Search
Nominate/Submit Case Studies
Job Board

Deprecated: parse_url(): Passing null to parameter #1 ($url) of type string is deprecated in /var/www/wp-wpodev/wp-includes/rewrite.php on line 504

Deprecated: preg_match(): Passing null to parameter #2 ($subject) of type string is deprecated in /var/www/wp-wpodev/wp-includes/rewrite.php on line 526

Deprecated: explode(): Passing null to parameter #2 ($string) of type string is deprecated in /var/www/wp-wpodev/wp-includes/rewrite.php on line 534

Deprecated: parse_url(): Passing null to parameter #1 ($url) of type string is deprecated in /var/www/wp-wpodev/wp-includes/rewrite.php on line 504

Deprecated: preg_match(): Passing null to parameter #2 ($subject) of type string is deprecated in /var/www/wp-wpodev/wp-includes/rewrite.php on line 526

Deprecated: explode(): Passing null to parameter #2 ($string) of type string is deprecated in /var/www/wp-wpodev/wp-includes/rewrite.php on line 534

FlowSure (2016)

Demonstrating value in self-learning intelligent event detection systems

Road flooding from burst pipe

An increasing UK population, concerns that climate change will lead to drier summers and tightening regulatory and public relations demands have seen UK water companies come under increasing pressure by Ofwat to minimise leakage and prevent interruptions to the clean water supply. As Government, regulators and water companies make pleas to consumers not to waste water, it is important that suppliers demonstrate this value too. Ageing infrastructure, unpredictable weather and increasing demand can place strain on a network, which inevitably leads to leaks, bursts and other major events that disrupt the supply so many rely on and take for granted.

The benefits of anomaly-detection software

When real-time data is combined with self-learning software, an additional layer of intelligence is added to the readily available network information, enabling water companies to optimise their business, and specifically:

  • Manage multiple network events occurring simultaneously.
  • Determine the severity of network events from the control centre via telemetry data.
  • Quickly find the source of bursts.
  • Prevent major instances of flooding and leakage.
  • Reduce costs of contact handling, compensation and regulatory penalties.
  • Meet and exceed regulatory targets.
  • Improve customer service levels.

Beating the burst

FlowSure, developed by Servelec Technologies, the leading provider of business optimisation solutions in the UK water industry, is more than a simple leakage detection system. It is a self-learning anomaly detection application that enables water suppliers to identify and predict events occurring on the network, rather than simply reacting to major incidents such as burst mains. This proactive approach to monitoring enables water companies to reduce costs, improve customer service and strengthen their position against regulatory targets.
Ensuring the UK’s water utility companies are able to minimise leakage and reduce the number of catastrophic burst events is key to minimising their expenditure and delivering improved customer service levels. FlowSure delivers a swift return on investment, but also ensures that the year-on-year performance enhancements and cost savings that are demanded by both shareholders and Ofwat are achievable.

What makes FlowSure different?

FlowSure’s self-learning capabilities provide autonomous network management support for operations, control centres and other key business areas.
FlowSure uses readily available, real-time data and smart algorithms operating in an Artificial Neural Network to automatically identify when and where a large burst or other significant event is happening, or is about to occur in a water network.
The application, which is initially ‘trained’ using existing data, continually monitors the water supply network until it detects abnormal behaviour. It then alerts the appropriate operational control and maintenance teams that a burst or other event is occurring. The configurable user interface provides guidance on the size of the event and maps of the affected area.
As supply network behaviour changes over time, the system is designed to keep itself up to date using the latest available telemetry data and learn about these changes automatically. The software can be hosted within company servers, use existing flow and pressure data, and be integrated with existing telemetry systems, such as Servelec Technologies’ SCOPE system.

Broken pipe

Broken pipe

Smart alarms

Smart alarm systems have the ability to learn what an acceptable level of tolerance is for any given signal, rather than simply working to pre-set thresholds.
FlowSure teaches itself from pre-installed data what is and is not usual and acceptable within a network’s telemetry data and then continuously adapts its own thresholds and reports anomalies against these behaviours. This enables accurate alarms to be presented visually, which help maintenance teams to react proactively with greater speed and accuracy therefore reducing costs, avoiding penalties and improving levels of customer satisfaction.

Other applications

In addition to the clean water applications of FlowSure, the principles of anomaly detection can also be applied to sewerage networks to detect blockages and hence reduce flooding and pollution. Abnormal trends in level monitors given the prevailing conditions can be identified and combined with information regarding the likelihood and consequences of asset failure to prioritise locations for investigation.
More generally, the approach can be applied to identify anomalies in a variety of signals monitored through telemetry systems. Companies have identified applications of FlowSure in turbidity monitoring and communications signal strength monitoring.
Performance of intelligent anomaly detection software FlowSure has been successfully trialled by a number of UK water companies that have demonstrated annual six-figure net savings can be achieved. These savings are achieved by avoiding network leakage, which reduces costs of contact handling and regulatory penalties as well as leading to improved customer service levels. Increased network visibility can also improve work management processes by prioritising jobs and optimising available resources, ensuring the right person is deployed with the right skills at the right time. Wessex Water is one water company that has benefitted from the application of FlowSure.

Wessex Water case study

Wessex Water made the decision to heavily invest in their data-capture and telemetry infrastructure in the Weymouth region in preparation for the nautical events of the 2012 London Olympics.
During the Olympics, data collected from 220 logging devices and 40 remote telemetry units (RTUs) installed across 33 district metered areas were manually monitored around the clock to ensure that any network events were responded to immediately. However, this was not a sustainable solution and therefore Wessex Water needed a method for automated monitoring of the data they were capturing and alerting control room staff to emerging events, such as large bursts, when necessary.
Servelec Technologies were approached by Wessex Water to trial a system which would analyse the flow and pressure data for the Weymouth area in near real-time and raise alerts in response to anomalous data, including an estimate of the size of the associated burst.
In addition, Servelec Technologies transferred the flow and pressure data from the data loggers to the software via the Scope telemetry system and processed the alerts raised.
For Wessex Water, the trial proved the usefulness of the approach for leakage and network management using readily available data.
Following the trial Chris Muscat, Distribution System Manager at Wessex Water concluded:
“In all, the trial was very successful. The system was used to monitor network state changes and a number of genuine leaks were detected. Once the self-learning had been established there were very few false alarms and the system even coped with the summer demand variation with a minimal increase in alarms.
“These positive outcomes confirm the potential in this arena to develop further and provide additional tools. With decision support and increased logic these tools will allow better targeted deployment of resources and in the future, intelligent anomaly detection tools will be the only effective way to manage water supply networks.”
Retrospective analysis by Servelec Technologies showed that over 90% of high confidence (90%+ reported confidence) flow alarms were genuine anomalies.

Further success

More recently, further trials have been completed by Sutton and East Surrey Water. On these trials, Daniel Woodworth, Infrastructure Asset and Leakage Manager at Sutton and East Surrey Water commented:
“Our six month trial of Servelec Technologies’ FlowSure software proved highly successful. The system was easy to use, included a user-friendly dashboard interface and the flow alarms met and exceeded our brief and alerted us to anomalies in the data analysed. We consider intelligent anomaly detection software to have a particularly strong capability as an operational tool identifying the emergence of burst events.”
With FlowSure recently installed for trial across the entire Portsmouth Water network during the second half of 2016, the opportunities for water companies to move to a more proactive standpoint in managing their network are being consistently demonstrated.
The positive outcomes of these trials are communicated on the Servelec Technologies website, www.servelectechnologies.com .

This paper was prepared by Alan Cunningham , Technical Director at Servelec Technologies
www.servelectechnologies.com


Warning: Undefined variable $sc_post_id in /var/www/wp-wpodev/wp-content/themes/wpo-2023/single-innovation.php on line 191