Experimental investigation of local scour around two piers in tandem arrangements for different nose shapes

Abstract

A popular configuration for studying complex flow characteristics and local scour around groups of piers by using piers arranged in tandem configurations or side-by-side. Alongside factors such as pier spacing and median sediment size, the nose configurations of piers play a key role in influencing scour patterns, and subsequently, the flow characteristics around the piers. In this thesis, an experimental investigation was conducted to analyze the effects of pier nose shape and pier spacing on turbulent characteristics and scour depths around the piers.Three different shapesnamely, round nose, 600 nose angle, and 900 nose angle pierswere placed in both single and tandem arrangements. In the tandem arrangements, two identical piers were positioned at center-to-center (C/C) spacings of 3L, 6L, and 9L (where L represents the pier length) within two different sediment beds: Sand-A (d50 = 0.23 mm) and Sand-B (d50 = 0.89 mm). Three-dimensional instantaneous velocity was measured using an Acoustic Doppler Velocimeter (ADV). From the experiments, it was observed that the magnitude of turbulent intensity and turbulent kinetic energy between the piers was found to be higher for smaller center-to-center (C/C) spacing, attributed to increased interference from the rear pier. The turbulence structure was also influenced by the shape of the pier as the 900 nose angle piers exhibited the maximum turbulence intensity and turbulent kinetic energy.The Reynolds Shear Stress value near the bed was higher at the upstream of front pier. In contrast, at downstream of the pier, an oscillating (positive-negative) pattern was observed due to flow separation and vortex shedding. Analyzing the measured scour depths data, it was observed that for a single pier, the maximum scour depth was 19.61% higher around the round nose pier compared to the 600 nose angle pier, and 5.89% higher compared to the 900 nose angle pier. This difference was attributed to the comparatively blunter nose of the round nose pier, which led to an increase in the magnitude of the downflow as well as the formation of a horse-shoe vortex. The position of maximum scour depths also varied as for round nose pier; it was formed at the front of the pier. However, for the 600 nose angle and 900 nose angle piers, the maximum scour was observed at the frontal edge. The pier shape factorwas calculated as 0.88 for the round nose pier, 0.73 for the 600 nose angle pier, and 0.775 for the 900 nose angle pier. The smaller shape factors for the 600 nose angle and 900 nose angle piers were a result of their sharp leading edges.

Experimental results also showed that the scour depth around the front pier always exceeded that around the rear pier. The highest scour depth around the front pier was recorded for 3 L C/C spacing compared with 6L and 9L C/C spacing attributed to the increased drag force of the front pier compared with other spacings. Notably, at a 9L C/C spacing, the scour depth around the front pier closely resembled that of a single pier configuration.Furthermore, it was observed that regardless of pier shape, the maximum scour depth was more pronounced for Sand-A in comparison to Sand-B. This observation was attributed to the formation of an armor layer at the base of the scour hole in Sand-A, which was influenced due to the accumulation of larger particles within the scour hole. Four existing scour depth predicting equations were used to compare predicted and observed scour depths around piers of different shapes. The results showed that the predictive equations consistently overpredicted scour depths. Among the four equations, Lacey's equation exhibited close values to the experimental results compared to the others.Lastly, a correction factor was introduced for incorporating the spacing between two piers in the CSU single pier equation, enhancing its ability to predict scour depths in tandem pier arrangements. Turbulence characteristics, as observed in the experimental results, were also compared with Computational Fluid Dynamics (CFD) model using ANSYS software. The simulated results exhibited patterns of turbulence characteristics similar to the experimental findings. However, the simulated results consistently overestimated the normalized streamwise velocity at mid-depth for all pier shapes. Furthermore, in all cases, the CFD simulations underestimated the normalized turbulent kinetic energy compared to the experimental findings. The findings of this study would be helpful in enhancing our understanding of the effects of pier nose shapes and spacing on turbulent flow characteristics and scour development around tandem piers.