Applied Mechanics and Materials Vol. 818 (2016) pp 156-165 · JANUARY 2016
Impact Factor: 0.15 · DOI: 10.4028/www.scientific.net/AMM.818.156

Transmission Line Using Discrete Wavelet Transform and Back-Propagation Neural Network Based on Clarke’sTransformation

Makmur SAINI1,3,a, Abdullah Asuhaimi Bin MOHD ZIN 1,b,
Mohd Wazir Bin MUSTAFA1,c, Ahmad Rizal SULTAN1,3,d, Rahimuddin 2,e
1. Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru,
2. Faculty of Mechanical Engineering, Universiti Teknologi Malaysia
3. Politeknik Negeri Ujung Pandang, South Sulawesi, Indonesia 90245
a. saini@fkegraduate.utm.my, b, abdullah@fke.utm.my, c. wazir@fke.utm.my,
d. ahmad.rizal@fkegraduate.utm.my, e. rahimnav@gmail.com

Keywords: 

Wavelet Transformation, back-propagation neural network, Fault Classification, Fault

detection, Clarke’s Transformation, PSCAD/EMTDC

Abstract. 

1. paper proposes a new technique of using discrete wavelet transform (DWT) and back-propagation neural network (BPNN) based on Clarke’s transformation for fault classification and detection on a single circuit transmission line. Simulation and training process for the neural network are done by using PSCAD / EMTDC and MATLAB. Daubechies4 mother wavelet (DB4) is used to decompose the high frequency components of these signals. The wavelet transform coefficients (WTC) and wavelet energy coefficients (WEC) for classification fault and detect patterns used as input for neural network training back-propagation (BPNN). This information is then fed into a neural network to classify the fault condition. A DWT with quasi optimal performance for preprocessing stage are presented. This study also includes a comparison of the results of training BPPN and DWT with and without Clarke’s transformation, where the results show that using Clarke transformation in training will give in a smaller mean square error (MSE) and mean absolute error (MAE). The simulation also shows that the new algorithm is more reliable and accurate.

Introduction

The transmission line is vital elements in power system since this electrical energy can be transferred from are placed to another. However, this transmission line part of power system has often impaired. Most of the disturbances on the power system come from the interference on thetransmission line. Therefore, speed and accuracy in the determination of fault detection and classification of disturbances in the transmission line have a very important role in electric power systems. This paper proposes a method of using discrete wavelet transform (DWT) and back-propagation neural network (BPNN) based on Clarke’s transformation to determine the fault detection and classification on single circuit transmission line. This study presents a different approach called alpha-beta transformation based on the Clarke’s transformation, which is also a transformation of a three-phase system into a two-phase system [1, 2], where the result of the Clarke’s transformation is transformed into discrete wavelets transform.. In recent years, several methods of fault classification have been proposed. Some of them are based on artificial neural network [3,4], wavelet transform [5,6] and also combination of these techniques [7-8]. This paper proposes a novel method for fault classification in transmission lines using discrete wavelet transform (DWT) and back-propagation neural network (BPNN). The key idea of the method is to use wavelet coefficient detail and the wavelet energy coefficient of the currents as the input patterns to create a simple multi-layer perception network (MLP). This paper presents the development of a new decision algorithm for use in the protective relay for fault detection and classification. To validate this technique, the fault conditions had been simulated using EMTDC / PSCAD [9].

Conclusion

This paper developed the technique which is the linking discrete wavelet transform (DWT) and back-propagation neural network (BPNN) based on the Clarke transformation for fault classification and detect on single circuit transmission lines. This study also includes comparison on the results of training BPPN and DWT with and without Clarke’s transformation, where the results show that using the Clarke’s transformation in training will produce smaller MSE and MAE  compared with without transformation Clarke’s, among the three structures, the Architecture result was the best, which was 12 – 10 – 20 - 4. This technique applies Daubechies4 (db4) as a mother wavelet using in this paper, the performance shows that the proposed technique gives good accuracy of transient classification.

Acknowledgment

The authors would like to express their gratitude to Universiti Teknologi Malaysia, The State Polytechnic of Ujung Pandang, PT. PLN (Persero) of South Sulawesi and the Government of South

Sulawesi, Indonesia for providing the financial and technical support for this research.