A Parameter-Free Approach for Fault Section Detection on Distribution Networks Employing Gated Recurrent Unit

Abstract

Faults in distribution networks can result in severe transients, equipment failure, and power outages. The quick and accurate detection of the faulty section enables the operator to avoid prolonged power outages and economic losses by quickly retrieving the network. However, the occurrence of diverse fault types with various resistances and locations and the highly non-linear nature of distribution networks make fault section detection challenging for numerous conventional techniques. This study presents a cutting-edge deep learning-based algorithm to distinguish fault sections in distribution networks to address these issues. The proposed gated recurrent unit model utilizes only two samples of the angle between the voltage and current on either side of the feeders, which record by smart feeder meters, to detect faulty sections in real time. When a network fault occurs, the protection relays trigger the trip command for the breakers. Immediately, the angle data are obtained from all smart feeder meters of the network, which comprises a pre-fault sample and a post-fault sample. The data are then employed as an input to the pre-trained gated recurrent unit model to determine the faulted line. The performance of this novel algorithm was validated through simulations of various fault types in the IEEE-33 bus system. The model recognizes the faulty section with competitive performance in terms of accuracy.

Machine Learning-Based Fault Location for Smart Distribution Networks Equipped with Micro-PMU

“…Each typ of fault was simulated with different fault resistances of 1, 5, 10, 15, 20, 25, 30, 35, 40, 45 The SVM classifier was trained for single-phase-to-ground (AG), two-phase-to-ground (ABG), three-phase-to-ground (ABCG), and phase-to-phase (AB) faults. Each type of fault was simulated with different fault resistances of 1,5,10,15,20,25,30,35,40,45, and 50 ohms. There were 10 sections in the test network, which resulted in 10 classes.…”

“…A gated recurrent unit block is a simplified model of a long short-term memory unit that eliminates some complex mathematical functions [24]. In [25], a new framework based on deep learning that employs a special type of recurrent neural network called the gated recurrent unit to locate faulty sections in the network is suggested. In this study, smart feeder meters are assumed to be installed in all network nodes, which is its primary disadvantage due to their high cost.…”

A Survey of Time-Series Prediction for Digitally Enabled Maintenance of Electrical Grids

“…The main objective is to ensure that the network can continuously deliver electricity. Despite its necessity in urgent situations, reactive maintenance can incur high expenses due to its let us fail and react nature [6][7][8][9]. It is crucial to implement a comprehensive maintenance strategy that encompasses planned, proactive, and predictive approaches to minimize costs, enhance reliability, and optimize grid performance.…”

Deep Learning-Based Fault Classification and Location for Underground Power Cable of Nuclear Facilities

“…Therefore, accurate and rapid fault classification and location schemes in the NPP underground power cable are vital for overcoming the entire electric network faults. Such faults might produce hazardous transients, fire, and explosions due to the cable's excessive heating, equipment failure, and power outages, reducing nuclear system reliability and increasing the possibility of national disaster due to radiation leakage [11]- [13]. Various approaches for measuring fault location in transmission networks have been proposed over the years.…”

“…For failure prediction in power systems, pattern recognition approaches and ML algorithms have become crucial [19]. In [13], a new framework based on Deep Learning (DL) that employs a particular type of Recurrent Neural Network (RNN) called Gated Recurrent Unit (GRU) to locate faulty sections in the network is suggested. This study installed intelligent feeder meters in all network nodes, which is its primary disadvantage due to their high cost.…”