On Nonlinear Parameter Estimation with Random Noise Signals

In the field of nonlinear dynamics it is essential to have well tested and reliable tools for estimating the nonlinear parameters from measurement data. This paper presents an identification technique based on using random noise signals, as initially developed by Julius S. Bendat. With this method the nonlinearity is treated as a feedback forcing term acting on an underlying linear system. The parameter estimation is then performed in the frequency domain by using conventional MISO/MIMO techniques. To apply this method successfully it is necessary to have some pre-information about the model structure and thus methods for nonlinear characterization and localization are studied. The paper also demonstrates the various ways the method can be formulated for multiple-degree-of-freedoms. The implementation of the method is illustrated with simulated data as well as a practical application, where the method is used to create a dynamic model of a test-rig with a significant nonlinearity