Continuously Scanning Laser Doppler Vibrometry and Its Application to Vibration-based Structural Damage Detection

Abstract

The objective of this research is to investigate a non-contact real-time moving sensor measurement method, i.e., continuously scanning laser Doppler vibrometry (CSLDV), and its application to vibration-based structural damage detection. In a conventional vibration measurement method, a sensor is fixed at a location on a structure surface and measures response of the structure. The spatial resolution of a vibration shape (VS) of the structure obtained by the conventional measurement method is low due to the limited number of sensors that can be attached to the structure. However, in this research, the laser spot from CSLDV continuously moves over a structure surface along a prescribed scan trajectory and measures velocities along the scan trajectory. Since the laser spot continuously moves, the velocities measured by CSLDV include both time- and spatial-domain vibration information of the structure, and a VS of the structure can be obtained from spatial-domain information. Advantages of this measurement method is that a VS can be obtained in a short measurement time and the spatial resolution of the VS can be very dense. An in-house CSLDV system, including a single-point laser Doppler vibrometer, a scanner and a control unit, is built to measure out-of-plane velocities of a structure. The first part of this research is to investigate signal processing methods to obtain a VS from velocities measured by the CSLDV system. Two methods, i.e., the demodulation method and the polynomial method, are developed to process CSLDV velocity output. The second part of this research is to investigate the application of CSLDV to vibration-based structural damage detection due to its dense measurement of a VS. Curvature-based damage detection methods are developed to identify damage. The CSLDV system is applied to detect different damage severity, different numbers of damage, and different damage types in beams and plates in this research. The third part of this research is to extend 1D-CSLDV to 3D-CSLDV to obtain 3D VSs of a structure, since 3D VSs can provide more structural dynamic information for numerical validation. In a 3D-CSLDV system, three laser spots from three vibrometers continuously move along the same scan trajectory and synchronously measure vibration of the same point on a structure surface. A kinematic model of CSLDV is developed and an improved method based on the least squares method and singular value decomposition is proposed to obtain positions of three CSLDVs with respect to a 3D VS measurement coordinate system. The methodology is validated by measuring 3D VSs of different structures using a 3D-CSLDV system.