Sensor technology finds leaks and tracks contaminants in drinking water

She says in cities' complex water distribution networks it is often difficult to determine what has happened in the pipes when leaks occur or water quality deteriorates.

By combining sensor data, mathematical models and machine learning, a team of researchers is now mapping the water’s pathways, with the goal of reducing drinking water losses and quickly stopping the spread of contaminated drinking water.

"Water is a critical societal resource that must be protected against both physical and cyber threats, so the monitoring of water systems needs to be strengthened," Fodor  says. She is research leader of the Digital Futures-project “Decision-making in Critical Societal Infrastructures".

The researchers want to answer two questions: “In which specific pipe in the water system has the leakage occurred that causes the water loss?” And “How does the pollution spread and where does it come from?”.

Water loss is a hot topic internationally . According to the OECD 's
calculations  from 2017, around 18 percent of Sweden's drinking water disappears on the way from source to consumer.
About 30 percent of treated water is lost worldwide  due to leakages in public water systems.

“In Stockholm, we are seeing major problems caused by the city's outdated water system . Our results show that by installing sensors in a few easily accessible places in the water network, and by using smart meters in the households, many of the problems can be detected quickly and accurately, Fodor says.

Thanks to the project's basic technology, which is based on 'big data', the researchers hope to offer municipalities more opportunities for cooperation in the future.
“We can hopefully help solve more problems than water leaks. For example, more efficient water drainage is needed for flooding, which is becoming more common with heavy rainfall and extreme weather.”

Katarina Ahlfort ( ahlfort@kth.se )
Photo: KTH/Jan Kansanen, Mostphotos

About the Democritus project

• The project Decision-making in Critical Societal Infrastructures (DEMOCRITUS) develops methods for monitoring and control of large-scale infrastructures using digitalization.
• The researchers design new methods for learning over large data sets, propose network solutions that support monitoring, learning and control, and construct data-driven models of the monitored physical processes.
• As an application, Democritus focuses on water distribution systems, which present many unsolved challenges for future societal systems.
• "An important outcome of the project is that we have initiated discussions on collaboration with the City of Stockholm, water treatment plants and the water industry," says research leader Viktoria Fodor.
   

Interdisciplinary collaboration

The team behind the research project "Decision-making in Critical Societal Infrastructures"  consists of researchers from the School of Electrical Engineering and Computer Science and the School of Engineering Sciences at KTH, along with researchers from Stockholm University and from RISE.

Experts have research experience in network design and optimization, learning and decision-making, security of cyber-physical systems and large-scale experimentation.