絮凝体沉降速度在线测量方法研究

Research on the Method of Online Measuring Floc Settling Velocity

絮凝体沉速可作为评价絮凝效果的指标,也可作为控制混凝剂加注量的依据。提出一种在线测量絮凝体沉速的方法,通过连续采集絮凝体图像,经计算机数据处理,结合多目标跟踪算法,跟踪絮凝体运动轨迹,计算每一个絮凝体的沉速。针对多目标跟踪中出现的目标瞬间丢失、交叉、合并或分离等异常情况,采用一种模糊推理关联算法与自适应卡尔曼滤波组合的多目标跟踪方法。实验结果表明,这种组合算法实现了絮凝体沉降速度的实时监测,可更精确方便地进行混凝剂加注的自动控制。

Floc settling velocity can be used as indicators for evaluation of flocculation,but also can be used as a basis for controlling the amount of coagulant notation. The method of on line measuring floc settling velocity is employing real seriate floc image which is analysised by computer to track the floc movement and calculate every floc ve- locities. Aiming at several problems occurred in multi-object tracking, such as the moving objects interleaving or over-lapping, a new multi-object tracking algorithm is proposed that combines the fuzzy reasoning algorithm with adaptive Kalman filter. Experimental results show that the combination algorithm can achieve the real-time monitoring of floc settling velocity and realize more accurate automatic control of coagulant filling.