itsowl-TT-APPiD: Assistenzsystem zur Prozessüberwachung in Produktionsanlagen der industriellen Druckweiterverarbeitung

The aim of this transfer project is to implement an intelligent assistance system for automated monitoring of the production process of a bookbinding machine for the printing industry. The assistance system collects and analyzes the data from the production plant and on the basis of machine learning algorithms, automatically creates a model of normal process behavior for the prodution plant. The model is utilized in the reliable detection of process anomalies and identification of their causes in case of errors. In the course of this, existing learning methods developed in the Cross-Sectional Project “Energy Efficiency” of the Leading-Edge Cluster Intelligent Technical Systems OstWestfalenLippe (it’s owl) are used. The following types of errors and anomalies can be detected:

• discrete signal errors (e.g. Sensor defect)

• erroneous continuous signal sequences (e.g. defective process)

• timing errors of the overall system (e.g. wearing of drives)