Experimental Evaluation of Performance of the New Design of Bed Protection Models (F-jacks) in Altering the Flow Pattern around Bridge Piers

Introduction
The various methods have been extensively studied by different researchers to reduce scouring around bridge piers, such as riprap, concrete blocks (CAU), collars, sacrificial piers, creating slot and roughness on the pier, and flow guide vanes, and their results generally relate to determining the size, location, and scope of installation and other geometric characteristics of the devices. Many countermeasures to prevent or reduce local scour around bridge piers, did not have the desired effective, and the concrete armor units (CAUs), which are made to protect shores from erosion caused by waves, have received very little attention in terms of bed armoring around the bridge piers.
Therefore, the aim of the present research is to experimentally investigate the hydrodynamics of flow around a new element designed to protect the bed around bridge piers from local scour. The F-jacks, a concrete armor unit (CAU), is introduced and its role on flow characteristics is evaluated for the first time in this study. This concrete element is a new design of the A-jacks concrete model, with one leg and five branches on top, and the angle between the branches surrounding the leg and the central branch is 30 degrees to ensure minimum contact between the legs of the element and the sediment surface. The selection of a 30-degree angle for the branches of the F-jacks element is due to its similarity to the diameter of the bridge pier, to provide complete coverage around the pier.
Methodology
The experiments of this research were performed in a 7-meter long, 50-centimeter wide, and 0.0001 slope flume with a rigid bed. A wooden cylinder with a diameter of D = 45mm and the same height as the flume was used as a model for a bridge pier and installed at a distance of 4.5 meters from the beginning of the flume (to develop the flow). The water depth (h) in the experiments was constant and equal to 15 centimeters, the flow rate (Q) was 0.021 m3/s, and the flow regime was fully turbulent and subcritical.
In order to understand the physics of flow in relation to three-dimensional velocity variations, a Nortek Acoustic Doppler velocimetry (ADV) with a frequency of 25 Hz and a sampling duration of 120 seconds was used. In order to evaluate the hydrodynamics of the flow around the protected pier with F-jacks units, three different placement patterns around the pier were considered: 1) a non-dense arrangement (P1), in which 24 F-jacks elements were placed next to each other around the pier, 2) a dense arrangement (P2), in which 22 F-jacks elements are interlocked around the pier, and 3) an SP arrangement, which refers to a situation where the single pier is placed in the central axis of the flume.
In addition to the measurement grid on the vertical XZ plane, a set of measurements was taken on the horizontal XY plane at Z/h=0.47 for each of the three selected patterns. To ensure the development of the flow in the test area, normalized mean velocity component profiles were shown to follow a similar trend for two 4 and 4.3 meter-long sections from the beginning of the flume, and the velocity data conforms to the logarithmic law of velocity distribution that confirms the validity of velocity data for developed flow conditions. Also, to verify the accuracy and sufficiency of measurements in the developed flow region under investigation, the power spectral density (PSD) of time series for all three velocity components shows that the slope of the power spectrum agrees well with the Kolmogorov -5/3 law in the inertial sub range.
Results and Discussion
Contour and vector plots of the time-average streamwise velocity component (u ̅) indicated that when the F-jacks elements were placed according to the P2 pattern around the pier, the flow pattern around the pier changes completely. Where, in the upstream of the pier, the average velocity significantly decreased from the water surface to the bottom, indicating the growth of the minimum and weakening of the downflow and horseshoe vortices. In the downstream of the pier, the high-velocity flow region at rear the pier disappeared, and the flow turbulence was significantly reduced, and the region of flow recirculation in the wake of the pier completely disappeared.
With the placement of F-jacks units around the pier, a strong upward vertical velocity (w ̅ ) is obvious around the pier compared to the single pier (SP pattern), which is stronger in the dense arrangement of the elements (P2 pattern). This factor refers to the positive effect of the presence of elements in reducing the growth of downflow (negative vertical velocity), reducing bed disturbances, and turbulence transfer away from the bed region in the wake area around the bridge pier.
The streamwise and vertical components of flow turbulence intensity (u_rms 〖,w〗_rms ) significantly decreased with the placement of F-jacks units around the pier according to the P2 pattern. Where, in the near-bed region around the pier, the turbulence intensity decreased by an average of about 93% compared to the SP pattern, indicating the high ability of F-jacks elements in controlling and reducing flow fluctuations and turbulence in this region and diverting flow fluctuations towards the water surface and away from the bed.
Comparison of Reynolds shear stress on XY plane at Z/h=0.47 for the three mentioned patterns revealed that -ρ(u^' w^' ) ̅ in the SP pattern is approximately 95% higher than P1 and P2 patterns, at the vicinity of the bridge pier. Furthermore, the magnitude of -ρ(u^' w^' ) ̅ has significantly decreased with the placement of F-jacks units as the P2 pattern around the pier.
Conclusion
The laboratory results presented in this paper provide a new understanding of flow behavior details around the bridge pier model with a new design of F-jacks armor units surrounding it on rigid bed conditions. The overall conclusion of this study showed that when F-jacks units are placed densely (P2 pattern) around the pier, the flow turbulence in this area is significantly reduced.