Fault location with the highest possible accuracy has a significant role in expediting the restoration process. This thesis provides a fault detection, classification and location methods using machine learning tools and advanced signal processing for radial distribution grid. The three phase current signals, one cycle before and one cycle after the inception of the fault are measured at the sending end of the grid. Discrete wavelet transform (DWT) is employed to extract useful features from the three phase current signal. Standard statistical techniques are then applied onto DWT coefficients in order to extract the useful features. Among many features mean, standard deviation, energy, skewness, kurtosis and entropy are evaluated and fed into the artificial neural network (ANN), Multilayer perceptron (MLP) and Extreme learning machine (ELM), to identify the fault type and its location. During the training process, all types with variations in the loading and fault resistance are considered. The performance of the proposed fault locating methods evaluated in terms the root mean absolute percentage error (MAPE), root mean squared error (RMSE), and Willmott’s index of agreement (WIA) ,coefficient of determination (R^2), and Nash-Sutciffe model efficiency coefficients (NSEC). The time it takes for training and testing are also considered. It is found that discrete wavelet transform with machine learning is very accurate and reliable method to for fault classifying and locating in both balanced and unbalanced radial system. 100% fault detection accuracy is achieved for all types of faults. Except the slight confusion of 3LG and 3L faults, 100% classification accuracy is also achieved. The performance measures shows that both MLP and ELM are very accurate and comparative in locating faults. The method can be further applied for meshed networks with multiple distributed generator.

Fault location with the highest possible accuracy has a significant role in expediting the restoration process. This thesis provides a fault detection, classification and location methods using machine learning tools and advanced signal processing for radial distribution grid. The three phase current signals, one cycle before and one cycle after the inception of the fault are measured at the sending end of the grid. Discrete wavelet transform (DWT) is employed to extract useful features from the three phase current signal. Standard statistical techniques are then applied onto DWT coefficients in order to extract the useful features. Among many features mean, standard deviation, energy, skewness, kurtosis and entropy are evaluated and fed into the artificial neural network (ANN), Multilayer perceptron (MLP) and Extreme learning machine (ELM), to identify the fault type and its location. During the training process, all types with variations in the loading and fault resistance are considered. The performance of the proposed fault locating methods evaluated in terms the root mean absolute percentage error (MAPE), root mean squared error (RMSE), and Willmott’s index of agreement (WIA) ,coefficient of determination (R^2), and Nash-Sutciffe model efficiency coefficients (NSEC). The time it takes for training and testing are also considered. It is found that discrete wavelet transform with machine learning is very accurate and reliable method to for fault classifying and locating in both balanced and unbalanced radial system. 100% fault detection accuracy is achieved for all types of faults. Except the slight confusion of 3LG and 3L faults, 100% classification accuracy is also achieved. The performance measures shows that both MLP and ELM are very accurate and comparative in locating faults. The method can be further applied for meshed networks with multiple distributed generator.

[1] Mirzaei M, Kadir MZAA, Moazami E, Hizam H. Review of fault location methods for distribution power system. Australian Journal of Basic & Applied Sciences 2009;3:2670.
[2] Shafiullah M, Abido MA. A review on distribution grid fault location techniques. electric power components and systems 2017;45:807–24. doi:10.1080/15325008.2017.1310772.
[3] Girgis AA, Zhu J, Lubkeman DL. Automated fault location and diagnosis on electric Power Distribution Feeders. IEEE Transactions on Power Delivery 1997;12.
[4] Mieee NY. Fault detection techniques for power transformer Nagireddy Ravi2 2007:1–9.
[5] Weber E. Traveling waves on transmission lines. Electrical Engineering 2013;61:302–9. doi:10.1109/ee.1942.6436308.
[6] Shafiullah M, Abido MA, Al-Hamouz Z. Wavelet-based extreme learning machine for distribution grid fault location. IET Generation, Transmission & Distribution 2017;11:4256–63. doi:10.1049/iet-gtd.2017.0656.
[7] Yan F, Liu W, Tian L. Fault location for 10kV distribution line based on traveling wave-ANN theory. PEAM 2011 - Proceedings: 2011 IEEE Power Engineering and Automation Conference 2011;2:437–40. doi:10.1109/PEAM.2011.6135094.
[8] Shafiullah M, Abido MA, Abdel-Fattah T. Distribution grids fault location employing ST based optimized machine learning approach. Energies 2018;11. doi:10.3390/en11092328.
[9] Ray P, Panigrahi BK, Senroy N. Extreme learning machine based fault classification in a series compensated transmission line. PEDES 2012 - IEEE International Conference on Power Electronics, Drives and Energy Systems 2012:1–6. doi:10.1109/PEDES.2012.6484297.
[10] Malathi V, Marimuthu NS, Baskar S. Intelligent approaches using support vector machine and extreme learning machine for transmission line protection. Neurocomputing 2010;73:2160–7. doi:10.1016/j.neucom.2010.02.001.
[11] Heidari Bafroui H, Ohadi A. Application of wavelet energy and Shannon entropy for feature extraction in gearbox fault detection under varying speed conditions. Neurocomputing 2014;133:437–45. doi:10.1016/j.neucom.2013.12.018.
[12] Malik H, Mishra S. Extreme learning machine based fault diagnosis of power transformer using IEC TC10 and its related data. 12th IEEE International Conference Electronics, Energy, Environment, Communication, Computer, Control: (E3-C3), INDICON 2015 2016:1–5. doi:10.1109/INDICON.2015.7443245.
[13] Zhang L, Yuan J. Fault diagnosis of power transformers using kernel based extreme learning machine with particle swarm optimization. Applied Mathematics and Information Sciences 2015;9:1003–10. doi:10.12785/amis/090251.
[14] Wong PK, Yang Z, Vong CM, Zhong J. Real-time fault diagnosis for gas turbine generator systems using extreme learning machine. Neurocomputing 2014;128:249–57. doi:10.1016/j.neucom.2013.03.059.
[15] Stetco A, Dinmohammadi F, Zhao X, Robu V, Flynn D, Barnes M, et al. Machine learning methods for wind turbine condition monitoring: A review. Renewable Energy 2019;133:620–35. doi:10.1016/j.renene.2018.10.047.
[16] Artificial Neural Network Building Blocks. TutorialspointCom n.d.
[17] Zhang D, Lin Z, Gao Z. A novel fault detection with minimizing the noise-signal ratio using reinforcement learning. Sensors (Switzerland) 2018;18:1–29. doi:10.3390/s18093087.
[18] Hastie T, Tibshirani R, Friedman J. Machine Learning Book. vol. 27. 2009. doi:10.1007/978-0-387-84858-7.
[19] A.Adeoti O, A. Osanaiye P. Effect of training algorithms on the performance of ANN for pattern recognition of bivariate process. International Journal of Computer Applications 2013;69:8–12. doi:10.5120/12085-8031.
[20] Kateris D, Moshou D, Pantazi XE, Gravalos I, Sawalhi N, Loutridis S. A machine learning approach for the condition monitoring of rotating machinery. Journal of Mechanical Science and Technology 2014;28:61–71. doi:10.1007/s12206-013-1102-y.
[21] Tang J, Deng C, Huang G Bin. Extreme learning machine for multilayer perceptron. IEEE Transactions on Neural Networks and Learning Systems 2016;27:809–21. doi:10.1109/TNNLS.2015.2424995.
[22] Guang-Bin Huang, Qin-Yu Zhu, Chee-Kheong Siew. Extreme learning machine: a new learning scheme of feedforward neural networks. 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat No04CH37541) 2014;2:985–90. doi:10.1109/IJCNN.2004.1380068.
[23] Ding S, Zhao H, Zhang Y, Xu X, Nie R. Extreme learning machine: algorithm, theory and applications. Artificial Intelligence Review 2015;44:103–15. doi:10.1007/s10462-013-9405-z.
[24] Nair NK, Asharaf S. Tensor Decomposition based approach for training extreme learning machines. Big Data Research 2017;10:8–20. doi:10.1016/j.bdr.2017.07.002.
[25] Jurado F, Valverde M. Applications of signal processing tools in a power systems Course. International Journal of Electrical Engineering & Education 2013;41:28–42. doi:10.7227/ijeee.41.1.3.
[26] Kristomo D, Hidayat R, Soesanti I. Feature extraction and classification of the Indonesian syllables using Discrete Wavelet Transform and statistical features. Proceedings - 2016 2nd International Conference on Science and Technology-Computer, ICST 2016 2017:88–92. doi:10.1109/ICSTC.2016.7877353.
[27] Ray P, Panigrahi BK, Senroy N. Hybrid methodology for fault distance estimation in series compensated transmission line. Transmission Distribution IET Generation 2013;7:431–9. doi:10.1049/iet-gtd.2012.0243.
[28] Marsh CP. Introduction to Continuous Entropy, 2013.
[29] Styvaktakis E, Bollen MHT, Irene I, Gu YH. A fault location technique using high frequency fault clearing transients. IEEE Power Engineering Review 1999;19:58–60. doi:10.1109/39.761821.
[30] Fourier Transform (FT). Questions and answers in MRI n.d. http://mriquestions.com/fourier-transform-ft.html (accessed July 25, 2019).
[31] Shafiullah M, Abido MA. A review on distribution grid fault location techniques. Electric Power Components and Systems 2017;45:807–24. doi:10.1080/15325008.2017.1310772.
[32] Continuous wavelet transform and scale-based analysis - MATLAB & Simulink n.d. https://www.mathworks.com/help/wavelet/gs/continuous-wavelet-transform-and-scale-based-analysis.html (accessed July 25, 2019).
[33] Rioul O, Vetterli M. Wavelets and Signal Processing. IEEE Signal Processing Magazine 1991;8:14–38. doi:10.1109/79.91217.
[34] Choudhury M, Ganguly A. Transmission line fault classification using discrete wavelet transform. 2015 International Conference on Energy, Power and Environment: Towards Sustainable Growth, ICEPE 2015 2016:21–4. doi:10.1109/EPETSG.2015.7510112.
[35] Singh GK, Sa’ad Ahmed SAK. Vibration signal analysis using wavelet transform for isolation and identification of electrical faults in induction machine. Electric Power Systems Research 2003;68:119–36. doi:10.1016/S0378-7796(03)00154-8.
[36] Shannon CE, Weaver W, Blahut RE. The mathematical theory of communication (TODO check jahr 48/49). Urbana: University of Illinois Press 1949;117:379–423. doi:10.2307/3611062.
[37] Ford C, Shalizi C. Is R-squared Useless? University of Virginia Library Research Data Services + Sciences 2015.