عنوان مقاله [English]
نویسندگان [English]چکیده [English]
The structure of turbulent flow over the scour hole is important for understanding the sediment particle entrainment and transport. In the present study, stochastic nature of turbulent flow around a bridge pier under a clear water regime has been investigated experimentally. Inside the scour hole, near the bed, the flow does not have enough energy to transport the particles; hence, the sedimentation occurs in this region. The particles in the scour hole near the bed are lifted up, however, they cannot move along the flow and finally they return to the bed. By comparing depth average values of the occurrence probabilities at the pier upstream, it was found that the sweep events probability is more than other events, which followed by ejection, outward and inward interaction. Since the sweep events are dominant phenomena at the pier upstream, the maximum scour depth takes place at this region and the eroded sediments are transported heading downstream.
Scour is a natural event caused by the erosive action of flowing water on the bed and banks of streams, which also takes place on region near the bridge piers and abutments. The stochastic nature of turbulent flow around the bridge pier has been investigated only in handful studies; therefore, the precise effect of pier on sediment transport is yet unclear. Kline et al. (1967) have introduced the concept of bursting phenomenon as a mean to describe the momentum transfer between the turbulent and laminar region near the boundary. Lu and Willmarth (1973) introduced the quadrant analysis for studying the the bursting phenomenon structure. The quadrant analysis was employed to determine the occurrence frequency of each individual event within a bursting process, i.e. outward interactions, ejections, inward interactions, and sweeps. Jafari and Keshavarzi (2010) investigated turbulence flow and sediment entrainment over the ripples by using quadrant analysis. However, despite the importance of coherent structures of turbulence and in particular the bursting events around the bridge pier; their characteristics have not been investigated in detail. Since the quadrant analysis is a powerful technique to recognize the structure of the bursting phenomenon and consequently to find the susceptible regions for sediment entrainment and deposition, this technique is considered in the present study to investigate the coherent structure of turbulent flow in scouring process around the bridge pier.
The experiments were carried out in a smooth rectangular flume with 8 m length, 0.4 m width and 0.6 m height. The channel entrance was filled with sand in order to generate fully developed flows. The working section, in which piers were located, was 1.6 m long with a 0.15 m recess on the bed and was located 4 m downstream from the flume entrance. The recess was filled with 0.72 mm mean particle size uniform sediment and the standard geometric particles deviation was 1.12 mm. Uniform sand, having the same size as that used for the scouring test, was glued over the false floor. The velocity profiles were measured around the bridge pier. In each vertical profile at least 20 points were measured within the flow depth from the bed. The ADV readings were taken when the scour hole was at the equilibrium condition.
The variation of the fractional contributions, |Si, H|, as a function of the hole size H for each of the four quadrants at z/h=0.2 (near the scour hole edge) at the up and downstream of the pier are shown in Fig. 1. At z/h=0.2, quadrant (II) and (IV) events are dominant and quadrant (I) and (III) events appear to contribute weakly to the Reynolds shear stress production. Generally, by increasing the hole size, the contribution of each event to the Reynolds shear stress generation becomes small. As indicated in Fig. 1, at the pier upstream at z/h=0.2, ejection (|S2,0|?0.8) is the dominant event which is followed by sweep (|S4,0|?0.6). The sweep contribution tothe Reynolds shear stress production becomes negligible for H>10, whereas the ejection contribution is significant even at H>20. On the other hand, the contributions of outward and inward interactions to the Reynolds shear stress are rather weak (|S1,0|=|S3,0|?0.2). These contributions vanish when H>6 and H>10 for quadrant (III) and (I), respectively.
At the pier downstream, the contribution of each event to the Reynolds shear stress production is more significant than pier upsream. In this region, at z/h=0.2, the contributions of ejection and sweep become comparable (|S2,0|=|S4,0|?1.4) and both are greater than the outward (|S1,0|?0.9) and inward interaction (|S3,0|?1). For each of the four quadrants at z/h=0.2 in the pier downstream the contribution of each event to the Reynolds shear stress is still significant for hole sizes as large as H>20. Therefore, at the pier downstream, although quadrant (II) and (IV) are more significant, each of the four quadrants has important effects on the Reynolds shear stress.