Fellner, D., Strasser, T., & Kastner, W. (2024). Chapter Twelve - Misconfiguration detection of inverter-based units in power distribution grids using machine learning. In R. Arghandeh & Y. Zhou (Eds.), Big Data Application in Power Systems (pp. 269–292). Elsevier Science. https://doi.org/10.1016/B978-0-443-21524-7.00009-8
Nowadays, electric power distribution grids increasingly incorporate distributed renewable generation with volatile power infeed along with new electrified loads such as heating systems or electric vehicles. These devices can introduce problems of grid overloading or voltage band violations. To cope with such problems, those units are usually equipped with grid-supporting control services. However, power system operators and energy utilities have no way of ensuring that these functions work as intended, usually due to a lack of sensory capacities in the field. Hence, additional monitoring capabilities are necessary. Therefore, this chapter introduces a framework for the misconfiguration detection of grid assets, especially inverter-based units, by handling operational grid data. Also, a linked monitoring application that uses data from the substation transformer and device levels for data mining and misconfiguration detection is introduced. The detection methods are merged with a disaggregation method to form an integrated diagnosis framework. The functionality of this integrated application is demonstrated on a selected grid-integrated photovoltaic system use case.
en
Research Areas:
Computer Engineering and Software-Intensive Systems: 25% Information Systems Engineering: 25% Climate Neutral, Renewable and Conventional Energy Supply Systems: 50%
