PSI - Issue 62

Tommaso Lazzarin et al. / Procedia Structural Integrity 62 (2024) 625–632 Lazzarin et al./ Structural Integrity Procedia 00 (2019) 000 – 000

630

6

The estimation of bed shear stresses at the bridges is crucial to assess the potential of sediment entrainment and to predict the occurrence of bed scours at the bridge structure (Chiew and Lim, 2000; Teruzzi et al., 2009). RANS simulations with proper turbulence models provide a reliable distribution of the mean shear stresses; however, previous studies demonstrated that bed shear stresses are generally underestimated if compared to corresponding simulations conducted with more advanced models such as LES or DES (Cheng et al., 2018). Eddy-resolving methods provide the instantaneous distribution of the bed shear stresses that allows obtaining more accurate predictions because, besides mean stress values, instantaneous fluctuations of the bed shear stress play a key role in determining the entrainment of bed sediments (Izadinia et al., 2013). Fig. 5 shows the instantaneous distribution of bed shear stresses,  , for the PF simulation conducted with the DES approach. The mutual effects of the pier wakes and of the orifice flow beneath the bridge deck, which are explicitly resolved by the model, induce strong temporal fluctuations of the bed shear stresses, especially just downstream of the bridge section where the flow expands. Immediately below the bridge, instead, the high-velocity induced by the PF generates a relatively steady region with high bed shear stresses. Time-averaged values of the bed shear stresses are about two times higher than those observed in the corresponding simulation in FS regime. The high bed shear stresses, which are expected to induce strong sediment entrainment in case of PF regimes, need to be properly evaluated to assess the bridge vulnerability. Indeed, the PF flow may cause strong erosion in the entire region beneath the deck, which poses serious concerns for the bridge stability, given that only areas around piers and abutments are generally protected against scour. The comparison of the results from RANS and DES simulations shows that the predicted sediment entrainment flux in the bridge region is comparable. Thus, for the engineering point of view, the two methods are comparable to study bed erosion in the region beneath the deck. This conclusion does not apply for the region downstream of the bridge and the case of PF regime, for which RANS simulation underestimates the potential sediment entrainment because it neglects the strong temporal fluctuations of bed stresses.

Fig. 5. Instantaneous bed shear stresses,  , for the PF regime simulation with DES at different times (a-c)

4. Conclusion Hydraulic modelling is a key element in bridge vulnerability analysis. Among the wealth of models nowadays available, 3D models based on the Detached Eddy Simulation approach and with deformable free surface capabilities represent the state-of-the art for such kind of analyses. The level of details and the information provided by such models far exceed those obtained using conventional hydrodynamic models (e.g., 2D depth-averaged models). Three-dimensional CFD models explicitly consider the complex interactions between the flow in natural rivers and the single components of the bridge structure, which is essential to obtain detailed and reliable results accounting for different specific factors, such as the angle of attack between the flow and the piers, the naturally-