Ensuring a steady supply of drinking water is crucial for communities, but predicting how much water will be needed is challenging because of uncertainties. As a part of Battle of Water Demand Forecasting (BWDF), this study delves into the application of Long Short-Term Memory (LSTM) networks for water demand forecasting in a city situated in the northeast of Italy. The focus is on forecasting the demand across ten distinct District Metering Areas (DMAs) over four distinct stages. To enhance the performance of the LSTM model, an evolutionary optimization algorithm is integrated, aiming to fine-tune the model’s hyper-parameters effectively. Results indicate the promising potential of this approach for short-term demand forecasting.

Rigorous system-wide aggregated water quality performance indices for water distribution networks are lacking in the literature due to complexities associated with high dimensional spatial and temporal water quality data. Water quality considerations unavoidably increase performance evaluation difficulties considerably. The formulation developed in this article addresses the post-extended period simulation high-dimensional data challenges. A system-wide joint-dynamic-response approach to water quality evaluation is introduced that accounts for spatial and temporal variations in nodal demands and the respective time-varying hydraulic and water quality properties of multiple service reservoirs. Effective comparisons of the water quality response of service reservoirs and their effects were achieved. This includes individual reservoirs and the combined effects of multiple reservoirs. Service reservoirs and the nodes they supply were particularly vulnerable from the standpoint of water quality. The role of the network’s topology considering water quality risks was revealed also. The correlation between the medians and flow-weighted daily means of the water quality parameters was very strong (R² ≥ 0.994) for the service reservoirs considered. Thus, the median could be useful as a practical performance surrogate in design optimization procedures. Finally, there seems to be an association between the flow-weighted daily means and overall hydraulic effectiveness of service reservoirs.

This study introduces a novel framework for conducting a comparative analysis of static and mobile sensing approaches for the collection of data to be used in network calibration. Two new algorithms that optimize deployment for both static and mobile sensors are proposed. The results indicate that deploying a single mobile sensor starting from various locations throughout the network for 24 h can yield pipe roughness calibration results as good as, or slightly superior, to those obtained using static sensors at approximately 90% of the potential monitoring nodes.