在水处理微涡流絮凝工艺研究中，絮体的识别、检测与跟踪方法运算复杂、效果欠佳，仅关注液相（水）而忽略固相（絮体）的单相流流场分布，存在无法准确反映其真实流态、内在规律不明等问题，拟采用 Canny及分水岭等算法进行边缘检测，准确获取絮体数量、分形维数等特征参数；采用基于BP神经网络的果蝇优化算法实现多目标跟踪，以快速准确得到絮体运动速度；通过絮体跟踪与数值模拟相结合，实现流场可视化，揭示微涡流絮凝工艺固液两相复杂流态下的流场分布规律，优化水力条件；获取流场流速、湍动能、能耗散和涡旋尺度等关键信息，建立融合絮体相对尺度、特征参数和出水浊度之间的微涡流絮凝工艺过程模型，进一步揭示絮凝内在规律。项目研究成果对优化微涡流絮凝工艺运行管理、节省絮凝剂投加量、提高出水水质、保障居民的饮用水安全具有重要的理论价值与实用价值。

There are lots of problems in micro-vortex flocculation process of water treatment. For example, the method of identification, detection and tracking of flocculation is complex and poor effectively, real flow cannot be reflected truly because of only considering liquid phase and ignoring solid phase, and micro-vortex flocculation inherent law is unclear. Aiming to the above problems, the project group intend to use Canny and Watershed Algorithm to proceed edge detection so as to gain accurately floc characteristic parameters including quantity and fractal dimension. And multi-target tracking is achieved by Fruit Fly Optimization Algorithm based on BP Nerve Network to obtain floc moving velocity quickly and accurately. Flow Visualization is acquired by numerical simulation and flocculation tracking to reveal distribution of flow field under solid-liquid two-phase complex flow, which could optimize hydraulic condition and obtain some key parameters e.g., flow velocity, tubulence kinetic energy, dissipated energy and vortex scale. The micro-vortex flocculation process model among the floc relative scale, characteristic parameters and effluent turbidity is established to further reveal the inherent law of flocculation. The research results have important theoretical value and practical significance to optimize the management of vortex flocculation treatment, improve water quality and ensure the safety of drinking water.