Meyer J, Dimopoulos I, Jäger J (2020)
Publication Type: Conference contribution, Original article
Publication year: 2020
Publisher: IEEE Computer Society
Book Volume: 2020-August
Conference Proceedings Title: IEEE Power and Energy Society General Meeting
ISBN: 9781728155081
DOI: 10.1109/PESGM41954.2020.9281758
The fast detection of a fault location is an important task in power grids to ensure system security at all. Todays protection relays are particular challenged by multivariate grid structures caused by volatile renewable infeeds. This paper presents an innovative strategy to use an Artificial Neural Network (ANN) for accurate fault location determination under all grid and fault conditions.The measurement results of several relays installed in the grid form a unique pattern for each fault condition, which an ANN is able to learn and recognize. This makes it possible to investigate faults on transmission lines that are not even protected by distance protection relays.The approach is validated with an adapted version of the IEEE 9 bus test power grid, where faults on various locations with different fault resistances were considered. The outcome is discussed and the effectiveness compared with the results of conventional protection relays.
APA:
Meyer, J., Dimopoulos, I., & Jäger, J. (2020). System-based fault locator based on a pattern recognition approach. In IEEE Power and Energy Society General Meeting. Montreal, QC, CA: IEEE Computer Society.
MLA:
Meyer, Janick, Ioannis Dimopoulos, and Johann Jäger. "System-based fault locator based on a pattern recognition approach." Proceedings of the 2020 IEEE Power and Energy Society General Meeting, PESGM 2020, Montreal, QC IEEE Computer Society, 2020.
BibTeX: Download