Detection of Gear Damage During Operation – HighSpeed Imaging and Machine Learning for Damage Progression Analysis and Condition Monitoring
Research Topic
| Short Title | HighspeedGearVision |
| Start of Project | Q3/2025 |
| Funding | DFG-Nr. 532790251, STA 1198/27-1 German Research Foundation, DFG |
| Contact | Dr.-Ing. M. Otto |
Project Description
Gear transmissions are a fundamental component of many drive systems. An unexpected failure therefore typically involves high risks for people, machinery, and costs. For this reason, a precise understanding of potential damage, its origin, and progression is essential. This information is used both in the design phase for optimal dimensioning, and during operation for condition monitoring through correlation with sensor signals.
However, current models and investigations lack data with high-resolution and continuous damage information, especially during the moments when damage progresses rapidly. This means that damage is often only recorded quantitatively at the beginning and end of gearbox test runs or field operations, and occasionally during brief interruptions. Continuous damage information over the entire service life, so-called run-to-failure data, is difficult to obtain and therefore not available to researchers.
This research project aims to explore an innovative approach to continuously capture the temporal and spatial progression of surface damage during ongoing test bench operation. Due to the often very high speeds involved, a high-speed imaging technology will be integrated into the test bench and data acquisition system. Using automated image recognition and machine learning algorithms, continuously quantifiable damage assessments will be derived.