Classification of Power Swing using Wavelet and Convolution Neural Network

This paper focuses on the fault classification between the power swing and the normal fault using the fault current at three phases by the help of CNN(Convolution Neural Network) architecture comprises of the inception , Residual Neural Network and their hybrid model which is the inception-v3 and inception resnet-v2 along with DWT (daubechies wavelet db-4) wavelet at eight levels of detailed coefficient decomposition at 12.5kHz sampling frequency along with the CWT filter for fault current decomposition in 3 machine 9 bus WSCC (Western System Coordinating Council) system for 300 km at different location and different fault inception angle of the transmission line during power swing in EMTDC/PSCAD platform along with Matlab 2018b.The architecture is more fast and can accurately classify the power swing and normal fault with minimal amount of time as compared to the conventional ANN (Artificial Neural Network) that is used before.

[1]  Ashok Pradhan,et al.  Differential Power-Based Symmetrical Fault Detection During Power Swing , 2012, IEEE Transactions on Power Delivery.

[2]  Watanyu Meesrisuk,et al.  Implementation of Real-Time Power Swing Detection Based on Support Vector Machine of Distributed Generation System Using LabVIEW MyRIO , 2018, 2018 21st International Conference on Electrical Machines and Systems (ICEMS).

[3]  A. K. Pradhan,et al.  A Fault Detection Technique for the Series-Compensated Line During Power Swing , 2013, IEEE Transactions on Power Delivery.

[4]  Jian Guo Zhu,et al.  A Novel Approach to Detect Symmetrical Faults Occurring During Power Swings by Using Frequency Components of Instantaneous Three-Phase Active Power , 2012, IEEE Transactions on Power Delivery.

[5]  M. Kezunovic,et al.  Detection of Symmetrical Faults by Distance Relays During Power Swings , 2010, IEEE Transactions on Power Delivery.

[6]  Bikash Patel,et al.  Detection of Power Swing and Fault During Power Swing Using Lissajous Figure , 2018, IEEE Transactions on Power Delivery.

[7]  Sean McGuinness,et al.  Performance of protection relays during stable and unstable power swings , 2018 .

[8]  Phasor Measurement Based Fault Detection and Blocking/De-Blocking of Distance Relay Under Power Swing , 2018, 2018 International CET Conference on Control, Communication, and Computing (IC4).

[9]  Sergey Ioffe,et al.  Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning , 2016, AAAI.

[10]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  David Dagan Feng,et al.  An Ensemble of Fine-Tuned Convolutional Neural Networks for Medical Image Classification , 2017, IEEE Journal of Biomedical and Health Informatics.

[12]  Mladen Kezunovic,et al.  Fast Distance Relay Scheme for Detecting Symmetrical Fault During Power Swing , 2010, IEEE Transactions on Power Delivery.

[13]  Azah Mohamed,et al.  Power swing and voltage collapse identification schemes for correct distance relay operation in power system , 2013 .

[14]  Ronald M. Summers,et al.  Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning , 2016, IEEE Transactions on Medical Imaging.

[15]  S. R. Samantaray,et al.  Spectral energy function for fault detection during power swing , 2011, 2011 International Conference on Energy, Automation and Signal.

[16]  Pei Liu,et al.  A Novel Scheme to Identify Symmetrical Faults Occurring During Power Swings , 2008, IEEE Transactions on Power Delivery.

[17]  Bhavesh R. Bhalja,et al.  New support vector machine-based digital relaying scheme for discrimination between power swing and fault , 2014 .

[18]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).