The part quality that can be achieved in forming and stamping processes strongly depends on the properties of the sheet metal material to be processed. However, since these material properties may fluctuate considerably and thus lead to the production of scrap, it is important to monitor such material fluctuations during part production. For this, the ongoing digitization of production processes provides new possibilities for part or quality monitoring. In this context, a novel AI-based method for the direct determination of material parameters from punching force curves measured in production was presented in a past study by the authors. This paper deals with the investigation of three further methods for extracting features from these recorded measuring data. In addition to domain knowledge-based feature engineering, statistical feature extraction (PCA) as well as a derivative-based method are analyzed and compared with each other and with the previously used AI (ANN) regarding their prediction accuracy of sheet metal properties.
CITATION STYLE
Görz, M., Schenek, A., Liewald, M., & Riedmüller, K. R. (2023). Evaluation of Feature Engineering Methods for the Prediction of Sheet Metal Properties from Punching Force Curves by an Artificial Neural Network. In Minerals, Metals and Materials Series (pp. 75–85). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-22576-5_8
Mendeley helps you to discover research relevant for your work.