Data-driven water quality prediction in chloraminated systems

Abstract

This paper proposes a data-driven method that provides water quality prediction within the entire Woronora delivery system in Sydney. Specifically, the key factors relating to water quality are identified through factor analysis. A Bayesian parametric decay model is formulated using the key factors to predict water quality. To estimate the water travel time, which links the upstream (reservoir) data to the downstream (resident) data, the hydraulic system is employed to capture the topology of the delivery system. Moreover, the uncertainties of both data and the model are analysed to define the boundaries of prediction for better decision making.