Tamer A. Kawady

Development and implementation of an ANN-based fault diagnosis scheme for generator winding protection

In this paper, the development and implementation of a new fault diagnosis scheme for generator winding protection using artificial neural networks (ANN) is introduced. The proposed scheme performs internal fault detection, fault type classifications and faulted phases identification. This scheme is characterized with higher sensitivity and stability boundaries as compared with the differential relay. Effect of the presence of nonsynchronous frequencies on the scheme performance is examined. Effect of different values of ground resistance on ground fault detection sensitivity is outlined. The scheme hardware is implemented based on a digital signal processing (DSP) board interfaced with a multi input/output (MIO) board. Test results of the proposed scheme corroborate the scheme stability and sensitivity.

ANN-based novel fault detector for generator windings protection

Summary form only given as follows. In this paper, an artificial neural network (ANN) based internal fault detector algorithm for generator protection is proposed. The detector uniquely responds to the winding earth and phase faults with remarkably high sensitivity. Discrimination of the fault type is provided via three trained ANNs having a six dimensional input vector. This input vector is obtained from the difference and average of the currents entering and leaving the generator windings. Training cases for the ANNs are generated via a simulation study of the generator internal faults using Electromagnetic Transient Program (EMTP). A genetic algorithm is employed to reduce training time. The proposed ANN algorithm is compared with a conventional differential algorithm. It is found to be superior regarding sensitivity and stability.

Performance evaluation of conventional protection systems for wind farms

Wind farms are characterized with some unique features during their normal and faulty operating conditions. Different factors participate usually into these conditions such as the distributed generation concept, the own behavior of the induction generator, varying wind speed, ... etc. This, consequently, arises different challenges regarding the behavior of their protection and control schemes. In this paper, the performance of the conventional protection schemes (currently in use with wind farms) is thoroughly investigated. The aimed investigation study is carried out on a 225 MW wind farm in AL-Zaafarana-Egypt as a simulation example using MATLAB. The package represents an ideal candidate for dynamic simulation purposes with its own highly performance graphical interface and built-in libraries. The major effective factors that may influence the fault characteristics are considered such as fault position, fault type, fault resistance and pre-fault loading. This explores the performance of wind farms during faulty conditions honestly. It also visualizes the performance of their protection elements and assists enhancing their performance in the future.

Bayesian Selectivity Technique for Earth Fault Protection in Medium-Voltage Networks

In Nordic countries, distribution networks are unearthed or compensated. Earth faults, particularly in compensated networks, provide fault currents that are low compared to the load currents. Identification of the faulty feeder is therefore difficult. A preliminary description of discrete wavelet transform (DWT)-Bayesian selectivity technique was introduced in reference 1 to identify the faulty feeder. It was dependent on a conditional probabilistic method applied to transient features extracted by using the DWT. However, a practical setting for this technique has not yet been presented. Furthermore, its sensitivity is limited to 1.5-k fault resistances, and is further reduced to 170 when considering current transformer and network noise. In this paper, the ratio between the absolute sums of the DWT detail level from each feeder is used as an input to the conditional probability approach, providing an enhanced selectivity decision. This input contributes to discriminating the faulty feeder during high resistance faults. The relay setting is introduced as a function of the number of feeders and their characteristic impedances, as the proposed algorithm is dependent on the discharges initial transients. The performance is evaluated taking into account current-transformer and network noises. A digital implementation is experimentally verified by using two digital signal processing boards.

Detection of Partial Saturation and Waveform Compensation of Current Transformers

Derivative-based approaches have been widely used for detecting current-transformer (CT) saturation. This letter proposes an efficient approach for detecting CT saturation with a generic discrimination index without the need for setting definition. For this aim, the simultaneous samples-based multiplication of the secondary current by its derivative is only used to differentiate between saturated and unsaturated wave portions. Reconstructing the detected saturated secondary current is performed by using Kalman filtering to efficiently extract the phasor quantities of the unsaturated current portion. The results provide evidence of the efficacy of the proposed detector and of the Kalman-based waveform reconstruction.

Wind Farm Protection Systems: State of the Art and Challenges

Owing to the rapid increase of the global population and their energy needs, traditional means to satisfy the burgeoning energy demands need careful reevaluation. Coupled with the uneven distribution of resources around the world, economic impacts of large-scale importation and the environmental impacts of continued dependence on nonrenewable fossil fuels, there is an imminent need to transfer, at least partly, the dependence on to renewable energy resources. Among these resources, wind electric conversion systems have emerged as the leader at the present time. According to the Global Wind Energy Council (GWEC) annual report, over 27 GW of new wind power generation capacity came on line worldwide in 2008 representing a 36% growth rate in the annual market, bringing the total global wind power capacity to over 120GW through the end of 2008 as shown in Fig. 1 [1]. This indicates that there is huge and growing global demand for emissions-free wind power, which can be installed quickly, virtually everywhere in the world.

Experimental investigation of high impedance faults in low voltage distribution networks

The aim of this paper is to visualize the general characteristics of downed conductor faults occurring in Low Voltage (LV) distribution networks. These faults when occur in rural areas are potentially life-threatening and may start fires with fault arcs. Moreover, these faults are usually characterized with very low fault currents resulting from the accompanied high fault impedance according to the surface of the faulted area. This consequently inhibits to detect these faults with conventional current-based relaying functions. Towards the paper goal, the properties of different experimental arcing fault cases are analyzed covering different fault sequences and different types faulted surface area. The obtained results assist to realize a better understanding of the prosperities of these faults and supervise the next steps aiming to get a practical, versatile and low-cost fault detector for such faults.

A Gabor Transform-based universal fault detector for transmission lines

In this paper, a new fault detection scheme is proposed for transmission systems using Gabor transform (GT) for signal processing purposes instead of the conventional Fourier analysis methods. The detector uniquely detects both low impedance faults and high impedance ones with the same efficacy providing universal and precise fault detection for transmission systems. It utilizes GT as a feature extractor and artificial neural networks for pattern recognition and classification processes. The use of GT is motivated by the fact that the Gabor elementary functions have distinctively an optimal localization property in the joint time and frequency domains, which leads to optimal extraction of features from signals that is hard to be detected using conventional detectors. The generalization capability of the system allows detecting fault signals with very low fault currents and associated with, sharp and abrupt changes, dc decaying, non integer harmonics and non stationary quantities. The alternative transient program (ATP) is employed for the investigation phase. All simulation tests corroborate the superiority of the proposed detector for transmission system protection.

Generalized 1-D Gabor Transform Application to Power System Signal Analysis

In this paper, the generalized 1-D Gabor transformation and possibility of application to power system signal analysis are presented. An artificial neural network based algorithm is employed to compute the Gabor coefficients for further decomposition of the power system signals. The time-frequency resolution analysis for sinusoidal, DC decaying, sudden change and non-integer harmonic signals is carried out. Contribution of this transform to power system applications is evaluated through various simulation test cases and scenarios. Recommendations for increasing the value of Gabor transform to protection area are outlined

Unsynchronized Fault-Location Technique for Double-Circuit Transmission Systems Independent of Line Parameters

In this paper, a new double-end fault-location technique is proposed for parallel transmission lines depending on unsynchronized measurements of both line ends without using line parameters. The proposed technique is based on profiling the correlation between the unknown synchronization angle and the fault distance to accurately estimate the fault point. The positive-sequence network with lumped parameter line model is first utilized in order to facilitate pinpointing the required synchronizing angle between both line ends. The least square error method is used to solve the equivalent positive-sequence network of the line using unsynchronized double-end measurements. Then, an accurate estimation of the fault distance can be realized for all fault types depending on the positive-sequence network of the faulty line. The performance of the proposed locator is thoroughly investigated using a detailed simulation of a selected 150-mi overhead parallel transmission system using the Alternate Transients Program/Electromagnetic Transients Program package. A wide variety of faulty conditions is covered in order to exploit the benefits of using double-end data even with very high fault resistance and untransposed transmission lines.