Detecting water leaks the smart way

| Information and Communication Technology

The once laborious and inefficient methods of water leak detection could soon be replaced by networked data logging monitored from an office desk.

There’s a new, convenient and cost-effective way of detecting and pre-locating leaks in water supply networks and you don’t even need to leave your desk to use it, says Craig Duthie of SebaKMT, part of the Megger Group.

The lift and shift approach

Leaks in the water supply network cost money – in many cases, a lot of money. So, in today’s cash-strapped business environment, there’s ever-increasing pressure to find and fix leaks quickly. The conventional approach is to wait until a leak is suspected or reported, then to despatch technicians to the area to confirm the problem and to determine its location.

The technique most often used is “lift-and-shift”. The technicians attach noise loggers to the pipes in the suspect part of the network and leave them in place for a period of time that can range from hours to days, depending on the nature of the problem under investigation. Then they retrieve the loggers and, typically after returning to their base, download the data from them, which they analyse using correlation and other techniques to confirm and pre-locate the fault.

If the results are inconclusive, the technicians may need to shift the loggers to a new location and repeat the whole process – hence “lift-and-shift”. This approach is labour intensive – at least two site visits are needed – and it can also take a long time to deliver the required results. In addition, it’s purely reactive – it won’t alert users to previously unknown problem, it is simply a way of investigating known or suspected leaks.

A better solution is to use noise loggers that can be interrogated wirelessly, install them at critical locations on the network and leave them permanently or semi-permanently in place. Then all that’s needed is for someone to drive round the logger locations periodically while wirelessly retrieving the data from them. This is much faster and more convenient than physically lifting the loggers for downloading, but it still requires a site visit and it still doesn’t provide a real time or even near real time overview of the network status.

Networked wireless leak monitoring

There is now however a solution that solves all of these problems – networked leak monitoring. Essentially this is an extension of the technique already described using wireless noise loggers but, instead of the loggers being interrogated from a passing vehicle, they automatically send their data to a central location at predetermined intervals – usually once a day – using the GSM mobile phone network.

The benefits are significant. The need for site visits to retrieve data is eliminated, which cuts costs and frees up technicians for more productive work. Possibly even more important, up-to-date information on the whole area monitored is available every day, which enables changes to be quickly detected and potential problems spotted before they develop into costly and disruptive major leaks.

Networked leak monitoring systems invariably provide tools for analysing data after it has been transmitted to the central location. These can, for example, be expected to include automatic calculation of enhanced spectrum analysis (ESA) values and history functions that make it easy to track the condition of the network over long periods, allowing critical sections to be readily identified. Well-implemented systems display the data from individual loggers overlaid on a map of the water network, making the information particularly easy and convenient to interpret.

The best systems go even further. If the loggers detect a potential leak, users can request the noise data from the affected measuring points, which is then sent to the central location as part of the next scheduled data transmission. Once this data has been retrieved, the users can listen to the recorded noises – still without needing to leave their desks – and thereby determine whether the alarm from the logger was caused by genuine leak noise or by spurious extraneous noise.

These top-end systems even allow data from multiple loggers to be accurately correlated, so that when a real leak is confirmed and the site team is despatched to pinpoint and repair it, they have a good, reliable indication of the leak location even before they set out. This elimination of uncertainty about the location saves both time and money.

Monitoring through the cloud

The most recent development in networked leak monitoring systems is cloud support. This means that the data available at the central location is also stored “in the cloud” – that is, on servers which can be accessed from any location where it’s possible to connect to the Internet.

Engineers and technicians can therefore instantly check the latest information using not only their laptop, but also with a tablet computer or even an Internet-enabled mobile phone. Fast, convenient access to this detailed information, even when working on site, is another major aid to efficiency and cost saving.

Hopefully, the benefits of networked leak monitoring are easy to appreciate, but there is sure to be a question about how much it costs. Undeniably there’s the up-front cost of buying noise loggers, GSM boxes and, for larger systems, wireless repeaters, but once the system is in place, the on-going costs are small. The equipment is reliable and has a long life, and the cost of transmitting the data to the central station is negligible.

And it’s always important to compare the costs with the savings, particularly in these days of expensive labour, limited availability of technicians and large financial penalties for wasting water through leakage. In reality, in even a modestly sized installation, a networked leak detection system will have a very short payback time, which makes it a very attractive – and very sound – financial investment.

Related news

Leave a Reply

Your email address will not be published. Required fields are marked *

5 × two =