a re-configurable testbed structure for system identification studies of uncertain nonlinear systems

A comprehensive analytical and experimental study was conducted to investigate and study some challenging problems related to the development of mathematical models and simulations capable of describing accurately the dynamic behavior of distributed, nonlinear systems with uncertain characteristics. a sophisticated, re-configurable test apparatus was designed and built for the investigation of generic types of mechanical components and subsystems including linear, and complex nonlinear phenomena widely encountered in the applied mechanics field. the apparatus was utilized as a major element of a global computer-controlled loop to automatically conduct different physical experiments to collect a statistically significant ensemble of measurements from the system, while its properties were modified. the physical structural properties were automatically adjusted (for each test loop) through a smart adaptive nonlinear component. the performed tests were subsequently used to build different reduced-order and classes of models to gage their utility to provide a better understanding of the physics of the underlying linear and nonlinear changes, and the associated uncertainties involving the model parameters. thus two different approaches: the nonparametric chain-like system identification approach (chainid) and a global identification approach (next/era) were implemented. the results of this study demonstrate that the structural health monitoring approach discussed is capable of accurately detecting, locating, and quantifying structural changes in the monitored systems. © 2022, iran university of science and technology.