High-performance beam finite element for predictive response in monitoring existing bridges

In recent years, several advanced technologies, such as Artificial Intelligence (AI) techniques, have been developed to automate inspections and monitoring processes of existing bridges. In this context, the efficiency of computational models is crucial in model updating for monitoring systems and training neural networks. Although the nonlinear structural response of the bridges can be efficiently analysed through two-dimensional and three-dimensional finite element (FE) models, these commonly require high computational efforts. This work adopts a high-performance beam finite element based on a damage-plasticity model, implemented in the OpenSees framework, for prestressed reinforced concrete girders. The beam FE relies on a force-based (FB) formulation which is more efficient than the classical displacement-based approach. The constitutive law of the concrete fibers is based on a plastic-damage model, which considers two different damage parameters for the compression and tensile behaviour to take into account the re-closure of the tensile cracks. Dynamic responses in both linear and nonlinear regime are simulated under white noise excitation. ANNs are trained in a subset of the predicted responses in the linear range and the trained network is used to simulate the high amplitude response in which nonlinear behaviour is experienced. Interesting results are acquired useful for further investigations.