Fraunhofer IDMT from Ilmenau shows reliable system for non-contact acoustic monitoring for production

Press Release /

The Fraunhofer Institute for Digital Media Technology IDMT will present a novel monitoring method for industrial production at the most important international industrial trade fair from April 17 to 21, 2023 in Hannover. At the booth of the Fraunhofer-Gesellschaft (Hall 16, Booth A12), the Ilmenau-based research institute invites all visitors to test the reliability of the acoustic monitoring method for themselves at an interactive demonstrator. Based on artificial intelligence (AI), acoustic monitoring can be used for the quality assurance of products and the optimization of processes along the entire value chain.

Air-hockey table at the Hannover Messe 2023
© Fraunhofer IDMT
Fraunhofer IDMT Ilmenau's air-hockey table demonstrates at the Hannover Messe how acoustic event detection can work in a real-world environment.
Diagram for error analysis in production
© iStock.com/Phuchit
For common individual problem analyses, such as incorrect machine settings, clamping of workpieces, or tool wear, the technical staff uses acoustic signals to identify faults

This year, the researchers at Fraunhofer IDMT chose a special demonstrator. Unlike classic industrial exhibits, the Fraunhofer IDMT Ilmenau booth is all about sport, fun, and interaction.

Air-hockey table as principle demonstrator for acoustic monitoring

Visitors can play air-hockey and see how acoustic event detection can work in a real environment - based on airborne sound analysis in combination with machine learning methods. Pucks made from different materials are recognized based on their typical sounds during the game - in real-time and robust against background noise. 

Project manager Peter Hofmann explains the idea behind the unique exhibit: "Our air-hockey table is a principle demonstrator. We are using our table to demonstrate the potential of acoustic monitoring for production. Within production environments, we encounter a wide variety of soundscapes composed of machine noise, process noise, and other interfering sounds. The challenge for us is to 'pick out' the specific acoustic events from challenging soundscapes and analyze them reliably. Our demonstrator recognizes these specific acoustic events - in this case the different pucks - very reliably thanks to machine learning methods. The acoustic distinction of the pucks’ different materials demonstrates the automated detection of predefined classes, i.e., error patterns or states." 

Avoid machine downtime and bad product quality

What will be presented as a hands-on exhibit at Hannover Messe is designed to help avoid problems with unexpected machine downtime and bad-quality or destroyed products in real production and manufacturing environments. Production efficiency can be increased, and costs can be reduced through acoustic monitoring.

An additional added value is that contact-free acoustic monitoring can be used where others, such as optical monitoring methods, reach their limits. AI-based industrial sound analysis can be used along the entire manufacturing chain: from process monitoring, in-line and end-of-line quality control of products, to predictive maintenance of machinery.  

The research and development work incorporates the long-standing and internationally recognized expertise in the fields of acoustic measurement technology and AI-based audio signal analysis of Fraunhofer IDMT Ilmenau. 

Visit our experts for industrial sound analysis at the Fraunhofer booth A12 in hall 16 and test our acoustic detection accuracy with a game of air-hockey!

Our expertise: Acoustic monitoring in production

Last modified: