Inter-turn short circuit fault is the main cause of dry-type air-core reactor faults. To quickly and accurately detect the inter-turn insulation condition of the reactor, a new method for online diagnosis of inter-turn short circuit faults in reactor coils is proposed. This method utilizes the terminal voltage and total current data of dry-type air-core reactors to construct characteristic graphs to reflect coil changes. The paper first studied the equivalent circuit models under normal and fault conditions and then established a two-dimensional simulation model based on the geometric parameters of the reactor. Finally, different inter-turn short-circuit fault scenarios were set up, and a series of verification analyses were carried out based on simulated voltage and current signals. The results show that this method exhibits significant variation patterns under different operating conditions of the reactor and has certain feasibility.

Dry-type air-core reactors (DAR) are critical components in power systems but are prone to inter-turn short circuit faults which interrupt the symmetry of the winding structure. Inspired by the online detection of transformer winding deformation, the V-I method has been adapted to diagnose short circuit faults in reactors. However, the diagnostic criteria and thresholds of V-I method remain unclear. This paper presents a novel method for determining the threshold for detecting inter-turn short circuit faults in DAR, integrating V-I analysis with machine learning techniques. Specifically, Gradient Boosting Regression (GBR) is used to compute a standard diagnostic criterion value, and curve fitting is also used to define the threshold for identifying inter-turn short circuit faults. The experimental results demonstrate that this method effectively identifies fault conditions in DAR.

针对变电站设备的红外图像特征提取和识别问题,提出了基于不变矩的图像特征提取和识别算法.基于统计学中矩函数的概念,提出了用于表征图像全局特征的图像矩函数,并以二维几何矩为基础,提出了用于图像识别的6 个矩不变量参数.通过对中心矩计算方法进行改进,消除了图像变换过程中的计算误差,实现了图像不变矩参数的准确计算.结果表明,所提出算法的综合识别率可达94％,能够满足实际应用需求.

针对电力通信网络频繁遭受Sybil攻击的问题,提出了一种基于K-means边聚类的Sybil攻击团体检测算法.通过优化边聚类和边介数的计算方法,提出了传统K-means聚类算法的改进方法,计算了通信网络中的聚类系数,根据合法用户的真实数量,建立更加精确的攻击边集合与真实边集合,从而初步检测出所有可疑的攻击边,并使用标签传播算法检测Sybil攻击行为所在的恶意团体.仿真结果表明,与经典的SybilLimit算法相比,在所有的攻击路径数量下,该Sybil攻击检测算法具有更加优秀的检测性能.

To solve the single time-frequency resolution problem of time domain reflectometry (TDR) and frequency domain reflectometry (FDR) in the localization of cable defects, a cable location method based on joint time-frequency analysis is proposed. Firstly, the basic principles of the joint time-frequency location method are briefly explained in terms of time-frequency distribution, reference signal selection, time-frequency offset, and time-frequency cross-correlation function. Then, the method proposed in this paper is used to achieve high-precision location of local defects, and the feasibility of the method is tested by modeling and simulation of 10 kV long cables with local defects. The simulation results show that the method can effectively locate cable local defects in different locations and degrees, and the error of the location results is only about 0.02%. Compared with the traditional TDR and FDR methods, this method utilizes the time limit and band limit characteristics of reference signals and introduces the time-frequency cross-correlation function to achieve higher precision location, which combines the advantages of TDR and FDR.

针对无线传感器网络的安全性问题,基于动态防御技术提出了一种具备多种安全策略的主动式防御安全模型.通过分析无线传感器网络的研究现状与安全需求,选取适当的安全策略和技术,建立基于动态防御技术的无线传感器网络的安全模型.利用NS2软件和MATLAB软件对基于动态防御技术的安全模型进行仿真,统计和分析无线传感器网络的节点在多种模式下接收数据量.仿真结果表明,与经典群密钥算法相比,所提出的安全模型具有更加优秀的安全性能和通信性能.

Partial discharge (PD) is one of the main causes of the interface insulation deterioration of cable accessories. Studies have shown that temperature has an important influence on the PD development, but the effect of temperature on the PD initiation is still not fully understood. This paper investigated the developing regularity of PD at the interface defects of the cable accessories under thermal cycling. An electro-thermal co-aging platform with a cable accessory having semiconductive-layer protrusion defect was firstly designed and introduced in the paper. The experiment loop in the platform was thermally cycled in air by using a current transformer to induce a current in the cable conductor. Cycle duration was 9.5 h with 8h/ON and 1.5h/OFF. Afterwards, PD numbers, PD amplitude and PRPD (Phase Resolved Partial Discharge) spectrum were collected and analyzed comparatively. Obtained results show that temperature change caused by current on-off control will result in temporary PD excitation. PD numbers and amplitude increase rapidly as the temperature rises or falls quickly. When temperature stabilizes, however, the increment slows down. Phenomenon above can be interpreted to the interface pressure change caused by thermal expansion and contraction of the interface insulation, namely “Breathing Effect”. The research revealed that PD signals in the interface defect are more easier to be detected when cable accessories are operating under high temperature gradient, which may provide practical references for PD field test of cable accessories.

针对变压器遭受短路冲击时的声信号,该文提出一种基于稀疏自适应S变换的变压器绕组状态检测方法.首先依据变压器短路冲击声信号的幅值谱确定了稀疏自适应S变换的窗函数,然后采用双向2维主成分分析对声信号时频特征矩阵进行了降维处理,据此定义了特征能量变化率来评判变压器的绕组状态.对某220kV变压器短路冲击声信号的计算结果表明:基于稀疏自适应S变换得到的声信号优化时频域特征具有分辨率高、能量聚集性强的优点,所定义的特征指标能有效反映变压器绕组状态变化.研究结果可为变压器短路冲击绕组状态检测提供重要依据.

随着电力大数据时代的到来,数据量的骤增使得维护决策者所面临的决策环境与决策问题变得愈发复杂.针对这一问题,在云计算技术和设备状态维护技术的基础上,对云平台下的设备状态维护系统进行研究,构建基于云计算的全维度设备状态维护系统整体框架,并建立系统维护模型,针对遗传算法在求解实际问题时,计算能力难以满足人们需要的不足,引入Map Reduce技术,结合遗传算法和Map Reduce技术进行并行设计,并通过对比实验分析系统性能,此研究工作为我国电力设备状态维护系统的发展提供了参考和借鉴.

The relationship between the open circuit fault and the three-phase current is summarized by analyzing the working state of the motor when the open circuit fault occurs in the power tube. An open circuit fault diagnosis method for power tube of brushless motor inverter is proposed for substation inspection robot. In this method, using the sampling values of the drive system Hall sensor and current sensor as raw data which are processed by the proposed fault diagnosis algorithm, then the fault diagnosis and location of single tube and multi-tube power tube are realized. Simulation test shows the power tube open circuit fault diagnosis method presented in this paper is suitable for fault diagnosis and location of power tube under different rotating speed and load of robot joint motor, it is not affected by system open loop, close loop or PWM modulation strategy. It does not need to increase hardware cost, it is convenient and reliable, and it is not affected by the change of working condition, can effectively realize the open circuit fault diagnosis of power tube in brushless motor inverter for substation inspection robot.

为准确识别变压器的工作状态,该文从运行中的变压器声音信号出发,提出一种基于压缩观测与判别字典学习的变压器声纹识别方法.首先采用稀疏随机矩阵对变压器声音信号进行压缩观测来获取观测信号,然后将变压器不同状态下的观测信号组成样本集,经迭代算法训练获得子字典、公共字典和判别字典.同时,为提升判别字典的分类性能,引入了Fisher判别约束项和公共字典低秩约束项来优化判别字典学习的目标函数,并对目标函数中的惩罚系数进行了优选.最后,求解待识别样本在判别字典上的稀疏表示系数,根据其在子字典上的重构误差最小原则进行识别.对某10kV干式变压器正常与典型故障下声音信号的分析结果表明,压缩观测可将声音信号数据量减小至原来的10％,大幅提升了运算效率.所构建的字典学习目标函数能有效增强字典的判别性能,对变压器不同工作状态的总体识别准确率可达96％,从而为变压器的声纹识别提供了一种新的思路.

运行中的变压器声信号由其机械结构振动经空气传播形成,易受周围环境影响,且所含干扰信号具有较大的不确定性,给基于声信号的变压器状态监测带来不利影响.为有效获取反映变压器运行状态的声信号,从特高压交流变压器声信号的声学特征出发,依据声信号的集合经验模态分解(ensembled empirical mode decomposition,EEMD)结果,提出了基于稀疏表示理论的特高压交流变压器声信号盲分离方法,即通过构建K-奇异值分解(K-singular value decomposition,K-SVD)字典和应用正交匹配追踪算法,得到了成分相对单一的特高压交流变压器声信号,有效抑制了特高压交流变压器声信号中的干扰成分.对某1000kV变电站主变声信号的计算结果表明:基于EEMD算法和稀疏表示理论能有效地对特高压交流变压器声信号中包含的环境噪声和人说话声等干扰成分进行盲分离,具有自适应性强、计算效率高和重构误差低的优点,可为基于声信号的特高压交流变压器状态监测提供重要数据支持.

In order to build an intelligent, unmanned and managed substation, the substation gradually adopts inspection robots instead of manual work. However, during the inspection process, the automatic identification of the pointer type meter always has the problem that the recognition accuracy is not high and is susceptible to illumination changes. In this paper, the problem of pointer instrument identification is studied. The method of instrument detection and localization algorithm, pointer and instrument scale fitting in complex environment is studied systematically. A pointer meter identification algorithm suitable for intelligent substation inspection robot is proposed. Using the positioning function of deep learning, the instrument classification algorithm based on Faster R-CNN is used to classify three types of tables: voltmeter, ammeter and digital table. Image processing of the positioned pointer meter image, such as tilt correction, extraction of scale segments, pointer line segments, line fitting, repairing the default tick marks, etc. is operated to calculate a more accurate pointer reading.

针对目前变电站数量规模大而人工巡检方式效率较低的问题,文中设计了一套面向变电站的机器人巡检控制系统.该系统分为基于巡检机器人及充电室等的移动站系统和由交换机、导航定位规划系统及C++语言开发的基站系统,具有巡检任务管理、信息显示、图像监控、机器人控制、报警输出等功能.经过实现与测试结果表明,该系统能够有效控制机器人完成定点巡检、定时巡检、遥控巡检等任务,检测结果准确且效率较高,为变电站无人巡检提供了一套完整的技术解决方案.

针对现有的语义分割算法存在分割结果空间不一致的问题,提出一种基于加权损失函数的多尺度对抗网络语义分割算法。在DeepLab v3基本框架的基础上,引入Pix2pix网络作为生成对抗网络模型,实现多尺度对抗网络语义分割。同时,为增加模型的泛化能力与训练精度,提出将传统的多分类交叉熵损失函数与生成器输出的内容损失函数和鉴别器输出的对抗损失函数相结合,构建加权损失函数。大量定性定量实验结果表明,该算法能够识别并分割细小的物体,其语义分割性能超过现有的深度网络,在保证语义分割空间一致性的同时提高了分割效率。

局部放电试验作为一种非破坏性试验,是电力设备绝缘检测和诊断的重要方法.人们对电力设备可靠性的要求越来越高,针对现在电力设备事故大部分是由设备绝缘局部放电造成的,文中通过对惰性气体、环氧树脂、空气、油四种介质中的局部放电的实测波形进行分析,得出了不同类型局部放电的波形时频域特征,为局部放电模式识别特征的提取提供了科学依据.

在智能电网中应用物联网技术,可以有效整合电网信息资源,实现电网信息的精细化和可视化.当前,物联网在电网中的应用相对简单,在电网设备运行监控和电力资产管理中尚无具体应用.针对输变电设备在电网系统中的重要性和智能电网的国家战略,研究基于全景数据的输变电设备状态监测系统,把信息模型引入到输变电设备的物联网中.根据输变电设备全寿命周期管理业务所需的设备全景信息,针对智能变电站设备,提出一个全面的全景信息建模方案,通过仿真分析监测网络的性能.该研究为我国输变电设备状态监测系统的发展提供参考和借鉴.

In this paper, we presented a novel solution using ensemble empirical mode decomposition and support vector machine joint approach to efficiently identify the multiple coexisting faults in the operational rotational machinery. To further improve the performance of SVM models, the EEMD is used to decompose the collected field vibration signals into a set of intrinsic mode functions (IMFs). The presented approach is validated using simulation experiments based on the benchmark vibration data. The numerical result confirmed the feasibility and effectiveness of the proposed solution.

随着科学技术的不断发展以及城市化水平的不断加快和人民生活水平的不断改善，人们对于供电的要求也在日益提高。新形势下，国家提出更好地加快城乡电网建设，更好地保障用电的安全和质量。针对高压变电站电气设计过程中存在的问题进行深刻的分析，积极探讨高压变电站在设计过程中需要注意的问题。