Shahed University
Fault detection and classification of an HVDC transmission line using a heterogenous multi?machine learning algorithm
Saber Ghashghaei | Mahdi Akhbari
URL :
http://research.shahed.ac.ir/WSR/WebPages/Report/PaperView.aspx?PaperID=158864
Date :
2021/10/05
Publish in :
IET Generation, Transmission & Distribution
DOI :
https://doi.org/ https://doi.org/10.1049/gtd2.12180
Link :
http://dx.doi.org/ https://doi.org/10.1049/gtd2.12180
Keywords :
Fault detection, HVDC transmission line, machine learning algorithm
Abstract :
This paper presents a novel integrated multi-Machine Learning (ML) system architecturefor the protection of bipolar HVDC transmission line in which different ML models ofSupport Vector Machine (SVM) and K-Nearest Neighbours (KNN) are used for faultdetection and classification. The KNN fault type classifier is designed as a dual-purposemodule, which not only detects the fault type but also acts as a redundant module forunsure fault declaration from the startup unit. Gradients and standard deviations of DCcur rent, voltage, harmonic current, and a correlation coefficient between the aerial andzero modes of DC current are appropriate feature vector extracted from single-end signalmeasurement. O verall, 154 training cases and 53 main test cases are obtained by simulatingvarious fault and non-fault states on a ±650 kV-1000 km Current Source Converter (CSC)–HVDC using an EMTDC/PSCAD platform. The ML modules are trained in MATLABand tested under different severe conditions with a total of 2220 test cases. Thanks to theappropriate feature vector and the proposed system architecture, the obtained results showthat the proposed algorithm is effective enough to detect and distinguish a variety of inter-nal faults and pseudo-faults/external faults. Also, it needs low training data requirements
Authors' Home page
Mahdi Akhbari