Desalination & Reuse Handbook - Flipbook - Page 83
SMART LEAKAGE REDUCTION MARKET PROFILE
Smart technologies in network loss reduction
Gathering and processing data on leaks in a distribution network can be divided into data production, event detection, and
asset management. The issue of integrating data collection from different sources with data analysis is key to the approach
of many companies, with acquisitions and partnerships increasing the breadth of many players’ technology portfolios.
District metered areas (DMAs) are hermetically sealed sectors of a network established by sectorisation works. Meter data
can then be used to compare the difference in the quantity of water entering and exiting a sector against that sector’s
consumption. A significant discrepancy between the two figures indicates a leak within the DMA. While these do not come
under smart technology, DMA installation is often a key part of introducing digital solutions into a network.
Sensors (sometimes known as loggers) take readings of flow and pressure levels or noise levels at a section of piping,
producing data that can be used to infer the emergence of leaks. Sensor suppliers increasingly offer software packages to
manage the data produced by their products, which in turn can sometimes integrate data from third-party products.
Event detection software aggregates data from multiple sources to show all readings relating to a given distribution
network section. Taken together, an unusally high flow reading combined with high noise levels and a drop in pressure is
a more reliable indicator of a leak than any one of these readings in isolation. While some software platforms provided by
sensor suppliers can play this role to a limited extent, they can rarely integrate as many inputs as event detection software.
Asset management software makes the jump from data management to data analytics. These platforms take on some
of the work of a network operator by making decisions based on inputs such as sensors and meter data, consumption
patterns, and hydraulic modelling. The key difference between these platforms and event detection software is the use of
advanced modelling, artificial intelligence-based decision making, and the potential for partial network automation.
NRW’s digital transformation
Uptake of smart water solutions is growing rapidly, and digital
transformation is a key focus for many utilities. Smart metering
is the most common first step in this process, with demand for
large-scale smart metering projects taking on an increasingly
global profile. In 2016, Veolia’s concession award in Lyon,
France, promised the installation of 400,000 smart water
meters, 2017 saw Ghana’s Ministry of Water and Sanitation sign
an MoU for 500,000 smart meters, and in August 2018 Saudi
Arabia’s National Water Company (NWC) announced that almost
900,000 of the 2.1 million smart meters it intends to install by
2020 were already in place. These are most commonly set up
for drive- or walk-by automatic meter reading (AMR), allowing
for data collection via a vehicle-mounted or hand-held receiver
remotely collecting meter data. A small but growing minority of
metering projects involve the installation of permanent advanced
metering infrastructure (AMI) to deliver real-time data to a utility.
Installing sensors onto underground assets requires fewer
measurement points than metering projects, but requires a
more pro-active approach to both network upgrade works and
data management. Although the technology’s initial uptake has
been slow compared to metering alone, regulator-specified NRW
reduction targets have driven adoption in Western Europe, and
similar regulatory restrictions coming into force in China and
Southeast Asia are likely to have the same effect. Meanwhile,
water-scarce countries in the Middle East – particularly Israel,
Saudi Arabia, and the UAE – have implemented a number of
promising pilot network monitoring projects. Smart technology
for physical NRW reduction has developed a track record of
successful pilot projects that is likely to drive a rapid increase in
uptake (see map opposite).
Installing district metered areas (DMAs) – dividing a network
into sectors with isolated entry and exit points – is often seen
as a fundamental step towards gathering NRW data effectively.
In its most common form, smart leakage detection involves
the installation of flow meters at DMA entry points, sometimes
with the addition of pressure sensors at critical points in the
network. This data most effectively informs network operator
decision-making when drawn from multiple sources: an increase
in flow accompanied by a steep drop in pressure is a more
reliable indicator of a leak than a flow measurement alone.
Sensor data can be imported directly into a SCADA system for
manual analysis or integrated into event detection software for
automatic processing. Such is the volume of data sensors can
return, the latter is certainly more effective. In the absence of
support from software, operators are able to monitor critical parts
of the network but active processing or aggregation of datasets is
impractical.
Integration of data into a utility’s operational procedures remains
a challenge, especially when the data is sourced from multiple
products from different vendors. The sector has adopted several
approaches to solve this. One solution is to offer leak detection
as a service, with installation and data management rolled into
one contract and leaks reported directly to the client. Visenti’s
comprehensive leak management system is one such example.
Once it has installed flow, pressure, and acoustic sensors onto
a network, the company can monitor the data produced from
its own control room, alerting the client to probable leaks. The
system also takes pressure measurements at a high frequency,
allowing it to pick up on potentially destructive pressure
transients caused by sudden changes in network pressure.
By monitoring readings from above-ground noise loggers,
subsurface pressure sensors and DMA flow meters, the system
can both highlight rising leaks and extend asset lifetimes by
informing better operation of pumping systems and pressure
reducing valves (PRVs). A city-wide rollout of the technology is
planned in Singapore, where Visenti processes 8 billion data
points each day in one of the largest smart water networks in the
world.
Another supplier of pressure monitoring solutions, i2O, takes
this a step further. By integrating network control into their oNet
product, i2O’s system automatically adjusts PRVs installed at
DMA entry and exit points to reduce pipe bursts and water losses
caused by unnecessarily high network pressure. CEO Joel Hagan
sees an intuitive link between monitoring systems and network
75
