Creation and analysis of a quality regulation
|Semester:||Winter Term 2017/18|
|Suitable as:||Systems Engineering Project|
|Group size:||3-6 persons|
As part of the MoQua project funded by the BMWi, the quality of a thermoforming process is to be increased and thus the reject numbers are reduced. The organic sheets made from this process (Fig. 1) are used in the aerospace and automotive industry. A quality enhancement is achieved by implementing an intelligent control which has several subordinate components. One of these components is a quality controller.
A quality control should change the process parameters so that defined quality characteristics correspond to the specifications (Fig. 2). This procedure is necessary since direct influencing of the quality features is not possible.
The main task of this project is the realization and validation of the quality control. For this purpose, an approximation of the process model by means of an artificial neural network (ANN) is required. Furthermore, on the basis of this network, a control of the quality features must be realized and checked for real-time capability. Finally, the generated controller is validated on the real process, with an evaluation of the control performance.
- Creation of a process model using artificial neural networks in MATLAB
- Realization of a real-time control of quality features
- Evaluation and validation of the created methods