Abstract
Adaptive modeling approaches have recently gained significant attention by the applied mechanics research community to perform real-time identification of nonlinear hysteretic structural systems under arbitrary dynamic excitations. Adaptive identification is critical for robust adaptive structural control and for tracking time-varying structural dynamic properties such as accumulating damage. This paper presents an overview of some of the authors’ previous work in this area, and also discusses some of the new issues being tackled with regard to this class of problems. The trade-offs between parametric based modeling and nonparametric modeling of nonlinear hysteretic dynamic system behavior are discussed. A new neural network based adaptive identification procedure is introduced. Both simulation and experimental results of the performance of the parametric and non-parametric methods are presented. A brief discussion of future directions and remaining technical challenges is presented.
