Baseform Tip: avoid water balance sectors with high pass-through

A common practice that severely undermines the reliability of water balance and NRW calculations

Flow pass-through – that is, water moving through a sector without being consumed – is a common feature in water distribution systems, as many network layouts are not sufficiently structured to provide a clear separation between the transport backbone and the distribution zones.

However, if the sectors are to be used for water balance calculations, that high pass-through flow can significantly compromise the validity of results. 

When the water that stays in the sector (the net input water, if you like) is much less than the volume passing through to downstream sectors, the combined error from the inlet and outlet flowmeters can become comparable in magnitude to that net input water volume. This will undermine the reliability of the water balance results (see here for some actual figures LINK), for example, when calculating NRW as a percentage of input volume. We frequently encounter this issue among the thousands of sectors monitored by utilities using our software.

How high is too high?

The combined (propagated) error of the flowmeter combination depends on how many inputs and outputs, and on their individual errors. Calculating those errors is relatively simple math: here's a good resource, among many.

To give a simple example – a sector with one inlet, one outlet, and a flowmeter error of 5% at both ends:

• for a pass-through flow 2 times greater than the water that stays in the sector (its net input water), the error estimating the latter would be ≈ 18%. 
• for a pass-through flow 3 times greater, the error would be ≈ 25%. 
• for a pass-through flow 4 times greater, the error would be ≈ 32%. 

Flowmeter measurement errors of 5% or higher are not unusual in the 'real world', as a result of oversizing, defective setup/calibration, lack of maintenance, or significant wear. 

This input water estimation is, of course, used as the denominator of the NRW indicator, and anchors all the component analysis behind standardized water balance. Needless to say that an NRW figure with a 20% or 30% error probably defeats the purpose.

What to do?

This problem is not insurmountable, but the first critical step is recognizing its existence. Ideally, the solution is to modify the network topology to reduce or eliminate pass-through – either by adjusting external boundaries or by further subdividing the sector. However, these options may not be immediately feasible. In any case, there remains the problem of making sense of data already recorded.

In such cases, the alternative is to minimize the effects of high pass-through. Start by determining the flowmeter error and its impact on results, then work to reduce that error – for example by correcting the flowmeter issues mentioned above.

Taking extra care in establishing zonal flows and interpreting results, especially when using DMA management software such as Baseform, can go a long way toward ensuring that water balance components remain reliable, repeatable, and sufficiently accurate. Error is an integral part of water balance, and can be factored in.

Investing resources and money in setting up a system of monitored sectors or DMAs can be rendered moot by forgetting to take care of the accuracy and reliability of the monitoring infrastructure. As always, it helps to keep present that the key objective is to get to reliable water balance figures that can provide the basis for effective water loss management.

Get in touch to find out how your utility can also benefit from Baseform.