STRUCTURAL DAMAGE IDENTIFICATION USING PHYSICS-BASED AND DATA-DRIVEN HYBRID MODEL METHODOLOGIES, MACHINE LEARNING AND DIGITAL TWIN
Damage identification; machine learning; digital twin; hybrid model; structural dynamics.
This work proposes a novel structural health monitoring (SHM) framework for structural damage identification, in the context of dynamical systems. The framework integrates a hybrid physics-based and data-driven model (that can result in a generalizable, accurate, interpretable and computationally efficient model), sensors signals and supervised machine learning methods, to construct a digital twin. The governing equations of structure motion discovered by hybrid modeling (corresponding to structure undamage) will be used to simulate the response of the system with different damage locations and corresponding intensities, from these simulations a dataset will be constructed to train the machine learning classifiers, considering the beam scenarios with damage and undamaged. The constructed digital twin will relate the inputs (sensors signals) of the physical twin to plausible damage scenarios to quickly warn if there is damage, where it is located and what is its severity, supporting engineering decisions. An example, considering a fixed-end steel beam model, was carried out for the purpose of application of the proposed framework, obtaining partial results that showed that the proposed approach is promising.