基于形态滤波和Hilbert变换的电压暂降检测研究

电压暂降已成为电能质量暂态问题中最为突出的一类.针对电压暂降检测中具有的准确性低、实时性较差的缺点,提出了一种基于Hilbert变化与形态滤波相结合的暂降检测方法.研究论述了Hilbert变化用于电压暂降检测的算法基本原理,介绍了数学形态学概念,并由此基础设计了形态学滤波器,同时阐述了滤波器参数的选取方法.仿真论证了该方法在单相接地故障引起的电压暂降中的检测效果,并与单相dq变换法进行了比较.从仿真结果可以看出该方法可以更高效精确地检测到暂降的幅值以及相位.     

基于概率神经网络的高压断路器故障诊断

高压断路器是最重要的电力设备之一,在电力系统中起控制和保护作用。为了提高高压断路器故障诊断的准确率,提出了一种基于概率神经网络(PNN)的高压断路器故障诊断方法。该方法在分析高压断路器的故障特性来确定特征信号的基础上建立了PNN故障诊断模型,该模型将采集的特征数据作为网络的输入,通过Parzen窗估计法得到类条件概率密度,进而按Bayes决策规则对特征数据进行分类。经仿真表明,概率神经网络故障诊断模型具有收敛速度快、故障诊断准确率高、容易训练等特点。因此,该方法是一种有效的故障诊断方法,具有良好的应用前景。     

基于Hilbert变换与Pisarenko谐波分解的电压闪变参数估计

提出了新的闪变参数估计方法,该方法在Hilbert变换提取闪变包络线的基础上,用Pisarenko谐波分解估计各调制分量的频率和幅值.在估计过程中,用FFT分析得到Pisarenko谐波分解所需的调制分量个数,解决了闪变调制分量个数不确定的问题.在设定噪声条件下,用所提方法分别对单一频率和多频率调制的闪变进行仿真,并将所得估计值与FFT频谱分析方法所得结果进行了比较,验证了所提方法可以更准确得到低频调制信号幅值和频率的优点.     

基于Hilbert变换与Pisarenko谐波分解的电压闪变参数估计

提出了新的闪变参数估计方法,该方法在Hilbert变换提取闪变包络线的基础上,用Pisarenko谐波分解估计各调制分量的频率和幅值。在估计过程中,用FFT分析得到Pisarenko谐波分解所需的调制分量个数,解决了闪变调制分量个数不确定的问题。在设定噪声条件下,用所提方法分别对单一频率和多频率调制的闪变进行仿真,并将所得估计值与FFT频谱分析方法所得结果进行了比较,验证了所提方法可以更准确得到低频调制信号幅值和频率的优点。     

径流预报的极点对称模态分解-Elman网络模型

针对径流序列非线性、非平稳的特点,将极点对称模态分解(ESMD)方法与Elman神经网络模型相结合,建立了ESMD-Elman神经网络组合模型,并应用于长江上游干支流8站的年、月径流预报.首先利用ESMD方法将径流序列分解为各模态分量和趋势余项;然后利用Elman神经网络模型分别预测各平稳序列;最后加和重构得到最终预测...     

基于正序电流故障分量相位比较的电压暂降扰动源分界

准确判断电压暂降扰动源的相对位置,对界定供用电双方责任以及制定治理措施提高供电质量具有重要意义。文中提出了基于正序电流故障分量相位比较原理的配电网电压暂降扰动源分界方法。通过建立不同位置发生不同类型故障时的正序故障分量等值网络,分析变电站内所有进出线的正序电流故障分量相位分布与故障位置间的关联关系,进而构建基于正序电流故障分量相位比较的电压暂降扰动源分界判据及其动作边界。该方法进行电压暂降扰动源分界时仅利用站端进出线的电流信息,测量信息获取方便,具有很好的工程应用前景。基于PSCAD建模仿真,验证了短路故障、相位跳变以及负荷扰动情况下,文中所提电压暂降扰动源分界方法效果较佳。     

基于改进S变换-TT变换的电压暂降特性研究

针对S变换的窗函数固定,以及在检测扰动信号的起止时刻、幅值变化、相位变化时的检测精度不高的问题,在S变换中引入两个调谐因子,得到一种改进的S变换,并对改进的S变换进行傅里叶反变换得到TT变换,将改进的S变换与TT变换相结合应用到主要引起电压暂降的3种暂降源中,并在MATLAB软件上对这3种暂降扰动源信号进行了仿真分析。仿真结果表明,所提的方法比使用S变换方法在检测电压暂降信号上具有更高的检测精度和更好的效果,为更好地提取暂降特征信息以及暂降源的识别打下基础。     

Hilbert变换在电压闪变检测中的应用

利用Hilbert变换(HT)对配电系统中的电压闪变水平进行检测,并用等纹波滤波器来实现HT。与文献中所用方法相比,该文所用方法在数学上更简单,实现更容易。HT能够检测配电系统中的电压闪变和系统频率的变化,精度高,使电压闪变补偿装置更容易控制。采用了不同的电压闪变信号验证该方法。通过Mat-lab仿真,简述了影响检测精度的不同因素。仿真结果验证了HT的检测能力,表明HT在检测电压闪变方面具有良好的性能.     

Hilbert变换在电压闪变检测中的应用

利用Hilbert变换(HT)对配电系统中的电压闪变水平进行检测,并用等纹波滤波器来实现HT.与文献中所用方法相比,该文所用方法在数学上更简单,实现更容易.HT能够检测配电系统中的电压闪变和系统频率的变化,精度高,使电压闪变补偿装置更容易控制.采用了不同的电压闪变信号验证该方法.通过Matlab仿真,简述了影响检测精度的不同因素.仿真结果验证了HT的检测能力,表明HT在检测电压闪变方面具有良好的性能.     

基于小波和神经网络的电能质量辨识方法

本文提出一种新的基于小波和神经网络技术的电能质量辨识方法。对各种电能质量信号进行时域、频域和幅值分析,并从小波变换结果中提取与信号时域、频域和幅值量相关的几个特征量来表征不同电能质量信号。将这些特征量作为神经网络的输入可以实现电能质量的辨识。计算结果表明该方法的有效性和准确性。     

Support vector classification and regression for fault location in distribution system using voltage sag profile

Fault location identification is an important task to provide reliable service to the customer. Most existing artificial intelligence methods such as neural network, fuzzy logic, and support vector machine (SVM) focus on identifying the fault type, section, and distance separately. Furthermore, studies on fault type identification are focused on overhead transmission systems and not on underground distribution systems. In this paper, a fault location method in the distribution system is proposed using SVM, addressing the limitations of existing methods. Support vector classification (SVC) and regression analysis are performed to locate the fault. The method uses the voltage sag data during a fault measured at the primary substation. The type of fault is identified using SVC. The fault resistance and the voltage sag for the estimated fault resistance are identified using support vector regression (SVR) analysis. The possible faulty sections are identified from the estimated voltage sag data and ranked using the Euclidean distance approach. The proposed method identifies the fault distance using SVR analysis. The performance of the proposed method is analyzed using Malaysian distribution system of 40 buses. Test results show that the proposed method gives reliable fault location.     

基于二维变分模态分解和Hilbert变换的局放信号特征提取方法

本文通过对二维Hilbert-Huang变换方法的改进,提出了一种基于二维变分模态分解(VMD)和Hilbert变换的局部放电灰度图像特征提取方法。首先,利用局部放电样本生成相应放电灰度图;其次,以二维VMD算法分解各放电灰度图像,获取各个不同中心频率的模态分量;然后,通过四元数Hilbert变换得到各模态函数对应的特征图,并提取灰度纹理特征,构成各放电样本对应的特征向量;最后,以BP神经网络分类器对提取出的局部放电特征量进行分类和识别。实验结果验证表明,同二维Hilbert-Huang变换和传统放电灰度图特征提取方法相比,基于本文方法所得特征量具有更高的正确识别率,验证了该方法的可行性。另外,本文所采用的二维VMD-Hilbert方法为局部放电信号的频谱分析拓展了新的思路。     

Detection and classification of voltage disturbances using a Fuzzy-ARTMAP-wavelet network

This paper presents a method for automatic detection and classification of voltage disturbances for problems related to power quality using signal processing techniques and intelligent systems. This support tool for decision making is composed of four modules. The first module continuously evaluates the system's operation state. The second module extracts the essential features from the three-phase voltage signal based on the discrete wavelet transform, multiresolution analysis and entropy norm concepts. The signal signature is processed via standardization and codification in the third module. The fourth module classifies the type of disorder using a Fuzzy-ARTMAP neural network. A total of 7023 power quality events, including voltage swell, voltage sag, outage, harmonics, swell with harmonics, sag with harmonics, oscillatory transient and flicker, were obtained through mathematical models and simulations using the ATP software. To demonstrate the performance of this method, an application is submitted considering a real electric energy distribution system composed of 134 buses with measurements performed on a 13.8 kV and 7.065 MVA feeder. The results indicate that the proposed method is efficient, robust and has high computing performance (low processing time), which allows, a priori, its application in real time.     

短时傅里叶变换和S变换用于检测电压暂降的对比研究

针对电压暂降的特征值检测问题,介绍了短时傅里叶变换(STFT)和S变换两种时频分析法,并对两种方法进行了对比分析。在STFT变换中选取不同窗宽的窗函数对同一电压暂降信号进行时频分解,分析不同窗宽对检测结果的影响。提出用时频等值曲线定位暂降的起、始时刻,用基频幅值曲线检测暂降幅值,用相位跳变曲线判定暂降发生时相位是否跳变。仿真结果表明,在STFT变换中窗口越小,检测结果越准确;与STFT变换相比,S变换的检测结果更准确,并且抗噪声能力强,有助于电能质量的治理。     

Wavelet based feature extraction of voltage profile for online voltage stability assessment using RBF neural network

Online voltage stability assessment is one of the vital requirements for intricate electric power systems. Due to the restructuring and liberalization, modern power systems tend to operate close to their stability limits with small security margin. In such environment, online voltage stability evaluation plays a significant role in secure operation of power systems. This paper presents a new approach for estimating voltage stability margin VSM, based on application of wavelet feature extraction method to network voltage profile. Voltage profile is adopted as the original input data for VSM estimation, because it contains sufficient information concerning network topology, load level, load-generation patterns and all system controllers. In this approach, in order to provide high discrimination between network voltage profiles, Multi-Resolution Wavelet Transform (MRWT) is utilized to extract the features of voltage profiles. Also, in order to eliminate the redundant features, principle component analysis (PCA) is used to select the most relevant features extracted by MRWT. Radial Basis Function (RBF) neural network is adopted to estimate system VSM using the dominant extracted features of the voltage profile by MRWT and PCA. Using voltage profile as the original data makes the proposed approach capable of estimating system VSM in both static and quasi dynamic conditions. The proposed approach has been implemented in New England 39-bus test system with promising results demonstrating its effectiveness and applicability.     

基于RBF神经网络的电压外环滑模控制的Vienna整流器

以Vienna整流器为研究对象,针对其传统电压外环滑模变结构控制不变性和对系统参数扰动敏感的问题,分析了以逼近率为基础的滑模变结构控制算法,提出了一种基于RBF神经网络的自适应电压外环滑模控制算法。该控制算法通过将RBF神经网络与滑模控制算法有效结合,同时将中点电位平衡控制加入到RBF神经网络自适应电压外环滑模控制算法的设计中,使用RBF神经网络对电压外环非线性系统进行自适应逼近,能够有效降低切换增益,削弱抖振,增强系统的抗干扰能力。最后,通过仿真分析与实验测试验证所提控制算法的有效性。将所提出的控制算法与传统滑模控制算法、PI控制算法进行比较,结果表明采用这种电压外环控制算法能够对直流输出电压目标值进行快速跟踪,平衡中点电位,改善了系统的动静态性能,提升了其抗干扰能力。     

基于卷积神经网络的轮胎缺陷X光图像分类

轮胎缺陷的类型直接决定着轮胎是否为残次品或废品,对于轮胎定级具有重要参考价值,探索高性能的轮胎缺陷分类方法至关重要.采用卷积神经网络技术,提出一个端到端的图像自动分类算法.首先,从国内某轮胎生产线上通过现场运行的轮胎X光射线缺陷检测系统采集胎侧异物缺陷、胎冠异物缺陷、气泡缺陷、胎冠劈缝、胎侧开根5种最常见缺陷类型和1种正常胎侧图像作为分类目标,并且依据缺陷图像的缺陷尺度,将每幅图像缩放到127×127像素的统一大小;然后,设计含有5个卷积层、3个池化层、3个全连接层的网络结构.最后,用采集的缺陷样本对所设计的深度网络进行训练学习与测试.并将该算法和大量传统分类算法进行实验对比,取得更好的分类效果,平均测试识别率高达96.51％.     

A neural network for signal decomposition problems

This paper presents the design of a neural network for signal decomposition problems with application examples. For this class of problems the proposed network has the same dynamics as the Hopfield net, but it is shown to realize the O(M²) connection paths among the M neurons with a number of wires and conductances increasing only linearly with increasing M, i.e. reducing this number by one dimension with respect to the quadratically increasing number of wires and conductances required in the Hopfield net. Other advantages of the proposed neural network are discussed in relation to classical examples of decomposition problems. In particular, a new architecture for an N-bit A/D converter is presented employing 4N conductances instead of the N(N + 1) Hopfield A/D conductances.     

面向配电网电压暂降治理的线路改造优化模型

线路改造是在电网侧进行电压暂降治理的主要方式之一。为了经济、合理地确定线路的改造位置和治理方式,建立了面向配电网电压暂降治理的线路改造优化模型。针对断路器重合闸和熔断器配合保护的配电网,考虑故障位置、故障类型等因素,讨论分析了10种典型的保护动作配合方式,评估了配电网电压暂降持续时间。结合传统电压暂降幅值的计算方法,计算线路改造治理方案的成本及效果。考虑用户电压暂降耐受能力,建立了用户生产中断次数评估模型。定义了电压暂降总支出,以总支出最小为目标,计及技术约束、经济约束、治理范围约束和风险域约束,建立了面向配电网电压暂降治理的线路改造优化模型,并基于人工蜂群算法进行模型求解。以IEEE 33节点系统为算例,考虑治理投资成本充足和不足2种场景,通过仿真验证了所提方法的正确性和实用性。     

Fast voltage contingency screening using radial basis function neural network

Power system security is one of the vital concerns in competitive electricity markets due to the delineation of the system controller and the generation owner. This paper presents an approach based on radial basis function neural network (RBFN) to rank the contingencies expected to cause steady state bus voltage violations. Euclidean distance-based clustering technique has been employed to select the number of hidden (RBF) units and unit centers for the RBF neural network. A feature selection technique based on the class separability index and correlation coefficient has been employed to identify the inputs for the RBF network. The effectiveness of the proposed approach has been demonstrated on IEEE 30-bus system and a practical 75-bus Indian system for voltage contingency screening/ranking at different loading conditions.