ABSTRACT
The physical state of a structure can be modeled as a function of its state variables (e.g., material and geometric properties). Deterioration processes cause changes in the state variables that in turn modify its ability to sustain given demands over time. Typically, deterioration processes are divided into gradual processes, where the deterioration evolves continuously over time, and shock processes, where the deterioration is modeled as discrete jumps. The damage accumulation within shock events is usually disregarded in simplified procedures that only model the total effect of the shock and use aggregated measures to quantify the shock intensity. However, this simplification can produce inaccurate estimates of the structure’s reliability. As a result, it is important to account for the specific evolution of the damage also in such processes. This paper models shock deterioration as a rapidly evolving but still gradual process. We use the formulation based on Stochastic Differential Equations proposed by Iannacone & Gardoni (2023) to model the effect of the time history of ground motion sequences on the deterioration of a bridge. This formulation allows obtaining the state variables together with their associated uncertainties during the shock, which in turn can be used to obtain the time-varying reliability of the system. The models are calibrated based on the results of advanced Finite-Element Models which record the state of the system in real time. We combine the proposed formulation with available methods for generating spectrum-compatible seismic excitations for a site of interest to predict the deterioration over time and the time-varying reliability of the bridge during the occurrence of the shock.
