Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Active learning-based adaptive control in low-runner processes

: Kaupp, Markus; Neher, Joachim

Univ. of Naples; International Academy for Production Engineering -CIRP-, Paris:
CIRP ICME 2010, 7th CIRP International Conference on Intelligent Computation in Manufacturing Engineering. Innovative and Cognitive Production Technology and Systems : 23 - 25 June 2010, Capri (Gulf of Naples), Italy
Capri, 2010
4 pp.
International Conference on Intelligent Computation in Manufacturing Engineering (ICME) <7, 2010, Capri>
Conference Paper
Fraunhofer IPA ()
LowRunner-Prozess; Prozessüberwachung; process control; process monitoring; classification; Prozeßautomatisierung; Prozeßüberwachung

Automated production processes require robust and reliable monitoring and control systems. Unfortunately, the effort needed to adapt classical process monitoring systems to a specific production process is very high. Thus, this adaption usually does not pay off for processes with small to medium batch sizes (so-called Low-Runner processes). This paper presents a novel approach for process monitoring and control that is based on active learning. The proposed approach significantly reduces the effort needed to initially adapt the monitoring system to a production process. Therefore, it is appropriate even for Low-Runner processes.