A generalized study of local scour downstream of a weir

• Main Author: Hiew, Wing Yian
• Other Authors: Lim Siow Yong
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Hydraulics; Scour; Scouring; Dimensional analysis; Regression analysis; Water resource; Weir; Apron
• Online Access: https://hdl.handle.net/10356/176252
• Institution: Nanyang Technological University

This research examined the critical parameters affecting the equilibrium scour depth at the downstream of an overflowing weir. A generalized prediction equation, based on the experiments in this study and data from previous FYP studies. has been proposed. The key variables involved in the formula are the weir height, flow rate, tailwater depth, and median sediment size. This study comprises two primary phases: the literature review and data collection phase followed by the experiment and data analysis phase. Phase 1: Past research works on weir scour such as Veronese’s early study in 1937, NTU’s extensive FYP data and relevant field data have been reviewed and analysed. Phase 2: Employing a hydraulic flume at the NTU laboratory, the real-world hydraulic conditions for a weir scour were studied through hands-on experiments. In this study, sediment diameters and weir height are selected as control variables, whereas tailwater depth and flowrate are the independent variables that are meticulously varied to investigate the gap and grasp the intricacies of the correlation and interplay among the parameters, which has not been examined closely in previous NTU experiments. This research found that the Shield parameters and eddies formation as well as secondary current observed in the scour hole should not be ignored for a more accurate scour prediction. The Shield parameter in the form of a densimetric Froude number is recognized as the most influential flow parameter from data analysis and the weir height is identified as the most effective length scale from the result of all researches conducted at NTU. Based on dimensional analysis and multiple regression analysis, a prediction equation was derived and was then compared using 36 previous non-NTU laboratory data with very small scour hole, 12 field data and 62 NTU obtained from well-controlled experiments followed by validation with very large-scale real-world data at Arno River with an overprediction of 43.9% and a satisfying correlation coefficient of R2=0.83. The reasonable rate of overprediction achieves a good balance between being safe and conservative in practice.