ZIM - SmartPress
Project: Transfer (R&D project)
Project participants
- Heger, Jens (Project manager, academic)
- Zein El Abdine, Mazhar (Project staff)
- Voß, Thomas (Project staff)
Description
In this cooperation project, an innovative data driven system is to be developed. This system aims to improve the production lines for cold and hot forming of metal sheets, rendering them more robust against changes in material properties and the surrounding. In the field of sheet metal forming, it is often the case that several stages of the process are arranged directly behind one another. Current systems employed operate as a ‘black box’, where it is only possible to observe the first and last stages of the process. This leads to an inefficient method of quality control, since no conclusions can be made regarding the causes of quality fluctuations. Consequently, if a faulty component is produced, the process still undergoes several pressing stages until the defect is discovered. All components that follow are similarly produced until the defect is detected later during the quality control and communicated back to the front of the process chain. The most common defect when deforming sheet metals is the tearing of the material and the formation of cracks. This occurs if, for example, there is scarcity of material in the process or the forces are incorrectly adjusted. Such defects depend on the current material and product properties as well as the system parameters, such as the pulling forces. With varying product properties (such as different degree of oiling), a corresponding adjustment of the process parameters (in this case the drawing forces) is necessary.
Status | Finished |
---|---|
Period | 01.12.17 → 31.05.20 |
Research outputs
Using data mining techniques to investigate the correlation between surface cracks and flange lengths in deep drawn sheet metals
Research output: Journal contributions › Conference article in journal › Research › peer-review
Smarte Anpassung von Presslinienparametern: Bildgebende Sensorik und maschinelles Lernen für robustere Blechumformprozesse im Automobilbau
Research output: Journal contributions › Journal articles › Research › peer-review