Nonlinear model updating of a reinforced concrete pedestrian cable‐stayed bridge

Abstract

Strong earthquakes often cause different levels of damage to reinforced concrete (RC) bridges. The identification of their damage state and the prediction of their future seismic performance are thus necessary for hazard rescue and damage retrofit. To achieve this objective, the nonlinear model and the nonlinear model updating of the bridges are needed. This paper presents a fiber frame element-based nonlinear model and a new nonlinear model updating method for a RC pedestrian cable-stayed bridge. The fiber frame element-based nonlinear model is established in OpenSees, accounting for both material and geometric nonlinearities. The evolution principles of envelop curves, together with hysteretic rules with respect to reloading and unloading paths for both concrete and reinforcement materials, are considered. The new nonlinear model updating method is developed in the time domain, utilizing both global and local structural responses. The new updating method can evaluate not only time-dependent strength degradation but also time-dependent stiffness deterioration so that the current damage state of the bridge can be identified and the future seismic performance and damage state can be predicted. Finally, shake table tests were carried out on the 1:12 scaled RC pedestrian cable-stayed bridge, and the comparison of the predicted global and local seismic responses of the bridge with the measured ones confirms the accuracy and effectiveness of the proposed nonlinear modelling and model updating methods. KEYWORDS fiber frame element, nonlinear modeling, nonlinear model updating, RC pedestrian cable-stayed bridge, shake table tests | INTRODUCTIONReinforced concrete (RC) bridges are the most vital and basic infrastructures playing a crucial role in earthquake crisis management, but earthquakes often cause different levels of damage to RC bridges because RC bridge structures have less redundancy and thus fewer alternative load paths compared with RC building structures. 1-3 How to identify the damage state of a bridge after an earthquake and predict its seismic performance to future earthquake are an important task in the seismic performance assessment of the RC bridges.To accomplish the damage identification and seismic performance prediction of a RC bridge, an accurate finite element (FE) model of the bridge is necessary, and a FE model-based damage identification method is needed. There are many FE model-based damage identification methods available, most of which use changes in dynamic characteristics

Collapse prognosis of a long‐span cable‐stayed bridge based on shake table test and nonlinear model updating

“…In the existing research on the nonlinear model updating of civil structures, the updating parameters are usually considered to be independent of each other 21–23,33,34 . However, in real conditions, the degradation in stiffness and strength is usually correlated in a damaged concrete structure, as shown in Figure 7A.…”

“…An effective way to reduce the modelling uncertainties of civil structures is to update the numerical model based on the measurement data acquired by the SHM systems. Existing studies have attempted to improve the accuracy of FE models through various model updating algorithms 18‐22 . Hasançebi and Dumlupınar 18 updated a reinforced‐concrete T‐beam bridge using an artificial neural network (ANN)–based method.…”

“…Xiao et al 19 built a multi‐scale FE model of the Stonecutters Bridge in Hong Kong and updated the modal frequencies and static responses of the model based on the response surface model (RSM) and multi‐start optimisation algorithm. Li et al 20 and Zheng et al 21 conducted the nonlinear model updating of reinforced concrete bridges based on the OpenSees‐SNOPT framework developed by Gu et al 23 Li et al 20 updated the modelling parameters of the high‐damping rubber bearings in a seismically isolated bridge and reconstructed the nonlinear seismic responses of the bridge during the 2010 Maule Chile earthquake. Zheng et al 21 conducted a shake table test on a 1:12 scaled pedestrian bridge and established the corresponding fibre beam‐column element‐based FE model in OpenSees.…”

“…Li et al 20 and Zheng et al 21 conducted the nonlinear model updating of reinforced concrete bridges based on the OpenSees‐SNOPT framework developed by Gu et al 23 Li et al 20 updated the modelling parameters of the high‐damping rubber bearings in a seismically isolated bridge and reconstructed the nonlinear seismic responses of the bridge during the 2010 Maule Chile earthquake. Zheng et al 21 conducted a shake table test on a 1:12 scaled pedestrian bridge and established the corresponding fibre beam‐column element‐based FE model in OpenSees. Both the global and local responses of the bridge were updated, and the results were in good agreement with the experimental measurement data.…”

Model updating‐based automated damage detection of concrete‐encased composite column‐beam connections

“…In addition to numerical finite element analyses and analytical calculation, nondestructive ambient vibration examinations are widely conducted to get the dynamic features in functioning condition without any external forces. A vast amount of literature can be found regarding the finite element model updating of steel structures, 13–20 reinforced concrete structures, 21–27 base isolated structures, 28 timber structures, 29,30 and masonry structures 31–36 . However, there are very limited studies about the ambient vibration test‐based system identification and FEM updating of concrete steel composite structures 37–39 …”