RT Conference Proceedings
T1 A nested decision tree for event detection in smart grids
A1 Turanzas, J.
A1 Alonso Martínez, Mónica
A1 Amarís Duarte, Hortensia Elena
A1 Gutierrez, J.
A1 Pastrana Portillo, Sergio
AB Digitalization process experienced by traditional power networkstowards smart grids extend the challenges faced by power gridoperators to the field of cybersecurity. False data injection attacks,one of the most common cyberattacks in smart grids, could leadthe power grid to sabotage itself. In this paper, an event detectionalgorithm for cyberattack in smart grids is developed based on adecision tree. In order to find the most accurate algorithm, twodifferent decision trees with two different goals have been trained:one classifies the status of the network, corresponding to an event,and the other will classify the location where the event is detected.To train the decision trees, a dataset made by co-simulating apower network and a communication network has been used. Thedecision trees are going to be compared in different settings bychanging the division criteria, the dataset used to train them andthe misclassification cost. After looking at their performanceindependently, the best way to combine them into a singlealgorithm is presented.
PB European Association for the Development of Renewable Energy, Environment and Power Quality (EA4EPQ)
SN 2172-038 X
YR 2022
FD 2022-09
LK https://hdl.handle.net/10016/36470
UL https://hdl.handle.net/10016/36470
LA eng
NO Procedings of: 20th International Conference on Renewable Energies and Power Quality (ICREPQ'22), 27-29 July 2022, Vigo, Spain.
NO This research was funded by Fundación Iberdrola España, within the 2020 research support scholarship program.
DS e-Archivo
RD 30 jun. 2024