Sundaravaradan Navalpakkam Ananthan

Bio: Sundaravaradan Navalpakkam Ananthan is an academic researcher from University of Texas at Austin. The author has contributed to research in topics: Fault (power engineering) & Electric power system. The author has an hindex of 7, co-authored 18 publications receiving 143 citations. Previous affiliations of Sundaravaradan Navalpakkam Ananthan include National Institute of Technology, Tiruchirappalli.

Real-time fault analysis of transmission lines using wavelet multi-resolution analysis based frequency-domain approach

Abstract: In a power system, transmission lines are prone to faults of different nature, which challenge the system stability and reliability. Thus system performance analysis under such fault conditions has drawn attention of researchers. Particularly, the advent of fast and efficient data acquisition using higher sampling rates combined with high speed digital signal processors has paved the way for efficient digital real-time simulations. Though sustained efforts have been made by different researchers to develop some good real-time digital simulators, this study is an attempt to implement a laboratory prototype model of a 20 V, 200 km transmission line, representing a 400 kV extra-high voltage transmission line, so as to improve the real time performance. In addition, efficient National Instruments based data acquisition system in conjunction with LabVIEW has been incorporated to acquire the best possible representative data with commensurate characterisation and transmission with fidelity. The unique contributions of this real-time fault analysis laboratory hybrid model are accurate fault detection and classification using a frequency-domain approach having immunity to fault impedance and fault inception angle, which affect the time-domain analyses severely. It is also equipped with visual displays so that even non-experts can use it for planning and decision-making purposes.

Fault Classification in Transmission Lines Using Wavelet Multiresolution Analysis

Abstract: Transmission lines are the vital links between generating stations and distribution substations, and it is essential to study, understand, and analyze the faults that occur in transmission lines. The isolation of the faulty line(s) is an important aspect of power system protection, as it not only prevents a calamitous situation but also protects the apparatus and increases its useful life.

Robust fault analysis in transmission lines using Synchrophasor measurements

Abstract: As more electric utilities and transmission system operators move toward the smart grid concept, robust fault analysis has become increasingly complex. This paper proposes a methodology for the detection, classification, and localization of transmission line faults using Synchrophasor measurements. The technique involves the extraction of phasors from the instantaneous three-phase voltages and currents at each bus in the system which are then decomposed into their symmetrical components. These components are sent to the phasor data concentrator (PDC) for real-time fault analysis, which is completed within 2–3 cycles after fault inception. The advantages of this technique are its accuracy and speed, so that fault information may be appropriately communicated to facilitate system restoration. The proposed algorithm is independent of the transmission system topology and displays high accuracy in its results, even with varying parameters such as fault distance, fault inception angle and fault impedance. The proposed algorithm is validated using a three-bus system as well as the Western System Coordinating Council (WSCC) nine bus system. The proposed algorithm is shown to accurately detect the faulted line and classify the fault in all the test cases presented.

A wavelet-adaptive network based fuzzy inference system for location of faults in parallel transmission lines

Abstract: In a power system, transmission line are exposed to the wrath of the nature, which makes it susceptible to the occurrence of faults. Such faults have to be cleared at the earliest to deliver utmost continuity of supply to the consumers. In this respect, the primary aim is to produce a methodology which supports numeric relays in precisely detecting, classifying and locating the fault in transmission lines within the minimum possible time. This paper uses a digital signal processing (DSP) based frequency domain approach to overcome the limitations of fault inception angle and fault location associated with conventional time domain relaying methodologies. The proposed methodology uses wavelet multi-resolution analysis to extract harmonic features from the current signal at the instant of occurrence of fault. It further aims at enhancing the performance of numeric relays in locating the faults in parallel transmission lines by using intelligent techniques. The paper proposes and validates a fault locating methodology using wavelet multi-resolution analysis and adaptive neuro-fuzzy inference system (ANFIS) on a two bus parallel transmission system using MATLAB. Various fault inception angles are considered in the case studies for proving the efficacy of the proposed methodology. This paper would support organizations developing numeric relays for accurate localization of transmission line faults.

Relay performance verification using fault event records

Abstract: Event reports recorded by intelligent electronic devices (IEDs) such as digital relays and fault recorders during disturbances depict the status and system parameters of the power system. Incorrect relay settings and unknown system parameters can lead to relay misoperation but information regarding these are available by performing a comprehensive analysis of fault records. Hence, it is necessary to regularly make a comprehensive assessment of the functioning of the relay to ensure reliable operation. The objective of this paper is to demonstrate various aspects of evaluating relay performance and verifying circuit parameters which are used in relay settings using field data in two case studies. While scrutinizing the relay’s operation, this paper also presents key insights on verifying circuit parameters using the same relay event records. The lessons learned from the case studies presented in this paper will equip a protection engineer to inspect the operation of a relay during a transmission line fault, help gain a better understanding of the fault event and take possible actions to prevent future occurrence of similar events.

Deep-Learning-Based Fault Classification Using Hilbert–Huang Transform and Convolutional Neural Network in Power Distribution Systems

Abstract: Fault classification is important for the fault cause analysis and faster power supply restoration. A deep-learning-based fault classification method in small current grounding power distribution systems is presented in this paper. The current and voltage signals are sampled at a substation when a fault occurred. The time-frequency energy matrix is constructed via applying Hilbert–Huang transform (HHT) band-pass filter to those sampled fault signals. Regarding the time-frequency energy matrix as the pixel matrix of digital image, a method for image similarity recognition based on convolution neural network (CNN) is used for fault classification. The presented method can extract the features of fault signals and accurately classify ten types of short-circuit faults, simultaneously. Two simulation models are established in the PSCAD/EMTDC and physical system environment, respectively. The performance of the presented method is studied in the MATLAB environment. Various kinds of fault conditions and factors including asynchronous sampling, different network structures, distribution generators access, and so on are considered to verify the adaptability of the presented method. The results of investigation show that the presented method has the characteristics of high accuracy and adaptability in fault classification of power distribution systems.

Protection scheme for power transmission lines based on SVM and ANN considering the presence of non-linear loads

Abstract: The proliferation of non-linear loads in recent times has a strong bearing on power quality due to the presence of harmonics in the power system. Harmonics lead to erroneous measurement and malfunctioning of protective relays, thus reducing the efficacy of transmission line protection. With the aim of improving the dependability of the protection scheme under varying non-linear loading condition, the study presents a hybrid support vector machine (SVM), artificial neural network (ANN) and Kalman filter based algorithm using voltage harmonics for the protection of three-phase transmission line. The post-fault voltage signals are processed by a Kalman filter to estimate the harmonic components, which serve as feature vectors for performing the fault detection and classification by SVM and zone identification as well as the location by ANN. The harmonic information discriminates faults from disturbances based on variations in the fundamental component, to improve the selectivity and accuracy. Considering the stochastic nature of fault occurrence in power systems, the efficacy of the proposed scheme is validated using Monte Carlo simulation. The experimental results confirm the superiority of using voltage harmonics for improving the dependability of transmission line protection.

ArcNet: Series AC Arc Fault Detection Based on Raw Current and Convolutional Neural Network

Abstract: AC series arc is dangerous and can cause serious electric fire hazards and property damage. This article proposed a convolutional neural network -based arc detection model named ArcNet. The database of this research is collected from eight different types of loads according to IEC62606 standard. The two most common types of arcs, including arcs from a loose connection of cables and those caused by the failure of the insulation, are generated in testing and included in the database. Using the database of raw current, experimental results indicate ArcNet can achieve a maximum of 99.47% arc detection accuracy at 10 kHz sampling rate. The model is also implemented in Raspberry Pi 3B for classification accuracy. A tradeoff study between the arc detection accuracy and model runtime has been conducted. The proposed ArcNet obtained an average runtime of 31 ms/sample of 1 cycle at 10 kHz sampling rate, which proves the feasibility of practical hardware deployment for real-time processing.