École de technologie supérieure
Fault classification in transmission lines utilizing imaging time-series and convolutional neural networks and adaptive relay protection
Abstract
dc:descriptionIn this research, a model is presented to fault classification in the transmission lines to detect and classify faults while keeping the coordination between the primary and the backup relays by adaptively changing the relay’s parameters accordingly. Furthermore, this research provides a method to select the directional overcurrent primary and backup relays in the transmission lines, as well as the mathematical model for coordinating these relays in a meshed power system. Additionally, a fault classification model based on deep learning as convolutional neural network to determine the fault type within the transmission lines. To comply with convolutional neural network, the voltages and the currents signals were transformed into images using Gramian Angular Field. The objective is to benefit from convolutional neural networks extract the relative temporal features from time-series signals, with the classification process performed using fully connected neural networks. Additionally, to ensure the synchronized operation of the primary and backup relay relays, the coordination between these relays are modelled as an optimization problem with constraints and objective function. Where the objective function is to minimize the total primary relay operating time by using GAMS software. The 9-bus test system is employed to determine optimal relay coordination according to the fault type and evaluate the results in comparison to existing literature.
Degree
thesis:*- Name thesis:degree_name
- Mémoire de maîtrise électronique
- Level thesis:degree_level
- masters
- Grantor dc:publisher
- École de technologie supérieure
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Khabaz, Baraa
Subjects
dc:subject × 1Rights
- Language dc:language
- en
Identifiers
dc:identifier.*- Identifier
- INSERT-YOUR-LIBRARY-ID-HERE3317