智控扫描式声学多普勒测流系统应用探讨

为加快水文监测新技术与装备在水文测报中的应用，在浙江嵊州站、江苏前垾村水文站分别建设智控扫描式声学多普勒流速仪测流系统示范应用站点。智控扫描式声学多普勒流速仪测流系统，通过水下智控转动扫描测流方式使传统的一维流速剖面数据扩展为二维扫描流速数据，并融合计算流体动力学、人工智能等多学科前沿技术，建模仿真河道流场以计算实时流量，原理上可有效提高测流精度和降低人工比测率定分析的工作量。通过对示范应用站点半年时间运行的稳定性和流量监测数据比测分析，得出示范应用站点测流系统稳定可靠，流量监测和人工监测数据趋势基本一致；在流态快速变化期间，测流结果存在一定迟滞性，可采用提升扫描速度的方式提高对流速变化的敏感度；流体及神经网络算法模型还需要进一步优化，结合多模型算法提高智控扫描测流系统的智能化能力，进一步提高流量监测的精度。

Demonstration and application analysis of intelligent scanning acoustic doppler flow measurement system

In order to speed up the application of new hydrological monitoring technology and equipment in hydrological forecasting, intelligent scan acoustic Doppler (ADCP) flow measurement system demonstration application stations are built in Shengzhou hydrological station in Zhejiang Province and Qianhan village hydrological station in Jiangsu Province respectively. The intelligent scan acoustic Doppler (ADCP) flow measurement system can expand the traditional one-dimensional velocity profile data into two-dimensional scanning velocity data by means of underwater intelligent rotating scanning flow measurement. The system integrates the computational fluid dynamics, artificial intelligence and other multidisciplinary cutting-edge technologies to model and simulate the river flow field to calculate the real-time flow. In principle, it can effectively improve the flow measurement accuracy and reduce the manual calibration and analysis workload. Through the stability analysis of half a year operation of the demonstration application site and comparison analysis of flow monitoring data, it is concluded that the flow measurement system of the demonstration application site is stable and reliable, and the trend of flow monitoring data is basically the same with that of manual monitoring data; During the rapid change of flow regime, the flow measurement results have a certain hysteresis, and the sensitivity to flow velocity changes can be improved by increasing scanning speed; The fluid and neural network algorithm model need to be further optimized, and the intelligence ability of intelligent control scanning flow measurement system need to be improved combined with multi-model algorithm, so as to further improve the accuracy of flow monitoring.