Model Validation of a Bolted Beam Using Spatially Detailed Mode Shapes Measured by Continuous-Scan Laser Doppler Vibrometry

Abstract

The mode shapes of a structure are typically more sensitive than its natural frequencies to local features such as cracks, joints or other anomalies. However, when updating a finite element model for a complicated structure, the analyst is typically only provided with mode shape measurements at a very small number of sensors, so they must rely primarily on the frequencies. With the advent of the Laser Doppler Vibrometer (LDV), one can obtain an unprecedented level of spatial resolution in an automated manner. However, there still may be limitations on the information that can be obtained due to variation in the input to the structure from point to point, the large time required to acquire measurements at each point sequentially for some structures, and logistical issues. Continuous-scan laser vibrometry has the potential to overcome some of these limitations. The continuous-scan approach involves sweeping the laser spot continuously over the structure during a measurement, and then processing the measurement to estimate the mode shapes over a line or surface from that single measurement. This work explores the impact of a spatially detailed measurement set on model validation of a simple structure comprised of a straight beam with a bolted lap joint. The natural frequencies of the system are found to change by 1-3% due to the joint, but when the experimental natural frequencies are compared with those of the model the expected trends are not observed, so it is difficult to ascertain whether the joint model is correct based on the natural frequencies alone. On the other hand, mode shapes measured by CSLDV clearly reveal the presence of the joint, showing a 16-26% reduction in a few of the shapes in a 10-20 centimeter region surrounding the joint. The observed changes in the mode shapes correlate well with those estimated by the model.