There is great potential for savings in compressed air, one of the most common and expensive forms of energy in German industry. Up to 30 percent of the energy used escapes unused from tiny leaks. The leakage add-on service for an intelligent flow sensor, which Fraunhofer IPA is developing with Sick AG, could make the tiresome search for leaks in the 60,000 compressed air systems in this country much easier: Self-learning algorithms evaluate the measurement data and thus get to the bottom of leaks.
Detecting holes, kinks or leaking connections used to take a lot of effort. This can change with the intelligent flow sensor: From the data on pressure, temperature and flow rate, it generates seamless curves. Since leakages are reflected in characteristic curve progressions, a self-learning algorithm can evaluate these curves. The user is then to be notified automatically via a small display and other interfaces such as MQTT and OPC-UA.