Machine learning for parameters diagnosis of spark discharge by electro-acoustic signal

Discharge plasma parameter measurement is a key focus in low-temperature plasma research.Traditional diagnostics often require costly equipment,whereas electro-acoustic signals provide a rich,non-invasive,and less complex source of discharge information.This study harnesses machine learning to decode these signals.It establishes links between electro-acoustic signals and gas discharge parameters,such as power and distance,thus streamlining the prediction process.By building a spark discharge platform to collect electro-acoustic signals and implementing a series of acoustic signal processing techniques,the Mel-Frequency Cepstral Coefficients(MFCCs)of the acoustic signals are extracted to construct the predictors.Three machine learning models(Linear Regression,k-Nearest Neighbors,and Random Forest)are introduced and applied to the predictors to achieve real-time rapid diagnostic measurement of typical spark discharge power and discharge distance.All models display impressive performance in prediction precision and fitting abilities.Among them,the k-Nearest Neighbors model shows the best performance on discharge power prediction with the lowest mean square error(MSE=0.00571)and the highest R-squared value(R^(2)=0.93877).The experimental results show that the relationship between the electro-acoustic signal and the gas discharge power and distance can be effectively constructed based on the machine learning algorithm,which provides a new idea and basis for the online monitoring and real-time diagnosis of plasma parameters.

Cycle-by-Cycle Queue Length Estimation for Signalized Intersections Using Multi-Source Data

In order to estimate vehicular queue length at signalized intersections accurately and overcome the shortcomings and restrictions of existing studies especially those based on shockwave theory,a new methodology is presented for estimating vehicular queue length using data from both point detectors and probe vehicles. The methodology applies the shockwave theory to model queue evolution over time and space. Using probe vehicle locations and times as well as point detector measured traffic states,analytical formulations for calculating the maximum and minimum( residual) queue length are developed. The proposed methodology is verified using ground truth data collected from numerical experiments conducted in Shanghai,China. It is found that the methodology has a mean absolute percentage error of 17. 09%,which is reasonably effective in estimating the queue length at traffic signalized intersections. Limitations of the proposed models and algorithms are also discussed in the paper.

Chromatic Mapping of Partial Discharge Signals

This contribution describes the use of chromatic techniques for quantifying signatures of partial discharges(PD).A brief description of the basis of chromatic monitoring is given along with explanations of how the approach may be applied for addressing signals produced by PDs.Laboratory test data from several sources for different types of PDs,operating conditions and geometries have been subjected to the chromatic procedures.These include a point-plane gap,a sphere in a liquid and treeing in a cable.Chromatic changes in PD signals have been investigated as a function of the amplitude of the alternating voltage producing the PD,the time duration before full electrical breakdown and for different forms of PDs.Results of the chromatic processing of these data are presented in the form of a number of chromatic maps relating to different quarter cycles of the alternating voltage producing the PDs.The results show the potential of the chromatic techniques for indicating the likelihood of full electrical breakdown and for distinguishing between certain forms of PDs.A summary of the chromatic processing procedures is presented for producing chromatic maps and for adaptation in further exploring PD signal features.

On-Line Measurement of Ion Density in Atmospheric Nitrogen Discharge Filaments via Radiation Signals from Plasma Oscillation

Diagnosis of the particle number density of plasma plays an important role in the understanding of plasma sources and processing.Regular radiation signals from plasma oscillation in filaments of atmospheric nitrogen discharge,which were excited by the injection of secondary electron beams during the propagation of the streamer,are employed to determine the ion density of plasma and its evolution in the filaments.Results show that the density of N4＋ in a filament of atmospheric nitrogen discharge is of the order of 1013 cm-3.It is also found that the recombination processes play a dominant role in plasma decay,and that the ion density decreases non-monotonically with time during streamer propagation.

Study on characteristics of acoustic signals generated by different DC discharge modes

Acoustic signals contain rich discharge information.In this study,the acoustic signal characteristics of transient glow,spark,and glow discharges generated through DC pin–pin discharge were investigated.The signals were analyzed in the time,frequency,and time–frequency domains,and the correlation between the electric and the acoustic signal was studied statistically.The results show that glow discharge does not produce measurable sound signals.For the other modes,with a decrease in the discharge gap,the amplitude of the acoustic signal increases sharply with mode transformation,the short-time average energy becomes higher,and the frequency components are more abundant.Meanwhile,the current pulse and sound pressure pulse have a one-to-one relationship in the transient glow and spark regimes,and they are positively correlated in amplitude.A brief theoretical analysis of the mechanism of plasma sound and the trends of signals in different modes is presented.Essentially,the change in the discharge energy is closely related to the sound generation of the plasma.

Extraction of Partial Discharge Acoustic Signal by Wavelet Transform with Teager′s Energy Operator

To develop a measurement system for monitoring partial discharge （PD） without the effect of external interferences,an algorithm of PD signal extraction based on wavelet transform with Teager＇s energy operators was presented. Acoustic signal generated by PD was selected to remove excessive interfering signals and electromagnetic interferences. Acoustic signals were collected and decomposed into I0 levels by wavelet transform into approximation and detail components. “Daubechies 25” was proved to be the most suitable mother wavelet for the extraction of PD acoustic signals. Compared with conventional wavelet denoising method, Teager＇s energy operators were adopted to the PD signal reconstruction and the signal to noise ratio was in creased by 20%-25% inthe experiment,without lost in energy and pulse amplitude.

A New Feature Space for Partial Discharge Signal Separation Based on DWT Coefficient Variance

In this paper, a new feature space for PD （partial discharge） signal separation is presented. Three typical PD defects were experimentally reproduced in a laboratory for obtaining independent PD sources. Signals were acquired with a digital storage oscilloscope and then post-processed with DWT （discrete Wavelet transform） for de-noising. The new feature space for PD source separation was constructed with the variance of each Wavelet coefficient vector and was compared with an established feature space for PD source separation; based on the energy of DWT coefficient vectors. After a space reduction by mean of PCA （principal components analysis）, the separation capability among them was measured by comparing the final classification error after training a neural network Results showed that with this new feature space it is possible to separate different sources of PD signals. Later, the feature space proposed was used to separate two PD sources from a real equipment tested. Further analysis on the reduced feature space has shown the band location of PD signals information for separating purpose.

Application of Complex Daubechies Wavelet in UHF Partial Discharge Measurements

On-line partial discharge(PD)detection still remains a very challenging task because of the strong electromagnetic interferences.In this paper,a new method of de-noising,using complex Daubechies wavelet(CDW)transform,has been proposed.It is a relatively recent enhancement to the real-valued wavelet transform because of tow important properties,which are nearly shift-invariant and availability of phase information.Those properties give CDW transform superiority over other real-valued wavelet transform,and then the construction algorithm of CDW is introduced in detail.Secondly,based on the real threshold algorithm of real-valued wavelet transform,complex threshold algorithm is devised.This algorithm take the different characteristics of real part and imaginary part of complex wavelet coefficients into account,it modifies the real and imaginary parts of complex wavelet coefficients respectively.Thirdly,to obtain a real de-noised signal,new combined information series is devised.By applying different combination of real part and imaginary part of de-noised complex signal,a real de-noised signal can be restored with higher peak signal-to-noise ratio(PSNR)and less distortion of original signals.Finally,On-site applications of extracting PD signals from noisy background by the optimal de-noising scheme based on CDW are illustrated.The on-site experimental results show that the optimal de-noising scheme is an effective way to suppress white noise in PD measurement.

Experimental analysis on the nonlinear behavior of DC barrier discharge plasmas

Nonlinear behavior of glow discharge plasmas is experimentally investigated.The glow is generated between a barrier semiconductor electrode,Chromium doped namely Gallium Arsenide（Ga As：Cr）,as a cathode and an Indium–Tin Oxide（ITO） coated glass electrode as an anode,in reverse bias.The planar nature of electrodes provides symmetry in spatial geometry.The discharge behaves oscillatory in the time domain,with single and sometimes multiperiodicities in plasma current and voltage characteristics.In this paper,harmonic frequency generation and transition to chaotic behavior is investigated.The observed current–voltage characteristics of the discharge are discussed in detail.

Study on Intermittent Discharge Characteristics of Typical Solid Insulation Defects in Gas Insulated Switchgear

In response to the problem of frequent leakage and false alarm of partial discharge insulation defects in GIS,this paper conducts experimental research on intermittent discharge characteristics of common solid insulation defects in GIS.Using true GIS to build amulti-source testing platform for intermittent discharge of solid insulation defects,and using pulse current method,ultra-high frequency method,ultrasonic method,and gas characteristic component detectionmethod to study the variation law of intermittent discharge characteristics of solid insulation defects.The results showthat:the intermittent discharge state of metal fouling defects on the solid insulation surface decreases with the extension of the discharge time,the discharge time interval as a whole shows an increasing trend,and the metal fouling defects on the solid insulation surface in the intermittent discharge state is difficult to develop into a breakdown discharge fault.The overall discharge amount of air gap defects inside solid insulation also shows the law of increasing first and then decreasing,and the discharge amount of air gap defects inside solid insulation stays at a large level for a long time during intermittent discharge,and there is a risk of breakdown,and the discharge will gradually evolve into continuous discharge;ultrasonic method and characteristic component detection method cannot realize the effective perception of intermittent discharge of metal fouling.The 50-period mapping information commonly used in the power field cannot reflect the intermittent discharge variation law of solid insulation,and the 200-period UHF mapping composed of four groups of 50-period UHF mapping data can initially derive the intermittent discharge distribution law of solid insulation defects.The research results of this paper are of great significance to expand the knowledge of intermittent discharge characteristics and the detection of GIS discharge defects in the field.

Modeling Rogowski Coils for Monitoring Surge Arrester Discharge Currents

Rogowski coils(RCs)are widely used to measure power or high frequency currents based on their design.In this paper,two types of RCs that are circular(traditional)and cylindrical shapes wound using wire covered by varnish are constructed.This construction is carried out to be suitable for monitoring the discharge current of the surge arrester installed in the distribution system.Concerning high frequency RC modeling for both types considering transfer function is introduced.Self-integrating for both types is attained.Therefore,the experimental tests using function generator for both coils are carried out to identify the parameters of the transfer function representing the introduced model.The measured signals for current and induced voltages are denoised for the parameter identification process.The denoised process is achieved using the MATLAB code‘wdenoise’while the parameters are estimated using the system identification toolbox.Verification of the proposed model is achieved using experimental results for the two coils.The sensitivity of the two coils is investigated based on the induced output voltage.The application concerning the two coils for monitoring the discharge current of the surge arrester is done.The results confirm the accuracy of the introduced RC model,as well as the performance of the cylindrical shape,is better than the traditional one.The simulation is carried out using MATLAB and ATPDraw programs.

变压器绝缘缺陷放电发展特性及检测方法研究现状及展望

变压器爆燃事故的发生大多源于其内部绝缘缺陷的局部放电,并逐渐发展至火花放电和电弧放电,造成巨大的经济损失。研究内部放电信号特征及其快速检测方法对保障电力变压器安全可靠运行具有重要意义。国内外学者围绕变压器内部绝缘缺陷放电特性及检测方法开展了大量研究工作。为此从局部放电、火花放电和电弧放电3个阶段综述了相关研究成果,对不同放电发展阶段的脉冲电流、电磁波、声、光以及化学产物等理化特性进行了总结分析,同时概述了基于不同信号的检测方法和对检测设备的要求,并对比了不同检测方法的优缺点,最后提出了对于变压器绝缘缺陷放电发展检测的研究展望。该文对进一步深入研究油中放电信号特征提取、快速检测以及变压器主动保护技术具有指导意义。

电力电缆局放在线监测神经网络自动识别精度的提升方法

为了尽可能降低电缆线路局放在线监测系统装置运行中对局放信号识别的误判误报的可能性,对局放的判读机理和识别精度进行了改进和提升。基于现场运行系统回收的数万个疑似电缆内部局放的告警记录数据进行了分析分类,对判别程序中沿用的局放自动逻辑判别程序追加了相间信号相关性过滤程序,对既有的神经网络在构造上作了优选优化,对神经网络学习数据的前置处理进行了改进。对改进提升后的局放判别程序采用了包括现场记录,模拟信号发生器和人工局放模型加电压实验的3种局放数据进行检验。结果表明,改进提升后的局放判别程序不但能够使现场的非局放告警数据的告警率降低到5%的期待水平,同时也提高了识别人工模拟产生的各种类型局放信号的准确性,大大降低了局放在线监测系统误判误报的可能性。

独立充放时序SIDO开关变换器电流型变频控制技术

单电感双输出(single-inductor dual-output,SIDO)开关变换器工作在共享充放时序下存在电感电流纹波大、输出支路间交叉影响严重以及电路参数宽范围变化下控制电路不能正常工作等问题.为此,提出一种独立充放时序电流型变频控制(current-mode variable frequency control,C-VF)技术.首先,具体描述变换器在连续导电模式(continuous conduction mode,CCM)下的工作原理,并推导主电路开环传递函数;进一步构建闭环小信号模型,推导闭环交叉阻抗,详细分析不同输出电压及负载电流下变换器的交叉影响特性;最后,通过仿真和实验进行验证.研究表明:相较于共享充放时序,独立充放时序C-VF CCM SIDO buck变换器减小了交叉影响,改善了负载瞬态响应性能;当两支路负载电压不等时,减轻某一支路负载可以降低该支路的交叉影响;当两支路输出电压相同但负载不同时,重载支路对轻载支路的交叉影响更小.

交流电压下油纸绝缘放电发展过程中电-磁-声特征参量的变化规律

沿面放电是变压器油纸绝缘结构最严重的放电故障类型之一。为了准确获取交流电压下油纸绝缘沿面放电发展过程中特征参量的变化规律,搭建了沿面放电发展过程模拟试验平台并建立了1套能够满足对放电发展过程中电、磁、声这3种物理量同步在线监测需求的测量系统。在此基础上,提取沿面放电发展过程中多物理参量的典型特征,通过对特征参量全过程幅值变化规律的分析对放电过程进行阶段划分,并给出相应的阶段特征及机理概述。试验结果表明,通过分析多参量典型特征及变化规律可以将沿面放电发展过程划分为4个阶段:微小能量放电阶段、低能量放电阶段、高能量放电阶段、临近击穿阶段,阶段特征与电荷分布及绝缘材料耐受强度有关。相比常规单参量监测手段,多参量同步检测方法为变压器沿面故障诊断和及时预警提供了更可靠准确的参考依据。

基于变分模态分解和稀疏表示的局部放电信号去噪算法

鉴于局部放电信号受各种噪声的干扰,文章提出一种基于变分模态分解和稀疏分解的局部放电信号去噪算法。以稀疏表示算法为核心,基于局部放电信号的特性构建其过完备字典,再采用匹配追踪算法在过完备字典中搜索出原信号的最佳匹配原子集合重构信号;为解决过完备字典维度过高而导致的搜索次数太多的问题,引进变分模态分解算法和峭度值筛选进行预处理和预重构;优化后的方法可以限制稀疏分解算法的搜索范围和字典参数,以减小计算复杂度。仿真验证以及对工程环境中实测信号的去噪结果表明:该方法具有更好的降噪效果,即使在极低信噪比的情况下,依旧能提取出有效的局部放电信号。

基于光纤传感检测的局部放电信号去噪方法

针对局部放电超声信号检测过程中存在噪声干扰的问题,提出一种基于变分模态分解和改进小波去噪的噪声抑制方法。首先对局部放电信号进行变分模态分解,得到若干频带分布不同的本征模态分量,结合模态的峭度剔除窄带干扰主导的分量;然后利用改进的小波去噪法,将其余有效模态分量中残留的白噪声进一步滤除;最后将剩余分量进行信号重构实现噪声抑制。对仿真局部放电信号去噪的结果表明,所提方法可提升信噪比约12.1 dB,且能够更加完整地保留原始信号波形。此外,在实验室环境下搭建了10 kV电压等级的局部放电模拟系统,采用Sagnac光纤传感检测局部放电超声信号。实验结果表明,所提方法具有良好的去噪性能,对实测信号的噪声抑制比达到了7.504。

基于改进小波阈值—CEEMDAN的变压器局部放电超声波信号白噪声抑制方法

为了有效去除变压器局部放电超声信号中的白噪声干扰,提高后续局部放电模式识别及定位的准确性,提出了一种基于改进小波阈值和自适应噪声完全集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)的变压器局部放电超声波信号去噪方法。首先,通过对放电信号进行CEEMDAN分解得到一系列由高频到低频的本征模函数IMF(intrinsic mode function);然后,利用多尺度排列熵(multi-scale permutation entropy,MPE)算法计算各阶IMF分量的排列熵PE(permutation entropy),根据各IMF的排列熵值确定信号的去噪阈值与有效阈值。对高于去噪阈值的IMF分量采用改进小波阈值法进行去噪处理,对低于有效阈值的IMF分量视为基线漂移进行剔除。最后,通过重构去噪分量与剩余分量来获得去噪后的超声波信号。仿真和实验结果均表明,文中所提出的去噪算法大大提高了信号的信噪比,并保留了原始超声波信号中的有效信息,对提高后续利用超声波信号进行局部放电模式识别及定位的精确性具有重要意义。

基于机-热-电多参量信号特征的GIS隔离开关状态评估方法

气体绝缘开关(gas insulated switchgear,GIS)设备中的隔离开关故障往往是机械、热、电气多物理量综合作用的结果,从微小缺陷发展至严重故障的过程中,会产生异常振动、温度、局部放电多种参量信号,同时测量并分析GIS隔离开关的多参量信息对于准确获取其运行状态具有重要意义。文中通过在实体220 kV GIS实验平台上模拟隔离开关最为常见的接触不良缺陷,研究不同接触状态时振动信号、温度变化和局部放电信号的演化规律,获得其与设备状态之间的对应关系,结合健康指数理论和改进层次分析法,提出基于多参量特征的GIS隔离开关综合评估方法。使用该方法开展现场测试,其检查结果与多参量综合评估方法的的分析结果相吻合。文中的研究为GIS隔离开关状态的带电检测提供了一种可行性较高的多参量融合分析新方法。

电子辐照下聚酰亚胺薄膜材料的表面充放电特性研究

地球同步轨道环境下电子辐照会使航天绝缘材料产生电荷积聚,可能引起介质静电放电,而静电放电也会影响电荷积聚情况。本文在电子辐照下对聚酰亚胺薄膜材料进行电位分布和静电放电电流联合测量实验,研究了地球同步轨道恶劣环境中电子辐照下聚酰亚胺薄膜材料的表面充放电过程。结果表明:15 keV、0.33 nA/cm^(2)的电子环境下聚酰亚胺薄膜表面电位呈负极性,电位分布模式包括单峰型、多峰型、火山口型和平顶型,其中单峰型和多峰型分布模式的放电次数最多。短时多次放电会转变电位分布模式,并使转变后的电位分布模式在长时间内保持。电位分布模式会影响放电电流信号的波前时间,而电位水平会影响放电电流信号的半峰值时间,低电位水平下的电流信号半峰值时间比高电位水平下大约高28%。