We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Deep Learning Method for Fault Diagnosis in High Voltage Transmission Lines: A Case of the Vietnam 220kV Transmission Line.
- Authors
Le Van Dai; Nguyen Nhan Bon; Le Cao Quyen
- Abstract
One of the significant obstacles in electrical power transmission is the incident of failures, most notably the fault in transmission lines. It can influence the quality of the supplied electrical power, and actually, the country’s security and economy are threatened. Researching an intelligent solution to diagnose the fault is the done thing. This paper integrates deep learning methods, namely the discrete wavelet transform (DWT), GoogleNet, and probabilistic neural network (PNN) approaches, to detect, classify, and locate the fault. The DWT-based multiresolution analysis (MRA) is used for fault identification to process energy distribution during the transient states. After feature extraction, the obtained energy of the subband signal is used as the feature vector. It is employed as an input of GoogleNet for the fault type classification. The effectiveness of the proposed method has been demonstrated on the 220kV transmission line between the Hoa Khanh and Hue substations of Vietnam via time-domain simulation using EMTP-RV and MATLAB softwares. The simulated results are compared to the accuracy of the line distance protection REL521 for considering even cases, including the normal case and ten different fault causes, and the conditions of the pre-fault load changes. As a result, it concludes that the proposed method is a favourable method applied to the problem of analyzing the power system stability in the field of power transmission.
- Subjects
VIETNAM; HUE (Vietnam); ELECTRIC lines; FAULT diagnosis; HIGH voltages; DISCRETE wavelet transforms; DEEP learning; ELECTRIC transients; DIAGNOSIS methods
- Publication
International Journal on Electrical Engineering & Informatics, 2022, Vol 14, Issue 2, p254
- ISSN
2085-6830
- Publication type
Article
- DOI
10.15676/ijeei.2022.14.2.1