Author(s): Li Gu
Linked Author(s): Li Gu
Keywords: No Keywords
Abstract: Short-term forecast of water demand is essential to the effective operation of water supply systems. This paper investigates the use of an artificial neural network (ANN) model to forecast short-term daily water consumption during the peak summer season in the Metro Vancouver region of western Canada. The selected ANN model considers the impact of prevailing weather conditions, including selected indicators of antecedent moisture conditions, along with other factors that indicate residential irrigation. It is shown that ANN outperforms some traditional methods of short-term demand forecasting, including regression and time series analysis.
Year: 2009