Training of a Classifier for Structural Component Failure Based on Hybrid Simulation and Kriging

Abstract

Hybrid simulation is a tool for investigating the dynamic response of a structural prototype subjected to realistic loading. Hybrid simulation is conducted using a hybrid model that combines physical and numerical substructures interacting with each other in a feedback loop. As a result, the tested substructure interacts with a realistic assembly subjected to a credible loading scenario. In the current practice, experimental results obtained via hybrid simulation support conceptualization and calibration of computational models for structural analysis. Instead, this paper extends the scope of hybrid simulation in constructing a safe/failure state classifier for the tested substructure by adaptively designing a sequence of parametrized hybrid simulations. Such a classifier is intended to compute the state of any physical-substructure-like component within system-level numerical simulations. It follows that the main contribution of this paper lies in the way experimental results are aggregated and integrated with structural analysis. The proposed procedure is experimentally validated for a three-degrees-of-freedom hybrid model subjected to Euler buckling.