Transmission Line Fault Research Articles

Bird's nest defect detection of transmission lines based on domain knowledge and occlusion reasoning

AbstractBird's nest defect is an important cause of transmission line faults. To achieve accurate detection of bird nest defects in complex scenarios, a bird nest defect detection model for transmission lines was proposed that combines domain knowledge and occlusion reasoning networks. On the one hand, the model utilized the domain knowledge of the location of the bird's nest, using edge detection to obtain tower area information to constrain the location of candidate frames. This helps to reduce the false detection caused by complex backgrounds. On the other hand, on the basis of analyzing the occlusion characteristics of bird nests, the model employed occlusion reasoning networks that randomly erase features at the feature level to simulate the occlusion of bird nests in real scenes and improve the model's detection capability for occluded targets. Additionally, a multi‐scale feature fusion algorithm was designed in this paper to adapt the model to the scale variations of bird nests in aerial images. Experimental results demonstrate that the model outperforms advanced target detection models and other bird nest defect detection methods, with an AP50 of 78.8% and an AR10 of 72.4% for defect detection.

Convolutional Neural Network for Transmission Line Fault Diagnosis Based on Signal Segmentation Approach

Convolutional Neural Network (CNN) is a Deep Learning algorithm that can take in an input signal, assign importance (learnable weights and biases) to various aspects of the signals and distinguish between them. The CNN algorithm trains a sample and obtains a CNN model capable of identifying different fault types to analyse transmission line faults. Fault analysis methods typically require feature extraction to synthesise the relevant and non-redundant information from the raw signals. However, these traditional methods are time-consuming and inconsistent, and they can produce biased results due to the reliance on human expertise and experience. Hence, this research focused on developing an intelligent system for fault detection in a three-phase transmission line using CNN. This research aims to develop a CNN model for automatic fault detection in a three-phase transmission line and evaluate the performance of the CNN model for analysing transmission line faults. The three-phase transmission line model was developed using MATLAB-Simulink. CNN was implemented to detect transmission line faults. The performance of CNN based on the signal segmentation approach was evaluated through three different types of data: ideal and noise-added signal data. The simulation result shows good performance accuracy of 99.11% for the ideal case, 99.36%, and 99.39% for 20 dB and 30 dB noise-added cases, respectively. The result shows that a higher noise value in transmission line fault current could increase the performance of CNN. In conclusion, the utilisation of CNN based on a signal segmentation approach for transmission line fault analysis has showed promising performance.

A comparative evaluation of Stacked Auto-Encoder neural network and Multi-Layer Extreme Learning Machine for detection and classification of faults in transmission lines using WAMS data

Smart grid is envisaged as a power grid that is extremely reliable and flexible. The electrical grid has wide-area measuring devices like Phasor measurement units (PMUs) deployed to provide real-time grid information and resolve issues effectively and speedily without compromising system availability. The development and application of machine learning approaches for power system protection and state estimation have been facilitated by the availability of measurement data. This research proposes a transmission line fault detection and classification (FD&C) system based on an auto-encoder neural network. A comparison between a Multi-Layer Extreme Learning Machine (ML-ELM) network model and a Stacked Auto-Encoder neural network (SAE) is made. Additionally, the performance of the models developed is compared to that of state-of-the-art classifier models employing feature datasets acquired by wavelet transform based feature extraction as well as other deep learning models. With substantially shorter testing time, the suggested auto-encoder models detect faults with 100% accuracy and classify faults with 99.92% and 99.79% accuracy. The computational efficiency of the ML-ELM model is demonstrated with high accuracy of classification with training time and testing time less than 50 ms. To emulate real system scenarios the models are developed with datasets with noise with signal-to-noise-ratio (SNR) ranging from 10 dB to 40 dB. The efficacy of the models is demonstrated with data from the IEEE 39 bus test system.

Research on Identification and Detection of Transmission Line Insulator Defects Based on a Lightweight YOLOv5 Network

Transmission line fault detection using drones provides real-time assessment of the operational status of transmission equipment, and therefore it has immense importance in ensuring stable functioning of the transmission lines. Currently, identification of transmission line equipment relies predominantly on manual inspections that are susceptible to the influence of natural surroundings, resulting in sluggishness and a high rate of false detections. In view of this, in this study, we propose an insulator defect recognition algorithm based on a YOLOv5 model with a new lightweight network as the backbone network, combining noise reduction and target detection. First, we propose a new noise reduction algorithm, i.e., the adaptive neighborhood-weighted median filtering (NW-AMF) algorithm. This algorithm employs a weighted summation technique to determine the median value of the pixel point’s neighborhood, effectively filtering out noise from the captured aerial images. Consequently, this approach significantly mitigates the adverse effects of varying noise levels on target detection. Subsequently, the RepVGG lightweight network structure is improved to the newly proposed lightweight structure called RcpVGG-YOLOv5. This structure facilitates single-branch inference, multi-branch training, and branch normalization, thereby improving the quantization performance while simultaneously striking a balance between target detection accuracy and speed. Furthermore, we propose a new loss function, i.e., Focal EIOU, to replace the original CIOU loss function. This optimization incorporates a penalty on the edge length of the target frame, which improves the contribution of the high-quality target gradient. This modification effectively addresses the issue of imbalanced positive and negative samples for small targets, suppresses background positive samples, and ultimately enhances the accuracy of detection. Finally, to align more closely with real-world engineering applications, the dataset utilized in this study consists of machine patrol images captured by the Unmanned Aerial Systems (UAS) of the Yunnan Power Supply Bureau Company. The experimental findings demonstrate that the proposed algorithm yields notable improvements in accuracy and inference speed compared to YOLOv5s, YOLOv7, and YOLOv8. Specifically, the improved algorithm achieves a 3.7% increase in accuracy and a 48.2% enhancement in inference speed compared to those of YOLOv5s. Similarly, it achieves a 2.7% accuracy improvement and a 33.5% increase in inference speed compared to those of YOLOv7, as well as a 1.5% accuracy enhancement and a 13.1% improvement in inference speed compared to those of YOLOv8. These results validate the effectiveness of the proposed algorithm through ablation experiments. Consequently, the method presented in this paper exhibits practical applicability in the detection of aerial images of transmission lines within complex environments. In future research endeavors, it is recommended to continue collecting aerial images for continuous iterative training, to optimize the model further, and to conduct in-depth investigations into the challenges associated with detecting small targets. Such endeavors hold significant importance for the advancement of transmission line detection.

Identification of Free Components during Non-Simultaneous Complex Faults in Overhead Lines: A Review

This study concerns the identification and interpretation of current and voltage signals during complex non-simultaneous disturbances in high-voltage overhead lines. It includes topics related to both the search for the optimal model of the system during complex disturbances as well as the analysis of the obtained results. The available EMT-like tools for the analysis of these phenomena are discussed. Models of high voltage overhead transmission lines are presented along with a discussion of their suitability for the study of electromagnetic transients during the analysis of complex non-simultaneous faults. Examples of transient waveforms of currents and voltages during non-simultaneous faults are presented, and the conditions for the occurrence of delayed current zero during short circuits in the system are discussed. The maximum values of the overvoltage factors and the conditions of their occurrence are described. Faults in transmission lines with various voltage levels suspended on the same supporting structures are discussed, together with an indication of the dangerous consequences of unusual intersystem faults. In the final part of the paper, the summary and recommendations for non-simultaneous fault studies are presented.

Study on forecasting model of transmission line failure rate and outage time

Given the complex problem of transmission line failure and outage and the difficulty of predicting the failure rate and outage time of the transmission line with certainty, the accident prediction information is provided for power grid managers. This paper proposes a three-order fundamental Fourier function fault rate prediction model based on a monthly time scale. This method does not rely on any strict assumptions and is driven by the real transmission line fault data in a province. The fitting results reflect the monthly bimodal fault characteristic distribution law of the transmission line fault rate. On this basis, the historical outage time is analyzed and counted, the probability density distribution characteristics of forced outage time are proposed, and the ExpGro function is used to fit it. The results show that the prediction results of the model proposed in this paper have good sharpness and reliability, which has a guiding significance for actual maintenance work.

Lightning Current Measurement Form and Arrangement Scheme of Transmission Line Based on Point-Type Optical Current Transducer.

Lightning strikes are the main cause of transmission line faults, and the accurate lightning current number is an important basis to guide scientific lightning protection. The use of sensors with excellent sensing performance to carry out lightning current monitoring on transmission lines is beneficial to the accumulation of key parameters of original lightning strikes, so it is necessary to study the lightning current measurement structure of transmission lines. In this paper, an optical current-sensing unit is used to monitor the lightning current on transmission lines. A measuring structure that can monitor key parameters of the lightning current under different types of lightning strikes is proposed. First, establish the lightning current return channel model and the equivalent model of the tower, study the influence of the transmission tower on the current in the lightning channel, and analyze the direct measurement position of the lightning current on the tower; establish the multi-wave impedance model of the tower, and build a multi-base tower. The simulation model of the transmission system analyzes the transmission characteristics of the lightning current on the transmission line and the lightning protection line in the case of different types of faults; from the perspective of the measurement of key parameters of the lightning current, the lightning current measurement structure of the transmission system is constructed to analyze different lightning strikes. The measurement effect of each monitoring position in the case of a lightning strike and the waveform characteristics of the fault current in the case of insulator flashover are analyzed.

Fault location in hybrid transmission lines on the basis of VMD energy entropy and SVM

Finding the head of a traveling wave is challenging because an overhead line-cable hybrid transmission line’s wave impedance is not constant. Because of this, it is difficult to measure the hybrid transmission line precisely. This leads to the proposal of a hybrid transmission line fault location algorithm, which combines VMD energy entropy for training support vector machines and traveling wave law graphs. The VMD energy entropy is first used to train SVM to identify the faulty section. The method of traveling waves with a single end is utilized to precisely fix the fault distance within the faulty segment when the type of traveling wave has been identified. The simulation results demonstrate that the method provides precise measurements and is effective for hybrid transmission lines with a range of fault distances and grounding resistances.

Deep learning techniques for transmission line fault classification – A comparative study

Despite advancements in technology, power system faults leading to electric power interruption remain a significant issue. Efficient restoration of the power system relies on the swift classification and clearance of faults. Among the various types of faults in transmission lines, open circuit and short circuit faults are commonly encountered. This study specifically focuses on the analysis of five types of short circuit faults: line-to-line, line-to-ground, double line-to-ground, triple line, and triple line-to-ground faults. Faults can cause both power failure and power loss in transmission lines. Once a fault occurs, it is crucial to restore electricity supply promptly to prevent further losses. Therefore, the development of a system capable of accurately and swiftly detecting and removing faults is essential. Traditionally, categorizing transmission line faults required sophisticated mathematical modeling, intricate signal processing techniques, and expert knowledge to interpret the output signals. In this paper, an alternative approach is proposed, utilizing deep learning techniques for transmission line fault classification. Specifically, the paper employs techniques such as artificial neural network (ANN), long short-term memory (LSTM), with and without window regression (WR). By implementing these deep learning techniques, automatic feature extraction and signal processing are achieved, streamlining the fault classification process. The experimental results obtained from this study demonstrate an accuracy of 42.98% for ANN, 99.98% for LSTM, and 99.99% for LSTM-WR. These outcomes underscore the effectiveness of the deep learning techniques employed in accurately classifying transmission line faults, with LSTM and LSTM-WR outperforming ANN in terms of accuracy.

Transmission line fault cause identification method based on transient waveform image and MCNN-LSTM

Power system operators commonly employ fault recorders to record and analyze transmission line faults. However, the length of transmission lines, the number and location of power sources and the sampling frequency of fault recorder will influence the characteristics of the fault waveform leading to fault identification error. This paper proposed a multi-head convolutional neural network with long short-term memory (MCNN-LSTM) using transient waveform image as input for transmission line fault type and fault cause classification. First, the characteristics of transmission line faults were analyzed. Second, MCNN-LSTM is designed by considering interaction between electrical signal. Finally, the modified IEEE-39 node power network was adopted to test the proposed method. The test results show that: this method can improve the identification accuracy of transmission line faults even if there is class unbalance in fault sample set, and shows strong adaptability when transferred to small-size fault samples of transmission lines.

Accuracy analysis using the EMD and VMD for two-terminal transmission line fault location based on traveling wave theory

This paper presents an accuracy analysis from the application of techniques for the extraction of characteristic data from current and voltage signals in transmission line fault location based on traveling wave theory. The techniques used were the empirical mode decomposition and the variational mode decomposition associated with Teager energy operator, which helps to identify the instantaneous energy of the first intrinsic mode function. The simulation of the power system was carried out using the MATLAB/Simulink® software, for a test system consisting of a 200 km long transmission line between two Thevenin equivalents. Data is obtained from both terminals. The fault resistance was fixed at 100 Ω, but the incidence angle, the sampling rate and the fault location were varied for all fault types. The numerical and graphical results proved that the techniques can extract the characteristic data of the current and voltage signals and estimate the fault distance to the terminal with high accuracy, depending only on the sampling rate adopted.

Adaptive Machine-Learning-Based Transmission Line Fault Detection and Classification Connected to Inverter-Based Generators

Adaptive protection schemes have been developed to address the problem of behavior-changing power systems integrated with inverter-based generation (IBG). This paper proposes a machine-learning-based fault detection and classification technique using a setting-group-based adaptation approach. Multigroup settings were designed depending on the types of power generation (synchronous generator, PV plant, and type-3 wind farm) connected to a transmission line in the 39-Bus New England System. For each system topology, an optimized pretrained ensemble tree classifier was used. The adaptation process has two phases: an offline learning phase to tune the classifiers and select the optimum subset of features, and an online phase where the circuit breaker (CB) status and the active output power of the generators are continuously monitored to identify the current system topology and to select the appropriate setting group. The proposed system achieved an average accuracy of 99.4%, a 99.5% average precision, a 99.9% average specificity, and a 99.4% average sensitivity of classification. The robustness analysis was conducted by applying several fault scenarios not considered during training, which include different transmission network configurations and different penetration levels of IBGs. The case of incorrect selection of the appropriate setting group resulting from selecting the wrong topology is also considered. It was noticed that the performance of developed classifiers deteriorates when the transmission network is reconfigured and the incorrect setting group is selected.

Transmission line fault classification based on spatiotemporal characteristic analysis with global and local discriminant analysis

To ensure the reliability of power supply, ultra-fast and reliable fault classification of the transmission line is the critical stage of the wide-area transmission network protection. As the distributed generation (DG) current injection to the transmission line, the amplitude characteristic of the break current protection reduces and the system reliability decreases. With the wide-area measurement generated to the power system, an online fault classification method with the global and local discriminant analysis (GLDA) is proposed to enhance the phase characteristic and decrease calculation load. In the single line to ground (SLG) fault process, the local current phase shift is analyzed to enhance the spatiotemporal characteristic. To fully measure the ordering degree of the spatiotemporal characteristic, the axiomatic definition for the characteristic entropy (CE) of the grid impedance in the individual bus in the wide-area transmission network is proposed according to the characteristic distribution. Based on the current spatiotemporal characteristic analysis, a multi-priority GLDA classifies the fault phase in a coordinating way using with the phasor measurement unit (PMU). The multi-priority GLDA-CE fault classification method reduces the computational load, ensures resilience against measurement noise and the transition resistance effect. Compared with the statistical entropy and neural network method, the classification accuracy and response speed of the GLDA-CE fault classification method are validated in the IEEE 39-bus system with 10 PMUs. Compared with the conventional statistical entropy and neural network methods, the GLDA-CE method is able to classify 226 entire power outages and 600 single phase outages within 1.621 s.

Neural Network based Transmission Line Fault Classifier and Locator using Sequence Values

Faults can easily occur at transmission line because the line is exposed to the environment. The fast fault location after a fault occurrence will minimize the time to repair the faulty part thus reduce the stress of power system due to long outage time. This paper demonstrates the development of fault classifier and fault locator using neural network. The positive, negative and zero sequence values of three-phase voltage and current during fault time were used as the inputs to train the neural networks. Various fault conditions which have different fault types, fault resistance values and fault locations have been simulated to generate the training data for both neural network based fault classifier and fault locator. From the results, the fault classifier and locator successfully classified and located all the simulated fault conditions. One of the advantages of using sequence values as the inputs for the neural networks is only one fault type need to be simulated and used as the training data for single line-to-ground fault (RG, YG and BG), line-to-line fault (RY, YB, BR) and double line-to-ground fault (RYG, YBG, BRG).

Adaptive Fault Locator for High-Voltage Transmission Lines Based on the Estimation of Power System Model Parameters

Abstract This paper presents an adaptive transmission line fault location method, which incorporates fault location devices at both line ends and utilises data import from a supervisory control and data acquisition system without strictly requiring data synchronisation. The developed method aims at achieving a higher degree of robustness, adaptiveness and accuracy. The adaptiveness is achieved by dynamic updating of mathematical models used on the basis of network-wide information, such as data on the state of circuit-breakers and apparent power at load and generation nodes. The robustness and accuracy are enhanced by incorporating two stages of identification of model parameters with the goal of reducing the decision variable space for the stage identifying fault parameters. Furthermore, in addition to utilisation of all measurements available at a particular substation, the developed method partially employs the measurements from the other end of the line by means of result cross-checks, but does not require a full data set, unlike deterministic-model-based methods. An optimisation-based approach, redundancy on the basis of extended measurement set, and cross-checks reduce the risk of fault location errors due to measurement errors or “voids” in the data available. Testing of the developed method demonstrates its accuracy and robustness in a wide range of pre-fault and fault regime scenarios, even when considering various pre-fault contingencies.

Extreme learning machine-based fault location approach for terminal-hybrid LCC-VSC-HVDC transmission lines

This paper proposes an Extreme Learning Machine-based fault location method to evaluate the DC transmission line faults of Terminal-hybrid LCC-VSC-HVDC systems. For this purpose, the voltage and current fault signals are firstly measured from the terminals of the hybrid HVDC system. Then, the high-frequency components of the input signals will be extracted and decomposed employing wavelet transform. In the next step, the energy of these extracted signals is calculated and provided to the extreme learning machine as intake train and test data. Finally, the exact location of the fault point will be determined using ELM. The proposed strategy is also applied to the same hybrid system using single-layer and double-layer neural networks, and the results have been thoroughly compared with ELM. The outcomes unequivocally indicate that by increasing the amount of DC filters in the transmission line, the average percentage error of fault location will be remarkably diminished. A ± 100 kV Terminal-hybrid LCC-VSC-HVDC system has been implemented and simulated via MATLAB software to assess the proposed method. It can be perceived that the ELM-based method not only works properly under various fault resistances, fault locations, and types of faults but also has significant superiority over the ANN-based approaches.

Improving Traveling Wave-Based Transmission Line Fault Location by Leveraging Classical Functions Available in Off-the-Shelf Devices

This letter presents a practical methodology to improve traveling wave (TW)-based fault location (TWFL) on transmission lines by leveraging classical single- and double-ended TWFL functions available in off-the-shelf relays and digital fault recorders. It eliminates negative effects of uncertainties in line propagation velocity and travel time, being free of time-consuming or complicated procedures. Real TW records are studied to demonstrate that the proposed solution can improve the accuracy and reliability of TWFL estimations when uncertainties about line propagation velocity or travel time exist.

Robust Fault Location Method for Transmission Lines Using PMUs

The random noise and abnormally greater numbers may affect the accuracy of traditional transmission line fault location. A robust fault location method for transmission lines based on PMUs (Phasor Measurement Unit) is proposed in this paper. The node voltage equations based on voltage fault components are adopted to establish the quadratic nonlinear overdetermined equations of fault location, only using limited PMUs adjacent to the faulted line. In order to quantify the impact of measurement errors, the residual errors are introduced into the fault location equations. Levenberg-Marquardt (L-M) algorithm has been used to solve the equations. Then the original fault location data at each time after the occurrence of one cycle is obtained. A filtering method is proposed to filter the possible interference such as noise and abnormally greater numbers in fault location sequence. The abnormal data that deviating from the mean can be detected by the probability distribution and replaced by smooth filtering data. The simulation results on IEEE 39-bus systems show that proposed algorithm can effectively improve the location accuracy, is not affected by the fault location, transition resistance, fault type, and has strong anti-interference ability and good robustness.

Localizing faults in power transmission line with applying signals of directional elements at both line ends

Localizing faults in power transmission line with applying signals of directional elements at both line ends

DEVELOPMENT OF A COMPREHENSIVE APPROACH TO DIAGNOSING THE OVERHEAD LINE FAULTS LOCATION IN SINGLE-PHASE GROUND FAULTS BASED ON A DIGITAL SUBSTATION DATA PROCESSING ALGORITHM

Link for citation: Sushkov V.V., Sukhachev I.S., Sidorov S.V. Development of a comprehensive approach to diagnosing the overhead line faults location in single-phase ground faults based on a digital substation data processing algorithm. Bulletin of the Tomsk Polytechnic University. Geo Аssets Engineering, 2023, vol. 334, no. 7, рр. 66-77. In Rus. Relevance. Emergency of electric complex of outgoing transmission lines is associated with the occurrence of damage to the line isolation and the resulting transients, marked by the presence of higher harmonic components of currents and voltage, which reduce the reliability and energy efficiency of electrical equipment of oil and gas companies: degrade the mechanical characteristics and efficiency of electric motors, accelerate aging of their isolation, disrupt the operation of control systems and emergency automatics, fault diagnostics systems. One of the reasons of high-level emergency situations, outages and emergency repairs in electric complexes of outgoing transmission lines of oil and gas companies is appearance of phase-to-ground faults. A large number of changing factors are known to influence the reliability of phase-to-ground fault location methods: parameters of transmission line operating mode, its structural and electrical parameters, fault type, transient resistance value, ground resistance and others. Inclusion in the transmission line fault location algorithms of averaged values of a number of the factors listed above independent of changes of humidity of soil and air, its temperature and other factors causes an error of methods of fault location up to 30 %. Accordingly, the research to develop a fault location method for overhead transmission lines with isolated neutral points during phase-to-ground faults based on a digital substation data processing algorithm, accounting for tower parameters and natural-climatic factors, is relevant. Objective: improving the reliability of power grids and providing consumers with power of the required standard by applying a digital fault location system to the outgoing transmission lines electrical complex. Methods: mathematical analysis of a power transmission line, taking into account its design features and natural-climatic factors on the basis of the method of mirror images; mathematical analysis of steady-state and transient modes of transmission line operation considering self and mutual parameters terms of wires; transmission line currents and voltages Fourier analysis; validation of developed method of phase-to-ground fault location in power grid of Tyumen region energy company. Results. The authors have developed the phase-to-ground faults location method for transmission lines in power grids with isolated neutral points, which differs in the following – first the faulty overhead line feeder branch is identified by correspondences between the negative sequence voltage values of the phase voltages at the transformer substation inputs, and then the distance to the location of the single-phase ground fault is determined along the branch line based on the resonant frequencies of the line taking into account its design features and natural and climatic factors.