Recently, the CTBU’s water resources management team of Intelligent Manufacturing Service International Science and Technology Cooperation Base has made two original achievements in “Full-view Intelligent Management” and “On-line Evolution of Artificial Intelligence Process Model” of urban sewage plants, which have been published in Environmental Research, (JCR I, the latest impact factor is 6.5), which is one of the latest achievements of the water resources management team’s cross-integration of environmental science + artificial intelligence, engineering + management and other disciplines.

A full-view operation and maintenance management method based on artificial neural network to deal with the high complexity of sewage treatment system was set forth in the paper titled as “A Full-View Management Method Based on Artificial Neural Networks for Energy and Material-Savings in Wastewater Treatment Plants”. By collecting and in-depth study of the related data of climate, population and economy related to sewage plants, the prediction method of influent water quality and quantity of sewage plants was established, and the accurate prediction was realized 14 days in advance. Then, with the simulation model of operation and maintenance process of sewage plants and genetic algorithm, the optimal operation and maintenance scheme was generated. Based on the research results of two real sewage plants in Chongqing, the fact that the intelligent decision-making system can save 11.20% and 16.91% of energy and material costs for sewage treatment plants had been proved.

In the paper titles as “Online learning-empowered smart management for A2O process in sewage treatment processes”, a set of online learning-empowered smart management for sewage treatment processes was set forth. In this study, the artificial neural network is used to simulate the wastewater treatment process, and then the online learning mechanism is added to enhance the dynamic adaptability of the model, so as to realize the online and real-time performance of the wastewater treatment process. Compared with the online monitoring data of sewage treatment engineering, the results show that the model method based on online learning stated in this study has higher prediction accuracy than the offline model and the conventional comparison model.