DeepNNet 15 for the prediction of biological waste to energy conversion and nutrient level detection in treated sewage water

Efforts toward sustainable environmental development encompass global initiatives to enhance water quality, sanitation, and wastewater management. This study addresses the energy-intensive nature of sewage water treatment, which is critical for curbing water pollution. We propose a 15-layer Deep Neural Network (DeepNNet-15) to analyse energy conversion and nutrient levels. Constructed with dense layers of up to 10 neurons each, featuring ReLU activation, DeepNNet-15 utilises a benchmark sewage water dataset for training. With 11 input and five output neurons, including energy and nutrient parameters, DeepNNet-15 predicts parameters with less than 3 % error. Its deep learning performance demonstrates over 90 % accuracy, precision, and above 97 % specificity. The practical significance of this research lies in the demonstrated efficacy of DeepNNet-15 in forecasting energy conversion efficiency and nutrient levels in treated sewage water. By harnessing advanced neural network architecture and machine learning techniques, this study offers a tangible contribution to sustainable environmental practices. It equips stakeholders with a robust tool to enhance sewage water treatment efficiency and effectiveness. It aligns with global sustainable development goals and promotes pollution mitigation, resource optimisation, and a cleaner, healthier environment.