A flocculation tensor to monitor water quality using a deep learning model

Abstract The increasing quantities of polluted waters are calling for advanced purification methods. Flocculation is an essential component the water process, yet flocculation commonly not optimal due to our poor understanding process. In particular, there little knowledge on mechanisms ruling migration pollutants during treatment. Here we have created first tensor diagram, a mathematical framework analyzed its properties with deep learning model, and developed classification scheme relationship pollutants. was constructed by combining pixel matrices from variety floc images, each particular period. Changing factors used make flocs such as coagulant dose pH, resulted in tensors, which were generate matrices, that diagram. Our algorithm employed diagram identify pollution levels. Results show map attributes over 98% sample images correctly classified. This approach offers potential reduce time delay feedback process categorization based clustering capabilities. advantage data improves efficiency speed response commercial