Neptune heralds a smarter approach to managing leaks

A massive collaborative effort is underway to come up with new tools to help water companies manage their distribution networks more effectively. Tim Door, ABB’s UK Water Industry Manager, explains.

Water companies in England and Wales manage distribution networks that total more than 335,000 kilometres, according to DEFRA. And that doesn’t include the almost 24 million connections to properties and associated customer supply pipes.

First there’s the bad news: Each and every metre of these pipes has the capacity to spring a leak. In 2008/9, water companies lost 2493 Megalitres per day (Ml/d) from the distribution network and a further 798Ml/d from supply pipes, adding up to a torrential 3,291Ml/d, according to the official figures from Ofwat.

The good news is that water companies have already made great inroads into the problems in their distribution systems, with leaks down by 5 percent in five years and by more than a third since they peaked in 1994/5.

A collaborative project involving industry and academics promises to provide companies with a whole new set of tools to ensure that they are targeting their ongoing leak reduction campaigns as effectively as possible.

Project Neptune is a £2.7 million, three-year initiative that aims to get companies working smarter by gathering and using real-time information on the pressures and flows in water distribution systems, not only to detect and repair leaks before they impact on customer supplies, but also to predict where failures are most likely to occur.

Launched in April 2007, Project Neptune is a strategic partnership between ABB, Yorkshire Water, United Utilities, the Engineering & Physical Sciences Research Council (EPSRC) and research teams at seven UK universities. The various tools being developed will help improve operator understanding about what’s going on in their distribution networks and enable them to take a more informed and integrated approach to three key areas: pressure and energy management, information management and decision support tools.

Data, data everywhere

The first step is to gather the data that the various models and alarms need to turn into useful information. This involves replacing traditional instrumentation with flow and pressure meters that can feed data back into the various Neptune applications.

Yorkshire Water had already started down this route before Neptune even started, with a programme known as RTNet which involved the replacement of existing manual logging equipment with systems equipped with onboard GPRS communications. Some 1,800 loggers were deployed by 2009, which is around 35% of Yorkshire Water’s entire logger population. The company plans to extend this to all 6,000 loggers in its 31,500 kilometre distribution network over the course of the AMP 5 investment cycle.

RTNet was initially trialled around the Harrogate and Dales area, with battery-powered devices transmitting data every 30 minutes. The resulting data trail successfully speeded up the detection and repair process during Yorkshire Water’s trial, with the company reporting that a burst that occurred before the system was implemented took around four hours longer to resolve than a similar incident during the trial.

Project Neptune takes the RTNet approach even further by using modelling to decide on the optimum placement of meters in order to maximise the sensitivity of the system using the fewest installed instruments.

Risky business
Another of the key innovations that looks set to emerge from Neptune is a risk-based decision support system, which is designed to ensure that the distribution system operator responds in the optimum way to any abnormal conditions and minimises the impact of any leaks on customer supplies.

There are two distinct aspects involved in assessing the risk of a burst. The first is analysing the chances of a particular burst occurring. The second is calculating the impact of a burst, should it occur. This combined information can then be used to prioritise alarms, so that those most likely to cause a big problem for customers get dealt with first.

For example, the likelihood of a pipe bursting is decided on the basis of three different models: the pipe burst prediction model, the hydraulic model and the customer contacts model. The first of these looks at the age and types of assets in the relevant section of the distribution system. Once this part of the model is defined it remains static. In contrast, the hydraulic model is constantly changing as the real time flow and pressure data enters the system. The customer contacts model also changes dynamically, according to the level of problems reported by customers.

The results can be presented in easy-to-use formats, such as risk maps over a given district metering area. Operators can also run manual “what if” scenarios to predict what impact different isolation strategies or other actions would have.

Essentially, the aim is to close the loop between the control room and the field:
• Collect data from the assets and transmit it to a central point
• Analyse the data and transform it into useful information
• Respond to the information and model potential responses from the control centre.

Pilot trials
Pilot tests of the new systems are currently being carried out in the distribution network of both Yorkshire Water and United Utilities. All the new tools and applications are deployed using one of ABB’s newest open platforms, the 800xA control system, combined with the closely integrated historian (PGIM) system.

Neptune is due to wind up in April 2010, soon after which ABB expects to commercialise the tools as a software package, subject to agreement with its partners on the project.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: