Single-station time-domain (SSTD) vibration testing technique: Theory and application

Abstract

The single-station time-domain (SSTD) technique is proposed as a vibration testing method for the identification of the dynamic characteristics of complex structures. The theoretical part of the study shows that the technique works, in principle, for lumped mass systems and continuous systems when a finite number of modes are included in the response. The idea behind the method is to make use of free vibration time response data—such as acceleration, velocity, displacement, or strain at a station on the structure— to determine unknown natural frequencies and associated viscous damping ratios. The analysis procedure is based on the construction of a mathematical model for the vibrating system from the experimental response data. This model has a number of degrees of freedom equal to the number of modes excited. The solution of the resulting eigenvalue problem identifies the natural frequencies and their damping ratios. Each pair of complex conjugate eigenvalues and eigenvectors corresponds to an underdamped mode, whereas each pair of real eigenvalues and eigenvectors corresponds to a critically or overdamped mode. Also included in the analysis is a study of how to optimize the effects of different parameters introduced in the method, such as sampling rates, frequency bandwidth and number of modes, on the accuracy of the results. An experimental investigation is described, for the identification of the dynamic characteristics of real structures by the SSTD method, in which a single channel was used to measure, filter and record the free acceleration responses at an arbitrary point on the structure after an adequate period of excitation; the excitation need not be measured. With the SSTD method, one can minimize the amount of instrumentation required and more accurate results have been obtained than from other vibration testing techniques under similar circumstances. The results obtained from this laboratory study have indicated that the SSTD method is not sensitive to measurement noise or round-off errors and therefore it shows promise of becoming one of the best techniques for structural identification, especially when the system is complex, has closely spaced natural frequencies, and/or is a heavily damped structure.