In the industry there is a need to:
- Design and manufacture gears and gear trains with extended rating life in a cost-efficient way;
- Monitor and predict the Remaining Useful Life of gearboxes and drive trains.
This is suitable for applications such as the condition monitoring of gears or the design & validation of gears. Unfortunately, we encounter technological barriers upon finding a solution:
- Direct (ground truth) measurements: Lack of technology to measure directly and continuously the initiation and evolution of a gear defect
- Today vision inspections are performed periodically and the inspected infrastructure must be stopped and secured.
- The environment is challenging (absence of light, dust, oil, etc.)
- Correlation of ground truth with indirect measurements: the correlation of the diagnostic indicators with the true evolution of the defect has not yet been achieved
- There is no correlation between the level of existing diagnostic indicators and the true degradation
- Robustness and genericity of diagnostic gear indicators: the influence of the type and size of gears (helical, spur) and the operating conditions (load, oil film thickness, speed, …) at the time and the accuracy of defect detection is not yet clear and validated.
With this project we aim to deliver:
- An accurate failure mode identification and a quantitative measurement of gear degradation based on vision which can be used as a ground truth.
- A camera-based inspection system
- Robust gear diagnostic indicators evaluated based on labelled data with quantified false alarms and missed detection rates
QED_SBO is a Strategic Basic Research (SBO) project. We are looking for companies to join the User Group and work with us on the valorisation of the project.
Interested? Please complete the form below and we will contact you as soon as possible: