Proceedings ofthe 5 Annual Paper Moet & 2" Civil Engineering Congsess, 29-30 July 2022, Dhak, Banglades ISBN: Manus, Serker Majeh Mure. Matsuda, Maker ad Alam orceeapn banat BRIDGE PIER $ ZOURING NUMERICAL MODELLING THROUGH EXPERIMENTAL, OBSERVATIONS R.R.Das',N. B. Nizam’ and B. Mahalder* Student, Dept. of Water Resources Engineering, BUET, Dhaka-1000, Bangladesh *Assistant Professor, Dept. of Water Resources Engineering, BURT, Dhaks-1000, Bangladesh ABSTRACT This study is formulated to investigate bridge pier scour through experimental and numerical modelling (HEC-RAS 2D) using cylindrical and rectangular pier at different angles of attack and flow velocity. The experiments were carried out in a Mobile Bed Visualization Tank located in the Hydraulics and River Engineering Laboratory, BUET with a 6cm thick bed of fine sand (450 = 0.3mm), Scouring was found prominent on the sides of piers supported by both experiment and model. The ‘erosion characteristics followed no specific pattern o be classified throughout the active area inthe experiment. The location of maximum scour depth was the same in case of both experiment and model. For rectangular pier, results from the experiment and the numerical model varied with different angles of attack and flow velocity. Scour depth varied in an increasing manner as the angle of attack was increased from 0° to 20° for the experiment, which also supported the ‘numerical model results. As far as flow velocity is concerned, the experimental and model results showed increasing pattern in scour depths, as velocity increased, maximum scour depth also decreased as flow velocity was reduced. For eylindrical pier bath experimental and numerical analysis data showed higher scour depths occurred around the cylinders, The extent ‘and impact of scour were higher for rectangular pier compared with eylindvical pier INTRODUCTION ‘The hydrodynamics of flow and their impact on the scour mechanism near bridge piers are of major interest to the engineers, ‘The interactions between the flow and flow obstructing structures (e.g., bridge piets, abutments, ete.) generate turbulence on. the sediment, resulting scour development which ean accelerate bridge failure. Scour and other hydraulic forces are thought to be responsible for about 60% of all bridge failures (Namace et al., 2018). Through erosion atthe bottom flow boundary, ‘an equilibrium state is gradually achieved between the erosive capability of flowing water and the resistance of bed materials ‘over time (Melville and Chiew, 1999), Due tothe presence of an unfavorable pressure gradient, the flow around an obstruction or pier forms horseshoe vortices near the obstacle, Wake vortices also form normal tothe planes of horseshoe vortices (Vijayasree et al, 2020). Horseshoe vortices are necklace like vortie structures which wrap around the pier and fold over its upstream portion and separate the upstream ‘entering boundary layer. These are results ofthe unfavourable pressure gradients brought on by the existence of pier (Kirk etal, 2005). To understand the flow field and scour around a single bridge pier, extensive research work has been cattied out by numerous researchers. Vijayasree et al. (2019) conducted studies to better understand the local scour around piers of various shapes. Laursen and Toch (1956) experimented the skew angle effect on the scour depth, According to Hassan etal (2016), the initial scour rate and equilibrium scour depth are affected by the pier shape. Since rectangular pier has a larger ‘maximum exposed area, substantially higher scour depths have been reported by the researchers, Noor et al. (2020) performed an experimental and HEC-RAS modelling of bridge pier scouring for square and rectangular piet. According to the study, for both experiment and mode, seour hole was larger for square pier than circular pier. However, ll these studies have been based on experiments and other numerical modeling. No significant attempt has been made to measure scour around piers with HEC-RAS 2D. Also, HEC-RAS ID provides scour depth through only one profile whereas HEC-RAS 2D. yyelds velocity data in every point of the 2D computational flow field and scour can be calculated from these velocities ‘Therefore, inthis study, experimental investigations were carried out and the results were compared withthe TIEC-RAS 2D analysis results using diferent pier shape and angle of atacks, 428METHODOLOGY Experimental Setup ‘This experiment was performed in a Mobile Bed Visualization Tank in the Iydraulies and River Engineering Laboratory, Department of Water Resources Engineering, BUET. The working section ofthe tank was 400 cm long, 61 cm wide with 0 sloping characteristics. A Gem thick sand bed having a median diameter of 0.3mm of the particles was equally distributed across the working region. Wooden rectangular pier of length, width and height of 1Sem, Sem, Sem respectively and @ slenderaess ratio of 1/3 was chosen and cylindrical pier of Sem diameter and 16cm height was selected for the experiment. Figure | and 2 illustrate the flume along with piers with respective dimensions, a0 mtb f= lowe we = hoe cos Poe SSS =i ‘Aas ev tcente @ ‘Seite £ be... | = ‘ ra 1 ma & P| SS =F ) Ad Vas ae centers Figure 1. Flume Dimensions & Location of Pier (a) Rectangular Pier; (b) Cylindrical Pier Experiments were run three times each for 0°, 10° and 20° angle of attacks at Sem, 6em and Jem water depths respectively. ‘The corresponding flow velocities were 2.1env, 2.74emvs and 3.3Semvs. Based on several trials, it was observed thet under these particular velocity values, scour development and propagation along with velocity variation was evident. Each experimental run was set for a duration of two and a half hours. Point gauge was used to measure bed level with respect to a preset datum before and aller the experiment in the defined mesh. The difference between the initial and final bed level ‘depicted the scour depth. HEC-RAS Modelling In addition, HEC-RAS 2D v6.1 was used to setup the numerical model. An appropriate terrin ofthe bed was generated and associated using RAS-Mapper. A bridge centerline was then drawn in Geometry Editor followed by entering bridge data, For setting boundary conditions, a 2D flow area was created witha roughness coefficient of 0.012 and with enough computation points for better accuracy. Steady flow data from the experiment was entered as unsteady flow dala as stage and flow hydrograph to maintain consistency upstream and downstream of the pier in the simulated results as flow parameters change ‘throughout the flume de tothe presence of the pier: The flow was then simulated with atime frame that corresponds to the experiment. Velocity profile was attained from simulation results in RAS-Mapper. The scour depths were then calculated using HEC-RAS ID model v6.1 using CSU equation (Guo etal, 2012) from the velocity values. The CSU equation is shown, in equation | Ye Kya y Ft wo where, y= Depth of scour in feet (on) IK, = Correetion factor for pier nose shape Kz = Correction factor for angle of atack of flow 29Ky = Correction factor for bed condition K4= Correction factor for armoring of bed material er wah in fet) ow depth directly upstream of the pier in feet (em). Fr, = Froude Number directly upstream ofthe pier. ‘A comparison was then drawn between experiment and model results. The new insights on flow structure reported here are expected to be helpful in simulating the scour process around uniform and compound piers. These will also be beneficial in refining the scour protection measures surrounding these structures RESULTS AND DISCUSSIONS ‘The vertical extent of scour around a bridge pier can be characterized by its scour profile, Figure 2 shows the scour profile in general for both rectangular and eylindeical pies. Localized scouring started ata distance from the pie, gradually increased and reached its peak atthe face of the pier. Maximum scour depth was found to be 5.8 em for rectangular pier at 20° angle of attack and at 3.35 emis velocity with 4.5 em being the maximum scour depth value for eylindrcal pier at 3.35 emis. Figure 2 Scour Profle at 2.13 em/s Velocity (a) Rectangular Pier at 10° Angle of Atack: (b) Cylindrical Pier Figure 3, 4 and S show generalized contours which were generated using ARC-GIS v10.7 for different cases in both experiment and model. These contours illustrate that scour depths are higher in the experiment compared to those in the ‘model, Model results produced about 10% lower values of maximum scour depth overall on average when compared with experimental results. This difference in scour depth occurred as a result of an underdeveloped flow pattern and flow undulation caused by the use of chute blocks upstream in the experiment. Scour was largely concentrated near the four comers of the rectangular pier although scouring occurs evidently at both upstream and downstream of the pier. Similar results were found by Vijayasree et al. 2019). Generally, seour pattern was more pronounced upstream of the pier than downstream. Downstream erosion occurs duc to formation of wake vortices. These wake vortices develop downstream due to flow separation at the sides of the pier. On both sides of the pier, a wake vortex grows each, sheds, and is eonvected downstream inducing erosion consequently (Stevens et al., 1991). With no precise pattern to depict, these contours follow a ‘general tendency of increased scour depths around the per G @ Figure 3. Contour for Reciangular Pier at angie of attack and 2.13 om/s Velocity for. (a) Experiment: (b) Model 430@ © Figure 4. Contour for Rectangular Pier at 10° angle of attack and 2.13 emis Velocity for (a) Euperiment; (b) Model ie 5 ConeurforReanuar Per a2 ange ef atack and? em's eo for Experiment) Mode ‘Scour was heavily focused in the area surrounding the cylindrical pier, The scour pattern was more apparent upstream than. downstream of the pier. Figure 6 demonstrates the contours for both experiment and mode! for the cylindrical pier. They follow the same course of changes as in the rectangular pier. These contours demonstrate a general rising trend of scour depths getting closer to the pier, but show no discernible pattera. Scouring was found maximum in the vicinity of the cylindrical pier. @ ) Figure 6 Contour for Cylindrical Per at 2.13 cm/s Velocity for (a) Experiment: () Model ‘Table 1 shows the maximum scour depths with regard to different flow velocities and angles of attack for both the rectangular nd the eylindtical pier and for both experiment and model, Comparing the results fram the model results to those found from. the experiment exhibits that maximum seaur depth inereases as angle of attack inereases. Similar trend was observed when. flow velocity was considered. As the flow velocity increased scour depths also increased. Experiment results show higher scour depths in all cases than those found from model results for both the rectangular and the cylindrical pier, Jalal and Hassan (2020) also found identical results in which the Flow-3D based model yielded lower values of scour depth than those ‘found inthe experiment. It can also be observed that scous depths are always larger for rectangular pier in case of experiment as compared with eylindrical pier which is similar to the findings by Al-Shukur and Obeid (2016). Cylindrical pier yields about 21% lower values of scour depth when compared to rectangular pier on average in the experiment, Model results also follow the same trend with the cylindrical pier producing 20% lower values of scour depth on average. 431Table I Maxinuon Scour Depth for Different Flow Velocities and Angles of Attack for Reclangular and Cylindrical Piers for both Experiment and Model ‘Masia Scour Dei (em) ‘Angle of Attack (Reetangular Pier) Flow Velocty = f me 7 Experiment | Medel_| Experiment] Model_| Experiment] Model 2 325, a [ao | as | 53 2.74.c1n's SL 42 [sa 48 St 32 faseme | sa a3 186 s sx bss CONCLUSIONS ‘The goal of this research was to compare experimental and numerical modeling (HEC-RAS) data in order to analyze bridge scout utilizing eylindrcal and rectangular piers at various angles of attack and establish scour depth at various flow, velocity, and discharge. The impacts of scour on a rectangular pier were measured in 80 nodes (Mesh size 810) at 0° and 10° angles of atack. The same were recorded at 88 nodes (Mesh size 8x11) for 20° angle of attack and a cylindrical pict Evidence by both experiment and model, discernible scouring was observed on the sides ofthe rectangular pier at every angle of attack and flow velocity. Incase of the cylindrical pie, maximum scour depths were found around the sides of the cylinder. The 2D model results yielded about 10% lower values of maximum scour depth overall on average comparing with experimental results. Maximum scour depths for different flow velocity were found substantially lower (about 21% on average) for eylindrical pier as compared to rectangular pier in the experiment, Similar results were found using the 2D ‘model results. Maximum scour depths for eylindrcal pier were found about 20% lower on average than those for rectangulae pier: The location of maximum scour was found identical in both experiment anid model. REFERENCES ALShokur, A. HLK., and Obeid, Z. H. (2016) “Experimental study of bridge pice shape to minimize local scour", International Journal of Civil Enginecring and Technology, 71), 162-171 Jalal, H. K., and Hassan, W. H (2020) “Effet of bridge pier shape on depth of scour” In IOP Conference Series: Materials Scienoe and Eagincering (Vol. 671, No. 1, p.012001),1OP Publishing Kirkil, G., Constantinescu, S.G., and Fitema, R. (2005) "The horseshoe vortex system around a etcular bridge pier ona at bed” In XXXIst Intemational association hydraulic research congress, Seoul, Korea Lausen, E. M, and Toch, A. (1956) "Scour around bridge piers and abutments (Vol. 4)", Ames, IA: Iowa Highway Research Boacd. Melville, B. W., and Chiew, Y. M. (1999) “Time scale for local scour at bridge piers”, Joumal of Hydraulic Engineering, 125(1), 59-66. Namaee, M.R., Li, ¥, Sui, J, and Whitcombe, T. (2018) “Comparison of three commonly used equations for calculating Tocal scour depth around bridge pier under ice covered flow condition”, World Journal of Engineering and Technology, 6(2), 50-62, Noor, M., Arshad, H., Khan, M., Khan, M.A., Aslam, M. S., and Ahmad, A. (2020) “Experimental and HEC-RAS Model of Bridge Pier Scouring", Journal of Advanced Research in Fluid Mechanies and Thermal Sciences, 74(1), 119-132. Stevens, M. A., Gasser, M. M., and Saad, M. B. (1991) "Wake vortex scour at bridge piers", Journal of Hydraulic Engineering, 117(7), 891-904, Vijayastee, B. A., Eldho, T. L, Mazumder, B. S., and Ahmad, N. (2019) “Influence of bridge pier shape on flow field and scour geometry”, International Journal of River Basin Management, 17(1), 109-129, Vijayasree, B.A., Eldho, T. 1, and Mazumder,B. S. (2020) "Turbulence statisties of flow causing scour around circular and ‘oblong piers", Journal of Hydraulic Research, 58(4), 673-686, 432