A Kalman filter-based method for displacement calculation of blade tip timing

Blade tip timing (BTT), as a non-contact measurement technique, is used to monitor the health condition of rotor blades in turbomachinery. The accuracy of the vibration displacement obtained by BTT is important for the reliability of blade health monitoring. In this paper, a novel BTT displacement calculation method based on Kalman filter (named K-BTT) is proposed to improve the accuracy of the calculated BTT displacement. K-BTT calculates displacements in the angular domain, where the angle relationship between blades can be used as prior information to reduce uncertainty. Compared with the traditional BTT (T-BTT) method, K-BTT makes full use of all blade positions rather than relying solely on the OPR sensor. To evaluate the reliability of K-BTT, the calculated displacement of K-BTT is compared with that of T-BTT in a simulation. The results demonstrate that root mean square error of the displacements calculated by K-BTT is less than that calculated by T -BTT. Moreover, the variance of the K-BTT error also performs better than T-BTT. In practical application, the influence of speed fluctuation on displacement calculation can be reduced by using K-BTT.