Modeling the effectiveness of ozone as a water disinfectant using an artificial neural network

An artificial neural network was used to analyze data from an experimental study on the use of ozone to inactivate a parvovirus in a synthetic and an actual industrial water source. The O sub(3) dose required to disinfect water was predicted as a function of specific environmental conditions. The network included time, alkalinity, organic carbon concentration, O sub(3) residual, initial virus concentration, and sonication as inputs; virus concentration was the output. The network effectively predicted the outcome of disinfection under conditions not previously encountered during training.