Filtern
Dokumenttyp
Gehört zur Bibliographie
- nein (3)
Schlagworte
- machine learning (3) (entfernen)
Institut / FIR-Bereiche
- FIR e. V. an der RWTH Aachen (3) (entfernen)
The development of renewable energies and smart mobility has profoundly impacted the future of the distribution grid. An increasing bidirectional energy flow stresses the assets of the distribution grid, especially medium voltage switchgear. This calls for improved maintenance strategies to prevent critical failures. Predictive maintenance, a maintenance strategy relying on current condition data of assets, serves as a guideline. Novel sensors covering thermal, mechanical, and partial discharge aspects of switchgear, enable continuous condition monitoring of some of the most critical assets of the distribution grid. Combined with machine learning algorithms, the demands put on the distribution grid by the energy and mobility revolutions can be handled. In this paper, we review the current state-of-the-art of all aspects of condition monitoring for medium voltage switchgear. Furthermore, we present an approach to develop a predictive maintenance system based on novel sensors and machine learning. We show how the existing medium voltage grid infrastructure can adapt these new needs on an economic scale.
The shop floor is a dynamic environment, where deviations to the production plan frequently occur. While there are many tools to support production planning, production control is left unsupported in handling disruptions. The production controller evaluates the deviations and selects the most suitable countermeasures based on his experience. The transparency should be increased in order to improve the decision quality of the production controller by providing meaningful information during his decision process. In this paper, we propose a framework in which an interactive production control system supports the controller in the identification of and reaction to disturbances on the shop floor. At the same time, the system is being improved and updated by the domain knowledge of the controller. The reference architecture consists of three main parts. The first part is the process mining platform, the second part is the machine learning subsystem that consists of a part for the classification of the disturbances and one part for recommending countermeasures to identified disturbances. The third part is the interactive user interface. Integrating the user’s feedback will enable an adaptation to the constantly changing constraints of production control. As an outlook for a technical realization, the design of the user interface and the way of interaction is presented. For the evaluation of our framework, we will use simulated event data of a sample production line. The implementation and test should result in higher production performance by reducing the downtime of the production and increase in its productivity.
Machine Learning
(2019)