Water is a precious resource, but one that is becoming increasingly scarce in many regions. How Siemens is using intelligent, self-learning systems to help water suppliers detect and locate new leaks online around the clock.
It is misleading to believe that water scarcity is mainly a problem of the global South. After all, the number of people without access to clean water in Africa alone will grow to over 300 million by 2050 – an increase of almost four times compared to 2016. The global demand for water is growing by 1 percent every year and the UN warns that “half of the world’s population already suffers from severe water shortages for at least one month a year”.
However, the effects of water scarcity are also becoming increasingly noticeable in industrialized countries. The WRI (World Resources Institute) water stress ranking lists six Western European countries among the 47 countries with high or extremely high water stress. Every year, almost a third of the world’s drinking water – equivalent to the total amount of freshwater abstracted annually in France, Germany, Italy, Spain and the UK – is lost during distribution due to outdated infrastructure. Water must therefore be managed more efficiently.
But that won’t be easy. In many parts of the world, water infrastructure – especially drinking water pipes and sewers – is ageing, but it is complicated and costly to replace. London, for example, is known for the fact that large parts of its 10,000-mile drinking water network still consist of pipes from the Victorian era. At current replacement rates, the pipes being laid in European countries today should be sufficient for the next 200 years. New ideas are therefore needed and this is where artificial intelligence (AI) comes into play.
Digital technologies are not yet very widespread in the water sector, but real-life examples show how AI can already have a profound impact on the industry. The Swedish water supply company VA SYD supplies drinking water to more than half a million customers in Malmö, Sweden’s third largest city, and in the Skåne region in the southwest of the country. The company used to lose around a tenth of its water. Although this was already much better than most water utilities (and although Sweden is not a dry country), VA SYD was determined to do more to further reduce leaks in the network it operates, which consists of 5000 kilometers of pipes.
© SiemensThe Swedish water supplier VA SYD was able to reduce losses to 8% by detecting small leaks.
The company introduced SIWA LeakFinder, an AI technology from Siemens that analyzes water flow data to detect leaks down to 0.25 liters per hour per kilometer of pipe length online. The ongoing collection and AI-supported evaluation of flow data in the network also results in ever more precise characteristic flow profiles, making leakage detection even more accurate. If current data deviates significantly from a characteristic profile, SIWA LeakFinder issues an alarm message that also indicates the rough location of the leak.
This allows leaks to be repaired much more efficiently, reducing overall water losses by up to 50 percent. This is because it is precisely the smallest leaks – water that escapes undetected for a long time through tiny cracks or gaps in pipe connections – that cause far greater losses over time than a burst pipe, which is visible relatively quickly and can be repaired.
This also applies to VA SYD. By finding and repairing these small leaks, VA SYD was able to reduce water losses in the drinking water network from 10 percent to 8 percent. The AI-supported project has been recognized as best practice in Sweden and has prompted other municipalities to adopt the same solution for their water supply. Numerous Siemens references worldwide also confirm this.
“Water suppliers in Central and Eastern Europe are also consistently improving their network monitoring. With SIWA LeakFinder, numerous small and medium-sized cities also want to significantly reduce their water loss rates,” explains Gilbert Schreiber, Sales Manager Water and Environment at Siemens Austria. The example of VA SYD shows how AI can help to optimize the operation of drinking water networks and reduce water losses without having to replace entire network sections at great expense. “SIWA LeakFinder is a solution that is used successfully worldwide. We are ready for use by customers in Austria and Central Europe,” says Schreiber.
If used in a targeted manner, AI can therefore have a decisive impact on water management. For waterworks that know how to integrate AI well into their operations and, for example, make optimum use of the self-learning properties of AI-supported tools, the technology could even become a comprehensive monitoring system for the entire water cycle.
