Power Line Fault Research Articles

Automatic classification of bird species related to power line faults using deep convolution features and ECOC‐SVM model

AbstractBird‐related outages greatly threaten the safety of overhead transmission and distribution lines, while electrocution and collisions of birds with power lines, especially endangered species, are significant environmental concerns. Automatic bird recognition can be helpful to mitigate this contradiction. This paper proposes a method for automatic classification of bird species related to power line faults combining deep convolution features with error‐correcting output codes support vector machine (ECOC‐SVM). An image dataset of about 20 high‐risk and 20 low‐risk bird species was constructed, and the feed‐forward denoising convolutional neural network was used for image preprocessing. The deep convolution features of bird images were extracted by DarkNet‐53, and taken as inputs of the ECOC‐SVM for model training and bird species classification. The gradient‐weighted class activation mapping was used for visual explanations of the model decision region. The experimental results indicate that the average accuracy of the proposed method can reach 94.39%, and its performance was better than other models using different feature extraction networks and classification algorithms.

Electrical line fault prediction using a novel grey wolf optimization algorithm based on multilayer perceptron

AbstractGrey wolf optimization algorithm (GWO) has achieved great results in the optimization of neural network parameters. However, it has some problems such as insufficient precision, poor robustness, weak searching ability and easy to fall into local optimal solution. Therefore, a grey wolf optimization algorithm combining Levy flight and nonlinear inertia weights (LGWO) is proposed in this paper. The combination of Levy flight and nonlinear inertia weight is to improve the search efficiency and solve the problem that the search ability is weak and it is easy to fall into the local optimal solution. In summary, LGWO solves the problems of insufficient precision, poor robustness, weak searching ability and easy to fall into local optimal. This paper uses Congress on Evolutionary Computation benchmark function and combines algorithms with neural network for power line fault classification prediction to verify the effectiveness of each strategy improvement in LGWO and its comparison with other excellent algorithms (sine cosine algorithm, tree seed algorithm, wind driven optimization, and gravitational search algorithm). In the combination of neural networks and optimization algorithms, the accuracy of LGWO has been improved compared to the basic GWO, and LGWO has achieved the best performance in multiple algorithm comparisons.

Bluetooth 5.0 Suitability Assessment for Emergency Response within Fire Environments

Natural disasters, such as wildfires, can cause widespread devastation. Future-proofing infrastructure, such as buildings and bridges, through technological advancements is crucial to minimize their impact. Fires in disasters often stem from damaged fuel lines and electrical equipment, such as the 2018 California wildfire caused by a power line fault. To enhance safety, IoT applications can continuously monitor the health of emergency personnel. Using Bluetooth 5.0 and wearables in mesh networks, these apps can alert others about an individual’s location during emergencies. However, fire can disrupt wireless networks. This study assesses Bluetooth 5.0’s performance in transmitting signals in fire conditions. It examined received signal strength indicator (RSSI) values in a front open-fire chamber using both Peer-to-Peer (P2P) and mesh networks. The experiment considered three transmission heights of 0.61, 1.22, and 1.83 m and two distances of 11.13 m and 1.52 m. The study demonstrated successful signal transmission with a maximum loss of only 2 dB when transmitting through the fire. This research underscores the potential for reliable communication in fire-prone environments, improving safety during natural disasters.

Improving the efficiency of medium voltage (MV) distribution lines using risk matrix for optimized maintenance

Medium Voltage overhead lines within North-West power utility companies in South Africa are confronted by higher volumes of deficiencies leading to frequent power failures. An essential quality of service factor considered in power distribution is the electricity supply to customers free of interruptions. Due to the higher number of electrical power line faults associated with the power system life span, achieving the desired level of reliability becomes an enormous task. The paper shares the findings of a study on performance improvement of MV distribution overhead lines using a risk matrix for deficiency analysis and clearing of such defects. A pilot study was conducted, and the risk matrix strategy developed in the study improved the reliability and key performance indicators by trimming down the overhead power line deficiencies. Defect clearing lowered outage frequency, durations, and power losses. The System Average Interruption Duration Index (SAIDI) was enhanced by lowering the outage duration by 108.64 h annually. The System Average Interruption Frequency Index (SAIFI) improved by reducing the outage frequency by 45.45 sustained occurrences annually and Reticulation Supply Loss Index (RSLI) by 8.31 MVA monthly.

Detection of High Impedance Faults in Power Lines using Empirical Mode Decomposition with Intelligent Classification Techniques

An uninterrupted and well-managed energy distribution system requires an effective protection system against various faults and uncertainties. Power system networks have a built-in issue called High Impedance Faults (HIFs), which can't be detected by standard protection mechanisms. This paper deals with the aforementioned issue by combining a deep learning technique (DLT) with a signal decomposition method known as Empirical Mode Decomposition (EMD). This approach utilizes the EMD for the extraction of multiple features from the input current signal, and LSTM neural network is employed for the accurate classification of HIF signals. The comparison results with various EMD methods and various switching events are also performed. Finally, this paper proposes an approximate location identification of HIFs. The simulation results demonstrate the effectiveness of the proposed method for detection in various fault scenarios.

Open-Circuit Fault Detection in a Single Conductor Transmission Line Using TM0 Mode Launcher

A novel surface wave (SW) launcher design for power line fault detection is presented in this paper. The launcher is capable of converting TE10 to TM0 mode and exciting SW signal at 5.5 GHz frequency with high efficiency, on a single conductor line. The average attenuation of 2 dB/m and front to back ratio of 19 dB are observed and verified by simulation and experiments. Effect of bend, nearby obstacles, and extra wire connection on SW propagation is also investigated by experiments, and corresponding S-parameter results are presented. The optimization of launcher design with dielectric window and conical back plate is also demonstrated. The proposed launcher is compact and lightweight and can be easily installed without damaging the conductor line. Close structure makes it capable of withstanding outdoor environmental conditions. Using this launcher, microwave signal can be efficiently transmitted and received over existing conductor lines such as live power lines, pipes, and HVAC ducts, and it can be used in applications such as power line communication, fault detection, and permittivity measurement.

Industrial Power Line Fault Detection System based on Miccrocontroller with GSM

The proposed project involves the creation of a model for detecting faults in industrial interfaces using microcontrollers. The aim of this project is to identify faults in an industrial electric line from a central station across various AC line components. The foundation of this project is based on the fundamental concept of Ohm's law. In the event of a fault, such as a short circuit, the current drop will vary depending on the length of the fault within the interface due to the current flow. To address this, a series of resistors are used to represent the interface, while a DC voltage is supplied to one end. The fault is detected by observing the voltage difference using an analog current sensing sensor, and a microcontroller is employed to perform the necessary calculations so that the fault distance can be displayed on a LCD and an SMS notification can be sent via the GSM module.

Traveling wave inversion method of power line faults

In substations, the fault signal at the measuring point includes a variety of signal aliasing. This is partly due to the short length of the substation lines and the exitance of multiple types of equipment that cause the reflection and refraction of the traveling waves. Compared with the incident wave, the distortion in the fault signal is often significant hence interfering with the fault identification and location. To address this issue, we propose an inversion method of the incident wave. To do this, we first investigate the impact of multiple outlets of the substation bus, equivalent stray capacitance of the bus to the ground, transformers, line trap, and other types of equipment on traveling wave transmission. Modeling the aliasing effect of the traveling wave measuring point signal we then propose a fault signal filtering algorithm based on the improved Empirical Mode Decomposition (EMD) and Fast Fourier Transform (FFT). Using this algorithm we then formulate an incident wave inversion model based on substation component parameters. The component parameters in the equivalent circuit model of substation are then accurately identified to achieve an accurate inversion of the incident waves. Identification of parameters is based on an optimization model of substation component parameters based on the Particle Swarm Optimization algorithm (PSO). Our results show that compared with the measured mixed wave signal, the fault characteristics of the inverted incident waves are more accurate, hence improving the fault location accuracy.

Fault Diagnosis in Power Line Inspection Using Normalized Multihierarchy Embedding Matching

For the maintenance of power lines, a core task is to diagnose the fault of different components from the aerial inspection images. Currently, deep-learning models trained on defect samples have achieved promising performances for automatic fault diagnosis. However, the slow accumulation process of fault data leads to a long-term challenge of data insufficiency in this field. In this article, a normalized multihierarchy embedding matching (NMHEM)-based anomaly detection method is proposed to inspect power line faults, which only utilizes defect-free samples during training. To impart the NMHEM with the ability to detect anomalous patterns in images, three main modules are introduced. First, the embedding generation module (EGM) is employed to extract deep hierarchical representations. Next, hierarchy-wise anomaly scores are calculated through the embedding matching module (EMM) to measure the anomalous degree, which can make the model more discriminative at different hierarchies. Finally, a normalizing module (NM) is developed and served as a credible scoring function indicating the probability of anomaly occurring, thus boosting the performance of the fault diagnosis. The proposed NMHEM adaptively aggregates local spatial and global semantic information which leverages the available nominal knowledge from normal data, achieving effective fault diagnosis of power lines. Experiments are conducted on the dataset that contains five key components. Results show that our method achieves 88.4% area under the curve (AUC) and 80.5% <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> -score, which outperforms other supervised and semisupervised methods.

A three-stage resilient dispatch of mobile emergency generators in a distribution system against hurricanes

Mobile emergency generators (MEGs) have been widely used to enhance the resilience of distribution systems. However, the traditional two-stage MEG dispatch strategies focus on only the prepositioning and reallocation before and after hurricanes, neglecting the resilient operation during hurricanes. Therefore, this paper proposes a novel three-stage MEG dispatch model to enhance the resilience of distribution systems. To improve the survivability of distribution systems during hurricanes, the fault isolation operations by remote-controlled switches (RCSs) and forced cut-off of wind power are coordinated with MEG dispatch, which is prepositioned as a preparedness measure before hurricanes. Based on the realization of fault scenarios during hurricanes, MEG reallocation and fault isolation by manual switches (MSs) are conducted to boost the restoration of the distribution system after hurricanes. Considering the uncertainties of wind power cut-off and line faults, the proposed method is incorporated into a stochastic dispatch model, which takes the conditional value-at-risk (CVaR) as an objective function to improve robustness in extreme cases. In addition, the computational burden with multiple scenarios is reduced by the modified progressive hedging algorithm (PHA). Finally, the proposed method is validated on a modified IEEE 33-bus distribution system. Simulation results show that the survivability and restoration of the distribution system can be enhanced by the proposed approach.

A framework for risk assessment and optimal line upgrade selection to mitigate wildfire risk

As wildfires in the United States are becoming more frequent and severe, mitigating the risk of wildfire ignitions from power line faults is an increasingly crucial effort. Long-term ignition prevention strategies, especially converting overhead lines to underground cables, are expensive. Thus, it is important to prioritize upgrades on lines that will reduce wildfire ignition risk the most. However, since so many factors contribute to ignition risk, it is difficult to quantify the wildfire risk associated with power lines. This paper examines how various risk definitions based on historical wildfire risk maps can be used to inform transmission upgrade planning. These risk metrics are evaluated using an optimization model that determines which overhead lines should be undergrounded such that the total wildfire risk in the network is minimized. The risk assignment and upgrade selection are tested on both a synthetic network and the actual transmission lines in California.

A comprehensive review of control techniques for compensating the fault current in resonant grounded distribution networks: From the perspective of mitigating powerline bushfires

Powerline faults are responsible for major bushfires around the world where arcs provoked from ground faults are common causes for igniting such fires. The grounding techniques (mainly resonant grounding) used in distribution substations and arc suppression devices play a crucial role for compensating fault currents in order to extinguish arcs so that the likelihoods of powerline bushfires are significantly reduced. Though passive arc suppression devices (e.g. Petersen coils) are extensively used for compensating the reactive component of the fault current, the active component of this current is still large enough to ignite the fire in bushfire prone areas for which active arc suppression devices are recently used. These active arc suppression devices incorporate residual current compensation inverters and the full compensation of the fault current rely on the control scheme of these inverters. This paper comprehensively reviews different control schemes that are used for compensating the fault current in resonant grounded power distribution systems. The existing control schemes are discussed in terms of the model used during the controller design process, loop structures, control block diagrams, and performance analysis frameworks (i.e. the type of fault impedances). It is worth mentioning that faults on resonant grounded power distribution networks exhibit the characteristics of high impedance faults and it important to consider this aspect for performance analysis of the control scheme. This paper also covers a brief overview of different grounding techniques used for mitigating fault currents and finally, the challenges with the existing controllers are identified in terms of extinguishing powerline bushfires. The comprehensive review motivates and guides future research activities on developing more efficient fault compensation techniques.

Research on power penetration cable line fault location based on distributed traveling wave location technology

The power penetration line of high-speed railway adopts all-cable lines, and its failure will seriously affect the safety and reliability of railway transportation. The accurate location of the fault will greatly reduce the range of finding the cable fault point and the workload of cable fault repair, shorten the maintenance time, and improve the reliability of high-speed railway power supply. This paper introduces the current situation of power line fault location, analyzes various traveling wave location principles and the traveling wave transmission and attenuation characteristics of all-cable penetration lines, and analyzes the adaptability of the traditional traveling wave location technology in all-cable penetration line fault location. It is proposed that the distributed traveling wave location method can greatly reduce the transmission distance of fault traveling waves and can effectively reduce the influence of attenuation and distortion in the traveling wave transmission process. For different types of fault traveling waves, different fault transition resistances, and different positions of fault points, the method can achieve the accurate location. The practical application case shows that the technology has high location accuracy and meets the requirements of a precise location.

An Anti-noise Fault Detection Algorithm for Distribution Network

Power fault directly threatens the safety of the power system. Aiming at the recognition ability of power line fault detection disturbed by noise, A fault detection method based on attentional mechanism combined with autoencoder neural network is proposed. The fault signal is decomposed by wavelet and the subband energy is used as the input of the neural network, which is composed of autoencoder and fault recognition network. The input signal is compressed by an encoder, represented by a small number of features, and transmitted to the fault detection network. By extracting features and training fault signals, the network structure can effectively identify fault categories. Simulation experiments based on IEEE34 show that the fault recognition rate of this method is over 98%, and it can detect distribution network faults effectively under the condition of noise, and has high anti-interference performance. It provides a new idea for fault detection of distribution networks and lays a foundation for a high robustness detection method.

EDLA-EFDS: A Novel Ensemble Deep Learning Approach For Electrical Fault Detection Systems

Early detection of electrical faults is a very essential research area due to its positive influence on network stability and customer satisfaction. Despite of the electrical fault detection problem has been researched during the last decade, the existence of an intelligent fault detection system is still rare in real-world applications. Therefore, this study proposes a novel Ensemble Deep Learning Approach for Electrical Fault Detection Systems (EDLA-EFDS) that resolves the limitations of existing systems such as automation, validation, and overfitting. The proposed approach benefits from two phases prior to the training phase, namely, data preprocessing and pre-training. Whereas the data preprocessing phase manages data by executing all elementary operations on the raw data, the pre-training phase selects both optimal features and hyperparameters by exploiting a double Particle Swarm Optimization (PSO) metaheuristic. Thereafter, a bagging ensemble system is deployed from three different deep learning paradigms, namely, Deep Neural Networks (DNN), Long Short-Term Memory recurrent Neural Networks (LSTM-RNN), and Deep Belief Networks (DBN). The ensemble system is followed by a majority voting engine to produce the final decision. Moreover, the performance of the ensemble system leveraged by the proposed approach is measured on the VSB dataset which is a modern and realistic dataset for power line fault detection. Finally, the analysis of the results using various scenarios and aspects such as the Receiver Operating Characteristic (ROC) curves and Friedman test is provided. The experimental results confirm the effectiveness of the proposed approach in solving the electrical fault detection problem.

Fault detection and classification on insulated overhead conductors based on MCNN‐LSTM

Insulated conductors are used widely in overhead power transmission due to the stability and reduced construction space. However, the ordinary protection devices are not able to detect the phase-to-ground faults without overcurrent. Insulated overhead conductor (IOC) faults are often accompanied by partial discharge (PD) phenomenon. Thus, PD monitoring and recognition plays an important role in evaluating the condition of insulation degradation or detecting power line faults. This paper presents a new approach based on a multi-channel CNN-LSTM (convolutional neural network, long short term memory) network for fault detection by determining whether there is local discharge phenomenon on the IOC, in which the three-phase voltage signals are processed with FFT to obtain low frequency and high frequency components, and then the two components together with the original three-phase signals are fed into three parallel CNNs having different filter lengths, and finally LSTM is used to compose those different-scale features sequentially. Then, the fault types are determined according to the result of fault detection. This proposed method is tested on the ENET public data set with eight types of faults, and simulation results indicate that the method can improve the detection and classification accuracy of IOC faults compared with other classification methods.

Edge–Cloud Collaboration Detection Approach for Small-Sample Imbalanced Faults in Power Lines

Fault detection in power lines is crucial to the reliable operation of power systems. Recent progress in the development of the artificial intelligence-based (AI-based) fault detection for power lines has been receiving an ever-growing attention. Existing AI-based fault detection models mostly rely on the assumption that the power line faults collected in a substation are adequate and balanced. However, obtaining massive and balanced line fault samples is difficult because of the complexity of practical environments and the inherent nature of faults. As a result, the performance of the models gradually deteriorates when the small-sample imbalanced degree of the power line fault set increases. This study proposes a novel edge–cloud collaboration detection method based on transfer and federated learning to address the above issue. Specifically, a fault detection model is established on the basis of a convolutional neural network and pretrained by introducing a new fault set with a sufficient number of labelled samples. The pretrained model is then deployed to each of the substations and fine-tuned in a federated learning manner. By performing fine-tuning in substations and global aggregation in a cloud platform, the proposed method updates the model on the basis of the small-sample imbalanced faults of power lines collected in substations. Case studies verify that after training by using the proposed edge-cloud collaboration detection method, the model can accurately detect faults in power lines even if the fault training set has a small number of samples and is imbalanced.

Resonant Grounded Isolation Transformers to Prevent Ignitions From Powerline Faults

Methods and research for preventing utility-caused wildfires have significantly advanced in the past few decades, including development of criteria on reduction of energy release in an electrical fault to mitigate the probability of ignitions. Technologies to minimize energy release during electrical faults and achieve these ignition benefits are being applied at the substation level. However, the benefits can be targeted to a high-risk area by resonant grounding an isolation transformer on a distribution circuit. By doing this it is possible to reduce single line to ground fault current beyond the transformer to a level unlikely to ignite fires. This system was constructed and shown capable of reducing the power into a phase-to-ground fault by a factor of more than 10,000 and detect ground faults smaller than 0.5 A. This allowed it to meet the ignition criteria mandated for distribution network fire safety in the Australian state of Victoria. It was also tested for typical ignition scenarios such as downed phase conductors and phase conductors contacting with a tree branch. In all cases the probability of ignition was reduced by 90% or more.

Box-Point Detector: A Diagnosis Method for Insulator Faults in Power Lines Using Aerial Images and Convolutional Neural Networks

Fault diagnosis of insulators in aerial images is an essential task of power line inspection to maintain the reliability, safety, and sustainability of power transmission. This paper develops a novel method for intelligent diagnosis of electrical insulators based on deep learning, termed Box-Point Detector, which consists of a deep convolutional neural network followed by two parallel branches of convolutional heads. These two branches are utilized to locate the fault region and estimate insulator endpoints, which presents a new representation for insulator faults. Endpoints of the faulty insulator string can provide detailed and correlative information for enhancing the diagnosis capability of component-dependent faults that occur on component bodies. The proposed Box-Point Detector implements all predictions including region and endpoint into one network thus forms an efficient end-to-end structure, and adopts a smaller downsampling ratio to generate high resolution feature-maps in order to preserve more original information for small size faults. Experimental results indicate that Box-Point Detector can accurately diagnose high-voltage insulator faults in real-time under various conditions. Compared with some previous works using Faster R-CNN, SSD, and cascading network, our Box-Point Detector shows more competitively capabilities with high accuracy and robustness.

Development of a Wavelet – ANFIS Based Fault Location and Identification System for Underground Power Cables

Transmission lines are the backbone of electrical power systems and other power utilities as they are used for transmission and distribution of power. Power is distributed to the end-user through either overhead cables or underground cables. In the case of underground cables, their propensity to fail in service increases as they age with time. The increase in failure rates and system crashes on older underground power cables now negatively affect system reliability and involve numerous losses. It is therefore easy to realize that the consequences of this trend need to be managed [3]. Identification of the type of defects and their locations along the length of the cables is vital to minimize the operating costs by reducing lengthy and expensive patrols to locate the faults, and to speed up repairs and restoration of power in the lines. In this study, a method that combines wavelets and neuro-fuzzy techniques for fault location and identification are proposed. For this methodology a power transmission line model was developed and different fault locations were simulated in MATLAB/SIMULINK, and, as an input to the training and development of the Adaptive Network Fuzzy Inference System (ANFIS), certain selected features of the wavelet transformed signals were used. Fault index obtained from wavelet transformation are used as input variable for fuzzy input block function. Different membership functions were observed within input block function. As per formulation of rules, for membership function, the output value of the defuzzifier component was decoded to give a crisp value of ANFIS output. ANFIS results were compared with actual values. A comparison of the ANFIS output values and the actual values show that the percentage error was less than 1%. Thus, it can be concluded that the wavelet-ANFIS technique is accurate enough to be used in identifying and locating underground power line faults. Which will help in solving this time taking and tedious problem more efficiently and thereby reducing human effort in finding the type of fault and its location.