It would be great if there was a single button a user could click in one of Bentley’s software products that says, “eliminate lost water” and the problem would be solved. However, the problem isn’t that simple and there are multiple facets to water loss. As such, Bentley provide an array of tools that help water systems reduce water loss.
Reducing pressure will reduce water loss from existing leaks and possibly reduce the number of new pipe bursts. With zero pressure, there will be zero water loss. However, there is a tradeoff between high and low pressures. Water suppliers want to provide acceptable but not excessive pressure. This can be accomplished using pressure management. This can involve reducing the downstream pressure setting on pressure reducing valves (PRV), reducing the target set point for variable speed pumps or adjusting pressure zone boundaries so that users receiving acceptable pressure.
Utilities should not attempt pressure management in the field without first testing them with the water distribution system's “digital twin”, the hydraulic model. Using WaterGEMS’ Pressure Dependent Demand (PDD) feature, water modelers can estimate the change in leakage due to pressure reduction. The modeler can simultaneously investigate potential adverse impact such as inadequate pressure or a reduction in fire flows.
WaterGEMS’ Pressure Zone Manager can be used to plan or adjust pressure zones to ensure a reasonable range of pressure is maintained. Pressure Zone Manager can display pressure zones and calculate their properties such as minimum and maximum pressure and elevations. If customers have excessive pressure, they can be moved to a lower zone.
Hydraulic transients (water hammer) are a common cause for many pipe breaks. Reducing transients can reduce pipe bursts and joint separation. The HAMMER model can be run to identify locations that can experience excessive pressure during transient events. Identifying the problems areas can be the first step in eliminating them. The next step involves using HAMMER to evaluate operational changes or waterhammer control equipment that can control transients. There is no need to find leaks if they don’t occur.
If a leak occurs in a large water distribution system, it’s effect on pressure and metered flows can often be overlooked. However, if the system can be subdivided into DMA’s where the inflow is known, a leak’s impact on the DMA can be noticed. This can lead to a focused search for the leak as opposed to randomly searching a large area.
DMA’s should be laid out so that there are one or a small number of meters that feed the area. This often involves closing valves to isolate the DMA and eliminate any “backdoor” feeds. Closing inlets into a DMA can have adverse effects on pressure or fire flows and as with pressure management, it is best to check the impacts of DMA creation on the system’s digital twin before implementing the DMA.
DMA inlets can also install pressure management equipment (as mentioned under pressure management above) to both account for water and reduce leakage simultaneously. DMA’s can also lead to more precise water audits in that flows will be known on a more granular level.
DMA creation also involves installation of water meters. Knowing the expected range of flow through a meter is essential in selecting the size and type of a meter. WaterGEMS can be used to predict the range of flows that a meter will encounter over its life.
Leakage reduction is a key element of water loss reduction. A leak causes an increase in pipe flows which results in additional head loss and a subsequent drop in pressure. If a leak is sufficiently large and the measurement of pressure is sufficiently accurate, pressure readings can be used to help leak detection and repair crews pinpoint the leak location.
This is where WaterGEMS’ Darwin Calibrator for Leakage can be valuable. The model can be loaded with known water demands from metering. (This is especially accurate with modern automated meter reading.) The model will usually predict pressures that are above those measured in the field. The Darwin Leak can use a Genetic Algorithm to move leaks around the system until the pressures in Darwin match those from field measurements. Those leak locations are the best places for the utility to start leak detection.
Darwin’s ability to find leaks is highly dependent on the quality of pressure readings from the field. The pressure reading, the elevation of the pressure gage and the expected demands when those readings are taken must be known accurately. To increase the pressure drop in the system due to leakage, sometimes it may be advisable to open a hydrant to increase head loss and, due to the nonlinear nature of head loss equations, the head loss due to the leak.
The model needs to be well calibrated in that the connectivity, roughness and boundary conditions need to be correct. However, it is not possible to know the exact location of the leaks before calibration. A good approach is to place non-revenue water evenly around the system so that the total demand in the model matches system total demand. This can be done for example by a global edit to the base demands in the Demand Control Center, or by using the Active Demand Adjustments calculation option, to add a fixed amount to all demands. Then, once major model discrepancies have been resolved (for example using Darwin Calibrator, or manual methods), remove the non- revenue demand and allow Darwin Calibrator to move the leak to the most likely leak location.
If there is a closed valve or a pump status is set wrong, Darwin Leak can send the solution to the wrong place. In particular, a closed valve can behave like a leak in that it increases head loss and lowers the pressure. When an area is identified as a possible leak location, it may be worthwhile to first check for inadvertently closed or partly closed valves in the area. Opening a closed valve may not reduce water loss but it will improve hydraulic performance.
Overall, there are three sizes of leaks. Those that
Darwin Leak is appropriate for the third category of leaks.
Performing Leakage Detection Using Darwin Calibrator
Water Loss Management
(PDF Whitepaper - must be signed in) Innovative Optimization Model for Water Distribution Leakage Detection
(PDF Whitepaper - must be signed in) What does it take to make automated calibration find closed valves and leaks?