Fault Classification

As environmental awareness has been culminating for the past few decades, people have been switching towards renewable sources of energy such as solar energy, wind energy, and hydropower. Among these, solar energy seems to be largely used due to the availability of solar energy in abundance, which in turn, has led to a mass increase in the number of solar power plants. Nevertheless, these solar panels are left unattended due to difficulty in the maintenance of solar power plants in large areas and the manual work required. It is important to find and replace these defective panels in time before any severe event occurs. Detecting hotspots, cracking and various other malfunctions in the photovoltaic cell can lead to an increase in the life of the solar panels by 5-10 years. In this paper, we propose a compact intelligent photovoltaic module diagnosis system based on a Deep Learning model that uses multilayer perceptron (MLP) to accurately identify and classify the faults in the photovoltaic module. We have achieved this by tweaking the VGG16 convolutional network architecture to detect 3 kinds of malfunctions -Hotspots, Cracking, Diode, and No-Anomaly for panels that are in good condition by using 2310 images for training, 665 for validation, and, 665 for testing. By using accuracy-score from sklearn.metrics, the model's accuracy was estimated to be above 90 percent for classifying photovoltaic modules.

The identification, categorization, and localization of faults play a crucial
role in maintaining the smooth operation of power systems. Distance relays
possess a significant capability to withstand power fluctuations, thereby
minimizing inadvertent disruptions in transmission lines. Addressing these
challenges involves the adoption of advanced fault analysis techniques to
enhance the accuracy and speed of relay operations. While modern machine
learning (ML) approaches are still nascent in fault analysis, the authors
propose a novel deep learning (DL) based long short term memory (LSTM)
method for precise fault detection, classification, and rapid fault location
estimation. The proposed approach is applied to the Kundur two-area 4
machine 11 bus system covering a distance of 220 km. The LSTM fault
detection (LSTM (FD)) module accurately detects and classifies faults, while the LSTM fault location (LSTM (FL)) module precisely estimates fault locations. The effectiveness of the proposed method is verified through a comparative assessment with various traditional ML and DL techniques. The protection modules are also tested under different fault locations, fault resistances, and noisy signals. The features taken into consideration for the operation of the protection modules are different bus voltages, bus currents, zero sequence voltage, zero sequence current, fault inception angle, and fault resistance.

This paper discusses a method for accurately estimating the fault location in multi-terminal High Voltage direct current (HVdc) transmission network using single ended current and voltage measurements. The post-fault voltage and current signatures are a function of multiple factors and thus accurately locating faults on a multi-terminal network is challenging. We discuss a novel data-driven Bayes Regression based method for accurately predicting fault locations. The sensitivity of the proposed algorithm to measurement noise, fault location, resistance and current limiting inductance are performed on a radial three-terminal MTdc network. The test system is designed in Power System Computer Aided Design (PSCAD)/Electromagnetic Transients including dc (EMTdc).

The degree of reliability is one of the main issues in having the uninterrupted operation of fault-tolerant measurement and control systems. This paper presents a technique for detecting and localizing current sensor fault in doubly fed induction generator wind turbine systems (DFIG-WT). For this purpose, a methodology divided into three steps has been carried out and is presented as follows: the first step is the application of the zero-sequence component as an indicator of the presence of the sensor fault. The second step is based on the application of Concordia transformation on the current signals to generate different criteria for localizing the faulty sensor. Finally, an artificial neural network model is developed to analyze the localization criteria and identify the faulty sensor. All simulations are carried out in MATLAB/Simulink environment. Results show that the proposed technique is effective and can be used in real-time fault detection and localization in the DFIG-WT.

Machine learning plays a crucial role in predicting and clas- sifying faults in electrical power systems. The complexity
and dynamic nature of these systems make them vulnerable to disturbances and elec- trical faults. Detecting faults in the circuit
can aid in maintaining the system by preventing potential damage from occurring.

As environmental awareness has been culminating for the past few decades, people have been switching towards renewable sources of energy such as solar energy, wind energy, and hydropower. Among these, solar energy seems to be largely used due to the availability of solar energy in abundance, which in turn, has led to a mass increase in the number of solar power plants. Nevertheless, these solar panels are left unattended due to difficulty in the maintenance of solar power plants in large areas and the manual work required. It is important to find and replace these defective panels in time before any severe event occurs. Detecting hotspots, cracking and various other malfunctions in the photovoltaic cell can lead to an increase in the life of the solar panels by 5-10 years. In this paper, we propose a compact intelligent photovoltaic module diagnosis system based on a Deep Learning model that uses multilayer perceptron (MLP) to accurately identify and classify the faults in the photovoltaic module. We have achieved this by tweaking the VGG16 convolutional network architecture to detect 3 kinds of malfunctions-Hotspots, Cracking, Diode, and No-Anomaly for panels that are in good condition by using 2310 images for training, 665 for validation, and, 665 for testing. By using accuracy-score from sklearn.metrics, the model's accuracy was estimated to be above 90 percent for classifying photovoltaic modules.

Induction motors are widely used in various industries for production processes due to their inherent characteristics like high reliability, robustness, low cost and ease of construction and operation. However, while in operation they face harsh and severe conditions imposed on them by the electrical, mechanical, thermal and environmental stresses which give rise to abnormalities in machine parameters such as voltages and currents. If the motor drive is allowed to continue to run under such situations, it will lead to catastrophic failure of the motor disrupting the production process and huge loss of revenue. Hence, an intelligent drive control method based on Hybrid Artificial Intelligence (AI) technique of Neuro-Genetic Algorithm is proposed in this paper for condition monitoring, fault diagnosis and evaluation of induction motor without any additional information. The hybrid architecture of Back Propagation Neural Network (BPN) to get the desired output fast and with Genetic Algorithm to overcome the computational complexity and time consuming process involved in BPN.

Distance relay is susceptible to maloperation during power swing. Power swing blocking function is integrated in distance relays to ensure security during power swing. However, if a fault occurs during power swing, it must be detected and cleared by the relay as quickly as possible. Since power swings and three-phase faults are both balanced phenomena, discriminating them is very challenging. This paper proposes a method to detect three-phase fault during power swing by computing the Hilbert envelope of the current waveform followed by zero-frequency filtering. The average rate of change in amplitude of the zero-frequency filtered current is used as the fault detector index. The discrimination capability between three-phase fault and power swing of the extracted feature is the main contribution of this paper. Results of various possible test cases simulated through EMTDC/PSCAD show that the proposed method is fast and reliable in detecting three-phase faults during power swings. KEYWORDS distance protection, power swing, power swing blocking, three-phase fault detection, zero frequency filtering 1 | INTRODUCTION Power system disturbances such as faults, switching on/off of large loads, disconnections of generators, line switching etc., cause oscillations in rotor angles among generators. This can result severe swings in the power-flow. 1 During power swing, the impedance measured by a distance relay may enter into its operating zones and can cause the relay to trip lines unnecessarily. Investigations on past blackouts worldwide (including the 2012 blackout in India) depict that Abbreviations: ANFIS, adaptive neural fuzzy inference system; DFT, discrete Fourier transform; DTFT, discrete time Fourier transform; FFT, fast Fourier transform; HE, Hilbert envelop; IDFT, inverse discrete Fourier transform; OST, out-of-step tripping; PSB, power swing blocking; SVM, support vector machine; TMI, transient monitor index; WT, wavelet transform; ZFF, zero-frequency filter; ZFFO, zero-frequency filter output Symbols: i(n), current signal measured at n th instant; i a (n), analytic signal derived from the measured current signal at n th instant; i h (n), Hilbert transform of measured current signal at n th instant; *, signal convolution; h(n), Hilbert transform operator at n th instant; H(ω), Fourier transform of Hilbert operator; I h (ω), Fourier transform of current signal; h i (n), magnitude of the analytic signal measured at n th instant; Φ(n), phase of the analytic signal measured at n th instant; x(n), first-difference of Hilbert envelope of current signal measured at n th instant; y(n), digital resonators output at n th instant; N, number of samples per cycle; b y n ð Þ, digital resonators output after trend removal at n th instant; a k , coefficients of digital resonators; y n ð Þ, local mean of digital resonators output at n th instant; Δ ZFFOph , average rate of change of zero-frequency filtering output; Δ ZFFO , fault detector index; Δ ZFFOth , threshold for fault detector index; ΔR, rate of change of apparent resistance; I sup , superimposed components of current; |ψ(I 2

A data science platform in a smart chemical plant enables simultaneity of multiple applications. However, in each application there is a wide variety of algorithms that can be implemented. Therefore, the objective of this work is to evaluate the performance of the algorithms to be implemented in a data science platform for applications related to fault classification, fault detection and virtual sensor in the Tennessee Eastman (TE) process. To achieve this, a synthetic data set obtained by simulating the TE process was built, where each algorithm was trained, optimized and evaluated according to the type of data science task associated with each application. The results show that the artificial neural network only achieved the best performance in fault classification. In the virtual sensor, gradient boosting (GB) and k-nearest neighbor (k-NN) achieved the best performance. Meanwhile, in fault detection, most of the evaluated algorithms achieved a fault detection rate around 88%. In conclusion, algorithms based on artificial neural networks did not achieve the best performance in all implemented applications, being surpassed by other non-linear algorithms.

This paper makes a contribution to the field of fault location finding in a new way that helps in the improvement of grid reliability. This paper proposes a study-based approach for fault allocation and fault type classification that uses the study of voltage and current frequency during the abnormal condition. Although, ideally frequency of voltage and current are the same in the abnormal condition they may differ from each other. This difference in frequency is separately measured by the phasor measurement unit (PMU) block at MATLAB/Simulink platform. The PMU (PLL-based, positivesequence) block is inspired by the IEEE Std C37.118.1-2011. In this approach, we measure the line voltage and current frequency variation with the help of installed PMU after this we present this measurement in characteristics form with the help of the scoping tool in MATLAB/Simulink and study them one by one, and proposed a conclusion for fault location identification and fault type classification. The proposed approach is able to identify the source side and load side fault location and also able to classify faults into two categories namely symmetrical and asymmetrical. The proposed approach is tested on two MATLAB/Simulink models and observed satisfactory.

This manuscript presents a novel micro-grid protection scheme based on S-transform (ST) and machine learning techniques. Initialisation of the proposed approach is done by extracting the current signals from the targeted buses of different feeders and pre-processing through ST to derive different needful differential features. The extracted features are further used as an input vector to the machine learning model to classify the fault events. The proposed micro-grid protection scheme is tested for different protection scenario, such as the type of fault (symmetrical, asymmetrical and high impedance fault), micro-grid structure (radial and mesh) and mode of operation (islanded and grid connected), etc. Three different machine learning models are tested and compared in this framework: naïve Bayes classifier (NBC), support vector machine (SVM) and extreme learning machine (ELM). The extensive simulated results from a standard IEC micro-grid model prove the effectiveness and reliability of proposed micro-grid protection scheme.

Support Vector Machine (SVM) is one of the most popular supervised learning techniques for classification and regression tasks. This study proposes and validates a novel, fast, and efficient primal SVM design using the aid of a piecewise linear approximation approach and separable linear programming optimization techniques. The GUROBI optimizer solver is employed to address this approximation-based SVM design challenge. The proposed SVM (we term it PLASVM) designing technique guarantees a global solution through the application of mixed integer linear programming and branch and bound algorithms. The linear programming (LP) method enables SVM training with large datasets in a short amount of time. The effectiveness of the suggested PLASVM algorithms is evaluated using a laboratory gas turbine engine. Using the fault classification technique, the performance of the gas turbine engine has been guaranteed against any faulty condition. The collected findings demonstrate that the PLASVM algorithms perform better than the existing techniques regarding training speed and accuracy.

This paper discusses how to use the warping information obtained after batch synchronization for process monitoring and fault classification. The warping information can be used for i) building unsupervised control charts or ii) fault classification when a rich faulty batches database is available. Data from realistic simulations of a fermentation process of the Saccharomyces cerevisiae cultivation are used to illustrate the proposal.

In recent times, inverters are considered as the basic building block in an electrical drive system used widely in many industrial drive applications. However, the reliability of these inverters is mainly affected by the failure of power electronic switches. Various faults in inverter may influence the system operation by unexpected maintenance, which increases the cost factor and reduce overall efficiency. In this paper, comparative study of three different fault detection and diagnosis systems for three phase inverter is presented. The basic purpose of these fault detection and diagnosis systems is to detect single or multiple faults efficiently. These techniques rely on the neural network for fault detection and diagnosis by using Clarke transformed two-dimensional features extraction, three-dimensional features extraction and features extraction using discrete wavelet transform (DWT) with a different number of features in each technique. Several features are extracted using different mechanisms and used in the neural network as input for fault detection and diagnosis. Furthermore, a simulation study is carried out to analyze the fault detection and diagnosis response of these techniques. Also, a comparative study has been performed by considering fault detection time and accuracy. Comparison results prove the supremacy of three-dimensional feature extraction technique over other two techniques as it can detect and diagnose single, double and triple faults in a single cycle with high accuracy as compared to other two techniques multi-cycles detection.

To emphasize the clarification of numberless botheration related with the distance security of a series capacitor compensated transmission line, an altercation has been done for the spot of fault on series compensated transmission line with the sustenance of discrete wavelet transform. Distance protection of series compensated transmission line has been an investigating grindstone. Wavelet analysis is a processing tool of signal which can be used for the analysis and location of fault occurring in transmission line. In this paper, MATLAB's simulink environment approach has been employed for carrying out simulation work. The fault current signal is scrutinized by adopting the discrete wavelet transform toolbox of MATLAB by selecting 'db5' as mother wavelet from wavelet family to access the coefficients of pre and post fault for fault locale on series compensated transmission line.

As environmental awareness has been culminating for the past few decades people have been switching towards renewable sources of energy such as solar energy, wind energy, and hydropower. Among these, solar energy seems to be largely used due to the availability of solar energy in abundance, which in turn, has led to a mass increase in the number of solar power plants. Nevertheless, these solar panels are left unattended due to difficulty in the maintenance of solar power plants in large areas and the manual work required. It is important to find and replace these defective panels in time before any severe event occurs. Detecting hotspots, cracking and various other malfunctions in the photovoltaic cell can lead to an increase in the life of the solar panels by 5-10 years. In this paper, we propose a compact intelligent photovoltaic module diagnosis system based on a Deep Learning model that uses multilayer perceptron (MLP) to accurately identify and classify the faults in the photovoltaic module. We have achieved this by tweaking the VGG16 convolutional network architecture to detect 3 kinds of malfunctions – Hotspots, Cracking, Diode, and No-Anomaly for panels that are in good condition by using 2310 images for training, 665 for validation, and, 665 for testing. By using accuracy-score from sklearn.metrics, the model’s accuracy was estimated to be above 90 percent for classifying photovoltaic modules.

This paper introduces a comparative study for fault detection and classification on parallel transmission line using cascade forward and feed forward back propagation. Both calculations were based on discrete wavelet transform (DWT) and Clarke's transformation. Daubechies4 mother wavelet (Db4) was applied to decompose coefficients of wavelet transforms coefficients (WTC) and wavelet energy coefficients (WEC) of high frequency signals. The coefficients were inputs for training of neural network back-propagation (BPNN). The results showed that the feed forward back propagation algorithm of Artificial Neural Network (ANN) models responded better than Cascade forward back propagation algorithm models, particularly in fault detection and classification on parallel transmission. The results showed that the proposed method for fault analysis was able to classify all the faults on the parallel transmission line rapidly and correctly. Streszczenie. W pracy przedstawiono badanie porównawcze wykrywania i klasyfikacji uszkodzeń równoległej linii przesyłowej z wykorzystaniem propagacji kaskadowej do przodu i do tyłu. Oba obliczenia oparto na dyskretnej transformacie falkowej (DWT) i transformacji Clarke'a. Falkę macierzystą Daubechies4 (Db4) zastosowano do dekompozycji współczynników przekształceń falkowych (WTC) i współczynników energii falkowej (WEC) sygnałów wysokiej częstotliwości. Współczynniki stanowiły dane wejściowe do szkolenia propagacji wstecznej sieci neuronowej (BPNN). Wyniki pokazały, że algorytm propagacji wstecznego sprzężenia zwrotnego modeli sztucznej sieci neuronowej (ANN) zareagował lepiej niż modele algorytmu kaskadowego propagacji wstecznej, szczególnie w wykrywaniu błędów i klasyfikacji w transmisji równoległej. Wyniki pokazały, że zaproponowana metoda analizy uszkodzeń była w stanie szybko i poprawnie sklasyfikować wszystkie uszkodzenia na równoległej linii przesyłowej. Wykrywanie błędów w równoległej linii przesyłowej z wykorzystanirem transformaty Clarke'a

Wavelet transform is a transform which provides both the time and frequency information simultaneously. CWT (continuous wavelet transform) provides highly redundant data which needs large computation time and resource, so DWT (Discrete wavelet transform )came into picture which provides sufficient amount of information for analysis and synthesis of the original signal with a reduced computation time. In DWT, filters of different cut-off frequencies are used to analyse the input signal sequence. The signal is passed through different high pass and half band low pass filters. Filtering a signal is done by convolution ofthe signal sequence with impulse response of filters. Haar wavelet coefficients are used here as impulse response of filters. Then the output of low pass filters are further passed through the high pass filter and half band low pass filters to get the 2nd level DWT coefficients. We have designed level 2 DWT using Vedic multiplier (for convolution) in Xilinx 14.7 softwar...

Harmonics are the major issues in Variable Frequency Drive systems. The objective of project is the analysis of harmonics in the signatures of motor current in various fault conditions of Variable Frequency Drive systems. The simulation is done on VSI fed induction motor at constant voltage/frequency and was carried out using MATLAB/ Simulink. Variable Frequency drives commonly uses "Insulated Gate Bipolar Transistor" (IGBT).To get sine wave at required frequency IGBT uses "Pulse Width Modulation" technique for simulation. Simulation of an induction was developed and by observing the current, torque and speed responses, its dynamic response was verified to set up the acceptability of the developed system. Simulation of three various fault conditions were carried out here: one IGBTs gate signal is open circuited, one IGBT in the inverter section is blown off, line to ground fault occurs in one phase of the motor terminal. The technique of frequency domain analysis is used as a method to discriminate various faulty conditions. For various faulted conditions frequency responses characteristics were analyzed and collated to setup the use of Fast Fourier Transform algorithm to distinguish the characteristics of fault. Frequency responses of three various faulty states were emphatically different and were indubitably shown.

Electrical power system has a high impact on the services of the society and in order to ensure healthy operation of the power system, adequate protection systems are required to mitigate the effect of faults as faults are inevitable in power systems. This research work majors on the application of a distance protection scheme to transmission lines in order to identify a fault and issue a trip command to clear the fault in the least possible time. In power systems, load flow analysis is regarded as the mother of all engineering studies, thus it is imperative to perform a load flow to determine the existing state of the power system network, that is, bus voltages, phase angles, active and reactive power flow through various branches and loads under steady state conditions as well as line losses. Maximum and minimum short-circuit analysis were also required to determine the fault currents at various buses thereby ensuring adequate selection of protective devices. The method of distance protection scheme used was a three-stepped distance protection scheme in which the transmission line was segmented into three zones; 70%, 90% and 100% and the discrimination between each zone was 200ms in order to ensure proper coordination of the protective devices. It was observed that some of the buses were suffering from the effect of under-voltage, thus, capacitor banks were employed for voltage improvement thereby preventing overload. The software used for modelling and analysis is known as Electrical Transient Analyzer Program (ETAP).

With an increasing capacity of inverter-based generation and with a 100% renewable energy power system on the horizon, grid forming converters have the potential to become the prevalent control mode in the grid. Thus, the correct performance of these devices is going to be crucial for system stability and security of supply. Most research related to the grid-forming control is focused on normal operating conditions, although significant effort has been devoted to current limitation strategies to ensure Low Voltage Ride Through (LVRT) capability. However, most contributions usually consider only balanced faults. This paper, proposes a new current limiting method based on the well-known threshold virtual impedance (TVI) that keeps the voltage source behaviour associated to the grid forming (GFM) capability, even when the current limit is reached, while reducing the voltage unbalance according to user-defined settings.

A new transmission line protection scheme, the Boundary Protection, based on wavelet transform algorithm and without communicotion channel is proposed. Simulation studies show that using the proposed algarithm any type af internal fauit even with high fault resistance or voitage approaching zero can be correctly Identified and discriminated from the directly grounded external ones with maximum vaitage. Besides having high reliability, this scheme displays satisfactory sensitivity. Furthermore, this algorithm can be easily implemented for real-time process with high performance hardware.

The occurrence of faults in any operational power system network is inevitable, and many of the causative factors such as lightning, thunderstorm among others is usually beyond human control. Consequently, there is the need to set up models capable of prompt identification and classification of these faults for immediate action. This paper, explored the use of artificial neural network (ANN) technique to identify and classify various faults on the 11 kV distribution network of University of Lagos. The ANN is applied because it offers high speed, higher efficiency and requires less human intervention. Datasets of the case study obtained were sectioned proportionately for training, testing, and validation. The mathematical formulations for the method are presented with python used as the programming tools for the analysis. The results obtained from this study, for both the voltage and current under different scenarios of faults, are displayed in graphical forms and discussed. The resu...

An approach to line protection that is based on fault characteristics extracted by means of wavelet transforms is presented. Analysis of the spectral energy of the phase voltage signals for different frequency bands enables faults to be detected, classified and faulted phases selected. The proposed approach points to the development of ultra-high-speed protective relaying and has been applied to a series-compensated line. The system has been simulated with MICROTRAN and the protection with MATLAB. The results show the feasibility of implementing this algorithm.

This paper proposed a new algorithm for fault location and classification using wavelet based on Clarke’s transformation to obtain the fault current. This novel method of fault current approach is studied by comparing the use of the glide path of the fault voltage. The current alpha and beta (Current Mode) were used to transform the signal using discrete wavelet transform (DWT). The fault location was determined by using the Clarke’s transformation, and then turned into a wavelet, which was very precise and thorough. The most accurate was the mother wavelet Db4 which had the fastest time and smallest error detection when compared with the other wavelet mothers. In this study, the Clarke’s transformation is also compared with the Karenbauer’s, which has produced results with similar error percentage. The simulation results using PSCAD / EMTDC software showed that the proposed algorithm could distinguish internal and external faults to get the current signal in the transformation of a...

Insulation failure within the windings of a power transformer arises due to over voltages, under voltages, notches, glitches, etc. A widely used technique to detect these power transformer insulation failures during an impulse test is the comparison of neutral current waveforms. Any shift in the recorded waveforms between reduced and full voltage confirms the existence of a fault in the windings. Hence, a proper analysis of neutral current waveforms is necessary to assess the condition of the insulation of power transformer winding. In order to carry out a wave shape analysis, a 61 MVA, 11.5/230 kV generator transformer is used and faults are created in the disks of high-voltage windings at specific locations. Neutral currents are recorded by applying a 100 V, low-voltage impulse. Noise inherent in the neutral current during recordings is isolated using a biorthogonal wavelet and the denoised signals are analyzed using the Shannon wavelet for identification of a fault in the winding.

The research article describes a system of machine architecture for protection of transmission lines of HVDC, in which multiple models of ML (KNN and SVM) are employed for fault classification and recognition. The K-Nearest Neighbor classifier is intended to serve two functions. It detects the type of fault as well which serves as a backup module for the starting unit's doubtful fault declaration. From a single-end single measurement, a feature vector consisting of standard deviations gradients and of DC and harmonic current, DC voltage, and correlation coefficients is retrieved. By simulating different states that are non-fault and fault states on a data set having training and test cases are obtained. The ML algorithm is trained in MATLAB and evaluated on a total of 2220 severe instances. The acquired results demonstrated the efficacy of the suggested method in detecting and distinguishing between various internal and external/pseudo problems.In this paper we will discuss how ...

With the rapid advancement of power electronic technologies and the reduction of photovoltaic cell price, the share of solar energy in the total power production has been booming recently. On the one hand, the increase in the amount of power delivered by solar energy can be beneficial in many economic and environmental aspects. On the other hand, this can cause various technical challenges to network operators. One of these issues is related to classifying faults located in distribution networks with high penetration of photovoltaic systems. Although many studies have paid significant attention to developing new algorithms applicable for a more active today distribution networks, there is still space for other improvements. Hence, after reviewing stateof-the-art researches, this paper was intended to develop a fault classification that is based on artificial neural networks. In particular, a technique so-called Multiplayer Perceptron Classifier was selected for the proposed algorithm. First, the authors generated a data set for the study by modeling and simulating a real distribution network with practical parameters provided by a local utility in the environment software PowerFactory/DigSILENT. Multiple fault scenarios were simulated. Second, a part of the generated data collection was used for network learning. Finally, the performance of the proposed methodology was demonstrated via testing on the remaining number of generated data.

Small and large disturbances in power system may sometimes lead to sustained oscillation in three-phase power, often termed as 'power swing' in the literature. Power swing is a balanced phenomenon and may be indistinguishable from a three-phase fault. Transmission lines being protected by distance relays may be incorrectly de-energized because of a power swing. So, three-phase faults should be discriminated from power swing with high reliability and dependability. In this study, a novel symmetrical fault detection technique has been proposed with dual use line relays for the first time. The proposed algorithm has been verified for a large number of fault scenarios including series compensated lines with single machine infinite bus and western system coordinating council-9 bus system. It has been seen that the proposed algorithm is giving fast detection of three-phase faults during power swing with high accuracy.

Application of two new ANN-based algorithms for arcing high impedance fault (HIF) detection in multigrounded medium-voltage (MV) distribution networks is presented in this paper. The paper provides an evaluation of two new structures of artificial neural networks (ANNs) that may be used for reliable HIF detection in multigrounded as well as isolated, compensated, and grounded via small resistance distribution grids. The results obtained by use of both neural nets are presented. The performance was tested using data obtained from staged HIFs in real MV network as well as from Electromagnetic Transients Program-Alternative Transients Program simulations. A small number of necessary neurons in developed ANNs, short measuring sliding data window, and easy interpretation of obtained output signals are the main advantages of the proposed approach. Satisfactory results of ANN performance were observed for all examined HIF cases in which the ground fault current was greater than 16 A. The selected ANNs of best performance show high reliability and immunity to transients resulting from switching operations in protected feeders and from capacitor bank switching. Index Terms-Artificial neural networks (ANNs), high impedance faults (HIFs), medium-voltage (MV) networks, transient analysis.

The research article describes a system of machine architecture for protection of transmission lines of HVDC, in which multiple models of ML (KNN and SVM) are employed for fault classification and recognition. The K-Nearest Neighbor classifier is intended to serve two functions. It detects the type of fault as well which serves as a backup module for the starting unit's doubtful fault declaration. From a single-end single measurement, a feature vector consisting of standard deviations gradients and of DC and harmonic current, DC voltage, and correlation coefficients is retrieved. By simulating different states that are non-fault and fault states on a data set having training and test cases are obtained. The ML algorithm is trained in MATLAB and evaluated on a total of 2220 severe instances. The acquired results demonstrated the efficacy of the suggested method in detecting and distinguishing between various internal and external/pseudo problems.In this paper we will discuss how ...

India's power sector is characterised by inadequate and inefficient power supply. Since the country's independence, consumers are confronted with frequent power cuts, and fluctuating voltages and frequencies. In addition, system losses are high throughout India's Transmission and distribution networks. Distribution Sector requires economical system to provide electrical energy at a suitable prize and at a minimum voltage drop to reduce the voltage regulation. So, we require the economical way to provide the electrical energy by State Electricity Boards to various consumers at minimum voltage drop and reduce the regulation of voltage required for load points, tie-points to select respective kVA capacity of transformers and hence the installation of suitable capacitor banks with proper locations for improvement of power factor and harmonics. This paper suggests the different methods of reduction of distribution losses in the 11kV urban distribution feeder to improve the voltage profile.

There is continuous variation in amplitude, phase, and frequency of voltage and current signals during power swing in a power system. Such variation makes the memorized pre-fault data less accurate to be used as non-fault component while computing superimposed quantity for protection decisions during power swing. In this paper, the voltage and current signal characteristics during power swing are analyzed and a technique is proposed to obtain correct non-fault components for superimposed quantity computation. The accuracy of the method in estimating the non-fault components is observed for different power swing conditions simulated in a 39-bus New England system and field data. The relay performance improvement as observed is also provided while applying the method to directional relaying, fault type classification, and fault location applications for faults during power swing. Comparative assessment reveals the superiority of the proposed method in improving the accuracy of the superimposed components computed during power swing.

The research presented is motivated by the demand for process improvement for companies active within software development. High demands on software quality are a reality. At the same time, short development time and low effort consumption are required. This stresses the necessity for process improvement. Empirical research methods and close cooperation with the industry partner addressed the research challenge. The research presented in this thesis shows how the analysis of faults through fault classification can be used to determine suitable and required process improvements. Two alternatives are investigated. First, a lightweight approach, and second a fault classification approach targeting all faults. The suitability of the fault classification is stressed as well as the importance of assigning the correct fault class. The latter is determined by classifier agreement calculations. Additionally, the research proposes that the appropriate occasion for a correct fault classificati...

— Proper detection of various faults occurring on the transmission line is very essential. In this paper, detection and classification of some these faults is done based on the information conveyed by the wavelet analysis of power systems transients. Maximum norm values, maximum detail coefficient, energy of the current signals are calculated from the Wavelet Toolbox in MATLAB/Simulink. Maximum norm value and energy of the signals detects the fault and threshold detail coefficient classifies the fault

The occurrence of faults in any operational power system network is inevitable, and many of the causative factors such as lightning, thunderstorm among others is usually beyond human control. Consequently, there is the need to set up models capable of prompt identification and classification of these faults for immediate action. This paper, explored the use of artificial neural network (ANN) technique to identify and classify various faults on the 11 kV distribution network of University of Lagos. The ANN is applied because it offers high speed, higher efficiency and requires less human intervention. Datasets of the case study obtained were sectioned proportionately for training, testing, and validation. The mathematical formulations for the method are presented with python used as the programming tools for the analysis. The results obtained from this study, for both the voltage and current under different scenarios of faults, are displayed in graphical forms and discussed. The resu...

This paper presents overview on methods which is used for locating faults on the HVDC transmission lines. Different methods like Distance Protection Method, Discrete Fourier Transform (DFT), Hilbert-Huang Transform Method, Independent Component Method, Fuzzy Logic Method, Wavelet Transform (WT), Natural Frequency Method, Artificial Neural Network (ANN) etc are used. This paper provides the description of all methods.

In order to meet the high demand for power transmission capacity, some power companies have installed series capacitor on power transmission line. The introduction of the capacitance in series with the line reactance allows reduction in line impedance, thus resulting increased power transmission capability. It provides the benefits of increased system capacity, reduced system losses and voltage regulation. The paper presents series compensated transmission line model by using PSCAD. Results are tested with different fault locations for compensated and without compensated transmission line. Performance of the developed method ensures that the proposed scheme gives the apparent effects when compensation is going to be added in the line.

Fault analysis (detection, classification and location) of transmission network is of great importance in power system. A Wavelet-Alienation-Neural (WAN) technique has been developed for the fault analysis of Unified Power Flow Controller (UPFC) compensated transmission network. The detection and classification of various outages are accomplished by alienation of wavelet based approximate coefficients computed from current signals. The precise location of faults is carried out by an Artificial Neural Network fed from estimated approximate coefficients computed from voltage and current signals of the same quarter cycle. The robustness of the algorithm is proved with the case studies of varying fault locations, sampling frequency, system parameters, effects of noise, fault incipient angle, different control strategies and fault path impedances. INDEX TERMS Fault detection, fault classification, fault location, unified power flow controller (UPFC).

This paper investigates the possibility of identifying classes by clustering. This study includes employing Self-Organizing Maps (SOM) in identifying clusters from EEG signals that could then be mapped to emotional classes. Beginning by training varying sizes of SOM with the EEG data provided from the public dataset: DEAP. The produced graphs showing Neighbor Distance, Sample Hits, and Weight Position are examined. Following that, the ground-truth label provided in DEAP is tested, in order to identify correlations between the label and the clusters produced by the SOM. The results show that there is a potential of class discovery using SOM-based clustering. It is then concluded that by evaluating the implications of this work and the difficulties in evaluating its outcome.

With an increasing capacity of inverter-based generation and with a 100% renewable energy power system on the horizon, grid forming converters have the potential to become the prevalent control mode in the grid. Thus, the correct performance of these devices is going to be crucial for system stability and security of supply. Most research related to the grid-forming control is focused on normal operating conditions, although significant effort has been devoted to current limitation strategies to ensure Low Voltage Ride Through (LVRT) capability. However, most contributions usually consider only balanced faults. This paper, proposes a new current limiting method based on the well-known threshold virtual impedance (TVI) that keeps the voltage source behaviour associated to the grid forming (GFM) capability, even when the current limit is reached, while reducing the voltage unbalance according to user-defined settings.

We describe an approach for testing a software system for possible security flaws. Traditionally, security testing is done using penetration analysis and formal methods. Based on the observation that most security flaws are triggered due to a flawed interaction with the environment, we view the security testing problem as the problem of testing for the fault-tolerance properties of a software system. We consider each environment perturbation as a fault and the resulting security compromise a failure in the toleration of such faults. Our approach is based on the well known technique of fault-injection. Environment faults are injected into the system under test and system behavior observed. The failure to tolerate faults is an indicator of a potential security flaw in the system. An Environment-Application Interaction (EAI) fault model is proposed. EAI allows us to decide what faults to inject. Based on EAI, we present a security-flaw classification scheme. This scheme was used to classify 142 security flaws in a vulnerability database. This classification revealed that 91 of the security flaws in the database are covered by the EAI model.

This paper discusses how to use the warping information obtained after batch synchronization for process monitoring and fault classification. The warping information can be used for i) building unsupervised control charts or ii) fault classification when a rich faulty batches database is available. Data from realistic simulations of a fermentation process of the Saccharomyces cerevisiae cultivation are used to illustrate the proposal.

Power systems are the largest and most complex human made systems, consisting of thousands of electrical sources, loads, transmission and distribution lines, power transformers, circuit breakers, etc. where faults always occurred. Faults can cause personnel and equipment safety problems, and can result in significant disruption to power supply and thus financial losses. In this paper we will present comprehensive mathematical suite to detect and classify fault dependent models of various types of power systems. This work will extract fault unique signatures by using polarization ellipse during the healthy condition and the polarization will be circular shape with radius equal the rated voltage of the system, but during the fault condition the polarization will be ellipse shape and the fault signature will be defined according the ellipse parameters major axis, minor axis, ellipticity and orientation angle, by using least squares criterion will define the ellipse parameters this system will identify and classify. This paper will be a milestone for extended paper based on the proposed mathematical modelling and applying it to identify, classify and localize with simulation model.

— Proper detection of various faults occurring on the transmission line is very essential. In this paper, detection and classification of some these faults is done based on the information conveyed by the wavelet analysis of power systems transients. Maximum norm values, maximum detail coefficient, energy of the current signals are calculated from the Wavelet Toolbox in MATLAB/Simulink. Maximum norm value and energy of the signals detects the fault and threshold detail coefficient classifies the fault

Tez (Yüksek Lisans) -- İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, 2002Thesis (M.Sc.) -- İstanbul Technical University, Institute of Science and Technology, 2002Bu çalışmada, dağıtım sistemlerinde hibrid “Dalgacık-Yapay Sinir ağı (YSA) tabanlı” bir yaklaşımla arıza sınıflama işlemi gerçeklenmiştir. 34.5 kV “Sağmalcılar-Maltepe” dağıtım sistemi PSCAD/EMTDC yazılımı kullanılarak arıza sınıflayıcı için gereken veri üretilmiştir. Tezin amacı, on farklı kısa-devre sistem arızalarını tanımlayabilecek bir sınıflayıcı tasarlamaktır. Sistemde kullanılan arıza işaretleri 5 kHZ lik örnekleme frekansı ile üretilmiştir. Farklı arıza noktaları ve farklı arıza oluşum açılarındaki hat-akımları ve hat-toprak gerilimlerini içeren sistem arızaları ile bir veritabanı oluşturulmuştur. “Çoklu-çözünürlük işaret ayrıştırma” tekniği kullanılarak altı-kanal akım ve gerilim örneklerinden karakteristik bigi çıkarılmıştır. PSCAD/EMTDC ile üretilen veri bu şekilde bir ön islemden geçirildikten sonra Y...