DMA fingerprinting: proof of concept - Executive Summary - Proof of concept for DMA fingerprinting, valve anomalies and leaks can be detected based on changes in pressure waves

Exploratory research has produced a proof of concept for the detection of closed valves in real water distribution networks using DMA fingerprinting: spectral analysis of pressure waves (transients) in District Metered Areas, as demonstrated theoretically by Richard Collins (University of Sheffield, UK). This research used a pilot DMA of the PWM water utility (at Velserbroek) with a single entrance (Figure 1.A). High-frequency measurements were registered over 22 days using pressure loggers at two strategic locations. Several disturbances, such as valve closures or the opening of a fire hydrant, were generated in the DMA. The spectrograms presented a stable picture during normal operational conditions and a visible change in the frequency response during the disturbances. Thus, a DMA does in fact have a fingerprint, and disturbances in a DMA can be detected. Important applications for drinking water utilities include the testing of (partial) closure of a valve and leak detection/localisation.

Interest: detection of anomalies in the distribution network. Previous research has shown that, on average, the position of 0.7 – 1.0% of the valves in drinking water distribution networks are other than expected. Typically, these anomalies come to light during flushing or repair work activities, resulting in additional CMLs (customer minutes lost). Greater understanding of anomalies in valve positions can help keep CMLs as low as possible. Theoretical research by Richard Collins (University of Sheffield, UK) demonstrates that pressure waves (transients) that flow through a DMA as a result of the opening or closure of taps in households, can provide information about valve position. Also, whenever someone turns on a tap (or takes a shower, uses the washing machine, etc.), a (small) pressure wave is transmitted through the DMA, which can be detected at the DMA’s entrance. The geometry of the DMA determines the path of these pressure waves and, combined with limited daily variations in drinking water demand, a frequency analysis of the transients produces a ‘fingerprint’ of the DMA in the spectral domain. Changes in geometry (associated, for example, with the closure of a valve or a leak) result in different paths and
therefore a different fingerprint. Although daily use varies a little every day, the frequency spectrum provides a reasonably stable picture as long as the path taken by the transient does not change significantly. When a valve is closed for example, the total volume that flows into the DMA and the pressure do not change. However, because the path of the pressure waves changes there is a visible change in the fingerprint of the DMA. The objective of this exploratory study is to verify whether transients can in practice actually produce a fingerprint of a DMA, and whether this makes it possible for instance to detect a change in the valve position.

Report – This Exploratory Research project is being followed up in DPWE2020, after which the outcomes will be described in the report KWR 2021.007 – “DMA Fingerprinting Fase II.