Utilizing Sophisticated Deep Learning Methods to Forecast Treatment Methods for Effluents Generated by the Paper and Pulp Industry

Industrialization has led to the contamination of water with industrial effluents and sewage, resulting in water scarcity issues in India and elsewhere. The effluents are characterized by a dark color, unpleasant scent, high organic content, and severe levels of COD, BOD, and pH. This study investigates the application of inexpensive absorbents and electrochemical oxidation to mitigate the environmental consequences of industrial wastewater treatment. The first procedure involves the usage of cabbage and betel leaf, while the secondary step utilizes Electrochemical Oxidation (EO), which achieves a COD elimination efficiency of 69.19%. This paper deals with the utilization of an artificial neural network model to provide a precise explanation of the EO process in wastewater treatment systems and effectively control and adjust it. The chemical oxygen demand (COD) of the effluent from an electrochemical oxidation wastewater treatment facility used in paper and pulp production was predicted using a neural network model. Testing demonstrates that the Back Propagation (BP) neural network simulation accurately predicts the changing tendency of the real value and possesses some predictive capability. Out of the 20 sample data groups used for simulation prediction, 9 sets have a prediction relative error that is less than 5%, with 45% of those falling within this range. Furthermore, 75% of the pairs had a prediction error of less than 10%, with 15 pairs within this range. The highest comparative error observed is 18.6%. The regression analysis yielded a correlation coefficient of 0.7431 between the real and projected values.