Journal of Water Resources Planning and Management

The loss of some amount of water through leaks is inevitable in a water distribution network (WDN). Usually, leakage losses are included as volume-based demands by increasing the nodal demands by a certain percentage in the design of WDNs. However, leaks are pressure dependent and increase with an increase in pressure in the pipeline. In this study, a methodology is proposed to include leakage as pressure-driven demands along with normal volume-based demands and both volume and pressure-based demands. Two cases for distribution of leakage in the network are considered during the design: (1) leakages that are distributed based on nodal demands, (2) those that are distributed based on the length of pipe. An iterative optimization methodology is proposed that considers the link diameters as continuous variables. It begins by assuming the nodal heads at all of the demand nodes are above the desired pressure heads. These assumed heads are successively corrected until the difference in cost in the two successive iterations is found to be negligible. The versatility of the methodology is shown by its application on various types of networks. The incorporation of pressure-driven leakage in the design process can be shown to generate optimal solutions that require incrementally larger diameters. However, when pressure-based demands are also considered with pressure-driven leakage, the increase in diameters is substantial.

It is well known that internal water pressure, including pressure transients, in water supply networks is a major contributing factor in many pipe failures. Reliable prediction of pipe failures needs accurate information about critical contributing factors such as internal water pressure. This paper reports on a numerical study of transient pressure development following a pressure-monitoring program. Two large water supply network sections were considered, and pressure transient hydraulic models were developed using a relevant computer program. Extended-period simulations (EPSs) were conducted for each section to understand its steady-state operation during normal operation. The pressure transient models were validated using field pressure data, which were measured using high-speed pressure-monitoring equipment installed at selected locations in both network sections. A methodology is proposed to integrate the results of these simulations in a water utility asset management database to improve pipe failure prediction.

Knowledge of when, how, and by whom water is being used is crucial for planning ways to conserve drinking water. The goal of this paper is to identify groups of similar households (whom) based on their regular high-magnitude behaviors (RHMBs) of water consumption (when and how). RHMBs are frequent recurrences of high water use with regular timing. Household RHMBs are promising targets for behavior change. A two-stage data analytics approach is proposed. First, smart meter data is analyzed to identify RHMBs automatically. Second, salient features of the RHMBs are used to group households with similar behaviors. The approach is evaluated on two contrasting towns from low-rainfall regions of Australia. RHMBs accounted for 2 to 10 times more water than the traditional water efficiency target of continuous flows. For one group of 220 households, 60% of peak-hour demand was RHMBs. This paper demonstrates how RHMBs can be used to pinpoint opportunities for tailored demand management. Targets for substantial reductions in water consumption and supply costs are identified.

Urban water demand forecasting is key to municipal water supply management. Short-term urban water demands are influenced by weather conditions. Thus, short-term urban water demand forecasting could be improved by using accurate weather forecasting information. This study explores the potential of using an analog approach with a newly developed retrospective weather forecast (reforecast) of a numerical weather prediction (NWP) for improving short-term urban water demand forecasting. The analog method derives an analog ensemble forecast resampled from observed data (analogs) based on the reforecast of a NWP: the Global Ensemble Forecast System (GEFS). The probabilistic and ensemble mean forecasts generated from analogs of weekly total rainfall (WeekRain), number of rainy days in one week (RainDays), number of consecutive rainy days in one week (CosRainDays), number of hot days in one week (HotDays), and daily mean temperature of the first seven lead days (T) from the reforecast were evaluated using in situ observations. The analog ensemble forecasts were used to drive seven water demand forecasting models based on autoregressive integrated moving average with exogenous inputs (ARIMAX) to make water demand forecasts in the Tampa Bay region of Florida. The GEFS-based analog forecast generally showed moderately high skill for WeekRain, RainDays, CosRainDays, and T but no skill for HotDays. The water demand forecasts driven by analog forecasts mostly showed higher skill than the original ARIMAX forecasts. Besides improving forecast accuracy, the analog-driven water demand forecasts accounted for the uncertainty of the weather forecasts, allowing for the assessment of demand forecast uncertainty. These results indicated that NWP-based analogs showed promising features for advancing the accuracy of short-term urban water demand forecasts.

This paper provides an insight into the selection of the most suitable configuration for closed distribution systems (i.e., systems with no storage capacity). The analysis is developed considering a wide ranges of operational scenarios in terms of flows and required heads. For each scenario, various design configurations are considered and compared. Results show that using more than a minimum of pumps can have lower operational and total costs thanks to the pumps operating closer to their best efficiency point. Small additional benefits in terms of operational and total costs may be obtained as a result of the introduction of the variable speed drive in at least one of the station pumps. Similar costs are obtained in the configuration where large pumps are flanked by a small jockey pump operating at low demand times. The design solution made up of large pumps fitted with a downstream hydropneumatic tank also represents a valid alternative option from the economical viewpoint for small flows.

A single-objective optimization model has been developed for water distribution system (WDS) pumping operations, considering five different types of pump operating regimes. These regimes use tank trigger levels, scheduling, and a combination of both to control pumps. A new toolkit development to alter rule-based controls in hydraulic simulation software has allowed more complex pump operating regimes than have previously been considered to be optimized. The performance of each of the regimes is compared with respect to two different objectives: cost and greenhouse gas (GHG) emissions, which were optimized separately to allow the comparison of regimes to be made more clearly. Two case study networks, including one that represents a segment of the South Australian WDS, illustrate the effectiveness of the model. Time-based scheduling operating strategies were found to perform better than the other types of pump operating regimes. Significant cost savings were achieved for the South Australian case study network compared with its current operation.

In this work, a novel methodology is presented to identify isolated DMAs (District Metered Areas) in a water distribution network having a large number of water sources, such as autonomous reservoirs, directly connected to the distribution network. The method proposed is based on graph theory. Initially, the network is subdivided to the least terms into elementary districts, areas of influence of each supply source. In a second step, the resulting elementary areas are conveniently coupled to form possible layouts of isolated DMAs solutions. The union process takes into account the size of districts and the overall network resilience. The method is effectively applied to a case study derived from a large real water distribution network.

Water scarcity and aquatic environment deterioration have become increasingly serious problems in Urumqi, China. To alleviate the conflicts arising from competitive water users and retard the environment degradation, an optimization model was developed for regional water resources management under considering the dual total amount of water quantity and pollutant emissions control. Water allocation schemes and guarantee rates that combine diverse water sources and economic activities were obtained through the solutions of the proposed model. The interactions among water allocation schemes, various activities benefits, and pollutant emissions were analyzed synchronously. Policies for industrial structure adjustment and pollutants emission reduction were acquired through the analysis of the results. The modeling results can be valuable for water resources planning and justification of the existing industrial structure.

There are significant computational requirements for assessing climate change impacts on water resource system reliability and vulnerability, particularly when analyzing a wide range of plausible scenarios. These requirements often deter analysts from exhaustively identifying climate hazards. This technical note investigates two approaches for generating a subset of stochastic climate realizations that efficiently explore a range of risk to water supply systems. In both methods, a large ensemble of stochastic weather time series is generated to simulate the natural variability of the local climate system, and a selected subset of these sequences is used in the impacts assessment. Method 1 selects the subset by first passing the entire ensemble through a rainfall-runoff model and then screening the hydrologic sequences using the sequent peak algorithm. Method 2 selects a subset of climate sequences based on climate statistics alone, prior to hydrological modeling. Both methods provide insight for identifying the climate statistics that best relate to the vulnerability of the water system and can be used to reduce the computational burden of modeling climate variability and change impacts.