Vision-Based Vibration Detection of Rotating Shaft with Assistance of Fringer Stripes

The shaft is an essential part of the rotor system, and the operation status of the rotor shaft will directly affect the operation of the rotor system. The vibration signal of the rotor shaft can respond to its operating status, and the current rotor shaft vibration measurement sensors are mainly eddy current and laser Doppler type, both of which can only realize the single-direction measurement of the rotor shaft, and when the need for multi-direction measurements, it is necessary to deploy the sensors in multiple directions, which is complicated and expensive for installation. In this paper, a vision-based rotary shaft vibration measurement method is proposed, which uses a monocular camera to capture the video of preset fringe stripes on the rotary shaft, and obtains the vibration information of the rotary shaft along the direction of the camera's optical axis by analyzing the variation of the density of the fringe stripes, while the vibration of the rotary shaft perpendicular to the direction of the camera’s optical axis is realized by tracking the symmetry center of the fringe stripes. The feasibility of the proposed method is preliminarily verified by simulation and experiment, which shows good potential for application.