Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Modelling the Grinding Process with Regression Models and Artificial Neural Networks

: Lange, D.; Schmidt, T.; Westkämper, E.

Production Engineering. Research and development in Germany. Annals of the German Academic Society for Production Engineering 3 (1996), No.1, pp.5-10
ISSN: 0944-6524
Journal Article
Fraunhofer IPA ()
artificial neural network; Fertigungsprozeß; fuzzy logic; Fuzzy-Logik; Schleifen

Increasing quality requirements, high process safety, low production costs and short production times are the contrasting items that influence grinding as a manufacturing method. The process setting parameters have to be chosen in an optimum way corresponding to the quality requirements to be able to fulfil these demands. It is necessary for this to describe the process in models. Regression approaches are one possibility of process modelling. Because of the high comlexity of the grinding process and the high number of influencing quantities, these approaches quickly reach their limits. By using artificial neural networds and fuzzy logic methods for process modelling, even these complex relations can be learned. Thus, a calculation of the process and quality characteristics is already possible before or during grinding.