Wavelet-based neural network for power disturbance recognition and classification

In this paper, a prototype wavelet-based neural-network classifier for recognizing power-quality disturbances is implemented and tested under various transient events. The discrete wavelet transform (DWT) technique is integrated with the probabilistic neural-network (PNN) model to construct the classifier. First, the multiresolution-analysis technique of DWT and the Parseval's theorem are employed to extract the energy distribution features of the distorted signal at different resolution levels. Then, the PNN classifies these extracted features to identify the disturbance type according to the transient duration and the energy features. Since the proposed methodology can reduce a great quantity of the distorted signal features without losing its original property, less memory space and computing time are required. Various transient events tested, such as momentary interruption, capacitor switching, voltage sag/swell, harmonic distortion, and flicker show that the classifier can detect and classify different power disturbance types efficiently.     

电能质量扰动在线辨识装置

设计了一种具有电能质量(PQ)扰动实时在线检测与分类功能的电能质量分析装置.该装置基于DSP和FPGA平台,实现了信号的采集、处理和显示.在算法上由于人工神经网络、专家系统、模糊逻辑、支持向量机等分类器过于复杂,故采用一种简单、高效的PO扰动分类和量化方法,即基于规则基的决策树RBDT(Rule-Based Decision Tree)模式识别方法,同时提取5个典型的PQ扰动时频特征量作为决策树的输入量,实现了9种典型PQ扰动的辨识.通过算法仿真及硬件平台验证,结果表明可以满足对PQ扰动分类的精度和实时性的要求.     

一种新的电能质量扰动信号压缩感知识别方法

针对现有电能质量扰动信号识别方法存在数据量大、准确率不高的不足,提出了一种基于压缩感知稀疏向量特征提取的电能质量扰动信号分类识别方法。该方法首先针对原始信号,利用压缩感知理论获取降维的测量信号,并基于?1范数正交匹配追踪算法获取稀疏向量。然后针对稀疏向量提取最大值、次大值、均方根、标准差、峭度和裕度因子等特征,作为神经网络的输入,实现电能质量扰动信号的分类识别。最后,针对六类典型电能质量扰动信号,开展仿真实验验证。仿真结果表明,现有识别方法需要处理的原始信号长度为1024,而所提方法特征提取时所处理的数据长度仅有30,从而大大减少了所需处理的数据量,并且由于实现了以非常少的数据量保存原有全部有用特征信息,因而更有利于提高识别准确率。通过与广泛采用的小波变换识别方法进行比较,所提方法的平均准确率高达98.71%,远远高于小波变换方法的92.86%。     

单一电能质量扰动的分类识别研究

电能质量扰动的分类识别对电能质量综合治理具有重要意义,为此提出了一种基于粒子群优化极限学习机的电能质量扰动分类新方法。利用小波变换将扰动信号做10层分解,提取有效区分扰动信号类型层数的能量差、能量差平均值及能量差的标准差作为特征向量,并将扰动信号与正常信号的均方根作为补充,减少输入向量维度。提出采用极限学习机训练误差作为粒子群的适应度函数来优化隐含层神经元个数,在提升分类速度的基础上保持较高的分类精度。经仿真验证表明,该方法能够准确有效地识别常见的7种扰动类型,相比于传统的BP神经网络具有较高的分类速度。     

基于Hoeffding Tree的电能质量在线扰动分类

为满足电能质量扰动事件的在线分类需求,提出了一种基于Hoeffding Tree的电能质量扰动在线分类方法。对电能质量在线扰动分类中的关键技术进行了研究,提出用小波变换和离散傅里叶变换相结合的判别方法检测电能质量扰动,该算法采用自适应滑动数据窗算法,能够根据扰动持续时间提取完整的扰动事件。以小波信号能量以及基波有效值构成特征向量,利用Hoeffding Tree算法构建增量式分类训练模型。仿真结果表明,所提方法的准确度和效率均满足电能质量扰动事件在线检测和分类的要求。     

Pattern analysis and classification for security evaluation in power networks

Security evaluation is a major concern in real time operation of electric power networks, exhibiting behavioral patterns under abnormal conditions. Security assessment and evaluation can be viewed as a pattern analysis task identifying abnormal patterns of the power system behavior under highly loaded conditions. Traditional method of security evaluation are highly time consuming and infeasible for direct on-line implementation. This paper presents application of pattern directed inference system for static and transient security evaluation and classification. A straightforward and quick procedure called Sequential Forward Selection method is used for feature selection process. The classifier model in the pattern directed inference system is designed using different pattern classifier algorithms, viz., conventional, neural network and machine learning classifiers. Support Vector Machine (SVM), one of the popular machine learning classifier, is recognized as a suitable pattern classifier for security evaluation problem. The generalization performance of SVM classifier is greatly influenced by the proper setting of its parameters. This paper also addresses different heuristic optimization techniques used in the selection of SVM parameters. The design, development and performance of different classifiers for power system security classification are presented in detail. Simulation work is performed on standard New England 39-bus benchmark system and the feasibility of implementation of the proposed SVM based classifier system for on-line security evaluation is also discussed.     

基于改进GSA-SVM算法的电能质量扰动分类方法

针对不同类型电能质量扰动信号分类准确率不高的问题,通过MATLAB/simulink搭建常见的9种不同的电能质量扰动信号的模型进行仿真分析,提出一种改进的万有引力搜索算法(improved gravitational search algorithm, IGSA)对支持向量机(support vector machine, SVM)的惩罚因子和核函数参数进行寻优的方法,通过优化SVM的惩罚因子和核函数参数,构建IGSA-SVM分类器,再把提取到的特征向量进行归一化之后输入到所构造好IGSA-SVM分类器中进行训练与分类。仿真结果表明,IGSA-SVM分类器的分类准确率比SVM和GSA-SVM这2种分类器都要好,可以实现对9种不同的电能质量扰动信号的快速准确分类,有利于解决实际的工程问题。     

基于优化DDAGSVM多类分类策略的电能质量扰动识别

针对电能质量扰动类型多样且识别率不高的问题,该研究的目的是如何将多类分类问题应用于支持向量机。首先通过S变换和FFT变换提取扰动信号特征量进行模型训练。其次将广义KKT判定条件与样本空间分布序列相结合引入类间识别度,将类间识别度最高的超平面函数作为分类器根节点,以此克服传统决策导向非循环图支持向量机分类器(DDAGSVM)在分类生成顺序上随机化的缺点,并将改进的DDAGSVM应用于电能扰动信号的识别分类。实验结果表明,所提算法较传统DDAGSVM算法有良好效果和更好的鲁棒性。     

An effective wavelet-based feature extraction method for classification of power quality disturbance signals

This paper presents a wavelet norm entropy-based effective feature extraction method for power quality (PQ) disturbance classification problem. The disturbance classification schema is performed with wavelet-neural network (WNN). It performs a feature extraction and a classification algorithm composed of a wavelet feature extractor based on norm entropy and a classifier based on a multi-layer perceptron. The PQ signals used in this study are seven types. The performance of this classifier is evaluated by using total 2800 PQ disturbance signals which are generated the based model. The classification performance of different wavelet family for the proposed algorithm is tested. Sensitivity of WNN under different noise conditions which are different levels of noises with the signal to noise ratio is investigated. The rate of average correct classification is about 92.5% for the different PQ disturbance signals under noise conditions.     

基于并行隐马尔科夫模型的电能质量扰动事件分类

为满足电能质量扰动准确分类的需求,提出了一种基于极大重叠离散小波变换(MaximalOverlapDiscrete WaveletTransform, MODWT)和并行隐马尔科夫模型(ParallelHiddenMarkovModel, PHMM)的电能质量扰动分类方法。首先利用MODWT提出一种实用的电能质量扰动检测算法,该算法无需设定检测阈值,可准确获取扰动时段的起止时刻。接着提取扰动时段的电压谐波成分并组成特征向量。然后用PHMM分类器对扰动信号进行分类识别。PHMM方法克服了人工神经网络方法收敛性较差、训练时间较长的缺陷,使分类器性能大大提升。通过应用于现场实测扰动数据表明,所提出的方法适用于多种类型的电能质量扰动检测,分类正确率高,训练速度快,具有良好的应用价值。     

Optically powered data link for power system applications

The practicability of an optically powered data link has been demonstrated. The link has moderate bandwidth (1 kHz), accuracy (1%) and dynamic range (>60 dB) over a useful range of ambient temperatures. The link uses commercially available components, including a photodiode array fabricated using the dielectric isolation process in silicon. An application to the measurement of current in a high voltage line by means of a linear coupler is described, and experimental results are presented. Power transmission efficiency is presently low, at about 0.3% overall (electrical-to-electrical) and 5% optical-to-electrical     

A signal processing module for power system applications

This paper presents a unified power signal processor (PSP) for use in various applications in power systems. The introduced PSP is capable of providing a large number of signals and pieces of information which are frequently required for control, protection, status evaluation, and power quality monitoring of power systems. The PSP receives a set of locally measured three-phase voltage and current signals and provides their fundamental components, amplitudes, phase angles, frequency, harmonics, instantaneous and stationary symmetrical components, active and reactive currents and powers, power factor, and the total harmonic distortion. Simplicity and integrity of its structure as well as its robustness with respect to internal parameters and external disturbances and noise render the proposed scheme very attractive for practical implementations.     

On distributed computing for on-line power system applications

This paper reviews the state-of-the-art of distributed computing (also called coarse-grained parallel computing) technology, which has rapidly evolved over the last two decades, with emphasis on the trend towards standardization that has occurred in the last few years. The review is focused on on-line power system applications, and excludes fine-grained parallel applications and planning applications. The applications are divided into two categories. The first category consists of applications where the motivation for distributed processing stems from geographical distribution. The second category is the rest of the on-line applications where the parallelism stems from the easily decomposable abstract (mathematical) model of the problem as opposed to being ‘physically based’. Some of the issues in such distributed computing are illustrated using the example of security-constrained optimal power flows. The paper concludes with some projections on the use of this technology in energy management systems (EMS) in the near future.     

永磁直驱风力发电系统最大功率跟踪非线性抗扰控制

针对永磁同步发电机的非线性、内部参数不确定以及外部扰动等问题,提出了一种直驱式永磁同步风力发电系统最大功率跟踪的非线性抗扰控制方法。该方法使用一种非线性光滑函数来设计非线性扩张状态观测器(NLESO)和非线性抗扰控制律。由NLESO来实现系统扰动及不确定性的估计,前馈到控制输入端对扰动进行补偿,从而有效提高了系统的抗扰能力。分析了NLESO的收敛性。仿真结果表明了该控制方法不仅具有响应速度快、控制精度高的特点,而且无超调无抖振现象,因而在风力发电系统最大功率跟踪控制领域具有较大应用价值。     

On-line detection of power system small disturbance voltageinstability

Small disturbance (SD) voltage stability (or instability) deals with a system's ability to maintain satisfactory voltages following a small disturbance. For an operating condition, a system's SD voltage stability depends on the proximity of the condition to the critical point (or voltage collapse point). A Q angle and Q directional derivatives are proposed for SD voltage instability detection and weak bus identification, respectively. The Q angle index can handle different kinds of loads, e.g., constant P and Q, constant impedance, and constant current, or a combination of them, and is effective in dealing with generator VAr limits. Moreover, the computation speed of the Q angle is fast, which makes it suitable for on-line application. Simulation results using two power systems are provided     

电力系统小干扰稳定分析的最小特征轨迹法

提出了一种基于Nyquist特征轨迹的多输入-多输出系统小干扰稳定分析方法,可以量化原动机-调速器回路及附加励磁控制回路对稳定裕度的贡献。首先通过框图变换得到了一种便于分析的简洁形式的Heffron-Phillips模型,然后根据广义Nyquist曲线和特征轨迹的关系定义了新的稳定裕度。基于前述模型推导了稳定裕度的线性解析表达式,该线性表达式包含了原动机-调速器回路以及附加励磁控制回路,从而能够清晰地看出二者对稳定裕度的影响。除此以外,所提方法还可以作为控制器参数的设置。最后通过多个算例证明了所提方法的准确性和有效性。     

Expert system for classification and analysis of power systemevents

This paper presents an expert system that is able to classify different types of power system events to the underlying causes (i.e., events) and offer useful information in terms of power quality. The expert system uses the voltage waveforms and distinguishes the different types of voltage dips (fault-induced, transformer saturation, induction motor starting) as well as interruptions (nonfault, fault-induced). A method for event-based classification is used, where a segmentation algorithm is first applied to divide waveforms into several possible events. The expert system is tested using real measurements and the results show that the system enables fast and accurate analysis of data from power quality monitors     

基于S变换和分类树的电网暂态电能质量扰动分类辨识

针对暂态电能质量的检测分析,分别在强弱两种噪声背景下运用S变换的不同方法对暂态多扰动信号进行定位检测.对于暂态多扰动的分类辨识,运用了基于S变换和分类树相结合的暂态电能质量多扰动分类辨识方法,首先运用S变换对暂态多扰动信号进行时频分析,然后提取扰动信号的特征量,最后生成用于对暂态多扰动信号进行分类的决策树分类辨识方法,以此来实现对暂态多扰动信号的分类辨识.仿真计算结果表明,该方法对暂态多扰动信号能够进行有效的分类辨识,准确度高且抗噪能力强.     

基于S变换和分类树的电网暂态电能质量扰动分类辨识

针对暂态电能质量的检测分析,分别在强弱两种噪声背景下运用S变换的不同方法对暂态多扰动信号进行定位检测.对于暂态多扰动的分类辨识,运用了基于S变换和分类树相结合的暂态电能质量多扰动分类辨识方法,首先运用S变换对暂态多扰动信号进行时频分析,然后提取扰动信号的特征量,最后生成用于对暂态多扰动信号进行分类的决策树分类辨识方法,...     

基于改进多层前馈神经网络的电能质量扰动分类

电能质量扰动分类是电能质量控制的重要工作之一,主要工作包括信号特征提取和分类器构造两个阶段。采用S变换与改进的多层前馈神经网络相结合,提出一种新的电能质量扰动分类方法。首先利用S变换对原始数据进行处理,提取具有代表性的4类典型特征以表征不同种类的扰动类型的特性,之后使用拟牛顿法和自适应因子改进传统的多层前馈神经网络,将特征作为改进的多层前馈神经网络的输入向量,实现自动的分类识别。实验表明,新方法减少了噪声对分类准确率的影响,学习能力强,能够有效的识别电压暂降、电压瞬升、电压中断、暂态震荡、谐波等5种电能扰动。