scholarly journals Deep Learning for High-Impedance Fault Detection: Convolutional Autoencoders

Energies ◽  
2021 ◽  
Vol 14 (12) ◽  
pp. 3623
Author(s):  
Khushwant Rai ◽  
Farnam Hojatpanah ◽  
Firouz Badrkhani Ajaei ◽  
Katarina Grolinger

High-impedance faults (HIF) are difficult to detect because of their low current amplitude and highly diverse characteristics. In recent years, machine learning (ML) has been gaining popularity in HIF detection because ML techniques learn patterns from data and successfully detect HIFs. However, as these methods are based on supervised learning, they fail to reliably detect any scenario, fault or non-fault, not present in the training data. Consequently, this paper takes advantage of unsupervised learning and proposes a convolutional autoencoder framework for HIF detection (CAE-HIFD). Contrary to the conventional autoencoders that learn from normal behavior, the convolutional autoencoder (CAE) in CAE-HIFD learns only from the HIF signals eliminating the need for presence of diverse non-HIF scenarios in the CAE training. CAE distinguishes HIFs from non-HIF operating conditions by employing cross-correlation. To discriminate HIFs from transient disturbances such as capacitor or load switching, CAE-HIFD uses kurtosis, a statistical measure of the probability distribution shape. The performance evaluation studies conducted using the IEEE 13-node test feeder indicate that the CAE-HIFD reliably detects HIFs, outperforms the state-of-the-art HIF detection techniques, and is robust against noise.

2020 ◽  
Author(s):  
Gabriela N. Lopes ◽  
Luiz H. P. C. Trondoli ◽  
José Carlos M. Vieira

Detection of high impedance faults (HIFs) in distribution systems is a challenging task, which has attracted the interest of the researchers for decades. The HIF current random behavior and its lowmagnitude cause difficulties for a reliable detection by traditional protection methods. Therefore, the hazards for grid devices, people and animals safety, associated with HIFs, motivate the research of new detection techniques. However, there is no fully efficient solution for this problem. In this context, this paper aimed to characterize HIFs by a set of real measurements considering different type of soils employing Fourier (FT), Wavelet (WT) and Stockwell Transforms (ST). The measurements were performed at the fault spot in a medium voltage test field specially built for this purpose. The idea is to highlight key characteristics of the HIF current waveforms pointed out by each of transform and assess which ones can be used as a promising tool for HIF detection. The results showed that the HIF current can be characterized by the interharmonic behavior during the fault, extracted by FT and by the high degree of energy variations at specific decomposition levels of WT and ST.


2021 ◽  
Vol 11 (24) ◽  
pp. 12148
Author(s):  
Gian Paramo ◽  
Arturo S. Bretas

High impedance faults present unique challenges for power system protection engineers. The first challenge is the detection of the fault, given the low current magnitudes. The second challenge is to locate the fault to allow corrective measures to be taken. Corrective actions are essential as they mitigate safety hazards and equipment damage. The problem of high impedance fault detection and location is not a new one, and despite the safety and reliability implications, relatively few efforts have been made to find a general solution. This work presents a hybrid data driven and analytical-based model for high impedance fault detection in distribution systems. The first step is to estimate a state space model of the power line being monitored. From the state space model, eigenvalues are calculated, and their dynamic behavior is used to develop zones of protection. These zones of protection are generated analytically using machine learning tools. High impedance faults are detected as they drive the eigenvalues outside of their zones. A metric called eigenvalue drift coefficient was formulated in this work to facilitate the generalization of this solution. The performance of this technique is evaluated through case studies based on the IEEE 5-Bus system modeled in Matlab. Test results are encouraging indicating potential for real-life applications.


Energies ◽  
2019 ◽  
Vol 12 (4) ◽  
pp. 598
Author(s):  
Tao Tang ◽  
Chun Huang ◽  
Zhenxing Li ◽  
Xiuguang Yuan

The identification of faulty feeder for single-phase high impedance faults (HIFs), especially in resonant grounding distribution system (RGDS), has always been a challenge, and existing faulty feeder identification techniques for HIFs suffer from some drawbacks. For this problem, the fault transient characteristic of single-phase HIF is analyzed and a faulty feeder identification method for HIF is proposed. The analysis shows that the transient zero-sequence current of each feeder is seen as a linear relationship between bus transient zero-sequence voltage and bus transient zero-sequence voltage derivative, and the coefficients are the reciprocal of transition resistance and feeder own capacitance, respectively, in both the over-damping state and the under-damping state. In order to estimate transition resistance and capacitance of each feeder, a least squares algorithm is utilized. The estimated transition resistance of a healthy feeder is infinite theoretically, and is a huge value practically. However, the estimated transition resistance of faulty feeder is approximately equal to actual fault resistance value, and it is far less than the set threshold. According to the above significant difference, the faulty feeder can be identified. The efficiency of the proposed method for the single-phase HIF in RGDS is verified by simulation results and experimental results that are based on RTDS.


Author(s):  
Kavaskar Sekar ◽  
Nalin Kant Mohanty

<p>High impedance faults (HIFs) present a huge complexity of identification in an electric power distribution network (EPDN) due to their characteristics. Further, the growth of non-linear load adds complexity in HIF detection. One primary challenge of power system engineers is to reliably detect and discriminate HIFs from normal distribution system load and other switching transient disturbances. In this study, a novel HIF detection method is proposed based on the simulation of an accurate model of an actual EPDN study with real data. The proposed method uses current signal alone and does not require voltage signal. Wavelet transform (WT) is used for signal decomposition to extract statistical features and classification of HIF into Non-HIF (NHIF) by Neural Networks (NNs). The simulation study of the proposed method provides good, consistent and powerful protection for HIF.</p>


2018 ◽  
Vol 2018 (15) ◽  
pp. 1120-1124 ◽  
Author(s):  
Vassilis C. Nikolaidis ◽  
Angelos D. Patsidis ◽  
Aristotelis M. Tsimtsios

Energies ◽  
2020 ◽  
Vol 13 (23) ◽  
pp. 6447
Author(s):  
Abdulaziz Aljohani ◽  
Ibrahim Habiballah

High-impedance faults (HIFs) represent one of the biggest challenges in power distribution networks. An HIF occurs when an electrical conductor unintentionally comes into contact with a highly resistive medium, resulting in a fault current lower than 75 amperes in medium-voltage circuits. Under such condition, the fault current is relatively close in value to the normal drawn ampere from the load, resulting in a condition of blindness towards HIFs by conventional overcurrent relays. This paper intends to review the literature related to the HIF phenomenon including models and characteristics. In this work, detection, classification, and location methodologies are reviewed. In addition, diagnosis techniques are categorized, evaluated, and compared with one another. Finally, disadvantages of current approaches and a look ahead to the future of fault diagnosis are discussed.


Sign in / Sign up

Export Citation Format

Share Document