Power Equipment Diagnosis Research Articles

Intelligent Fault Diagnosis Method for Transformer Driven by Multiple Vibration Data

Abstract Early monitoring and early warning of transformer failure are crucial to ensure the safe and dependable functioning of the power system, as transformers are essential equipment within the system. On multivariate vibration data, this work presents a sophisticated approach for diagnosing faults in transformers which integrates entropy model and machine learning. Firstly, a new coarse-grained method is introduced into the improved multiscale fuzzy entropy to improve the disadvantage of the traditional entropy model with large entropy fluctuation under high-scale factors. Furthermore, the features of transformer multivariate vibration signals are extracted by the entropy model. Finally, the extreme learning machine is utilized to achieve efficient detection of transformer defects. This method provides a useful method reference for fault diagnosis of power equipment with multi-vibration signal, and has good reference value.

Fault diagnosis of power equipment based on variational autoencoder and semi‐supervised learning

SummaryThe issue of fault diagnosis in power equipment is receiving increasing attention from scholars. Due to the important role played by bearings in power equipment, bearing faults have become the main factor causing the shutdown of wind turbines units. Therefore, this paper takes bearing equipment as an example for research. In order to solve the problem of insufficient and unbalanced fault sample data of wind turbines bearings, a fault diagnosis (FD) method based on variational autoencoder and semi‐supervised learning is proposed in this paper. Firstly, based on Label Propagation‐random forests (LP‐RFs) and a small number of labeled fault samples, a semi‐supervised learning algorithm is proposed to label the original data samples. Secondly, a small number of training samples are preprocessed by the variational autoencoder to reduce the imbalance of the fault samples. Then, the RFs‐based method is adopted to train the processed fault samples to obtain a mature FD classifier. Finally, the proposed method is applied to FD for bearings, and the results show that the proposed method can realize bearings fault diagnosis (BFD). And meanwhile, the proposed method can also be applied for fault diagnosis in power transmission and transformation systems.

Wafer-Scale Field-Effect Transistor-Type Sensor Using a Carbon Nanotube Film as a Channel for Ppb-Level Hydrogen Sulfide Detection.

Sulfur hexafluoride is widely used in power equipment because of its excellent insulation and arc extinguishing properties. However, severe damage to power equipment may be caused and a large-scale collapse of the power grid may occur when SF6 is decomposed into H2S, SOF2, and SO2F2. It is difficult to detect the SF6 concentration as it is a kind of inert gas. Generally, the trace gas decomposed in the early stage of SF6 is detected to achieve the function of early warning. Consequently, it is of great significance to realize the real-time detection of trace gases decomposed from SF6 for the early fault diagnosis of power equipment. In this work, a wafer-scale gate-sensing carbon-based FET gas sensor is fabricated on a four-inch carbon wafer for the detection of H2S, a decomposition product of SF6. The carbon nanotubes with semiconductor properties and the noble metal Pt are respectively used as a channel and a sensing gate of the FET-type gas sensor, and the channel transmission layer and the sensing gate layer each play an independent role and do not interfere with each other by introducing the gate dielectric layer Y2O3, giving full play to their respective advantages to forming an integrated sensor of gas detection and signal amplification. The detection limit of the as-prepared gate-sensing carbon-based FET gas sensor can reach 20 ppb, and its response deviation is not more than 3% for the different batches of gas sensors. This work provides a potentially useful solution for the industrial production of miniaturized and integrated gas sensors.

ConvNeXt-based anchor-free object detection model for infrared image of power equipment

Infrared thermography has been widely used in the monitoring and diagnosis of power equipment to ensure the reliable functioning of power systems because of its non-contact and rapid capabilities. With the promotion of intelligent detection schemes in daily patrol inspection of power equipment by utilities, the amount of infrared images of power equipment (IIOPE) is growing geometrically. The research community has begun using deep learning technology to automatically detect objects in the IIOPE. However, the performance of existing detection methods is still not satisfactory because of the inherent attributes of power equipment infrared images, such as low contrast, complex background, and multi-scale object distribution. This research proposes an anchor-free detection model based on ConvNeXt to address these issues. An enhanced feature pyramid network (FPN) is presented to address the issue of multi-scale object distribution in the IIOPE. In addition, a dynamic soft label assignment strategy is proposed for precise object localization. Extensive experimental results on the self-built IIOPE dataset and the PASCAL VOC 2007 dataset demonstrate that the proposed method yields positive results. When compared to the current anchor-based as well as anchor-free infrared image detectors for power equipment, the proposed model offers improved accuracy while ensuring the detection speed that may satisfy the practical requirements.

A multi-stream multi-scale lightweight SwinMLP network with an adaptive channel-spatial soft threshold for online fault diagnosis of power transformers

Fault diagnosis of power equipment is extremely crucial to the stability of power grid systems. However, complex operating environments, high costs and limitations of single-modal signals are the biggest bottlenecks. To this end,a multi-tream, multi-scale lightweight Swin multilayer perceptron (MLP) network (MLSNet) with an adaptive channel-spatial soft threshold is proposed in this paper. First, a Res2net-based feature-enhanced method is used to learn the correlated features of vibration and voltage multi-modal signals. Second, a novel MLSNet is designed to combine the benefits of Swin transformers with an MLP with a lightweight convolutional neural network and employs a staged model to extract various scale features. Third, an adaptive deep fusion approach employing a channel-spatial soft threshold module is used to integrate and recalibrate staged information at different scales. The overall accuracy of the proposed method can reach 98.73% in various experiments, potentially making it an effective method for online fault diagnosis of power transformers.

A Portable Industrial Acoustic Detection System With Multiband Resonant Amplification

Detection of acoustic signals is an effective approach for online-monitoring of the power equipment. However, the signal-to-noise ratio of the acoustic detection system is limited by the noisy and harsh environment around the power equipment. In this study, an acoustic detection system based on resonant-enhanced acquisition is proposed. By utilizing a horn-shaped acoustic resonator with multiple resonant frequencies, the proposed system can realize long-distance acoustic sensing with high sensitivity in both the audible and ultrasonic ranges. The sensitivity of the system in the designated frequencies is enhanced by 28 dB, while the signal-to-noise ratio increased by 14.1 dB. Integrated with a control terminal, the system can perform complex functionality, e.g., record and analyze the acoustic information and provide visual/audible feedback for human operators. The proposed system can be considered as a technological platform for the acoustic fault diagnosis of power equipment.

Transmission Line Equipment Infrared Diagnosis Using an Improved Pulse-Coupled Neural Network

In order to detect the status of power equipment from infrared transmission line images under the spatial positioning relationship of the transmission line equipment, such as corridor, substation equipment, and facilities, this paper presents an improved PCNN model which merges an optimized parameter setting method. In this PCNN model, the original iteration mechanism is abandoned, and instead, the thresholding model is built by the maximum similarity thresholding rule. To ensure similarity during classifying neighboring neurons into cluster centers, a local clustering strategy is used for setting the linking coefficient, thus improving the efficiency of the method to detect the power equipment in infrared transmission line images. Finally, experimental results on transmission line infrared images show that the proposed method can provide the basis for the diagnosis of power equipment, preventing the casualties and property damage caused by the thermal damage of power equipment, and effectively improving the safety risk identification and operation control ability of power grid engineering.

Fault Diagnosis Algorithm of Transformer and Circuit Breaker in Traction Power Supply System Based on IoT

Transformers and circuit breakers are essential equipment in traction power supply systems. Once a fault occurs, it will affect the train’s regular operation and even threaten passengers’ personal safety. Therefore, it is essential to diagnose the faults of the transformers and circuit breakers of the traction power supply system. At present, power companies have made many achievements in fault diagnosis of power equipment, but there are still problems with real-time and accuracy. The Internet of Things (IoT) is a technology that connects different types of terminal devices for information exchange and communication to achieve intelligence. It includes data acquisition and transmission, information interaction, processing, and decision-making from bottom to top. It uses sensor terminals to obtain real-time status information on electrical equipment. Moreover, it conducts real-time monitoring and intelligent processing of the equipment status of the traction power supply system. In this paper, the multi-data fusion technology of the IoT combines the real-time information of electrical equipment with fault diagnosis to realize the fault diagnosis of transformers and circuit breakers. First, we built an equipment fault diagnosis system based on the multi-terminal data fusion technology of the IoT. Secondly, the transformer fault diagnosis model is established. We adopt the BP neural network algorithm based on particle swarm optimization (PSO) to realize transformer fault diagnosis and use PSO to optimize the feature subset to improve the diagnosis performance. Finally, the fault diagnosis model of the vacuum circuit breaker is established. We select the current change and time node as typical fault feature quantities and use the PSO–BP neural network algorithm to realize the fault diagnosis of the circuit breaker.

An infrared image detection of power equipment based on super‐resolution reconstruction and YOLOv4

Infrared imaging technology is widely used in the fault diagnosis of power equipment. The power equipment images obtained by infrared thermal cameras generally have the characteristics of low resolution, low contrast, similar geometric features, and complex image backgrounds. The traditional infrared image detection methods have problems of low detection accuracy, low real-time performance, and missed and fault detection, all of which will lead to great difficulties in the automatic identification process of infrared images of power equipment. This paper adopted an algorithm based on You Only Look Once (YOLO) v4 to realize the detection of electrical equipment under infrared images. First of all, the authors utilized the Fast Super-Resolution Convolutional Neural Network (FSRCNN) algorithm to perform super-resolution reconstruction on the infrared images of power equipment to achieve image enhancement. Moreover, the lightweight network Mobilenetv3 was selected as the backbone network of YOLOv4 to improve the running speed by reducing network parameters. Finally, the performance of the YOLOv4 algorithm was compared to verify the effectiveness of the algorithm proposed in this paper. The experimental results showed that the proposed method had strong generalization ability and reliable detection of infrared targets of various types of power equipment, with the mean average precision (mAP) reaching 91.3%.

Intelligent Diagnosis Method of Power Equipment Faults Based on Single‐Stage Infrared Image Target Detection

With the rapid expansion of the scale of the power grid, the efficiency of fault diagnosis has been severely challenged by the large amount of inspection image data generated by intelligent devices such as drones and inspection robots. In order to improve the efficiency of fault diagnosis for power equipment in substations, a new method for intelligently diagnosing different types of faults in power equipment is proposed. For circuit breakers and insulators, YOLOv4 is selected as the target detection model. To improve the detection performance of the YOLOv4 model, this paper improves it: the Cross Stage Partial (CSP) structure is introduced in the Spatial Pyramid Pooling (SPP) module of the neck of the YOLOv4 model. The experimental results show that after using the optimal learning rate decay strategy, the mAP and frames per second (FPS) of the improved YOLOv4 model are better than the original YOLOv4 and PP‐YOLO model. Finally, an intelligent diagnosis terminal system for power equipment faults is developed. Through the target recognition and rapid extraction of equipment temperature, the intelligent diagnosis of thermal faults of equipment is realized. This method is especially suitable for accurate fault diagnosis of more power equipment, and has potential huge applicability in the field of power equipment diagnosis. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

Fault diagnosis method of substation equipment based on You Only Look Once algorithm and infrared imaging

Aiming at the problems of local temperature increase of cable, heating of insulator porcelain disc, local heating of lightning arresters and transformers in the process of abnormal heat generation fault diagnosis of power equipment. This study proposes a target detection method that YOLOv4 (You Only Look Once version 4), combined with the infrared image data of power equipment to locate and diagnose the faulty heating area of power equipment. A dataset is established, and image annotation is made by detecting infrared image data of typical electrical equipment collected by the inspection department using a thermal imaging camera. Then this study established a target detection model to calculate the degree of overlap between the power equipment area and the fault area, which is used to determine whether the power equipment to be measured has abnormal heat generation. Finally, the optimal detection model was established by analyzing 6000 test samples in the four kinds of power equipment image dataset. The detection results are compared with Faster Region Convolutional Neural Network algorithms, Single-Shot multibox Detectors, and You Only Look Once version 3. The results show that YOLOv4 has an average accuracy rate of 92.2%. The YOLOv4 model established in this study can effectively detect anomalies in the infrared image of power equipment can be effectively applied to infrared detection of power equipment in substations.

Power Interference Suppression Method for Measuring Partial Discharges under Pulse Square Voltage Conditions

Partial discharge (PD) is an important metric for the insulation diagnosis of power equipment. However, its detection is affected by the strong electromagnetic interference generated by pulse square voltage. We therefore propose a power interference suppression method for partial discharges under pulse square voltage based on a quadratic measurement method. We conduct analysis of the topology circuit when partial discharge occurs in the insulation test sample and introduce the basic principle of the secondary measurement method according to the superposition principle and the linear relationship between the square voltages at different peak values. We verify the feasibility of this method by simulating a PD signal with power interference. Subsequently, we use the successive interception comparison method to solve the non-correspondence of the two initial measurement points problem and design and manufacture the transformer turn-to-turn oil-paper insulation test sample and experimental tank. By measuring the PD starting voltage of the insulation test sample under the power frequency voltage, we determined the first measurement voltage under the pulse square voltage and obtained the signal x1(t) to subsequently measure the PD signal x2(t). According to the proposed successive interception comparison method, the signal x1(t) is processed, and the secondary measurement method suppresses the power interference of the measured signal x2(t). We demonstrate that the proposed method effectively suppresses the power interference in PD detection under a pulse square voltage.

Compressed Sensing Super-Resolution Method for Improving the Accuracy of Infrared Diagnosis of Power Equipment

The infrared image of power equipment plays a crucial role in identifying faults, monitoring equipment condition, and so on. The low resolution and low definition of infrared images in applications contribute to the low accuracy of infrared diagnosis. A super-resolution reconstruction method of infrared image, based on compressed sensing theory, is proposed. Firstly, by analyzing the variation of high-frequency information in infrared images with different blurring degrees, the image gradient norm ratio is introduced to estimate the blur kernel matrix in the degradation model a priori. Then, in the process of image reconstruction, we add the full variational regularization term to the traditional compressed sensing model, and design a two-step full variational sparse reconstruction algorithm. Experimental results verify the effectiveness of the method. Compared with the existing classical super-resolution methods, this method offers improvement in subjective visual effect and objective evaluation index. In addition, the final image recognition and infrared diagnosis experiments show that this method is helpful to improve the accuracy of infrared diagnosis of power equipment.

Infrared Image Segmentation for Power Equipment Using Linear Spectral Clustering and Maximal Similarity-based Region Merging

<p>The diagnostic method of power equipment based on infrared images is widely used because it has the advantages of non-contact and does not affect the online operation of power equipment. However, in actual using, the power equipment diagnosis method based on infrared image still relies on manual judgment, that is, the detection personnel can judge the fault according to the obtained infrared image of power equipment by experience. This process consumes a lot of time, and the subjectivity is strong, misjudgment rate is higher, which cannot meet the requirements of modern smart grid development. Infrared image of power equipment contains a lot of noise, and the edge is fuzzy. In this paper, we propose a new infrared image segmentation method for power equipment by using linear spectral clustering and maximal similarity-based region merging under complex backgrounds. In this method, the linear spectral clustering algorithm (LSC) is used to segment the image into super-pixels, and the pixels with similar color and distance are clustered to the same center. The calculated OTSU threshold based on the global image is used to pre-label the background of each super-pixel block. The maximum similarity-based region merging algorithm (MSRM) is utilized to merge the super-pixel blocks. Meanwhile, it obtains the target equipment, the over-segmentation and under-segmentation rates are reduced effectively. Finally, the mathematical morphology algorithm is used to post-process the image. Experimental results show that, compared with other algorithms, this new method can obtain more accurate and complete target equipment under complex background. </p> <p> </p>

A space hybridization theory for dealing with data insufficiency in intelligent power equipment diagnosis

Absence of fault samples of power equipment is a big bottleneck that limits the development of diagnostic method. This paper proposes a Space Hybridization Theory for dealing with diagnostic data insufficiency. First, the experimental datasets of transformer winding failure are obtained, constituting the real space. Then, the transformer digital space models are constructed similarly to obtain the simulation datasets, constituting the virtual space. After that, the features of the samples in two spaces are extracted through windowing feature calculation, and the real-virtual space is obtained through feature hybridization. The final samples in the space are taken as the auxiliary dataset during the intelligent diagnosis process. The diagnostic accuracy with the proposed method increased by several percent than without. While the MAPE between the real space and the virtual space is less than 9.4%, it can comprehensively improve the diagnostic effect.

Decomposition Characteristics of SF6 under Arc Discharge and the Effects of Trace H2O, O2, and PTFE Vapour on Its By-Products

The research on decomposition characteristics of SF6 and its by-products have great significance to the operation, maintenance, condition assessment and fault diagnosis of power equipment. In this paper, the particle composition models of SF6, SF6/polytetrafluoroethylene (PTFE), SF6/PTFE/O2, SF6/PTFE/H2O, and SF6/PTFE/O2/H2O were established by using Gibbs free energy minimization method, and the effects of trace H2O and O2 impurities and PTFE vapour on SF6 by-products were studied by the models. In order to verify the correctness of the simulation results, a series of breaking experiments were carried out on a 40.5 kV SF6 circuit breaker, and a gas chromatograph was used to detect and analyse the SF6 by-products. It was found that when PTFE vapour is involved in the arc plasma, the main by-product after arc quenching is CF4, and the molar fractions of C2F6 and C3F8 are very low. When O2 is involved, the main by-products are SOF2, SO2 and SO2F2, and a small amount of CO and CO2 was also produced. When H2O is involved, the main by-products in simulation are SOF2, SO2 and HF, and a small amount of SO2, CO2, CO, SO2F2 and H2 was also produced. The experimental results are in good agreement with the above results.

An Adaptive Pulse Separation Strategy for PD Detection in Frequency-Tuned Resonant Tests

Frequency-tuned resonant (ACRF) test is one of the most widely used technologies for power equipment diagnosis. However, the intense interference generated by the variable-frequency power source hinders its further application in partial discharge (PD) detection. For this reason, an adaptive PD separation method based on improved clustering by fast search and find of density peaks (CFSFDP) is proposed in this article to separate PD signals and disturbances under the ACRF test. First, the characteristics of the PD signals and the interference signals generated from the ACRF test system are analyzed and compared. Then, an improved CFSFDP algorithm is proposed to realize the automatic cluster center selection and signal separation, and the working principle of the PD separation strategy based on the improved CFSFDP algorithm is introduced. Finally, the effectiveness of the proposed method is validated by performing the ACRF test on a defective cable. The experimental results show that this method can realize the separation and detection of PD signals in ACRF tests. The proposed method provides a feasible solution for performing PD detection using the ACRF test system without the restriction of the test system structure.

Automatic Fault Diagnosis of Infrared Insulator Images Based on Image Instance Segmentation and Temperature Analysis

As an onsite condition monitoring method, an infrared inspection can help to discover and analyze abnormal temperature increases in power equipment. For improving the efficiency of the onsite diagnosis of insulators in substations, this article proposes an automatic diagnosis method using instance segmentation and temperature analysis of infrared insulator images. For developing this method, thousands of infrared images from field inspection databases were collected to establish an annotated data set of insulator images. With the aid of the Mask R-convolutional neural network (CNN), it was possible to extract multiple insulators automatically in the infrared images. Transfer learning, as well as the dynamic learning rate algorithm, were then employed to realize the training process of Mask R-CNN with the annotated image data set. The result of the testing experiment showed that the mean Average Precision (mAP) of the model is 0.77, and the frame per second (FPS) is 5.07, which indicated great identification accuracy and computing speed of the proposed model. Next, function fitting was realized to extract the temperature distribution of each insulator. Finally, to evaluate the condition of each insulator, rules, which are based on the related standards, were established using machine language. This is the first time that the machine could independently realize fault analysis of multiple insulators in the infrared images, which is a great attempt to adapt the development of the Internet of Things and the tendency of predictive maintenance. Moreover, because of the universality of the model algorithm used, automatic infrared fault diagnosis for other power equipment could also be performed in a similar manner, which has significant potential applicability in the area of power equipment diagnosis.

Quantitative aging assessment method for cellulose pressboard based on the interpretation of the dielectric response mechanism

Cellulose pressboard is used as a key element in various power equipment insulation structures and determines the operational safety of electric equipment and power systems. This paper aims to increase the application of a pressboard aging assessment method based on frequency domain spectroscopy, a promising nondestructive method for power equipment diagnosis, through in-depth theoretical analysis and accurate data fitting. To achieve this purpose, first, based on classic theoretical models in the field of dielectric physics, a mathematical deduction is proposed to separate each dielectric process in the overlapping frequency band. Second, a novel analysis approach that considers both the pressboard dielectric response and the thermally stimulated current is proposed and demonstrated to be effective in determining the physical meaning of pressboard microscopic dielectric processes. Third, a function model based on a specific microscopic mechanism is established to fit the frequency domain spectra of pressboard samples with different aging degrees. Based on this approach, the variations in charge carrier motion characteristics in the frequency domain with increasing pressboard aging degree are clarified. Moreover, the function model provides characteristic parameters that are clearly related to the degree of polymerization of the material, which represents a quantitative method for evaluating the aging degree of cellulose pressboard by indirectly obtaining its frequency response curve.

Research on automatic location and recognition of insulators in substation based on YOLOv3

With the development of a smart grid, the automatic location of power equipment is becoming a trend. In this study, a method for automatic location identification and diagnosis of external power insulation equipment based on YOLOv3 is proposed. This deep learning algorithm is used to extract the characteristics of image data under the visible light channel of the insulator. It learns and trains the collected data to realise the rapid location identification and frame selection of the external insulation equipment and extract discharge characteristics of the target box under the ultraviolet channel. According to the number of photons and the spot area information, the operating status of the equipment is determined. The results show that the YOLOv3 algorithm with a training rate of 0.005 achieved a fast convergence of the location recognition model. The average recognition accuracy was 88.7% and the average detection time was 0.0182 s. The combination of visible light path insulator target recognition and ultraviolet light path diagnosis can realise a lean and intelligent diagnosis of power equipment. This method had good real-time performance, accuracy, and robustness to the background. It provides a new concept for intelligent diagnosis and location analysis of power equipment.