Prerpints.org logo
Preprint
Review

Machine Learning Advances in Transmission Line Fault Detection: A Literature Review

Altmetrics

Downloads

170

Views

74

Comments

0

Submitted:

04 May 2024

Posted:

06 May 2024

You are already at the latest version

Alerts
Abstract
Fault detection in transmission lines plays a role in maintaining the dependability and steadiness of power networks. Traditional methods for identifying faults often struggle to handle the diverse nature of real world fault situations. Machine learning (ML) algorithms offer a data centered approach that can adjust and learn from datasets potentially overcoming the limitations of traditional approaches. This document presents a review of progress in using ML for detecting faults in transmission lines. By drawing insights from a variety of studies we explore how ML algorithms have evolved in fault detection, including techniques like networks, recurrent neural networks featuring Long Short Term Memory and convolutional neural networks. We delve into the spectrum of applications where ML is used for fault detection across fault scenarios and operational settings. Additionally we discuss the obstacles and prospects linked to putting ML based fault detection systems into practice such as challenges with data quality, model interpretability and integration with existing grid monitoring systems. Lastly we outline future research paths focused on pushing forward the boundaries of fault detection, in power transmission systems through approaches and collaborative endeavors involving academia, industry players and policymakers. In general, this review highlights how machine learning has the power to revolutionize fault detection methods enhancing the resilience and dependability of power systems.
Keywords: 
Subject: Engineering  -   Electrical and Electronic Engineering

Introduction

Transmission line fault detection is a crucial aspect of maintaining the reliability and stability of electrical power systems. With the increasing complexity and interconnectivity of modern power grids, traditional methods of fault detection face significant challenges in terms of accuracy, speed, and adaptability to evolving grid conditions. In response to these challenges, there has been a growing interest in leveraging machine learning (ML) techniques to enhance fault detection capabilities.
Machine learning offers the potential to automatically learn complex patterns from data, enabling more accurate and efficient detection of faults in transmission lines. In recent years, there has been a surge in research exploring the application of machine learning algorithms for transmission line fault detection [1]. Studies have demonstrated the effectiveness of various ML approaches, including artificial neural networks (ANNs), support vector machines (SVMs), decision trees, and ensemble methods, in analyzing electrical signals and identifying fault patterns. [2,3]
For example, Li et al. [4] utilized a deep learning-based approach for fault detection in power transmission systems, achieving high accuracy rates even in the presence of noise and disturbances. Similarly, Wong et al. [5] proposed a hybrid machine learning model combining convolutional neural networks (CNNs) and long short-term memory (LSTM) networks for fault diagnosis in power systems, showcasing significant improvements in fault detection performance compared to traditional methods.
The adoption of machine learning techniques for transmission line fault detection holds promise for enhancing the resilience and reliability of power grids. ML-based approaches have the potential to overcome the limitations of rule-based systems and signal processing techniques by autonomously learning from large volumes of data and adapting to changing grid conditions. Moreover, machine learning algorithms can provide insights into the underlying patterns of fault occurrences, facilitating proactive maintenance and grid optimization strategies. [6]

Purpose of The Literature Review

In this literature review, we aim to provide a comprehensive overview of recent advances in machine learning for transmission line fault detection. By synthesizing findings from recent studies and examining methodologies, results, and implications, we seek to elucidate the potential of machine learning to revolutionize fault detection in power transmission systems. Through a critical analysis of the literature, we aim to identify key trends, challenges, and future research directions in this rapidly evolving field.

Materials and Methods

A comprehensive search of electronic databases such as IEEE Xplore, ScienceDirect, Elsevier, and Google Scholar was conducted to identify relevant studies published between 2020 and 2024. Keywords including "machine learning," "transmission line," "fault detection," and related terms were used to retrieve articles. The inclusion criteria encompassed studies that specifically addressed the application of machine learning techniques for transmission line fault detection. After screening titles and abstracts, relevant articles were selected for full-text review. Data extraction included details on the machine learning algorithms used, datasets employed, evaluation metrics, and key findings.

Results

A total of 25 were finally included in this review literature after careful and thorough screening. Table 1 shows the fault identified and tested in these studies. Table 1 also presents the components of the transmission lines affected by a specific fault as well as the causes and effects of these faults.
Table 2 shows the fault detection and diagnosis using machine learning. The type of machine learning used was identified and presented in the table.

Machine Learning Algorithms

Recent advancements in transmission line fault detection have been driven by the application of various machine learning algorithms. Researchers have explored a range of techniques, including Bayesian neural networks (BNN), multi-layer perceptron neural networks (MLP), recurrent neural networks (RNN) with Long Short-Term Memory (LSTM), and convolutional neural networks (CNN). These algorithms offer unique advantages in capturing different aspects of fault data, from temporal dependencies to spatial features. For instance, LSTM networks are well-suited for capturing long-term dependencies in sequential fault signals, making them effective for detecting high impedance faults. On the other hand, CNN architectures excel at extracting spatial features from fault data, making them suitable for tasks such as short-circuit fault detection. Additionally, specialized approaches like capsule networks with sparse filtering (CNSF) and deep pyramid feature learning networks (DPFL) have emerged to address specific challenges in fault detection, such as hierarchical feature extraction and discriminative feature learning. By leveraging the capabilities of these machine learning algorithms, researchers aim to enhance the accuracy, efficiency, and adaptability of fault detection systems in power transmission networks, ultimately improving the reliability and resilience of electrical grids.

Dataset Characteristics

The datasets used in these studies vary in terms of fault types, operating conditions, and signal characteristics. Fault scenarios include line-to-ground (LG), line-to-line (LL), double line-to-ground (LLG), and triple line-to-ground (LLLG) faults, as well as high impedance and short-circuit faults. Some studies focused on specific fault types, while others considered a broader range of fault scenarios. The availability of diverse datasets enabled researchers to train and evaluate machine learning models on representative data, contributing to the robustness and generalizability of the developed models.

Performance Metrics

Performance evaluation metrics such as accuracy, precision, recall, and F1-score were commonly used to assess the effectiveness of machine learning models in fault detection. Aker et al. [7] reported high accuracy rates for fault classification using BNN and MLP models. Similarly, Fahim et al. [8] demonstrated the superior performance of self-attention CNNs in detecting short-circuit faults compared to traditional methods. These metrics provide insights into the model's ability to correctly identify faults while minimizing false alarms, enabling researchers to quantitatively evaluate model performance.
Table 1. Summary of different types of faults with connected information: affected components, causes and effects.
Table 1. Summary of different types of faults with connected information: affected components, causes and effects.
Preprints 105589 i001
Preprints 105589 i002
Preprints 105589 i003
Table 2. Fault detection and diagnosis via machine learning.
Table 2. Fault detection and diagnosis via machine learning.
Authors Year Fault Method Used in Detection and Diagnosis [7]
Aker et al. [7] 2020 LG, LL, LLG and LLLG Bayesian neural network (BNN), multi-layer perceptron neural network (MLP)
Anand et al. [9] 2020 LG, LL, LLG and LLL Empirical mode decomposition (EMD)
Belagoune et al. [10] 2021 High Impedance Long Short-Term Memory (LSTM) neural network
Dr. Bindhu V. el al. [11] 2021 Short Circuit Fault ZigBee communication protocol
Biswas et al. [12] 2019 LLLG, Unbalanced, Critical, Faults Producing Voltage/Current Inversion UPFC Unified power flow controller
PSCAD Power system computer aid design
Doria-García et al. [13] 2021 High Impedance, Nonlinear arcing Gauss-Newton method
(DPFL) Deep Pyramid Feature Learning Network
Fahim et al. [8] 2020 Short Circuit Fault Self-attention convolutional neural network (SAT-CNN) model
Fahim et al. [14] 2021 Short Circuit Fault Capsule network with sparse filtering (CNSF)
Ferreira et al. [15] 2020 Short Circuit Fault Feedforward neural networks (FNN)
Godse et al. [16] 2020 Short Circuit Fault Artificial Neural Network (ANN)
Agrawal et al. [17] 2020 Short Circuit Fault IoT diagnosis
Haq et al. [18] 2020 Three-Phase (LLLG) Fault Db4 wavelet
Leh et al. [19] 2020 Line-to-ground fault Feedforward neural networks (FNN)
Li et al. [4] 2020 Single-pole grounding fault VSC-HVDC
Liang et al. [20] 2020 Short Circuit Fault Region-based Convolutional Neural Network (R-CNN)
Liu et al. [21] 2021 Insulator Faults Region-based Convolutional Neural Network (R-CNN)
Lu et al. [22] 2020 Short Circuit Fault Time domain model based methods
Mukherjee et al. [23] 2020 Short Circuit Fault Artificial Neural Network (ANN)
Rafique et al. [6] 2021 LG, LL, LLG, and LLL Recurrent Neural Networks (RNN)
Teimourzadeh et al. [24] 2020 single-phase to ground short circuit Convolutional Neural Network (CNN)
Tong et al. [25] 2020 Short Circuit Fault, Three-phase (LLLG) Fault IEEE 39 bus system
Wang et al. [26] 2020 Three-phase (LLLG) Fault Wavelet noise Reduction, Clarke transform, Stockwell transform and Decision Tree (WRC-SDT)
Wong et al. [5] 2021 Short Circuit Fault Convolutional Neural Network (CNN)
Zhang et al. [27] 2021 Internal and External Fault Stationary wavelet transform (SWT)
Zheng et al. [28] 2021 Short Circuit Fault Region-based Convolutional Neural Network (R-CNN)

Discussion

Comparison with Traditional Methods 

Several studies compared the performance of machine learning-based approaches with traditional fault detection methods. Belagoune et al. [10] demonstrated the effectiveness of LSTM networks in detecting high impedance faults compared to conventional methods. Similarly, Zhang et al. [27] showcased the advantages of using CNNs combined with SWT for fault detection over traditional signal processing techniques. These comparisons highlight the superiority of machine learning models in accurately detecting faults and overcoming the limitations of rule-based systems and signal processing methods.
Machine learning-based fault detection approaches demonstrated robustness to different operating conditions, noise levels, and fault types. For example, LSTM networks used by Belagoune et al. [10] exhibited adaptability to high impedance faults, while CNN models employed by Fahim et al. [8,14] showcased resilience to noise and disturbances in signal data. The ability of machine learning models to generalize well to unseen data contributes to their adaptability in real-world applications and ensures reliable fault detection under varying conditions.
Compared to traditional fault detection methods, machine learning-based approaches demonstrated superior performance in terms of accuracy, efficiency, and adaptability. Traditional methods, such as rule-based systems or signal processing techniques, rely on predefined rules or features, which may lack flexibility and robustness in handling complex fault scenarios. In contrast, machine learning models autonomously learn from data, enabling more accurate and timely fault identification without the need for explicit rule definitions.

Limitations and Challenges 

Despite their effectiveness, machine learning-based fault detection approaches face several challenges. These include the availability of labeled training data, the interpretability of complex models, and the integration of ML-based solutions into existing grid monitoring systems. Furthermore, the deployment of machine learning models in real-time monitoring systems requires careful consideration of computational resources and infrastructure compatibility. Addressing these challenges is essential to ensure the practical applicability and reliability of machine learning-based fault detection solutions in power transmission systems.

Conclusion

The study of new machine learning ideas for finding faults on transmission lines shows a growing field that can make power systems more reliable and efficient. Experts have looked at many machine learning methods, like supervised, unsupervised and hybrid ways, to find faults on transmission lines more accurately, quickly, and reliably. By using advanced algorithms like artificial neural networks, support vector machines, decision trees, and deep learning models, they have made good progress in detecting and classifying faults on transmission lines. Also, using different datasets, ways to extract features, and optimization techniques has helped machine learning-based fault detection systems work better. The studies show that using machine learning methods to address the complex challenges of finding faults on transmission lines is important. These methods can automate fault detection processes, reducing downtime, lowering operational costs, and improving overall system reliability. However, challenges such as data quality, scalability, and interpretability remain significant areas of concern that warrant further investigation.
In conclusion, the literature review highlights the transformative impact of machine learning on transmission line fault detection, paving the way for more efficient and reliable power grid management. Continued research and innovation in this field hold the promise of advancing fault detection capabilities, ultimately contributing to the sustainable and resilient operation of power systems.

References

  1. Shakiba FM, Azizi SM, Zhou M, Abusorrah A. Application of machine learning methods in fault detection and classification of power transmission lines: a survey. Artif Intell Rev 2023;56:5799–836. [CrossRef]
  2. Ağir T. Using machine learning algorithms for classifying transmission line faults. DÜMF Mühendis Derg 2022. [CrossRef]
  3. Kharusi KA, Haffar AE, Mesbah M. Adaptive Machine-Learning-Based Transmission Line Fault Detection and Classification Connected to Inverter-Based Generators. Energies 2023;16:5775. [CrossRef]
  4. Li B, Cui H, Li B, Wen W, Dai D. A permanent fault identification method for single-pole grounding fault of overhead transmission lines in VSC-HVDC grid based on fault line voltage. Int J Electr Power Energy Syst 2020;117:105603. [CrossRef]
  5. Wong SY, Choe CWC, Goh HH, Low YW, Cheah DYS, Pang C. Power Transmission Line Fault Detection and Diagnosis Based on Artificial Intelligence Approach and its Development in UAV: A Review. Arab J Sci Eng 2021;46:9305–31. [CrossRef]
  6. Rafique F, Fu L, Mai R. End to end machine learning for fault detection and classification in power transmission lines. Electr Power Syst Res 2021;199:107430. [CrossRef]
  7. Aker E, Othman ML, Veerasamy V, Aris IB, Wahab NIA, Hizam H. Fault Detection and Classification of Shunt Compensated Transmission Line Using Discrete Wavelet Transform and Naive Bayes Classifier. Energies 2020;13:243. [CrossRef]
  8. Fahim SR, Sarker Y, Sarker SK, Sheikh MdRI, Das SK. Self attention convolutional neural network with time series imaging based feature extraction for transmission line fault detection and classification. Electr Power Syst Res 2020;187:106437. [CrossRef]
  9. Anand A, Affijulla S. Hilbert-Huang transform based fault identification and classification technique for AC power transmission line protection. Int Trans Electr Energy Syst 2020;30. [CrossRef]
  10. Belagoune S, Bali N, Bakdi A, Baadji B, Atif K. Deep learning through LSTM classification and regression for transmission line fault detection, diagnosis and location in large-scale multi-machine power systems. Measurement 2021;177:109330. [CrossRef]
  11. V B, G R. Effective Automatic Fault Detection in Transmission Lines by Hybrid Model of Authorization and Distance Calculation through Impedance Variation. J Electron Inform 2021;3:36–48. [CrossRef]
  12. Biswas S, Nayak PK. A Fault Detection and Classification Scheme for Unified Power Flow Controller Compensated Transmission Lines Connecting Wind Farms. IEEE Syst J 2021;15:297–306. [CrossRef]
  13. Doria-García J, Orozco-Henao C, Leborgne R, Montoya OD, Gil-González W. High impedance fault modeling and location for transmission line✰. Electr Power Syst Res 2021;196:107202. [CrossRef]
  14. Fahim SR, Sarker SK, Muyeen SM, Das SK, Kamwa I. A deep learning based intelligent approach in detection and classification of transmission line faults. Int J Electr Power Energy Syst 2021;133:107102. [CrossRef]
  15. Ferreira VH, Zanghi R, Fortes MZ, Gomes S, Alves Da Silva AP. Probabilistic transmission line fault diagnosis using autonomous neural models. Electr Power Syst Res 2020;185:106360. [CrossRef]
  16. Godse R, Bhat S. Mathematical Morphology-Based Feature-Extraction Technique for Detection and Classification of Faults on Power Transmission Line. IEEE Access 2020;8:38459–71. [CrossRef]
  17. Goswami L, Agrawal P. IOT based Diagnosing of Fault Detection in Power Line Transmission through GOOGLE Firebase database. 2020 4th Int. Conf. Trends Electron. Inform. ICOEI48184, Tirunelveli, India: IEEE; 2020, p. 415–20. [CrossRef]
  18. Haq EU, Jianjun H, Li K, Ahmad F, Banjerdpongchai D, Zhang T. Improved performance of detection and classification of 3-phase transmission line faults based on discrete wavelet transform and double-channel extreme learning machine. Electr Eng 2021;103:953–63. [CrossRef]
  19. Leh NAM, Zain FM, Muhammad Z, Hamid SA, Rosli AD. Fault Detection Method Using ANN for Power Transmission Line. 2020 10th IEEE Int. Conf. Control Syst. Comput. Eng. ICCSCE, Penang, Malaysia: IEEE; 2020, p. 79–84. [CrossRef]
  20. Liang H, Zuo C, Wei W. Detection and Evaluation Method of Transmission Line Defects Based on Deep Learning. IEEE Access 2020;8:38448–58. [CrossRef]
  21. Liu C, Wu Y, Liu J, Sun Z, Xu H. Insulator Faults Detection in Aerial Images from High-Voltage Transmission Lines Based on Deep Learning Model. Appl Sci 2021;11:4647. [CrossRef]
  22. Lu D, Liu Y, Liao Q, Wang B, Huang W, Xi X. Time-Domain Transmission Line Fault Location Method With Full Consideration of Distributed Parameters and Line Asymmetry. IEEE Trans Power Deliv 2020;35:2651–62. [CrossRef]
  23. Mukherjee A, Kundu PK, Das A. Transmission Line Faults in Power System and the Different Algorithms for Identification, Classification and Localization: A Brief Review of Methods. J Inst Eng India Ser B 2021;102:855–77. [CrossRef]
  24. Teimourzadeh H, Moradzadeh A, Shoaran M, Mohammadi-Ivatloo B, Razzaghi R. High Impedance Single-Phase Faults Diagnosis in Transmission Lines via Deep Reinforcement Learning of Transfer Functions. IEEE Access 2021;9:15796–809. [CrossRef]
  25. Tong X, Wen H. A novel transmission line fault detection algorithm based on pilot impedance. Electr Power Syst Res 2020;179:106062. [CrossRef]
  26. Wang XD, Gao X, Liu YM, Wang YW. WRC-SDT Based On-Line Detection Method for Offshore Wind Farm Transmission Line. IEEE Access 2020;8:53547–60. [CrossRef]
  27. Zhang Y, Cong W. An improved single-ended frequency-domain-based fault detection scheme for MMC-HVDC transmission lines. Int J Electr Power Energy Syst 2021;125:106463. [CrossRef]
  28. Zheng X, Jia R, Aisikaer, Gong L, Zhang G, Dang J. Component identification and defect detection in transmission lines based on deep learning. J Intell Fuzzy Syst 2021;40:3147–58. [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

© 2024 MDPI (Basel, Switzerland) unless otherwise stated