Author(s): Lu Wang; Zihan Wang; Xudong Ma; Xingnian Liu; Ruihua Nie
Linked Author(s): Lu Wang, Xingnian Liu
Keywords: Coustic Doppler Current Profiler bed load transport acoustic bed velocity machine learning
Abstract: Accurate measurement of bedload movement is of great significance for hydropower development, river disaster prevention, and watershed ecological protection. The Bottom-Tracking (BT) function of the Acoustic Doppler Current Profiler (ADCP) allows for immediate determination of the acoustic bed velocity, which is strongly linked to the bedload particle velocity and has been investigated for high temporal and spatial resolution measurements of the bedload movement. Deviation in ADCP-measured bedload particle velocity can be attributed to differences in acoustic reflection properties between bedload particles and the bed surface and no study has yet investigated this bias and how to correct it. In this study, a sequence of flume experiments is conducted, in conjunction with machine learning methods, to examine the origins of inaccuracy in the measured bedload particle velocity acquired by ADCP. Furthermore, a new method for quantifying bedload transport rates using ADCP is devised, which can be applied to different bedload thicknesses.
Year: 2025