1.3 - Smart Sensor for predictive maintenance of a milling machine
- Event
- iCCC2026 - iCampus Cottbus Conference
2026-05-05 - 2026-05-07
Cottbus - Band
- Vorträge
- Chapter
- Condition Monitoring
- Author(s)
- J. Liebermann, M. Lehmann, D. Mayer, A. Schneider, A. B. Gowri - Fraunhofer IIS, Dresden
- Pages
- 42 - 45
- DOI
- 10.5162/iCCC2026/1.3
- ISBN
- 978-3-910600-10-2
- Price
- free
Abstract
We present a wireless, retrofittable smart sensor for continuous vibration-based condition monitoring and predictive maintenance of milling machines. The work is motivated by the need to detect specific conditions of the machine, like tool wear, to protect surface quality in high-throughput machining of large, costly parts. Our goal is to evaluate a lowcost hardware solution that can be quickly applied to existing milling machines against a state-of-the-art system. Following a methodical development cycle, we build a smart sensor including hardware and software components, acquire ground-truth data of milling with a sharp and a blunt tool, develop an edge-deployable AI model, implement the AI model on the smart sensor, and validate the implementation. Final deployment demonstrates reliable classification and actionable alerts, with comparative tests indicating monitoring quality comparable to a high-cost reference at reduced cost and integration complexity. Thus, the smart sensor can operate as a milling assistant, signalling worn tools with minimal installation effort and maximal data privacy.