A novel method has been presented in this paper for the diagnostics of nonlinear systems using the features of the nonlinear response and capabilities of computational intelligence. Four features of the phase plane portrait have been extracted and used to characterize the nonlinear response of a nonlinear pendulum. An artificial neural network has been created and trained using the numerical data for the estimation of parameters of a defective nonlinear pendulum setup. The results show that, with appropriately selected features of the nonlinear response, the parameters of the nonlinear system can be estimated with an acceptable accuracy.

This content is only available via PDF.
You do not currently have access to this content.