Image: Sensor Edge Cloud chain from the sensor to the cloud<\/figcaption><\/figure>\n<\/div>\n\n\nUse Case Metering<\/h2>\n\n\n\n
Together with an industrial partner from the metering sector, the scientists at Fraunhofer IIS are investigating the possibilities of reducing the CO2<\/sub> footprint of a metering solution. Today, metering solutions are used to automatically read out consumption values for energy (electricity, gas, water, or heat) remotely. They can thus contribute to a change in consumption behavior and reduce household energy demand by improving the timeliness of consumption values.<\/p>\n\n\n\nIn a project phase lasting around six months, scientists at Fraunhofer IIS will use a sample product to investigate the extent to which different communication technologies and optimization of the energy supply can contribute to reducing CO2<\/sub>. As millions of smart meters are already being installed today, even small increases in efficiency can have a significant impact.<\/p>\n\n\n\nUse Case Condition Monitoring<\/h2>\n\n\n\n
In collaboration with industry partner Marposs Monitoring Solutions GmbH and Ruhr West University of Applied Sciences, Fraunhofer IMS is investigating the environmental impact of AI-based solutions for condition monitoring, in this case for wear monitoring of machines in machining technology. As supposedly \"old iron\", machining production processes still occupy a key position in mechanical and plant engineering. Innovative technologies such as condition monitoring enable potential savings of up to 37% of CO2<\/sub> emissions in these applications1<\/a><\/sup>, however, this requires additional computing power and sensor technology, and thus, especially in the case of mass distribution, a significant use of energy and materials.<\/p>\n\n\n\nA team of scientists from Fraunhofer IMS, together with Marposs and the Institute of Mechanical Engineering at Ruhr West University of Applied Sciences, is comparing various AI hardware modules that enable process monitoring directly on the machine tool without resorting to computing power in the cloud or at the edge. A comprehensive analysis of the energy consumption of milling machines is intended to identify potential for resource optimization through the use of embedded AI. Over a period of three months, the hardware module with the best environmental balance for this application will be determined from among those available. The further development of these processes and AI methods could lead to significant energy savings and improved sustainability in industry, such as through more efficient, targeted shutdown of machine components and energy-optimized process control.<\/p>\n\n\n\n
\n\n\n
![\"Bild](\"https:\/\/greenict.de\/wp-content\/uploads\/2024\/01\/grafik.png\")