Data-driven Machine Condition Monitoring in Rotating Machinery

By Peter Karsmakers – KU Leuven

Description:

As the integration of sensors and IoT devices grows, vast amounts of operational data have become accessible, unlocking new possibilities for predictive maintenance and failure prevention. In this talk we will focus on AI techniques that leverage acoustic sensor data to detect anomalies, diagnose faults, and predict failures. Two case studies concerning rotating machinery will demonstrate how these methods can effectively reduce downtime and enhance maintenance strategies.

Presenter bio:

Peter Karsmakers is an Associate Professor within the Computer Science Department in the Declarative Languages and Artificial Intelligence (DTAI) section at KU Leuven and is a member of the Leuven.AI institute. Since 2022, he has been a principal investigator of Flanders Make. His research interests include designing machine learning algorithms that consider application-specific constraints like the computing platform, need for physical consistency, and limited availability of annotated data. He worked on diverse industrial collaboration projects that involve monitoring applications using microphones, accelerometers, and radars.