基于神经网络的距离保护训练算法的研究

对神经网络式距离保护做了简要概述,并且为了使该新型保护更加完善,对其存在的训练算法问题做了深入的研究.对8种改进的BP(Back-Propagation)算法训练神经网络进行了比较,通过反复训练,得出8种算法在权值调整和误差收敛过程中所遇到的共同点和差别,并得出初始权值、激励函数和神经元个数等因素的选取对神经网络学习的影响.实验证明:LM算法和尺度共轭梯度反向传播算法训练网络收敛速度快,受权值影响小,但占用较大的内存;贝叶斯正则化算法有较好的泛化能力.     

Fault impedance estimation algorithm for digital distance relaying

Fault impedance is one of the major parameters that must be estimated accurately in digital distance relaying application. In this paper, a new algorithm is proposed based on symmetrical components theory. The proposed algorithm has computational advantage over previously suggested symmetrical components based algorithms. A procedure for applying shunt fault conditions to the sequence equations to estimate fault impedance of the protected transmission line is discussed. The Alternative Transient Program (ATP) that is available on personal computers was used in evaluating the proposed algorithm. ATP models a power system and simulates many fault conditions on a selected transmission line. Fault data obtained were used in calculating fault impedance using the proposed algorithm. Fault impedance estimates were inserted in relay characteristics to determine suitability of the proposed algorithm for first zone distance protection. Sample results of these studies which show stable fault distance estimates are presented and discussed in the paper     

A new distance relaying algorithm based on complex differential equation for symmetrical components

This paper presents a new digital impedance measuring technique for transmission lines that combines symmetrical components and the complex differential equation of an equivalent fault loop circuit. The phase voltages and currents at the relaying point are transformed into symmetrical components using Fourier filters of short window length. Depending on fault type, an appropriate fault loop circuit is formed, signals of which are the appropriate symmetrical components, while a parameter of which is the positive sequence impedance being a geometrical measure of the distance from the relaying point to a fault. The impedance, however, is measured very fast by on-line solving the complex differential equation originated for this fault loop circuit. Consequently, this approach combines frequency domain estimation of symmetrical components (accurate filtration) and time domain measurement of positive sequence impedance (high speed response).

The presented method suits well the protection of parallel lines against high-resistance faults occurring very close to the far end of a line. A new method is proposed for detecting high-resistance faults and deciding which line out of two parallel lines actually suffers a fault.

The included EMTP test results demonstrate the efficiency of the proposed relaying algorithm.     

Numerical distance relaying algorithm based on Mathematical Morphology and Least-Squares Curve Fitting method

A new numerical distance relaying algorithm based on Mathematical Morphology and a Least-Square Curve Fitting has been proposed. It makes use of morphological operators to detect and classify the faults and a Curve Fitting method to estimate the fault distance. To show the effectiveness of the proposed technique, computer simulations have been carried out on a test system, considering several fault types in different locations, loading angle and fault incidence angle. Due to simple mathematics operations used, the protection scheme proposed displays secure and effective high speed fault detection and tripping.     

A novel fuzzy neural network based distance relaying scheme

This paper presents a new approach to distance relaying using fuzzy neural network (FNM). The FNN can be viewed either as a fuzzy system, a neural network or fuzzy neural network. The structure is seen as a neural network for training and a fuzzy viewpoint is utilized to gain insight into the system and to simplify the model. The number of rules is determined by the data itself and therefore a smaller number of rules is produced. The network is trained with the backpropagation algorithm. A pruning strategy is applied to eliminate the redundant rules and fuzzification neurons, consequently a compact structure is achieved. The classification and location tasks are accomplished by using different FNN's. Once the fault type is identified by the FNN classifier the selected fault locating FNN estimates the location of the fault accurately. Normalized peaks of fundamental voltage and current waveforms are considered as inputs to all the networks and an additional input derived from the DC component is fed to fault locating networks. The peaks and DC component are extracted from sampled signals by the EKF. Test results show that the new approach provides robust and accurate classification/location of faults for a variety of power system operating conditions even with resistance in the fault path     

Multi-layer neural networks applied to distance relaying

This paper demonstrates the use of Artificial Neural Networks (ANNs) as pattern classifiers for a distance relay operation of transmission lines. Two different types of ANN architectures, concerning the input data, are taken into account. One approach utilises the first five post-fault samples as inputs. The other one employs the magnitudes of the three-phase voltage and current phasors (including the zero sequence) as inputs. A comparison of how well the schemes performed is carried out. An improvement concerning the use of ANNs for protection purposes is found.     

Application of a distance relaying scheme to compensate fault location errors due to fault resistance

This paper proposes a distance relaying scheme based on the current phase jump behavior during fault conditions to improve the apparent impedance estimated by the distance relay. For a nonpilot protection scheme, the measured impedance is affected by error due to the combined effects of fault resistance and prefault load. An experimental relation between the current phase jump introduced with fault inception and the X/R ratio seen by the distance protection is deduced. The phase jump correction factor obtained is an exponential function of the X/R ratio of the line. This factor is applied to the apparent impedance measured by the relay and it allows mitigating the adverse effect of prefault power. The relaying scheme improves significantly the accuracy in estimation of the resistive fault location. The application of this scheme does not require communication links from the remote end of line and is applicable to all types of fault.     

基于准同步算法的谐波分析方法

介绍了一种采用准同步算法的谐波分析量方法，在不需要采样严格同步的条件下实现高精度的测量。仿真结果表明该方法的可行性及优越性。     

An efficient algorithm to calculate intrinsic thermoelectric parameters based on Landauer approach

The Landauer approach provides a conceptually simple way to calculate the intrinsic thermoelectric (TE) parameters of materials from the ballistic to the diffusive transport regime. The approach relies on the calculation of the number of propagating modes and the mean free path for each mode. The modes are calculated from the energy dispersion (E(k)) of the materials, which require heavy computation and are often restricted to energy relations on sparse momentum (k) grids. Here an efficient method to calculate the distribution of modes (DOM) from a given E(k) relationship is presented. The two main features of this algorithm are: (i) its ability to work on sparse dispersion data, and (ii) creation of an energy grid for the DOM that is almost independent of the dispersion data therefore allowing for efficient and fast calculation of TE parameters. The effect of K-grid sparsity on the compute time for DOM and on the sensitivity of the calculated TE results are provided. The algorithm calculates the TE parameters within 5% accuracy when the K-grid sparsity is increased up to 60% for all the dimensions (3D, 2D and 1D). The time taken for the DOM calculation is strongly influenced by the transverse K density (K perpendicular to transport direction) but is almost independent of the transport K density (along the transport direction). The DOM and TE results from the algorithm are bench-marked with, (i) analytical calculations for parabolic bands, and (ii) realistic electronic and phonon results for Bi₂Te₃.     

An algorithm to estimate mean traffic speed using uncalibratedcameras

We present a novel approach to estimate traffic speed using a sequence of images from an uncalibrated camera. We assert that exact calibration is not necessary to estimate speed. Instead, we use: 1) geometric relationships inherently available in the image, 2) some common-sense assumptions that reduce the problem to a one-dimensional geometry, 3) frame differencing to isolate moving edges and track vehicles between frames, and 4) parameters from the distribution of vehicle lengths to estimate speed     

Transmission line distance relaying using a variable window short-time Fourier transform

This paper presents a new approach for transmission line protection using a variable window short-time Fourier transform known as S-transform. The S-transform (ST) is a time–frequency spectral localization method, similar to short-time Fourier transform (STFT), but with a Gaussian window whose width scales inversely, and whose height scales linearly with the frequency. The change in spectral energy of the ST of the current and voltage signals provide the information regarding fault detection. After the fault detection, the impedance to the fault point is calculated using the estimated phasors of the faulted current and voltage signals which provide accurate results even with noisy conditions. Also, the fault location is calculated using polynomial curve fitting technique with a devised index obtained from the ratio of spectral energy of the voltage and current signals, respectively.     

An efficient ordering algorithm to improve sparse vector methods

The authors present a novel node-ordering algorithm to enhance sparse vector methods in power system analysis. The proposed technique locally minimizes the number of nonzero elements of the inverse of the table of factors. It uses the cardinality of the set of nodes which precede each node in the path graph as a tie-break criterion in the minimum degree elimination process. Test results are included showing that the method performs better than previously published methods     

An approach for explicitly modeling the protective relaying systemin substation reliability evaluation studies

The authors present an approach for evaluating the reliability of a power system substation taking into account its associated protective relaying system. The method employs the technique of minimum cutsets to compute the frequency and duration of a fault with respect to the criterion of continuity of electric service. The protective relaying system is accounted for by explicitly incorporating its elements into a new functional model developed for their associated circuit breakers. The approach was implemented in the form of a computer program and used to compare the reliability of several substation arrangements. These comparisons as well as some details on the practical implementation of the approach are also presented     

最小二乘法在距离保护中的应用初探

电压和电流信号中存在的非周期分量和高次谐波分量会影响到距离保护测量阻抗的真实性.通过滤除非工频信号而获得的基波阻抗值,可以确保距离保护动作的正确性.基于最小二乘原理的参数估计是应用相当普及的方法,将最小二乘法应用于距离保护中,构建了基于微分方程的系统模型,提出通过测量电感参数获得测量电抗,为解决基于微分方程测量电阻参数估计模型误差大的问题,提出基于有功功率概念构建的测量电阻参数估计模型,并就所提出方法进行了仿真验证,表明通过最小二乘法进行的测量阻抗参数估计是可以应用于距离保护的.     

Neural network approach to fault classification for high speedprotective relaying

This paper presents a new approach to fault classification for high speed protective relaying and show its effectiveness in computer simulations on parallel transmission lines. The scheme is based on the use of neural network architecture and implementation of digital signal processing concepts. We begin by classifying several fault types like 1-phase-to-ground, 2-phase-to-ground and 3-phase-to-ground faults. We proceed with classification of arcing and nonarcing faults in order to obtain a successful automatic reclosing. Encouraging results are shown and indicate that this approach can be used for supporting a new generation of very high speed protective relaying systems     

A new algorithm for relaying and fault location based on autocorrelation of travelling waves

The technique employs correlation of the incident and reflected wave signals at one end of the line. Since the reflected wave is very similar in shape to the incident wave causing it, except for a constant time delay and attenuation, autocorrelation results. The property of ‘evenness’ of the autocorrelation is used to determine the delay time, which is a function of fault position, employing numerical integration which is suitable for a digital computer. It is shown that fast relaying as well as accurate fault location is accomplished on a real-time basis, even in the presence of a considerable amount of high frequency line transients. The technique has been tested by numerical simulations using the standard Electromagnetic Transient Program (EMTP).     

Evaluation of uncertainties in double line distance relaying. Aglobal sight

Uncertainty sources, when relaying double lines with distance relays, during simple or intercircuit faults, are evaluated. Tendencies on errors and their magnitude are easily viewed in a forward reach settings vs. real fault distance covered plane. Contributions of different factors are separated and relations among them are established using the mentioned plane. Desired situations to be evaluated include any type of fault, for any previous state and any fault resistance, as well as any type of distance unit. The intent of the analysis presented is not only to evaluate known tendencies of distance relay responses, in double or single lines, but also to identify the causes and to show the usefulness of the employed methodology for any other type of study to be performed when calculating settings, analysing or designing the best protection philosophy in double or single lines     

基于改进Hausdorff距离的运动目标自动跟踪算法

针对视频运动目标的跟踪问题,提出了一种基于改进Hausdorff距离的运动目标自动跟踪算法。首先对视频图像作预处理,使用Canny算子进行边缘提取;然后采用对称差分算法建立运动目标的初始模板;接着采用改进的Hausdorff距离和新的匹配搜索策略,实现后续帧中运动目标与初始模板的匹配;最后,采用一种新的模板更新算法———4步法对模板进行更新。实验结果表明,该算法不仅能准确跟踪运动目标,而且能有效提高目标跟踪效率。     

A simple algorithm for the symmetrical components relaying and monitoring

Contents An algorithm suitable for computation of symmetrical components is presented in the paper. The algorithm is based on the Four-Sample Method and it extracts the rms values of the symmetrical components. The proposed algorithm is very simple and suitable for the application in the microprocessor-based metering and power system protection and monitoring.

---

Ein einfacher Algorithmus zur Erfassung symmetrischer Komponenten für Schutz und Überwachung

Übersicht In dem Beitrag wird ein geeigneter Algorithmus zur Berechnung symmetrischer Komponenten vorgestellt. Dieser beruht auf der Erfassung von vier Abtastwerten pro Grundperiode und liefert Effektivwerte der symmetrischen Komponenten. Der vorgeschlagene Algorithmus ist sehr einfach und anwendbar zur Meßwerterfassung mit Hilfe von Mikroprozessoren für den Schutz und die Überwachung von Energiesystemen.