical area-based model. A local probabilistic low-cycle fatigue model analyzes the resulting stress distributions accounting for statistical size effects. Via Maximum Likelihood Estimation of these probabilistic low cycle fatigue results, a probabilistic model for crack nucleation of forging flaws is obtained. This proposed probabilistic model is based on experimental data for realistic heavy duty gas turbine rotor temperature and stress conditions. It can be utilized in the energy sector for component life time quantification. Our suggested approach can support component assessment under flexible gas turbines operation conditions driven by increased availability of intermittent renewable energy sources.