Potential problems with mechanical equipment are often revealed by sound and vibration. The acoustic signals therefrom are often used to diagnose defects and monitor the performance of the equipment. The technology is not fully developed and can benefit from further study. There are economic incentives to perfect such diagnostic systems since they promise to provide the capability of rapid inspection of large machines or structures with minimal interruption of operations. Improving the sensitivity of such systems and minimizing their potential for false alarms would represent a major step in advancing the technology. This project addresses that goal. It could lead to new ways of processing acoustic signals in order to extract the maximum amount of useful information. The beam is chosen as the object of study because it is a basic structural component and because there is an extensive theoretical literature on its vibrational modes, including the effect of cracks. There is a variety of uncertainties in the condition of beams in service which can cause variations in their acoustic signatures. These uncertainties may arise from manufacturing variations in dimensions and materials, wear, miscellaneous nicks and dents, small delaminations, corrosion, dirt or internal stress. If changes in the acoustic signatures of the beam caused by these harmless effects are great enough, they may be the cause of false alarms in a detection system. This project's goal is to be able to predict the extent of such effects and to compare them with the effects of flaws of serious concern. The research program concentrates on the modeling of unflawed vibrating beams and the construction of an experimental fixture for laboratory data acquisition. The system modeling effort focuses on the prediction of the response of unflawed beams, the prediction of the response of cracked beams, and the preliminary prediction of the effects of various other conditions specified above. The experimental fixture activity focuses on the design of a test fixture, its fabrication and assembly, data acquisition from unflawed beams, data acquisition from flawed beams, data acquisition from beams of uncertain condition, and comparison of signals from these experiments.
