Abstract
Blade tip timing (BTT) measurement has been used to measure the vibration of turbomachinery blades in recent years. People researches focus on the development of processing methods. Newly developed BTT processing methods are usually tested by numerical simulations. Because blades will deform when loaded, the detected points by sensors relative to the blade tip will deviate from their expected positions. A novel simulated BTT model is proposed to identify the movement of the detected points. Finite element method (FEM) is used, and the blade tip profile is discretized into series of segments, each of which is composed of adjacent nodes from the blade tip grid to avoid the error possibly caused by coarse girds. There are two factors which will deviate the detected points from their initial positions: static deformation and vibration. This method is validated by BTT datasets from three different simulations. In addition, an improved sine fitting method and a displacement modification method are proposed to process the deceptive signal of the detected point. The effect of the sensor configuration on the parameter identification is also studied. The result shows that when the static point cannot be determined, it would be better if the sensor could be mounted at a place where it is far away from the nodal line and the static deformation there is small.
