Session: 26-02 Probabilistic Fatigue Crack Nucleation and Growth Lifing Applications

Paper Number: 151763

Probabilistic Gas Turbine Rotor Disk Forging Flaw Crack Nucleation Model Based on Experimental Data and Plasticity-Corrected Stress Intensity Factor 

A probabilistic model for quantifying the number of load cycles for crack nucleation of forging flaws in turbine rotor disks has been improved. In our previous work [1-2], Coffin–Manson–Basquin (CMB) formulation and hazard density function describe the probabilistic crack nucleation life. The fracture mechanics approach achieves an adequate description of the crack nucleation phase to improve the model with a more correlative and essential relationship [3]. In this paper, the model takes the plasticity-corrected stress intensity factor (∆KJ) as the crack driving force parameter and correlates it with the experimental load cycle data for crack nucleation.
Two different approaches for the calculation of ∆KJ are implemented and compared. For the analytical solution, the stress intensity factor (K) and plastic limit load are calculated based on flaw morphologies, types, boundary conditions, and material properties. The failure assessment diagram (FAD) is considered to calibrate K with the plasticity effect. For the numerical solution, we determine the ∆KJ by calculating the J integral numerically surrounding the defect front. The elastic-plastic analysis with the finite element approach is conducted to include the plasticity. A mesh convergence study has been performed to reduce the effect of elemental shapes and sizes on the numerical solution. The comparison reveals the difference between analytical and numerical solutions. Loading conditions and geometry of defects and specimens strictly constrain the analytical approach. In some cases, the orientations of defects need to be adjusted to acquire valid solutions. In addition, the plasticity calibration with FAD is highly sensitive to material properties and the calculated limit load. The numerical approach is more applicable and leads to an accurate solution, which is preferable in industry.
The improved probabilistic crack nucleation model with the numerically determined ∆KJ proposed in this paper is considered more rational and applicable. This model can be expanded in future studies by examining the impact of various materials and different defect patterns.

[1] Radaelli, F., Amann, C., Aydin, A., Varfolomeev, I., Gumbsch, P., & Kadau, K. (2022). A probabilistic model for forging flaw crack nucleation processes for heavy duty gas turbine rotor operations. Journal of Engineering for Gas Turbines and Power, 144(12). https://doi.org/10.1115/1.4056044 
[2] Radaelli, F., Amann, C., Aydin, A., Varfolomeev, I., Gumbsch, P., & Kadau, K. (2021). A probabilistic model for forging flaw crack nucleation processes. Journal of Engineering for Gas Turbines and Power, 143(10). https://doi.org/10.1115/1.4051426 
[3] Aydin, A., & Varfolomeev, I. (2021). Entwicklung eines Rissinitiierungsmodels für Schmiedefehler in Rotorstählen unter zyklischer Belastung: Abschlussbericht. https://publica.fraunhofer.de/entities/publication/da90b2b0-9bc9-42a2-a85d-2962d8f9e7f7/fullmeta

Presenting Author: Yanqiao Yang Siemens Energy

Presenting Author Biography: Doctoral Candidate in Probabilistic Fracture Mechanics

www.linkedin.com/in/yanqiao-yang-584a24224

Authors:

Yanqiao Yang Siemens Energy
Christian Amann Siemens Energy
Igor Varfolomeev Fraunhofer IWM
Peter Gumbsch Fraunhofer IWM
Kai Kadau Siemens Energy