Real-Time Predictive Maintenance – Model-Based, Simulation-Based and Machine Learning Based Diagnosis

David Kaufmann, Florian Steffen Klück, Franz Wotawa, Iulia-Dana Nica, Hermann Felbinger, Adil Mukhtar, Petr Blaha, Matus Kozovsky, Zdenek Havranek, Martin Dosedel

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in Buch/BerichtBegutachtung


Predictive maintenance focuses on forecasting faulty or unwanted behaviour and defines appropriate countermeasures to be taken. Diagnosis, i.e., the detection of failures, the identification of faults, and repair provides useful foundations for predictive maintenance. In this article, we show how diagnosis, and in particular model-based, simulation-based and machine learning based diagnosis, can be used in practice. For this purpose, we introduce a simplified DC e-motor simulation model with the capability of fault injection to be used to show the efficiency of the introduced diagnosis methods based on the model’s behaviour. A simulation run of the system under test with pre-defined injected faults during runtime is used to validate the results obtained by the diagnosis methods. The results outline a promising application of these diagnosis methods for industrial applications, since each algorithm shows a time efficient and reliable diagnosis in relation to find the root cause of an observed faulty behaviour within the model. Further, the root cause analysis, performed with the introduced diagnosis methods, offers an excellent starting point for future development of self-adapting systems.
TitelArtificial Intelligence for Digitising Industry
Redakteure/-innenOvidiu Vermessen, Reiner John, Cristina De Luca, Marcello Coppola
Herausgeber (Verlag)River Publishers
ISBN (elektronisch)9788770226639
ISBN (Print)9788770226646
PublikationsstatusVeröffentlicht - Sept. 2021

Dieses zitieren