A self-configurable fault detection system for Industrial Ethernet networks
-
Stefan Windmann
Stefan Windmann is postdoctoral researcher at the Fraunhofer Application Center Industrial Automation (IOSB-INA) in Lemgo since 2012. He studied electrical engineering and technical computer sciences at the University of Paderborn where he finished his PhD in 2008. From 2008 to 2012 he was working as software developer for image processing systems. Currently he is working in the fields of embedded software engineering, information retrieval in production systems and the optimization and diagnosis of industrial automation systems.
Fraunhofer IOSB-INA, Langenbruch 6, 32657 Lemgo, Germany
and
Oliver Niggemann
Oliver Niggemann received his PhD in Computer Science from the University of Paderborn. For 7 years he worked in various positions in industry, both in development and in leading management positions. In 2008 he accepted the professorship for technical computer science at the University of Applied Sciences OWL in Lemgo, Germany. He is an executive board member of the Institute of Industrial IT (inIT). Since 2009 is also the deputy director of the Fraunhofer Application Center for Industrial Automation in Lemgo. He is heading the graduate school Intelligent Systems for Automation (ISA) and he is the scientific director of the Graduate Center GZ.OWL in Lemgo. His research interests comprise methods and applications for Artificial Intelligence in production systems.
Fraunhofer IOSB-INA, Langenbruch 6, 32657 Lemgo, Germany
Abstract
In this paper, a self-configurable fault detection system for automated production systems with Industrial Ethernet is proposed. The scope of the proposed fault detection system are process variables, i.e., the observed actuator and sensor signals. Self-configuration of the fault detection system is enabled by recording and analyzing the link connection of the Ethernet network during system start. In a subsequent training phase, a knowledge base is automatically built from the observed process variables. Knowledge-based fault detection is accomplished once the knowledge base is established. Fault detection has been evaluated for a glue production process. In this application case, the knowledge-based fault detection method yielded a balanced accuracy of 99.81%, while a model-based method, which has been used as reference, produced a balanced accuracy of 93.11%.
Zusammenfassung
In diesem Beitrag wird ein sich selbst konfigurierendes Fehlererkennungssystem für automatisierte Produktionssysteme mit Industrial Ethernet vorgestellt. Das Fehlererkennungssystem wurde insbesondere für die Erkennung fehlerhafter Prozessvariablen, d. h. fehlerhafter Aktor- und Sensorsignalen, entwickelt. Die Selbstkonfiguration des Fehlererkennungssystems wird durch das Mitschneiden und die Analyse des Verbindungsaufbaus im Ethernet Netzwerk während des Systemstarts ermöglicht. In einer nachfolgenden Trainingsphase wird eine Wissensbasis automatisch aus den beobachteten Prozessvariablen generiert. Sobald die Wissensbasis aufgebaut ist, wird eine wissensbasierte Fehlererkennung durchgeführt. Die systematische Auswertung der Fehlererkennung wurde für eine Klebstoffproduktion vorgenommen. In diesem Anwendungsfall konnte mittels der wissensbasierten Fehlererkennung eine Genauigkeit von 99, 81 % erzielt werden, während mit einer modellbasierten Methode, die als Referenz verwendet wurde, nur eine Genauigkeit von 93, 11 % erreicht werden konnte.
About the authors
Stefan Windmann is postdoctoral researcher at the Fraunhofer Application Center Industrial Automation (IOSB-INA) in Lemgo since 2012. He studied electrical engineering and technical computer sciences at the University of Paderborn where he finished his PhD in 2008. From 2008 to 2012 he was working as software developer for image processing systems. Currently he is working in the fields of embedded software engineering, information retrieval in production systems and the optimization and diagnosis of industrial automation systems.
Fraunhofer IOSB-INA, Langenbruch 6, 32657 Lemgo, Germany
Oliver Niggemann received his PhD in Computer Science from the University of Paderborn. For 7 years he worked in various positions in industry, both in development and in leading management positions. In 2008 he accepted the professorship for technical computer science at the University of Applied Sciences OWL in Lemgo, Germany. He is an executive board member of the Institute of Industrial IT (inIT). Since 2009 is also the deputy director of the Fraunhofer Application Center for Industrial Automation in Lemgo. He is heading the graduate school Intelligent Systems for Automation (ISA) and he is the scientific director of the Graduate Center GZ.OWL in Lemgo. His research interests comprise methods and applications for Artificial Intelligence in production systems.
Fraunhofer IOSB-INA, Langenbruch 6, 32657 Lemgo, Germany
©2017 Walter de Gruyter Berlin/Boston
Articles in the same Issue
- Frontmatter
- Editorial
- Machine vision in automation technology
- Methoden
- Real-time motion prediction using the chromatic offset of line scan cameras
- Geometrical and spectral data stitching for combining hyperspectral imaging systems
- A self-configurable fault detection system for Industrial Ethernet networks
- Anwendungen
- Modellbasiertes Lernen in der Oberflächeninspektion
- Signalverarbeitung zur Inline-Prozessüberwachung für koaxial integrierte Messtechnik in der Lasermikrostrukturierung
- CAD-basierter Workflow für den semi-automatischen Entwurf optischer Inspektionssysteme
- Persoenliches
- Herrn Prof. Dr. rer. nat. Henning Tolle zum 85. Geburtstag
Articles in the same Issue
- Frontmatter
- Editorial
- Machine vision in automation technology
- Methoden
- Real-time motion prediction using the chromatic offset of line scan cameras
- Geometrical and spectral data stitching for combining hyperspectral imaging systems
- A self-configurable fault detection system for Industrial Ethernet networks
- Anwendungen
- Modellbasiertes Lernen in der Oberflächeninspektion
- Signalverarbeitung zur Inline-Prozessüberwachung für koaxial integrierte Messtechnik in der Lasermikrostrukturierung
- CAD-basierter Workflow für den semi-automatischen Entwurf optischer Inspektionssysteme
- Persoenliches
- Herrn Prof. Dr. rer. nat. Henning Tolle zum 85. Geburtstag
