Fault Detection for Covered Conductors With High-Frequency Voltage Signals: From Local Patterns to Global Features

The detection and characterization of partial discharge (PD) are crucial for the insulation diagnosis of overhead lines with covered conductors. With the release of a large dataset containing thousands of naturally obtained high-frequency voltage signals, data-driven analysis of fault-related PD patterns on an unprecedented scale becomes viable. The high diversity of PD patterns and background noise interferences motivates us to design an innovative pulse shape characterization method based on clustering techniques, which can dynamically identify a set of representative PD-related pulses. Capitalizing on those pulses as referential patterns, we construct insightful features and develop a novel machine learning model with a superior detection performance for early-stage covered conductor faults. The presented model outperforms the winning model in a Kaggle competition and provides the state-of-the-art solution to detect real-time disturbances in the field.

[1]  E. M. Lalitha,et al.  Wavelet analysis for classification of multi-source PD patterns , 2000 .

[2]  S. Misák,et al.  A complex classification approach of partial discharges from covered conductors in real environment , 2017, IEEE Transactions on Dielectrics and Electrical Insulation.

[3]  R. Bartnikas,et al.  Trends in partial discharge pattern classification: a survey , 2005, IEEE Transactions on Dielectrics and Electrical Insulation.

[4]  Hazlee Azil Illias,et al.  Partial discharge classifications: Review of recent progress , 2015 .

[5]  Xiandong Ma,et al.  Interpretation of wavelet analysis and its application in partial discharge detection , 2002 .

[6]  Mehdi Vakilian,et al.  Partial Discharges Pattern Recognition of Transformer Defect Model by LBP & HOG Features , 2019, IEEE Transactions on Power Delivery.

[7]  Stanislav Mišák,et al.  A Novel Method for Detection of Covered Conductor Faults in Medium Voltage Overhead Line Systems , 2018, IEEE Transactions on Industrial Electronics.

[8]  Sanyam Shukla,et al.  Analysis of k-Fold Cross-Validation over Hold-Out Validation on Colossal Datasets for Quality Classification , 2016, 2016 IEEE 6th International Conference on Advanced Computing (IACC).

[9]  A. Savitzky,et al.  Smoothing and Differentiation of Data by Simplified Least Squares Procedures. , 1964 .

[10]  Ivan Zelinka,et al.  A Complex Network Based Classification of Covered Conductors Faults Detection , 2016, ECC.

[11]  M. Lehtonen,et al.  Calibration of on-line partial discharge measuring system using Rogowski coil in covered-conductor overhead distribution networks , 2011 .

[12]  R. Schafer,et al.  What Is a Savitzky-Golay Filter? , 2022 .

[13]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[14]  Tomáš Vantuch,et al.  Towards the Character and Challenges of Partial Discharge Pattern Data Measured on Medium Voltage Overhead Lines , 2019, 2019 20th International Scientific Conference on Electric Power Engineering (EPE).

[15]  Ronald W. Schafer,et al.  What Is a Savitzky-Golay Filter? [Lecture Notes] , 2011, IEEE Signal Processing Magazine.

[16]  Gian Carlo Montanari,et al.  Partial Discharge Phenomenology and Induced Aging Behavior in Rotating Machines Controlled by Power Electronics , 2014, IEEE Transactions on Industrial Electronics.

[17]  L. Satish,et al.  Can fractal features be used for recognizing 3-d partial discharge patterns , 1995 .

[18]  S. Sriram,et al.  Signal denoising techniques for partial discharge measurements , 2005, IEEE Transactions on Dielectrics and Electrical Insulation.

[19]  Tie-Yan Liu,et al.  LightGBM: A Highly Efficient Gradient Boosting Decision Tree , 2017, NIPS.

[20]  Hien M. Nguyen,et al.  Borderline over-sampling for imbalanced data classification , 2009, Int. J. Knowl. Eng. Soft Data Paradigms.

[21]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[22]  Ivan Zelinka,et al.  Dynamical properties of partial-discharge patterns , 2018, Int. J. Parallel Emergent Distributed Syst..

[23]  Matti Lehtonen,et al.  Effects of Rogowski Coil and Covered-Conductor Parameters on the Performance of PD Measurements in Overhead Distribution Networks , 2008 .

[24]  A. Rodrigo Mor,et al.  New clustering techniques based on current peak value, charge and energy calculations for separation of partial discharge sources , 2017, IEEE Transactions on Dielectrics and Electrical Insulation.

[25]  D. Chicco,et al.  The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation , 2020, BMC Genomics.

[26]  F. Garnacho,et al.  A clustering technique for partial discharge and noise sources identification in power cables by means of waveform parameters , 2016, IEEE Transactions on Dielectrics and Electrical Insulation.

[27]  Michal Prilepok,et al.  Partial Discharge Pattern Classification Based on Fuzzy Signatures , 2017 .

[28]  Anna Veronika Dorogush,et al.  CatBoost: unbiased boosting with categorical features , 2017, NeurIPS.

[29]  S. Misak,et al.  Testing of a Covered Conductor’s Fault Detectors , 2015, IEEE Transactions on Power Delivery.

[30]  J. Friedman Greedy function approximation: A gradient boosting machine. , 2001 .

[31]  Balaji Srinivasan,et al.  Identification and localization of partial discharge in transformer insulation adopting cross recurrence plot analysis of acoustic signals detected using fiber Bragg gratings , 2017, IEEE Transactions on Dielectrics and Electrical Insulation.

[32]  M. Lehtonen,et al.  Modeling and experimental verification of on-line PD detection in MV covered-conductor overhead networks , 2010, IEEE Transactions on Dielectrics and Electrical Insulation.

[33]  G. J. Paoletti,et al.  Partial discharge theory and technologies related to medium-voltage electrical equipment , 2001 .

[34]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

[35]  Seetha Hari,et al.  Learning From Imbalanced Data , 2019, Advances in Computer and Electrical Engineering.