Tip-Timing Data Analysis for Mistuned Bladed Discs Assemblies

In turbomachine industry, bladed assembly vibration measurements are very important for blades behaviour estimation. These measurements are generally obtained from strain gauges. However, one of the most promising methods for the analysis of blade vibrations in rotating bladed assemblies is the Blade Tip Timing or Optical Blade Vibration Measurement method. A set of optical sensors is mounted on an engine casing, in front of a disc, and measures the times of arrival of each blade. These timings are then used to compute the vibrations of the blades. However the fundamental problem for spectral analysis of blade tip timing data is that the signals are undersampled and aliased. We propose here a new method for spectral estimation of blades responses from tip timing data that overcome these difficulties. The method proposed in this communication is based on the use of a minimum variance filter associated with an iterative updating of the autocorrelation matrix. That allows to process correctly a signal even if the number of known signal samples is less than equivalent Nyquist criterion. This approach is suitable for spectral analysis of undersampled and aliased signals. Perfomances of the spectral estimator have been evaluated for one simulated and one experimental test cases. The method seems very promising for the monitoring of mistuned bladed discs.