Predicting the onset of filamentous bulking in biological wastewater treatment systems by exploiting image analysis information

The performance of the activated sludge process is limited by the ability of the sedimentation tank to separate the activated sludge from the treated effluent and to concentrate it. Apart from bad operating strategies or poorly designed clarifiers, settling failures can be mainly attributed to filamentous bulking. Image analysis is a promising technique that can be used for early detection of filamentous bulking. In this work, correlations between image analysis information, i.e., the total filament length per image, the mean form factor, the mean equivalent floc diameter, the mean floc roundness, the mean floc reduced radius of gyration and classical measurements (such as the Sludge Volume Index (SVI)) have been sought and the potential of exploiting this information in ARX type black box models to predict the onset of filamentous bulking is presented.