Conventional methods for water and wastewater treatment are energy-intensive, notably at the stage of coagulation–flocculation, calling for new strategies to predict pollutant reduction because the amount of energy consumed is related to how much of the pollutant is treated. Here we developed a model, named Bio-logic, inspired by ecosystems, where pollutants represent organisms, coagulants are food, and the wider environmental conditions are the living environment. Artificial intelligence was used to learn the biological behavior, which enabled an accurate prediction of the amount of pollutant reduction. Results show that pseudo-biological objects that have a strong affinity for biological food, such as turbidity, total phosphorus, ammonia nitrogen and the potassium permanganate index, induced a strong correlation, between measured pollutant consumption capacity and predicted values. For instance, R2 correlation coefficients are 0.97 for turbidity and 0.92 for the potassium permanganate index in the laboratory; and 0.99 for turbidity, 0.90 for total phosphorus, 0.75 for ammonia nitrogen and 0.63 for the potassium permanganate index in water treatment plants. Overall, our findings demonstrate that artificial intelligence can use the water Bio-logic model to predict the pollutant consumption capacity.
相似文献The increasing quantities of polluted waters are calling for advanced purification methods. Flocculation is an essential component of the water purification process, yet flocculation is commonly not optimal due to our poor understanding of the flocculation process. In particular, there is little knowledge on the mechanisms ruling the migration of pollutants during treatment. Here we have created the first tensor diagram, a mathematical framework for the flocculation process, analyzed its properties with a deep learning model, and developed a classification scheme for its relationship with pollutants. The tensor was constructed by combining pixel matrices from a variety of floc images, each with a particular flocculation period. Changing the factors used to make flocs images, such as coagulant dose and pH, resulted in tensors, which were used to generate matrices, that is the tensor diagram. Our deep learning algorithm employed a tensor diagram to identify pollution levels. Results show tensor map attributes with over 98% of sample images correctly classified. This approach offers potential to reduce the time delay of feedback from the flocculation process with deep learning categorization based on its clustering capabilities. The advantage of the tensor data from the flocculation process improves the efficiency and speed of response for commercial water treatment.
相似文献