Simulation of Blade-Tip Timing signals inTurbomachinery

The bladed disks in aircraft engines are subject to vibration during operation, excitations shorten their lifespan and raise the possibility of crack formation, both of which can result in catastrophic damages. The conventional method for vibration measurement on turbomachinery blades is to apply strain gauges on the blade surface. In recent years, a non-intrusive alternative technique for measuring blade vibration using a non-contact method has gained popularity. The Blade-Tip Timing (BTT) technique works based on the concept of time of arrival (TOA) of the blades passing in front of stationary sensors, placed on the casing around the blades. In absence of any structural vibration, the TOA would depend only on the rotational speed of the blade. Meanwhile, when the tip of the blade is vibrating, the TOA will depend on the amplitude, frequency and phase of vibration. Hence, it is possible to associate these data to the vibration amplitudes and thus study the vibratory behavior of the blades. The application and the type of data that are required, determine the number, type, and locations of the probes. The starting point of this research is to achieve a numerical simulation of Blade-Tip Timing measures in order to avoid the experimental constraints associated with this method and evaluate the blade-tip timing data processing algorithms. The Vibration and Acoustic Analysis Laboratory of the "Ecole Polytechnique de Montréal" has drawn up a Python code based on the principle of time integration and has been written to be able to simulate the amplitude of blade vibrations in a linear and non-linear framework. One of the new themes of the laboratory is to generate BTT signals, i.e. to digitally simulate the measurement of the vibration amplitude by a sensor to be able to study more precisely the characteristics of these signals. In the first instance generate “tip timing” signals from linear vibratory signals obtained in the laboratory, drawing attention to possible improvements in the TOA computation. Subsequently analysis algorithms dedicated to signal reconstruction were implemented and validated.