Life time extension of turbine rotating components under risk constraints: A state-of-the-art review and case study

• Published in: International journal of fatigue Vol. 129; p. 104799
• Main Authors: Guan, Xuefei, He, Jingjing
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.12.2019; Elsevier BV
• Subjects: Crack propagation; Fatigue crack growth; Fatigue failure; Fatigue life; Flaw detection; Fracture mechanics; Inspection; Life assessment; Life extension; Life time extension; Materials fatigue; Non-destructive evaluation; Nondestructive testing; Probabilistic methods; Probability of detection; Risk; Rotation; Sizing; State-of-the-art reviews; Statistical analysis; Steam turbines; Ultrasonic testing; Uncertainty analysis; Fatigue crack growth; Life time extension; Risk; Non-destructive evaluation; Probability of detection
• ISSN: 0142-1123; 1879-3452
• DOI: 10.1016/j.ijfatigue.2018.08.003

•A method for life extension of steam turbine rotor is developed.•Method is centered on fatigue life evaluation with ultrasonic inspection data.•Probabilistic approach is employed to incorporate uncertainties.•Demonstration with a realistic steam turbine rotor is made. This paper presents a methodology for life time extension of turbine rotating components under risk constraints subject to fatigue crack damage. A state-of-the-art review is given to reveal the historical perspective and the current practice of life time extension. The evaluation of the fatigue life of these components is highly important to mechanical integrity, and in many cases is centered on the ultrasonic inspection data and fracture mechanics. Due to the inherent uncertainty and stochastic nature of the fatigue crack propagation, the remaining life assessment needs to incorporate various uncertainties for risk mitigation. Among the uncertainties, the sizing uncertainty of an embedded crack is significant because the crack cannot be directly accessed and need to be assessed indirectly using ultrasonic inspection. The assessment result is uncertain owing to the unknown orientation of the crack and inspection competency. This uncertainty can be quantified by correlating the actual flaw size and the reported flaw size. This study investigates the method of extending the fatigue life by scientific quantification and minimization of this uncertainty. By analyzing the sizing performance based on the previous sizing data and the present sizing data using the method of model-assisted probability of detection, the life extension can be realized without compromising the risk requirement in a probabilistic framework. The overall method is demonstrated using a realistic steam turbine rotor with actual ultrasonic inspection data.