Search

In this paper, a protection scheme of 500 kV, 50 Hz, and 220 km double end transmission system based on wavelet transform (WT), high-speed decision tree (DT), and artificial neural network (ANN) are proposed. The scheme is implemented using a microcontroller co-simulated with MATLAB/Simulink and it is validated in real-time. The proposed scheme performs fault classification, location estimation, and fault zone identification using both DT and ANN in a comparative study. The scheme calculates the wavelet transform approximation coefficients of the current and voltage signals and then sends its energy to a specific decision tree/neural network to recognize the fault type, its zone, and its location. It was found that the accuracy of the decision tree classifier outperforms the ANN classifier, in addition, achieves a significantly higher classification speed of about 0.4 msec inference time and it is easier to be applied in real-time. In contrast, the neural network is more accurate in calculating fault zone and locations. DT and ANN are trained and tested using different sets of data obtained from simulated faults that differ in type, resistance, and location. Modeling fault scenarios and data processing are carried out using the MATLAB/Simulink software package. The hardware implementation uses an 8-bit ATmega microcontroller that is interfaced with the simulated model using MATLAB support package for Arduino. Simulation results confirm that the proposed scheme using DT for fault classification and ANN for zone and location estimation is accurate and reliable even with varying fault resistance, type, and distance, and is therefore applicable in practice on a digital platform. [ABSTRACT FROM AUTHOR]

This paper introduces a novel approach to high-speed fault protection strategy for bipolar Current Source Converter (CSC) based high voltage DC transmission line. The Bagged-Tree Ensemble-Classifier (BTEC) method is employed to detect the presence of a fault and categorizes the various types of faults in the DC transmission line. The implementation of the Wavelet Transform (WT) technique assists in diminishing the effect of deviation in the fault resistance value, fault location, and disturbances created due to noise on the performance BTEC method. The development of the predictive model through concatenating individual models is the prime objective of the ensemble schemes. Ensemble procedure assists to enhance the performance accuracy of the fault protection strategy. The local measurement at the rectifier end side is enough for the proposed strategy to realize the protection of the entire line with no use of communication channels. The AC sinusoidal voltage, DC current and DC voltage are recorded at the rectifier end terminal and sampled at 1 kHz, which is very less as compared to the other existing techniques. The proposed strategy detects the fault in the least possible time of 4 ms and able to classify all the tested faults with reasonable accuracy. The results convey that the proposed strategy offers high-speed detection of fault and preciseness in the classification of all types of faults, despite variation in the fault resistance up to 180 Ω. The proposed scheme is applicable for DC faults and also detects the presence of AC faults in either rectifier end as reverse fault and in inverter end as sub-sequent forward section fault. [ABSTRACT FROM AUTHOR]

The fault component power and energy based relaying schemes have limitation during double-phase to ground fault due to the mutual coupling effect. This article presents a superimposed reactive power coefficient (SRPC) based pilot relaying scheme for the protection of shunt compensated transmission line. SRPC of a phase is defined as the ratio of integrated superimposed reactive power (SRP) of that phase to the maximum magnitude of integrated SRPs among all three phases. If SRPC of any phase is found less than −0.5, an internal fault is detected and that phase is considered as a faulty phase. If no phase possesses SRPC less than −0.5, the fault is an external fault. The algorithm will start only if the superimposed differential current of any phase is found more than 50% of pre-fault differential current. The results are obtained using PSCAD/EMTDC software for different conditions, such as different fault resistances, fault locations, static VAR compensator locations, and source impedances. The proposed scheme is found robust and accurate for all the cases considered. The performance of the proposed scheme is superior to the recently proposed similar type of scheme and is unaffected from capacitive coupled voltage transformer and current transformer measurement errors and synchronization error. [ABSTRACT FROM AUTHOR]

The Thyristor-Controlled Series Capacitor plays an important role in high voltage power transmission. However, due to non-linearities introduced by the protective equipment of the Thyristor-Controlled Series Capacitor, fault location in the series compensated line becomes a difficult task. The kernel extreme learning machine-based method for fault location in series compensated line is proposed in this article. Results of the kernel extreme learning machine are compared with other popular single-hidden layer feedforward network-based techniques, such as extreme learning machines, support vector machines, and relevance vector machines. Performances of these single-hidden layer feedforward networks are verified on two power systems: (1) Two-area equivalent system, and (2) 12-bus system modeled in detail. Simulation studies are performed with wide variation in system and fault parameters, such as compensation level, load conditions, fault resistance, fault location, and fault inception angle. The kernel extreme learning machine achieved better accuracy in lesser training time and parameter-tuning time. [ABSTRACT FROM PUBLISHER]

This article presents a fault detection and classification scheme for a series-compensated transmission line that is based on post-fault superimposed energy. The derivations of the scheme presented in this article are obtained with consideration of both the real and reactive components of the power system. However, the criterion depends only on the real components of the power system. For a forward fault, superimposed energy is negative, whereas it is positive for a reverse fault. If the relays of both ends detect forward fault, it is an internal fault; else, it is an external fault. The magnitude of superimposed energy depends on the fault type, location, and resistance, which makes it difficult to classify the type of fault as a fixed threshold cannot be set. Therefore, to classify the type of fault, energy coefficients have been introduced that depend on the superimposed energy measured at the relay. To test the capability of the superimposed energy based scheme, the test system has been simulated in PSCAD/EMTDC (Manitoba HVDC Research Centre) and an algorithm has been implemented in MATLAB (The MathWorks, Natick, Massachusetts, USA). Results proved that the scheme is accurate and robust against different system conditions and uncertainties. [ABSTRACT FROM AUTHOR]

In this article, a novel protective relaying scheme based on a finite-state machine is proposed to detect fault in transmission lines, classify the fault, and identify the faulty phase. The three-phase fundamental component of current and the zero-sequence current signals measured at one end of the double-circuit line are used as inputs. The finite-state machine based relaying scheme relies on time-series analysis of current signals only and is built upon the concepts from the finite-state automata theory. The finite-state machine works by transition of one state to another by following certain conditions. The proposed relay is tested during different shunt faults—inter-circuit and cross-country faults—with wide variations in fault parameters. The proposed method is adaptive to variation in fault type, fault resistance, fault inception angle, fault location, power flow angle, different line length, transient faults, current transformer (CT) saturation, and no-fault events. The relay performed correctly for 99.9% of test cases, proving the effectiveness of the proposed method. Furthermore, the proposed method can provide faster, more reliable protection against all shunt faults, inter-circuit and cross-country faults, with wide variations in parameters, and the protection range is effectively extended and greatly improved, which contributes to system safety and stability. [ABSTRACT FROM PUBLISHER]

The proposed work develops a decision tree-induced fuzzy rule base intelligent protection scheme for fault detection and classification in a microgrid with multiple distributed generation interfaces. The proposed protection scheme retrieves one cycle post-fault current signal samples of each phase from fault inception at bus ends of the respective feeder to derive some differential features. The retrieved current samples are pre-processed using S-transform to obtain a time–frequency contour. The statistical features, such as energy, mean, standard deviation, and entropy, are computed from the time–frequency contour, which is further used to calculate the differential features. The differential features are used to build the fault classification tree. From the decision tree classification boundaries, the fuzzy membership functions are drawn, and further, the corresponding fuzzy rule base is generated for the final relaying decision. The proposed scheme is developed on a MATLAB/SIMULINK (The MathWorks, Natick, Massachusetts, USA) platform, including wide variations in faulted conditions, and the extensive test results indicate that the proposed intelligent relaying scheme can reliably provide protection measures for microgrids with different modes of operation. [ABSTRACT FROM PUBLISHER]

This article demonstrates a technique for diagnosis of fault type and faulty phase on overhead transmission lines. A method for computation of fault location is also incorporated in this work. The proposed method is based on the multi-resolution S-transform, which is used for generating complexS-matrices of the current signals measured at the sending and receiving ends of the line. The peak magnitude of the absolute value of everyS-matrix is noted. The phase angle corresponding to every peak component is obtained from the argument of the relevantS-matrix. These features are used as input vectors of a probabilistic neural network for fault detection and classification. Detection of faulty phase(s) is followed by estimation of fault location. The voltage signal of the affected phase is processed to generate theS-matrix. The frequency components of theS-matrices for different fault locations are used as input vectors for training a back-propagation neural network. The results are obtained with satisfactory accuracy and speed. All the simulations have been done in MATLAB (The MathWorks, Natick, Massachusetts, USA) environment for different values of fault locations, fault resistances, and fault inception angles. The effect of noise on both the current and voltage signals has been investigated. [ABSTRACT FROM PUBLISHER]

Abstract—This article presents a fault-classification method for transmission lines based on voltage phasors using classification and regression trees. The proposed method is intended to aid system operators in understanding the outputs of a phasor measurement unit only state estimator. Faults are classified into four categories when the estimator is positive sequence and into ten categories when the estimator is three phase. The fault data are generated in PowerWorld® (PowerWorld Corporation, Champaign, IL, USA) and DSA Tools® (Powertech Labs Inc., Surrey, British Columbia, Canada). The pre-fault state consists of a variety of operating conditions and loading angles of faulted lines. The fault condition comprises different fault types, fault locations, fault impedances, and fault incidence angles. Fault classification is done using MATLAB® (The MathWorks, Natick, Massachusetts, USA).The approach is successfully tested on the IEEE-118 bus system. The results demonstrate that the technique developed here is effective and robust, irrespective of the pre-fault and fault conditions. [ABSTRACT FROM PUBLISHER]

A new transmission line fault-classification algorithm based on half-cycle post-fault current data is presented for an advanced series-compensated transmission line equipped with a thyristor-controlled series compensator. The proposed scheme was developed with the signal feature enhancement tool of discrete wavelet packet entropy measures. The Chebyshev neural network is presented as network-growing technique for protective classification, the single-layer structure of which is a more powerful classifier that eliminates the need for complicated network design. A comparative implementation study of the multi-layer perceptron and Chebyshev neural network authenticates benefits gained by the Chebyshev neural network. To demonstrate the advantage gained by Chebyshev neural networks compared to support vector machines, a comparative study is presented with a support vector machine based classification technique. The fault datawere obtained by dynamic simulation of a sample system using the real-time power system simulator PSCAD (Manitoba HVDC Research Centre, Winnipeg, Manitoba, Canada). Extensive testing reveals the effectiveness of the Chebyshev neural network for fault classification; a comparative study brings out the superiority of the Chebyshev neural network for neural network design and implementation against the multi-layer perceptron. The Chebyshev neural network proved advantageous against support vector machines as being insensitive to the classification parameter. [ABSTRACT FROM AUTHOR]

This article presents a classification methodology based on probabilistic neural networks. To automatically select the training data and obtain the performance evaluation results, the “K-fold” cross-validation method is used. Then, the probabilistic neural network is compared with the feed-forward neural network and the radial basis function network. The goal is to propose a classifier that is capable of recognizing 11 classes of three-phase distribution system faults to solve the complex fault (three-phase short-circuit) classification problem for reducing the multiple-estimation problem to estimate the fault location in radial distribution systems. The data for the fault classifier is produced by DigSilent Power Factory, Integrated Power System Analysis Software on an IEEE 13-node test feeder. A selection of features or descriptors obtained from voltages and currents measured in the substation are analyzed and used as input of the probabilistic neural network classifier. It is shown that the probabilistic neural network approach can provide a fast and precise operation for various faults. The simulation results also show that the proposed model can successfully be used as an effective tool for solving complicated classification problems. [ABSTRACT FROM AUTHOR]

This article proposes an approach for the protection of transmission lines with flexible AC transmission systems based on artificial neural networks using the total least square estimation of signal parameters via rotational invariance technique. The required features for the proposed algorithm are extracted from the measured transient currents and voltages waveforms using the total least square estimation of signal parameters via rotational invariance technique. Since these transient waveforms are considered as a summation of damped sinusoids, the total least square estimation of signal parameters via rotational invariance technique is used to estimate different signal parameters, mainly damping factors, frequencies, and amplitudes of different modes contained in the signal. Those features are employed for fault detection and faulted phase selection using artificial neural networks. Two types of flexible AC transmission system compensated transmission lines, namely the thyristor-controlled series capacitor and static synchronous compensator, are considered. System simulation and test results indicate the feasibility of using neural networks with the total least square estimation of signal parameters via rotational invariance technique in the fault detection, classification, and faulted phase selection of flexible AC transmission system compensated transmission lines. [ABSTRACT FROM AUTHOR]

This article presents a new approach to classify various types of power system faults using wavelet transforms and the adaptive resonance theory. The key idea underlying the approach is to decompose a given disturbance signal into other signals, which represents a smoothed and detailed version of the original signal. The proposed technique consists of a preprocessing unit based on discrete wavelet transform in combination with adaptive resonance theory. Through wavelet analysis, faults are decomposed into a series of wavelet components, each of which is a time-domain signal that covers a specific octave frequency band. The parameters selected for fault classification are the detailed coefficients of all phase current signals that are collected only at the sending end of a transmission line. The information is then fed into adaptive resonance theory for classifying the faults. The study is performed on a sample power system network. Extensive simulation studies carried out using MATLAB (http://www.mathworks.com) shows that the proposed algorithm not only provides an accepted degree of accuracy in fault classification under different fault conditions, but it is also reliable, fast, and computationally efficient tool. [ABSTRACT FROM AUTHOR]

This article presents a fast, hardware efficient logic for fault detection and classification in transmission lines. The proposed logic employs only one-terminal current samples and is based on multi-resolution wavelet analysis. First-level high-frequency details of the modal currents in the range of 5 to 10 kHz are extracted. Depending on the amount of high-frequency components in the transformed current signals after processing, the faults are classified into ten types. An adaptive threshold value is chosen, rather than a fixed threshold in the case of faults involving the ground, to make the classification reliable and accurate. The validity of the proposed logic was exhaustively tested by simulating various types of faults on a system modeled in Alternative Transient Program (ATP)/Electromagnetic Transient Program (EMTP).The proposed logic was found to be acceptably reliable and accurate even in the presence of fault resistance and current transformer (CT) saturation. [ABSTRACT FROM AUTHOR]

The proposed algorithm for fault location, different from conventional algorithms that are based on deterministic computations on a well-defined model to be protected, employs wavelet transform together with fuzzy inference system (FIS) and the adaptive neuro-fuzzy inference system (ANFIS) to incorporate expert evaluation so as to extract important features from wavelet multi-resolution analysis (MRA) coefficients for obtaining coherent conclusions regarding fault location. Simulation results indicate that both the classification and localization algorithms are immune to the effects of fault inception angle, impedance and distance. The most significant contribution of this article is that the proposed ANFIS approach has superiority over FIS for location of transmission line faults and thus can be used as an effective tool for real-time digital relaying purposes. [ABSTRACT FROM AUTHOR]

Two methods for classification of transmission line faults are presented and compared, one based on a wavelet analysis and the other on an artificial neural network (ANN). While the wavelet analysis based approach requires consideration of wavelet multiresolution analysis (MRA) level-1 details of the three phase currents and delta currents, the ANN based approach requires the samples of three phase currents as inputs. Simulation studies using EMTP and MATLAB and considering wide variations in fault location, fault inception angle, and fault point resistance for different types of faults have shown that the wavelet analysis based method has an edge over the ANN based method. [ABSTRACT FROM AUTHOR]

This article presents a new method for identification and classification of faults based on wavelet multiresolution analysis (MRA). Daubechies eight (D-8) wavelet transforms of the three phase currents on a transmission line are used. The peak absolute value, the mean of the peak absolute values, and summation of the third level output of MRA detail signals of current in each phase extracted from the original signals are used as the criterion for the analysis. The effects of fault distance, fault inception angle and fault impedance are examined. Extensive simulations are carried out to generate time domain input signal using EMTP (Microtran) on a 400 kV, 300 km long line and simulation results show that the proposed method is a simple, effective and robust method suitable for high impedance faults also. [ABSTRACT FROM AUTHOR]

This article presents an Artificial Neural Network (ANN) approach to simulate a complete scheme for distance protection of transmission lines. The protection technique is based on a modular approach whereby different neural network modules have been adopted for fault detection, fault classification and fault location. Three-phase voltage and current samples were utilized as inputs for the proposed scheme. The Alternative Transients Program (ATP) software was used to generate data for the transmission line (400 kV). The results obtained show that the global performance of the ANN architectures is highly satisfactory under a wide variety of different fault conditions. [ABSTRACT FROM AUTHOR]