| This research addresses the parameter identification problem of a linear dynamic system from forced acceleration responses without any knowledge of input forces. Such a problem is attributed to the added effect of an applied force in its impulse acceleration response function. When using the random decrement technique for such acceleration responses, the random decrement signature of the acceleration response has the added effect resulting from the random decrement signature of the applied force. This result has greatly restricted the practical value and applicability of such a random decrement acceleration signature in the time domain identification.; The random decrement based method has been proposed in this study to provide a solution to the identification problem. The random decrement based method elaborates response data in the time domain and, then, frequency domain by combining the random decrement technique with the fast Fourier transform based algorithms. The theoretical derivation of the random decrement based method initially develops a new frequency response function approach for the acceleration responses, by means of a small modification of the traditional approach. The adoption of the modified frequency response function establishes a theoretical basis defined for the frequency response functions of the random decrement signatures, if the dominant vibration modes are significantly excited, well separated, and lightly damped.; Numerical and experimental simulations have been implemented to identify the modal parameters of building models from forced acceleration responses using the random decrement based method. The numerical simulation is based on various dynamic systems subjected separately to white noise excitations and fluctuating wind forces. The experimental simulation employs a five-story aeroelastic building model tested in a wind tunnel. These simulations are particularly critical to manifest the characteristics of the random decrement acceleration signatures when using the random decrement based method. The applicability of the random decrement based method has been verified by both numerical and experimental results. The performance, limitations, and difficulties encountered by the proposed random decrement based method are also discussed. |