Self-Operational Wastewater Treatment Plant using Supervised Learning Algorithm

Abstract

Treatment of wastewater is a significant step towards reducing pollution and enhancing water quality. The dynamic design, dominant shock and treatment technologies contribute to instability and variability in the wastewater treatment system. Such risks contribute to variability in the quality and operational costs of effluent water and external pollution from water. Artificial intelligence has been an important method to reduce wastewater treatment problems and complications.The integrated Q-learning algorithms (model-free reinforcement learning) have a supervised learning mechanism that optimizes the control of hydraulic retention time(HRT) and internal recycling ratio (IRR) strategies of the AAO (Anaerobic-Anoxic-aerobic) system. To improvise the operational strategies of the anaerobic Anoxic Oxic tank under different load conditions Q matrixes where build for hydraulic retention time and internal recycling ratio. On the suggested supervised Q learning algorithm performed properlyunder smart analysis and sustainable ideal control techniques were fully deployed under oscillating influential loads in waste management.