Modal parameter identification for vibration control

For the active control of flexural vibrations the parameters in the mathematical model of a vibrating structure must be determined. In this paper a stochastic modeling method is applied to the identification of modal parameters including undamped natural frequencies and damping ratios. Autoregressive and autoregressive-moving average models are used and compared to extract significant features of the method. A simulation study using multidegree of freedom vibration systems shows that the accuracy of estimation depends on data-length, sampling interval and measurement noise. An application of the method to the experimental data of the transverse vibration of a cantilever beam demonstrates that the identified results of the natural frequencies are quite consistent and those of the damping ratios are reasonable to the first digit. The results are compared with those obtained by other methods to discuss the practicality and reliability of the proposed method.