The increasing use of renewable energy sources to produce electricity requires additional operational flexibility from fossil-fuel gas and steam turbines. To compensate for renewable energy fluctuations in the electrical grid, a gas turbine (GT) engine needs to be more flexible, operating in peaking and partial loading modes as well as the base-load operation mode. Understanding how these different modes affect the lifetime of turbine components is critical to ensuring favourable RAM (Reliability, Availability, and Maintainability).
Component lifetimes in peaking modes are limited by the number of thermo-mechanical cycles that a component can experience before crack initiation. The useful lifetime of some components can be increased by basing the predicted lifetime on the number of cycles for crack initiation plus the number of cycles for the crack to reach its maximum allowable length based on the fracture toughness K1C criterion for linear elastic fracture mechanics (LEFM). This is usually accomplished by using the Paris law to predict the rate of crack growth. Once cracks are formed, further propagation depends on the states of stress and strain near the cracks. These factors, which drive crack growth, can be quantified by the energy release rate. The Paris law predicts crack growth as a function of the energy release rate under linear elastic conditions, commonly for load controlled tests with load ration R>0. However, large thermal and mechanical loading can result in plastic deformation under cyclic loading conditions.
Most GT components operate under strain controlled conditions generated by thermal loading. In this paper, a novel method is used to characterize crack growth under cyclic strain conditions in regions under plastic strain. The experimental data reveal that the rate of crack growth changes under plastic conditions in comparison with the linear elastic case. Especially compared to very high stress intensities ΔK of load controlled tests, here the allowable displacement limiting strain control matters. Applying experimental data from material tested under cyclic loading and elastic-plastic material response, component lifetime has been reliably predicted. Hereafter the developed method is referred to as elastic plastic fracture mechanics (EPFM) lifetime assessment. The EPFM approach more closely predicts the observed rate of crack growth than linear elastic fracture mechanics. LEFM over-predicts component lifetime for cracks growing in plastic regions under cyclic loading and could lead to catastrophic failure of a component. Therefore, the lifetime of a highly loaded component is more reliably assessed using the EPFM approach, as demonstrated for two alloys in this paper.