Monitoring and control of UV and UV-TiO2 disinfections for municipal wastewater reclamation using artificial neural networks.

The use of ultraviolet (UV) irradiation as a physical wastewater disinfection has increased in recent years, especially for wastewater reuse. The UV-TiO(2) can generate OH radicals, which is highly effective to inactivate microorganisms in wastewater disinfection. However, both UV and UV-TiO(2) disinfections create multiple physical, chemical, and bio-chemical phenomena that affect their germicidal efficiency. It is difficult to build a precise control model using existing mathematic models. This study applies artificial neural network (ANN) models to control UV and UV-TiO(2) disinfections. Experimental results indicate that the ANN models, which precisely generate relationships among multiple monitored parameters, total coliform counts in influent and effluent, and UV doses, can be used as control models for UV and UV-TiO(2) disinfections. A novel ANN control strategy is applied to control UV and UV-TiO(2) disinfection processes to meet three total coliform count limits for three wastewater reuse purposes. The proposed controlled strategy effectively controls UV and UV-TiO(2) disinfection, resulting in acceptable total coliform counts in effluent for the three wastewater reuse purposes. The required UV doses for UV-TiO(2) disinfection were lower than those for UV disinfection, resulting in energy saving and capacity reduction of 13.2-15.7%.