Design and Development of Neural Network based Controller for the Fault Detection in 33kV-0.4kV Transmission Line

Abstract

The relentless growth of electrical power systems mandates the development of efficient fault detection mechanisms to ensure the reliability and stability of the grid. This research presents a novel approach to three-phase fault detection using a neural network controller implemented within the MATLAB Simulink environment. The proposed model leverages the capabilities of neural networks to accurately identify and classify faults in real-time, contributing to the robustness of power system operation. The heart of the developed solution is a meticulously designed neural network architecture, trained on a comprehensive dataset generated through meticulous simulations of various fault scenarios. Leveraging the advantages of deep learning, the neural network demonstrates its proficiency in discriminating between healthy and faulted system states with high precision. The model\'s adaptability and capability to handle noise and dynamic variations in fault characteristics underline its efficacy in practical deployment. To assess the model\'s performance, extensive comparative analyses are conducted against existing fault detection methods. The results underscore the superiority of the proposed neural network controller, showcasing its ability to detect and classify three-phase faults swiftly and accurately. Furthermore, the controller\'s generalization potential is evaluated through rigorous cross-validation procedures, affirming its reliability in diverse operating conditions.

Introduction

I. INTRODUCTION

Any part of a power system, such as generating units, transformers, the transmission network, and/or loads, might experience frequent breakdowns. Defects can substantially disrupt supply, unsettle the whole system, and even cause crew deaths, as is well acknowledged. Defect identification is therefore crucial from an operational and financial point of view. Defects should be identified as soon as feasible, ideally in real time, to allow for speedy remedial action before substantial power supply disruptions happen. Neural networks are built on neurophysical models of human brain cells and their connections. A distinguishing trait of such networks is their exceptional pattern recognition and learning abilities. The primary advantage of neural networks is their ability to learn on their own.

Simple neurons stacked in layers and often connected together make form an Artificial Neural Network (ANN), which is modelled after biological structures. It shows the feed-forward ANN structure, often known as the perceptron. The inputs to the Ni number of neurons in each ith layer are connected to the neurons in the layer below. The input layer receives the excitation pulses. To put it simply, an elementary neuron is like a processor that produces an output by performing a simple non-linear operation to its inputs. An ANN may be trained by altering the weights in accordance with the training set. Every neuron has a weight associated with it. An Artificial Neural Network may be taught to respond based on inputs by altering the node weights. Therefore, we need a collection of data known as the training data set in order to train the neural network.

Neural networks are built on neurophysical models of the interactions between human brain cells. These networks are exceptionally good at recognising patterns and picking up new information. One of the key benefits of neural networks is their ability to learn on their own. The network is initially provided with the appropriate input and output values. An Artificial Neural Network (ANN) is a collection of fundamental neurons that are frequently connected in topologies that draw inspiration from biology and organised in a number of layers.

Conclusion

In this study, a comprehensive investigation into three-phase fault detection utilizing a neural network controller within the MATLAB Simulink framework was undertaken. The aim was to enhance the reliability and precision of fault detection mechanisms in electrical power systems. The model\'s remarkable performance, as evidenced by a Mean Squared Error (MSE) of 0.170 and a high Regression Coefficient of 0.9868, attests to its efficacy in accurately identifying and classifying faults. The key strength of the developed neural network controller lies in its ability to learn complex fault patterns and generalize its knowledge to diverse fault scenarios. By leveraging a meticulously generated dataset encompassing a wide array of fault types and system conditions, the neural network demonstrated a commendable aptitude for swiftly discerning between healthy and faulted states.The reported results substantially surpass existing fault detection methodologies, underscoring the model\'s practical utility in real-world scenarios. The high regression coefficient reflects the strong correlation between the model\'s predictions and the actual fault occurrences. These outcomes validate the potential of the proposed approach to bolster the resilience and reliability of power systems against three-phase faults.While this study attains satisfactory performance metrics, there remain avenues for further exploration. Fine-tuning the neural network architecture and incorporating advanced training strategies may yield incremental enhancements. Additionally, testing the model on diverse datasets from various power system configurations and network sizes could validate its adaptability across a spectrum of operational contexts.The developed three-phase fault detection model, empowered by a neural network controller, offers a substantial leap forward in power system fault management. The convergence of machine learning and power engineering showcased in this research has the potential to revolutionize fault detection methodologies, ensuring the steadfast operation of electrical grids in the face of anomalies.

References

[1] Carrasco, J. M., Franquelo, L. G., Bialasiewicz, J. T., Galván, E., PortilloGuisado, R. C., Prats, M. M.,& Moreno-Alfonso, N. (2006). Power-electronic systems for the grid integration of renewable energy sources: A survey. IEEE Transactions on industrial electronics, 53(4),1002-1016. [2] Ganechari, S. M., & Kate, S. (2005). Alternative energy sources. Alternative Energy Sources, 5 [3] Madeti, S.R. and Singh, S.N., 2017. Online fault detection and theeconomic analysis of grid-connected photovoltaic systems. Energy, 134,pp.121-135 [4] Pinto Moreira de Souza, D., da Silva Christo, E., & Rocha Almeida, A.(2017). Location of faults in power transmission lines using the ARIMAmethod. Energies, 10(10), 1596. [5] Kumar Panigrahi, Basanta, et al. \"Islanding detection in a hybrid power system using continuous wavelet transform.\" 2017 International conference on Circuit, Power and Computing Technologies (ICCPCT). IEEE, 2017.. [6] Mir Mohammad Taheri, Heresh Seyedi, Behnam Mohammadi-ivatloo, \"DT-based relaying scheme for fault classification in transmission lines using MODP\", Generation Transmission & Distribution IET, vol. 11, no.11, pp. 2796-2804, 2017 [7] A. M. Abeid, H. A. A. El-Ghany and A. M. Azmy, \"An advanced traveling-wave fault-location algorithm for simultaneous faults,\" 2017 Nineteenth International Middle East Power Systems Conference (MEPCON), Cairo, 2017, pp. 747-752. [8] Gururajapathy, S. S., Mokhlis, H., & Illias, H. A. (2017). Fault location and detection techniques in power distribution systems with distributed generation: A review. Renewable and sustainable energy reviews, 74, 949-958 [9] https://journals.sagepub.com/doi/full/10.1177/0020294013510471 [10] Hui Hwang Goh, Transmission Line Fault Detection: A Review, Indonesian Journal of Electrical Engineering and Computer Science, Vol. 8, No. 1, October 2017, pp. 199 ~ 205 [11] Avagaddi Prasad, J. Belwin Edward, K. Ravi, “A review on fault classification methodologies in power transmission systems”: Part—I,Journal of Electrical Systems and Information Technology,Volume 5, Issue 1,2018,Pages 48-60, ISSN 2314-7172,https://doi.org/10.1016/j.jesit.2017.01.004. (https://www.sciencedirect.com/science/article/pii/S2314717217300065) [12] N Rosle, “Fault detection and classification in three phase series compensated transmission line using ANN”, ICE4CT 2019 Journal of Physics: Conference Series 1432 (2020) 012013 IOP Publishing doi:10.1088/1742-6596/1432/1/012013 [13] Sahilbhai Vhora, Underground Cable Fault Detection Using Matlab, International Journal of Innovative Research in Science, Engineering and Technology Vol. 7, Issue 12, December 2018 [14] Majid Jamil, Sanjeev Kumar Sharma, D. K. Fault Classification of Three-Phase Transmission Network using Genetic Algorithm. nternational Journal of Engineering and Applied Sciences (IJEAS) ISSN: 2394-3661, Volume-2, Issue-7, July 2015 [15] M. Sanaye-Pasand, H. Khorashadi-Zadeh, “Transmission Line Fault Detection & Phase Selection using ANN”, International Conference on Power Systems Transients – IPST 2003 in New Orleans, USA

Copyright

Copyright © 2023 Harshit Choubey, Arun Pachori. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.