Around 60,000 compressed air systems are in operation in Germany, which together consume 16.6 terawatt hours every year - 7% of the total electricity consumption of domestic industry. However, up to 30 % of the energy used escapes unused through tiny leaks. Detecting these holes, kinks or leaking connectors has so far involved a great deal of effort.

"The products and methods available on the market for detecting leaks are not worthwhile for many users," says Christian Dierolf from the Industrial Energy Systems department at the Fraunhofer Institute for Manufacturing Engineering and Automation IPA. "To use them, you either have to repeatedly detect leaks with an ultrasonic device or retrofit new valves for individual monitoring of the pneumatic actuators." Many companies are therefore forced to live with waste.

Researcher Dierolf has tackled this problem in close cooperation with Sick and is developing an additional leakage service for an intelligent flow sensor from Sick. This service continuously records pressure, temperature and flow rate and generates seamless curves. These curves are evaluated by a self-learning algorithm. The trick here? "Clustering: leaks are reflected in characteristic curves. The algorithm recognizes these and sounds the alarm," explains Dierolf.

Implementation should be simple: The sensor, which Fraunhofer IPA and Sick are jointly developing from concept to series production, does not need to be connected to the machine control of the compressed air system or to an industrial PC. Instead, the flow sensor itself has a small display and other interfaces such as MQTT and OPC-UA, which allow the user to be notified automatically. The user can also access the sensor via a web interface. The algorithm will also teach itself. With this so-called unsupervised machine learning, a human only has to check at the end whether the algorithm has drawn the right conclusions from the available information.

"For us, the results from the joint development project are reason enough to make clustering available to our customers as an intelligent service as standard in future generations of flow sensors," says Thomas Weber, Head of Development in the Industrial Instrumentation division at Sick.

So far, the additional leakage service is a prototype. A compressed air demonstrator system was recently built at Sick's headquarters in Waldkirch and delivered to the Fraunhofer IPA in Stuttgart. It contains the sensor prototype, which is now being tested and further developed.