A Fuzzy Logic to Predicting Velocities in Primary Sedimentation Tanks

It is essential to have a uniform and calm flow field for a primary settling tank with high performance. In other words reducing the kinetic energy of the flow inside the settling tank can improve significantly the efficiency of these tanks and give opportunity to particles to settle in the settling zone of the tank. So determination of the velocity magnitude in various positions can be a good index of the sedimentation tanks efficiency. So this study applies a new soft computing technique, i.e. an adaptive neuro-fuzzy inference system (ANFIS), to better prediction of measured velocity magnitude inside the sedimentation tank. Validation of the developed network (ANFIS) was performed using a set of velocity data collected at a sedimentation tank. The results confirm that the suggested network can more accurately predict the measured velocity data when compared to an equation based on a regression method. The proposed ANFIS approach produces satisfactory results R = 0.953 for x-velocity and R =0.94 for z-velocity compared to the experimental results. 2 2