Dissolved Oxygen Intelligent Optimization Control System in the Aeration Tank of Wastewater Treatment

A new control method for dissolved oxygen is presented.According to influent water quality,the method uses online MIMO-LSSVM(multiple input multiple output - least squares support vector machine) soft sensor model to predict effluent water quality parameters,and then takes these parameters as a feedback signal of water quality.The method uses fuzzy neural networks to optimize the DO(dissolved oxygen) settings corresponding to the influent water quality dynamically.At last,inverse control based on neural network is used to track the optimized dissolved oxygen settings.Under the premise of achieving the water targets,this method not only can ensure the stability of effluent water quality,but also can eliminate aeration redundancy effectively.In this way,the aeration can be optimized dynamically and power consumption can be reduced significantly.