Optimising network usage
Langford Water Services reveals its new leakage-reducing concept and explains how it can reduce water loss in an unconventional way
Leak reduction, in traditional terms, means using a variety of techniques, painstakingly searching out leaks by use of either low-tech (listening stick) or high-tech (leak noise correlator) devices, checking fluctuations in pressure and varying levels of usage (night lines) when customer consumption is at a minimum. A newcomer on the scene, Langford Water Services, is taking a different approach that appears to be enjoying success in the most challenging situation in the country – Thames Water’s vast, complex London network. The company is concentrating on reducing night lines and the amount of water put into supply, which are conventional proxies for measuring leakage. But it is doing it in an unconventional way.
The concept that Langford Water Services works by is known as virtual water, or ‘Virwatt’. Langford’s managing director Gareth Jones explained: “Water companies throw water into the network and, if someone says there is a low pressure problem, for example, another pump is turned on to increase the supply of water and hence the pressure in the pipeline.
“We found that, if we looked at a network with our algorithms we could predict how much water a network should be using and compare this figure to how much water was actually being used. If you can then reduce the hydraulic gradient, you can reduce the amount of water being used in the network and can then reduce night lines – reduce the amount of water being put into the pipes – by anything up to 20%.”
This then will influence perceived leakage losses by up to 50% at a fraction of the cost of conventional strategies, he says. Gary O’Donaghue, Thames Water’s area network manager for Sewarstone in north-east London, the area where Virwatt was trialled, said: “[Langford] concentrated on making the supply more stable, improving pressure in areas where it was low and decreasing it in areas where it was too high. They also reduced demand for the zone, which is good.”
Like many areas throughout the UK, this particular network has been subjected to urban expansion – new housing, redeveloped industrial areas and shopping/leisure facilities placing an ever increasing strain on local water resources and the underground infrastructure over the past few years. Over a similar period, the water into supply levels have increased by some 20% which has, until now, been put down to urban expansion and increased leakage levels.
The work undertaken by Langford Water Services in optimising the network has effectively reduced the water into supply levels by a similar amount simply by improving the network’s efficiency. One quick win in the initial eight-month contract was achieved by examining valve operation in order to optimise flows. “The whole project was nothing to do with new assets but working with the existing ones,” O’Donaghue explained. “They looked at the way the network was working and reconfigured it so it was flowing in the best possible way.”
Langford looked at the lay of the land, throttling some valves so some flow directions in the mains were altered. The point of this was to use the area’s natural gradients to facilitate flows . This meant some booster pumps did not need to be used (an additional energy-saving bonus), and the whole network became more stable. O’Donaghue noted: “Before, we may have been pushing water to high points from lower ones.”
Some mains were also dedicated to particular pumping stations, so when Thames started these pumps, water was not drawn off from other areas that sometimes suffered from low pressure. “They have done some good work – they have reduced customer impact and we are finding it easier to fill our reservoirs”, added O’Donaghue. The preliminary work finished at the end of October 2004 and the company is currently working in another of Thames’s zones, hoping to achieve a similar success. Jones initially worked in the industry as a general instrumentation and telemetry subcontractor, specialising in flow metering during AMP2.
Forming Langford Water Services, he carved out a niche as a flow measurement expert, focussing on the entire process control loop and, crucially, where to site the flow meter. His work led him to question why water companies were finding it increasingly difficult to keep on top of the leakage problem, despite their huge investment in the physical assets. Jones teamed up with Geoff Taylor, a Thames Water engineer involved in leak investigation, and they came up with a new observation – conventional wisdom states that, in a pipe full of water, water will only flow if someone is using it or there’s a leak. Jones and Taylor realised the only reason water flows at a particular point is due to the hydraulic gradient at that point – something that had never been stated before.
Traditional network modelling works by assuming a steady flow and a static head, because it is assumed water is not compressible and that the pipes are full of water. The reality of any network is that the pipes are not simply full of water – varying amounts of chemicals and air are also present. This led Jones and Taylor to redefined the compressibility of water and prove it to Dr Hoi Yeung, of Cranfield University’s School of Water Science, with a table-top 3D model demonstration. The company has embarked on a two-year programme with Cranfield University’s Knowledge Transfer Partnership to refining the Virwatt theory.