Abstract
This contribution focuses on the validation of a numerical strategy developed jointly by Safran and Polytechnique Montréal for the simulation and the analysis of blade-tip/casing contact interactions in low-pressure compressor stages. A large experimental campaign provided data (including strain measurements on the blade and wear profiles) for several contact configurations involving four distinct blades and one type of abradable coating. The numerical strategy is here improved by introducing a new cut-off criterion to ensure the physical relevance of the presented results, specifically by keeping the maximum stress within the blade below the material's yield stress. Similarly to previous publications involving a single contact configuration, the numerical model is first calibrated for one of the four blades of interest. It is seen that the results using the numerical model?critical speed, relative wear depth between leading edge and trailing edge, maximum stress levels within the blade?are in good agreement with the experimental observations. Using the same calibration, numerical simulations are then blindly run for the three other blades. The results demonstrate that numerically predicted key quantities align well with experimental data. Additionally, the numerical model provides an accurate relative assessment of a blade?s sensitivity to contact in agreement with experimental observations. This paper thus presents the first blind validation of a numerical strategy dedicated to blade-tip/casing contact interactions. Simultaneously, it also demonstrates that this model may be considered for the early discrimination of blade profiles depending on their sensitivity to contact.
