Predictive Maintenance on the Edge Device and PLC – from algorithm development to an industrial application

This presentation covers the entire predictive maintenance workflow using a digital twin model of a packaging motor. Machine learning algorithms for predictive maintenance are developed. In particular, healthy and faulty data is generated through a Digital Twin model. Shown is the use of apps to rapidly get started with feature extraction, algorithm development, and testing, using batch data generated by the digital twin model. For such algorithms, C code is generated and imported into a PLC, which is then tested in real time against a baseline RealTime HW machine onto which the plant model is deployed.


- Basic Machine Learning
- Basic automatic code Gen knowledge


How to develop a Machine Learning Algorithm using Data from digital twins and the automatic code generation of these ML Algorithm to run on embedded devices like PLCs




Dr. Rainer Muemmler works as a Senior Application Engineer at MathWorks and covers topics such as data analytics, artificial intelligence, predictive maintenance, hardware connectivity, image processing and the Internet of Things. Before joining MathWorks, he worked for various aerospace companies.



Newsletter Industrial IoT Conference

Sie möchten über die Industrial IoT Conference
auf dem Laufenden gehalten werden?