The exhibit shows how an AI-based acoustic monitoring system analyzes machine conditions, detects faults, and relies on cross-location learning. Identical machines are in use at each location of the system, and their operating noises are analyzed by AI. The pre-trained models classify three different conditions.
Since errors occur rarely, the amount of data available for AI training at a single location is limited. This is where distributed learning (federated learning) comes in: instead of exchanging confidential audio data directly, the AI models share only learned knowledge in the form of model parameters. This improves error detection across locations without compromising data security.
The demonstrator shows an innovative combination of intelligent acoustic condition monitoring and distributed learning – in this example, specifically for classifying engine noises.
Condition monitoring using airborne sound analysis and AI is conceivable for many applications in industrial production – whether for continuous monitoring of motors and gearboxes or for monitoring individual production steps, such as welding battery boxes. Thanks to the optimal selection of acoustic sensors and pre-trained AI models, deviations and errors can be reliably detected even in noisy industrial environments.
This technology sets new standards for efficient and secure AI-supported quality assurance in production – at every location.
Your contact at the stand:
Mareike Helbig
Fraunhofer IDMT
