Abstract
Metal forming processes in general can be characterised as repetitive processes; this work will take advantage of this characteristic by developing an algorithm or control system which transfers process information from part to part, reducing the impact of repetitive uncertainties, e.g. a gradual changes in the material properties. The process is highly non-linear and the system plant is modelled using a non-linear finite element and the gain factors for the iterative learning controller is identified solving a non-linear optimal control problem. The optimal control problem is formulated as a non-linear least square problem where the system response is evaluated using a non-linear finite element model of the process.
Original language | English |
---|---|
Journal | International Journal of Advanced Manufacturing Technology |
Volume | 88 |
Issue number | 1-4 |
Pages (from-to) | 3–18 |
Number of pages | 16 |
ISSN | 0268-3768 |
DOIs | |
Publication status | Published - Jan 2017 |
Keywords
- Machine learning
- Iterative learning control
- In-process control
- Feedback control
- Metal forming
- Deep drawing
- Finite element method
- Process robustness