Abstract: Aiming at the problems of slow deployment, high wading risk, and data lag of traditional hydrological monitoring equipment in the closure of water conservancy projects, this paper takes the second-phase cofferdam closure project of Jintangchong Reservoir as the research object to carry out the on-site application and evaluation of the Wuhan University AiFlow handheld visual flowmeter. Three temporary test sections (upper, middle, and lower) were arranged along the flow direction in the main flow area of the closure gap, and a total of 57 sets of valid flow velocity data were collected, which were cross-validated with UAV flow measurement data. The results show that the equipment adopts a pure visual non-contact operation mode, can be quickly deployed in safe areas, and realizes "immediate measurement and immediate reporting" relying on edge computing technology. It can accurately capture the dynamic changes and transverse distribution characteristics of flow velocity in the closure gap, and the monitoring data is stable and highly consistent with the trend of UAV flow measurement data. The successful application of this equipment solves the monitoring problems in complex closure scenarios, provides key data support for the optimization of construction schemes and dynamic decision-making, and promotes the transformation of construction decision-making from experience-based judgment to data-driven mode.