The application of artificial neural network (ANN) models for short-term (15 min) urban water demand predictions is evaluated. Optimization of the ANN model's hyperparameters with a Genetic Algorithm (GA) and use of a growing window approach for training the model are also evaluated. The results are compared to those of commonly used time series models, namely the Autoregressive Integrated Moving Average (ARIMA) model and a pattern-based model. The evaluations are based on data sets from two Canadian cities, providing 15 minute water consumption records over respectively 5 years and 23 months, with a respective mean water demand of 14,560 and 887 m3/d. The GA optimized ANN model performed better than the other models, with Nash-Sutcliffe Efficiencies of 0.91 and 0.83, and Relative Root Mean Square Errors of 6 and 16% for City 1 and City 2, respectively. The results of this study indicate that the optimization of the hyperparameters of an ANN model can lead to better 15 min urban water demand predictions, which are useful for many real time control applications, such as dynamic pressure control.

No abstract

No abstract

Pressure control is recognized as an efficient measure to reduce leaks from water distribution systems. The effectiveness of various pressure control modes, by means of pilot operated diaphragm pressure reducing valves (PRVs), is evaluated in this paper taking into account the sensitivity of the valve to various settings. First, the response of a PRV to consecutive pressure settings variations was experimentally evaluated in the hydraulic simulation laboratory of National Institute for Scientific Research (INRS). These experiments revealed that the studied valve reacts only when the pressure setting variation corresponds to at least 1/6 turn of the pilot valve. Second, a real case study from Quebec City, Canada, was simulated in order to evaluate the impact of the PRV response on three pressure control modes: fixed control, time based control, and real time control (RTC). The results show that RTC of pressure leads to leakage rate reduction on the studied network but that the PRV operational constraints limit the expected performance of RTC.