The advancing digitization opens up new possibilities in the value chain. With the increasing integration of sensors, relevant parameters of systems can be continuously recorded and stored in databases. This enables, for instance, continuous monitoring of technical systems with regard to their health state. In addition to conventional approaches based on physical modelling, databased methods in the field of machine learning have emerged in recent years. Algorithms can be used to train models on the available database in order to interpret new data and transfer it into a state of health information.
These approaches are not only interesting from an information technology point of view, but can also benefit from the expert knowledge of engineers. For instance, mechanical engineers are required to select meaningful measurement parameters that can be converted into databased models.
In this lecture, the students will gain application-oriented insights into the basics of machine learning using examples from the current research of the involved institutes. The lecture will cover relevant areas of statistics, data mining and algorithm development. Based on the presented application cases the students learn to acquire interesting data from an engineering point of view, to filter the data, to extract relevant features and to build models for diagnosis and prognosis using methods of machine learning. Common process models are taught as well as the final evaluation and assessment of the developed methods and models.
Machine Learning Applications is a 6 CP lecture from the “Wahlpflichtbereich Ib” in the Master Mechanical Engineering.