Improved strain-energy based mistuning models Part I: Motivation, Approach, and Understanding
One of the grand challenges in turbomachinery forced response is accurately predicting individual blade responses in a system. This paper helps address these challenges and describes an improved FMM model for computing the mistuned forced response of an embedded compressor rotor. Specifically, the paper concentrates on coupling aerodynamic and structural dynamics into the mistuning model and incorporating a strain energy parameter that can help improve the accuracy of the mistuned prediction, especially in the "veering" region. Previously, the authors published results highlighting a wide range of mistuned predictions and the impact of multi-row interaction on these responses. One gap in the literature is the absence of a robust strain energy-based model, which can help improve the accuracy, particularly for cases in which the system mode family is not isolated, i.e., the mode family lies in the veering region, which in turn leads to a significant amount of strain energy in the disk. The current paper starts from an existing mistuning formulation under the FMM framework and describes two approaches to incorporating the strain energy parameter. The key conclusions are: 1) the non-dimensional mistuning parameters significantly improve prediction by including this strain energy-based parameter. 2) the improved strain energy formulations lead to a significant improvement in the mistuning amplification factor for a non-isolated mode family and help couple the physics better in addition to better predictions for a non-isolated family of modes, 3) The coefficient based mistuning model shows promising results with the calibrated coefficient value being the same at multiple resonant crossings for this configuration.
