Traditional engine maintenance, repair and overhaul (MRO) are geared toward fixed schedules. However, with online condition monitoring, assessments and prognosis, it is required that MRO be adaptive to the life consumption with respect to the actual usage of the engine to realize the benefit of prognosis. Shifting to this new paradigm, there are several challenges:
1. How exactly the life is consumed in components under complex usage profiles that may involve a combination of low and high cycle fatigue, thermomechanical fatigue and creep, for regular usage (aside from incidents)?
2. What are the physical failure mechanism(s) in components under the above conditions, the understanding of which may help to select the most appropriate detection, repair and replacement (including material insertion) strategy, for life renewal/extension and cost reduction?
3. Understanding the limitations of repairs for the particular failure mechanism.
To overcome these challenges, one needs physics-based material failure models that can reflect the true failure mechanisms on the component level under actual (complex) usage profiles. Conventional, empirical test-data correlations for isolated conditions fall short of this requirement because pure fatigue and pure creep are not experienced by real components in service. In addition, to ensure the repaired component’s structural integrity, the material database would have to be comprehensive enough to cover the effect of intrinsic repair defects and microstructures that are not present in the original component material. While models and data are integral parts of digital twins, it is proposed that physics-based models are needed to cover the entire application domain continuously. This paper will introduce a physics-based method of life consumption evaluation and discussion through past experience in line with the development of Digital Twin concepts for sustainment.