Predictive maintenance is increasingly being adopted, as it can reduce unplanned downtimes and maintenance costs when industrial equipment breaks. In this video series, you will see how you can use simulation models of industrial systems along with Model-Based Design to cover the entire predictive maintenance workflow. The workflow spans from data acquisition and preprocessing to design and deployment of the predictive maintenance algorithm onto a PLC and as standalone executable or web application.
Part 1: Data Generation Learn how physical modeling can help you generate synthetic failure data necessary for the development of your predictive maintenance algorithm.
Part 2: Feature Extraction Learn how MATLAB can help you manage your data and extract useful condition indicators of your system.
Part 3: Training a Machine Learning Model See how the Classification Learner app enables you to train and validate your condition monitoring algorithm.
Part 4: Code Generation and Real-Time Testing Learn how to automatically generate code from your machine learning model and test it on real-time hardware (e.g., on a B&R PLC).
Part 5: Development of a Predictive Model Learn how to build a model to predict the remaining useful life (RUL) of your system.