摘要：

净水处理工艺中，混凝过程是一个大滞后、非线性的复杂系统。针对其特点并结合目前国内外加药控制研究现状，开发了一种基于絮凝体图像处理、模糊逻辑控制的前馈-串级智能加药系统。通过水样图像采集与处理提取絮凝体特征关键参数，计算絮凝体等效直径，建立投加模型确定絮凝剂投加量，再通过模糊PID控制及前馈补偿实现絮凝剂的全自动调节和精准投加。系统监控及性能分析显示，与传统手工投加对比，智能投加系统沉淀后出水浊度标准偏差降低27%，与设定值的相对误差控制在10%以内，絮凝剂的投加量节省11.04%，在絮凝剂投加量和处理效果上具有显著优势，表现出良好的可行性和推广潜力。

关键词: 图像处理, 模糊PID, 絮凝剂投加

Abstract:

In the process of water treatment, the coagulation process is a complex system characterized by significant lag and non-linearity. To address its characteristics and based on the current research status of dosing control, a forward-cascade intelligent dosing system based on floc image processing and fuzzy logic control was developed. Through the acquisition and processing of water sample images, key parameters of floc characteristics were extracted, and the equivalent diameters were calculated to establish a dosing model to determine the coagulant dosage. Subsequently, automatic adjustment and precise dosing of coagulants were achieved through fuzzy PID control and feedforward compensation. System monitoring and performance analysis showed that compared with traditional manual dosing, the intelligent dosing system reduced the standard deviation of the sedimentation effluent turbidity by 27%, with the relative error controlled within 10% of the set value. The dosage of coagulants was reduced by 11.04%. The intelligent dosing system demonstrated significant advantages in coagulant dosage and treatment effects, which showed promising feasibility and potential for dissemination.

Key words: image processing, fuzzy PID, coagulant dosing

中图分类号: 

• TN911.73

• TU991.21