Comprehensive evaluation of the transformer oil-paper insulation state based on RF-combination weighting and an improved TOPSIS method

The accurate identification of the oil-paper insulation state of a transformer is crucial for most maintenance strategies.This paper presents a multi-feature comprehensive evaluation model based on combination weighting and an improved technique for order of preference by similarity to ideal solution(TOPSIS)method to perform an objective and scientific evaluation of the transformer oil-paper insulation state.Firstly,multiple aging features are extracted from the recovery voltage polarization spectrum and the extended Debye equivalent circuit owing to the limitations of using a single feature for evaluation.A standard evaluation index system is then established by using the collected time-domain dielectric spectrum data.Secondly,this study implements the per-unit value concept to integrate the dimension of the index matrix and calculates the objective weight by using the random forest algorithm.Furthermore,it combines the weighting model to overcome the drawbacks of the single weighting method by using the indicators and considering the subjective experience of experts and the random forest algorithm.Lastly,the enhanced TOPSIS approach is used to determine the insulation quality of an oil-paper transformer.A verification example demonstrates that the evaluation model developed in this study can efficiently and accurately diagnose the insulation status of transformers.Essentially,this study presents a novel approach for the assessment of transformer oil-paper insulation.

Identification of XLPE cable insulation defects based on deep learning

The insulation aging of cross-linked polyethylene(XLPE)cables is the main reason for the reduction in cable life.There is currently a lack of rapid and effective methods for detecting cable insulation defects in power-related sectors.To this end,this paper presents a method for identifying insulation defects in XLPE cables based on deep learning algorithms.First,the principle of the harmonic method for detecting cable insulation defects is introduced.Second,the ANSYS software is used to simulate the cable insulation layer containing bubbles,protrusions,and water tree defects,and the effects of each type of defect on the magnetic field strength and eddy loss current of the cable insulation layer are analyzed.Then,a total of 10 characteristic quantities of the total harmonic content and 2nd to 10th harmonic currents are constructed to establish a database of cable insulation defects.Finally,the deep learning algorithm,long short-term memory(LSTM),is used to accurately identify the types of insulation defects in cables.The results indicate that the LSTM algorithm can effectively diagnose and identify insulation defects in cables with an accuracy of 95.83%.

Chaotic Characteristic of Time Series of Partial Discharge in Oil-Paper Insulation

The chaotic characteristics of time series of five partial discharge （PD） patterns in oil-paper insulation are studied. The results verify obvious chaotic characteristic of the time series of discharge signals and the fact that PD is a chaotic process. These time series have distinctive features, and the chaotic attractors obtained from time series differed greatly from each other by shapes in the phase space, so they could be used to qualitatively identify the PD patterns. The phase space parameters are selected, then the chaotic characteristic quantities can be extracted. These quantities could quantificationally characterize the PD patterns. The effects on pattern recognition of PRPD and CAPD are compared by using the neural network of radial basis function. The results show that both of the two recognition methods work well and have their respective advantages. Then, both the statistical operators under PRPD mode and the chaotic characteristic quantities under CAPD mode are selected comprehensively as the input vectors of neural network, and the PD pattern recognition accuracy is thereby greatly improved.

The Research of Connection between Degree of Polymerization and Frequency Domain Dielectric Spectroscopy Characteristics during Oil-Paper Insulation Aging

In order to study the linkage effects between degree of polymerization and frequency domain dielectric spectroscopy characteristics of oil-paper insulation, the frequency domain dielectric response test platform of oil-paper insulation is set up. Complex permittivity of oil-paper insulation respectively composed by new or aged oil and insulation paper with different DP are tested, and complex permittivity of oil-paper insulation respectively composed by insulation respectively composed by new oil and insulation paper with different DP and low or high moisture content are tested. The test results are analyzed, and the analysis results show that the degree of polymerization of insulation paper has an influence on complex permittivity of oil-paper insulation though influencing the distribution of moisture and acids between oil and paper.

A Simple and Effective Surface Defect Detection Method of Power Line Insulators for Difficult Small Objects

Insulator defect detection plays a vital role in maintaining the secure operation of power systems.To address the issues of the difficulty of detecting small objects and missing objects due to the small scale,variable scale,and fuzzy edge morphology of insulator defects,we construct an insulator dataset with 1600 samples containing flashovers and breakages.Then a simple and effective surface defect detection method of power line insulators for difficult small objects is proposed.Firstly,a high-resolution featuremap is introduced and a small object prediction layer is added so that the model can detect tiny objects.Secondly,a simplified adaptive spatial feature fusion(SASFF)module is introduced to perform cross-scale spatial fusion to improve adaptability to variable multi-scale features.Finally,we propose an enhanced deformable attention mechanism(EDAM)module.By integrating a gating activation function,the model is further inspired to learn a small number of critical sampling points near reference points.And the module can improve the perception of object morphology.The experimental results indicate that concerning the dataset of flashover and breakage defects,this method improves the performance of YOLOv5,YOLOv7,and YOLOv8.In practical application,it can simply and effectively improve the precision of power line insulator defect detection and reduce missing detection for difficult small objects.

Photon count technique as a potential tool for insulation microdefect detection:Principles and primary results

The requirements for the construction of a new power system inevitably pose significant challenges and changes to the operation and maintenance of the power grid.To ensure the safe and stable operation of ultra-high voltage(UHV)transmission equipment,this work reports on the principles and preliminary results of using electroluminescence(EL)-based photon counting(PC)methods for early detection of micro-defects in GIS/GIL insulation spacer.In this study,the impact of voltage,gas pressure,and gas composition on the photon response of insulation is examined.Furthermore,the corresponding relationship between defect status and photon response characteristics is explored,along with the discussion of the EL mechanism and its evolution induced by defects.The research results demonstrate that PC measurement exhibits high sensitivity to variations in millimeter-scale defect size,position,and morphology at lower electric fields before partial discharge(PD)initiation.With this regard,this paper reveals promising prospects for the early detection of micro-defects in UHV transmission equipment using PC measurement-based methods.

Intelligent identification method of insulation pull rod defects based on intactness-aware Mosaic data augmentation and fusion of YOLOv5s

The authors introduce the intactness-aware Mosaic data augmentation strategy,designed to tackle challenges such as low accuracy in detecting defects in insulation pull rods,limited timeliness in intelligent analysis,and the absence of a comprehensive database for information on insulation pull rod defects.The proposed strategy incorporates the YOLOv5s algorithm for detecting defects in insulation pull rods.Initially,the YOLOv5s network was constructed,and a dataset containing photos of insulation pull rods with white spots,fractures,impurities,and bubble flaws was compiled to capture images of defects.The research presented a data enhancement approach to improve the images and establish a dataset for insulation pull rod defects.The YOLOv5s algorithm was applied for both training and testing purposes.A comparative analysis was conducted to assess the detection performance of YOLOv5s against a conventional target detector for identifying defects in insulation pull rods.Furthermore,the utility of Mosaic's data augmentation technique,which incorporates intactness awareness,was evaluated to enhance the accuracy of identifying insulation pull rod defects.The research findings indicate that the YOLOv5s algorithm is employed for intelligent detection and precise localisation of flaws.The intactnessaware Mosaic data augmentation strategy significantly improves the accuracy of detecting faults in insulation pull rods.The YOLOv5s model used achieves a performance index mAP@0.5:0.95 of 0.563 on the test set,distinct from the training set data.With a threshold of 0.5,the mAP@0.5 score is 0.904,indicating a substantial improvement in both detection efficiency and accuracy compared to conventional target detection methods.Innovative approaches for identifying defects in insulation pull rods are introduced.

A New Method for the Aging Evaluation of Oil-Paper Insulation Using η-Butanol and Methanol

To improve the accuracy and efficiency of the aging life prediction and assessment of transformer oil-paper insulation,and to make up for the deficiencies of traditional characterizers,such as 2-furfural,carbon monoxide,and carbon dioxide,a method for the simultaneous determination of methanol,ethanol,n−propanol,and η-butanol in oil with a single injection is established by headspace-gas chromatography-mass spectrometry.The measured results show that the determination limits of the four alcohol characterizers can be controlled to 10μg/kg level.Based on this method,the change patterns of the above four alcohols with thermal aging time and degree of polymerization are obtained through thermal aging experimental research.Ethanol,η-propanol,and η-butanol in oil indicate nearly linear correlations with thermal aging time and degree of polymerization,similar to that of methanol.By analyzing 52 sets of measured data of 500 kV EHV transformers in operation,η-butanol is found to have excellent performance,and a new method to evaluate the aging state of oil-paper insulation employing η-butanol and methanol is proposed along with the aging attention value model.The measured data of 500 kV EHV transformers in operation indicate that the combination of η-butanol and methanol as the preferred characterizers can effectively compensate for the shortcoming of traditional characterizers in the early stages of aging,and the feasibility of the method is verified.Two possible pathways for the generation of η-butanol by cellulose cleavage during the aging of oil-paper are proposed from the chemical structure of cellulose.

An Experience of On-site PD Testing for Condition Monitoring of an 11 kV PILC Cable Insulation System

A cable circuit of a substation in the United Kingdom showed high level of PD activities during a survey using hand hold PD testing equipment. The authors were invited to carry out on-site PD testing experiment to further diagnose and locate the potential problem of the cable system. This paper presents the experience of the present authors carrying out the cable test. Following a brief introduction to the experiment equipments and physical connections, the paper analyses the data collected from the testing, including PD pulse shape analysis, frequency spectrum analysis and phase resolved PD pattern analysis. Associated with PD propagation direction identification, PD source diagnosis and localisation was made. Four different types of sensors, which were adapted during the testing, are shown to have different frequency bandwidths and performed differently. Aider comparing the parameters of the sensor and the PD signals detected by individual sensor, optimal PD monitoring bandwidth for cable system is suggested.

Photon counting technique as a potential tool in micro-defect detection of epoxy insulation pull rod in gas-insulated switchgears

The micro-defects in epoxy-based insulation materials generate a local high electric field which results in continuous degradation,seriously endangering the insulation system of gas-insulated switchgears.A highly sensitive detection technique is reported for micro-defects of insulation pull rods based on the photon counting(PC)technique.The re-sults demonstrated that for an epoxy-based insulation pull rod,the photons released during electroluminescence and ionisation at 2 kV,which is less than the partial discharge inception voltage,can be clearly detected.The findings presented a strong correlation between photon counts and defect severity.Discourse has been conducted to elucidate the mechanism behind defect-induced PC,employing the amplification of ionising luminescence through electric field distortion induced by micro defects and the augmentation of electroluminescence through the aggregation of trap charge.In this regard,the authors verified that PC can serve as a potential tool in the detection of micro-insulation defects,which also has huge potential in online insulation condition monitoring.

Research on the feature extraction of DC space charge behavior of oil-paper insulation

As the ultra high DC voltage grid comes into an important stage in the ‘Twelfth Five-year Plan’ of China, the requirement on the operation safety of power equipment needs to be improved. Oil-paper insulation plays a significant role as the main insulation material of many key power equipments. The space charge behaviors of oil-paper insulation not only affect the dielectric strength, but also relate to the stability and security of power transmission facilities. Until now, the research on the space charge behaviors of oil-paper insulation remains on the stage of objective description and phenomenon explanation, thus it is very urgent to make some feature extractions on the test results in order to further investigate the DC space charge behaviors of oil-paper insulation. In this paper, based on the large numbers of test data in the first stage, data mining and feature extractions were performed on the DC space charge behavior of oil-paper insulation. The variation laws of some key characteristic parameters obtained in different experiment conditions such as the total charge injection amount, apparent charge mobility, and trap energy distribution were analyzed. In addition, the distribution of trapping level inside oil-paper was calculated by using the PEA test results based on the mechanism of the isothermal decay current, which realized the approximate calculation of trap depth by charge decay characteristics. What’s more, the exponential injection and decay equations of DC space charge changing with time were acquired for the first time, and the physical meanings of related parameters were defined. This has laid a foundation for the prediction and simulation of space charge in the oil-paper insulation.

Space Charge Characteristics of Oil-paper Insulation in the Electro-thermal Aging Process

High electric field and temperatures can result in electrical property degradation of oil-paper insulation and accelerate its aging process.This paper describes the regulation of space charge in the process of electro-thermal aging.Electric field distortion and differences in characteristic parameters during electro-thermal aging are analyzed,along with the effects of electro-thermal aging on space charge from the perspective of trap energy.High temperature and an electric field are used to accelerate the aging of oil-paper,and space charge characteristics are measured using the pulsed electro-acoustic(PEA)method throughout the electro-thermal aging.Based on results obtained,it can be concluded that homo-charge injection occurs at the anode and the type of the charge injection at the cathode varies throughout the electro-thermal aging;the decrease of permittivity during aging allows for space charge injection to take place;growth of trap depth makes the space charge accumulate in the middle of the sample;and space charge accumulation after electro-thermal aging results in inner electric field distortion,which leads to large current and decreased breakdown in voltages.

Electrical Properties of Semi-insulating GaAs Grown from the Melt Under Microgravity Conditions

A technologically important undoped semi-insulating(SI)GaAs single crystal was successfully grown in the Chinese recoverable satellite as far as we know for the first time by using a similar growth configuration described previously.The experimental results proved that the space SI GaAs crystals have a lower density of defects and defect-impurity complexes as well as a better uniformity.

Electron states scattering off line edges on the surface of topological insulator

We study the local density of states （LDOS） for electrons scattering off the line edge of an atomic step defect on the surface of a three-dimensional （3D） topological insulator （TI） and the line edge of a finite 3D TI, where the front surface and side surface meet with different Fermi velocities, respectively. By using a S-function potential to model the edges, we find that the bound states existed along the step line edge significantly contribute to the LDOS near the edge, but do not modify the exponential behavior away from it. In addition, the power-law decaying behavior for LDOS oscillation away from the step is understood from the spin rotation for surface states scattering off the step defect with magnitude depending on the strength of the potential. Furthermore, the electron refraction and total reflection analogous to optics occurred at the line edge where two surfaces meet with different Fermi velocities, which leads to the LDOS decaying behavior in the greater Fermi velocity side similar to that for a step line edge. However, in the smaller velocity side the LDOS shows a different decaying behavior as x-1/2, and the wavevector of LDOS oscillation is no longer equal to the diameter of the constant energy contour of surface band, but is sensitively dependent on the ratio of the two Fermi velocities. These effects may be verified by STM measurement with high precision.

基于轻量化网络与增强多尺度特征融合的绝缘子缺陷检测

随着无人机搭载目标检测算法在输电杆塔绝缘子巡检领域的发展,针对绝缘子缺陷检测速度较低,网络复杂度高且缺陷小目标难以准确检测的问题,提出一种基于轻量化网络与增强多尺度特征融合的YOLOv5-3S-4PH模型进行绝缘子缺陷实时检测。首先将重构的ShuffleNetV2-Stem-SPP(3S)网络作为YOLOv5的主干网络,显著减小了网络的参数量和计算量;其次引入针对小目标的增强多尺度特征融合网络以及4个预测头,来增强网络对绝缘子缺陷的感知能力,并结合Mosaic-9数据增强、CIoU损失函数进一步补偿轻量化导致的检测精度损失;最后将其应用到自制绝缘子数据集进行验证。实验结果表明,该文所提出的模型相对于未改进的YOLOv5,全类平均精度提高了3%,检测速度提高了81.8%,参数量、计算量分别压缩了82.4%、67%。因此,所提出的模型更适合部署在无人机平台上进行绝缘子缺陷的实时监测。

基于局部特征深度信息的绝缘子小样本缺陷检测

基于深度学习的目标检测技术已广泛应用于绝缘子缺陷检测中,然而现有目标检测算法主要基于大量缺陷样本训练网络模型,无法对少样本缺陷进行准确识别。针对绝缘子缺陷检测过程中缺陷样本量不足的问题,该文提出了一种基于局部特征深度信息的绝缘子小样本缺陷检测方法。首先采用旋转目标检测网络改进Faster R-CNN(faster region-based convolutional neural network)模型提取绝缘子串区域,然后对绝缘子串特征进行划分,提取绝缘子串局部特征并基于深度推土距离(deep earth mover’s distance,Deep EMD)网络实现小样本缺陷检测。实验结果表明,在玻璃绝缘子自爆缺陷检测中,所提出方法采用2张训练样本可取得与现有目标检测方法 200张训练样本相同的效果,采用10张训练样本的绝缘子自爆检测在与真值框的交并比阈值为0.5至0.95之间的平均精度(mean average precision,mAP)达到0.65,该方法为小样本电力设备缺陷智能化检测提供了新的方法和思路。

Effect of As Interstitial Diffusionon on the Properties of Undoped Semi-insulating LECGaAs

Annealing was carried out at 950 and 1120 degreesC under various As pressure for undoped (ND) semi-insulating (SI) LECGaAs. The effects of annealing on native defects and electrical properties were investigated. Experimental results indicate that, after an annealing at 950 degreesC for 14 h under low As pressure, the Hall mobility decreases and the resistivity increases dramatically for the samples. These changes in electrical properties are due to the generation of intrinsic acceptor defects, and the generation of the intrinsic acceptor defects originates from the outdiffusion of As interstitial at high temperature. The generation of the intrinsic defects and these changes in electrical properties can be suppressed by increasing the applied As pressure during annealing. The concentration of the main donor defect E12 (AsGaVGa) can be decreased by about one order of magnitude by an evacuated annealing at 1120 degreesC for 2-8 h followed by a fast cooling. The decrease in E12 concentration can also be suppressed by increasing the As pressure during annealing.

IDD-YOLOv7:一种用于输电线路绝缘子多缺陷的轻量化检测方法

YOLO目标检测算法是当前基于图像的输电线路绝缘子缺陷检测的主流方法,然而现有模型复杂度较大,亟需合理有效的参数压缩方法作为前提条件,来为解决无人机边缘设备部署的困境问题奠定基础;同时,无人机航拍的绝缘子缺陷图像背景复杂、缺陷尺寸较小,容易出现误检、漏检等问题。为此,提出了一种用于输电线路绝缘子多缺陷检测的Insulator Defect Detection-YOLOv7(IDD-YOLOv7)模型,以降低模型复杂度,提高模型鲁棒性。首先,在多尺度特征融合的过程中加入坐标注意力(Coordinate Attention)机制,抑制复杂背景的干扰,提升模型对小目标的全局感知能力;之后,设计C3GhostNetV2模块,用于捕获不同空间像素之间的远程依赖性,在增强模型表达能力的同时降低模型的参数量和浮点运算量;最后,提出Focal-CIoU损失函数,提高模型高质量anchor的贡献,加快模型的收敛速度。实验结果表明,本文方法与基线模型相比mAP^(50)提升了3.8%,查准率和召回率分别提升了1.7%和7.6%,参数量和浮点运算量分别下降了18.3%和14.0%,绝缘子自爆、破损、闪络缺陷的AP^(50)分别提升了0.8%、4.5%、6.3%。

基于深度学习的输电线路绝缘子故障检测方法

针对航拍巡检高压输电线路上绝缘子目标易受复杂背景和部分遮挡影响,造成传统算法难以准确检测的问题,建立了一种基于改进YOLOv5的输电线路绝缘子检测模型。首先,利用GA+K的检测框优化算法对选择的模型进行改进,来提高识别精度;然后,在YOLOv5算法框架中融合CBAM模块来提升图像中故障目标区域的显著度;其次,采用高斯函数改进YOLOv5中的非极大值抑制方法,提高对遮挡目标的识别准确率;最后利用辽宁某电网公司提供的无人机巡检图像制作数据集,并将所提算法与4种经典目标检测算法进行比较。试验结果表明,相比于4种对比算法,该算法能够在保证较高检测精度的同时具有较好的实时性,并且平均检测精度可以达到95.1%,每张图片的检测时间为0.04 s,兼具目标检测的准确性和实时性。

TSV阵列在温度循环下的晶格变化对电学性能影响的研究

针对TSV(Through Silicon Via,硅通孔)热机械可靠性较差且其结构特性变化对电学性能影响情况不明的问题,基于一种TSV阵列叉指电极对样品开展了-55~125℃的温度循环试验,测试了温循后阵列电极的电学性能与边界绝缘性能变化,跟踪测试了TSV铜柱的几何尺寸变化,对比分析了试验前后TSV结构的形貌和晶格特性等衍化规律。结果表明,温度循环载荷导致了TSV-Cu晶粒向上生长、铜柱体积变大、微观结构缺陷产生以及电学性能退化;此外,还定量地构建了晶格特性变化对TSV阵列电极电学性能影响的关系架构,具有一定的工程意义。