Domain Adaptation for Ageing State Recognition of Cables used in Power Systems

As power cable systems play a critical role with respect to the reliability of power grids, it is necessary to adopt predictive maintenance techniques to prevent their damage. Especially for very expensive equipment used in power generation or smart grids, failure must be avoided. Partial Discharge (PD) phenomenon is considered as one of the main cause of deterioration of power cables insulation and reduce the cable lifetime. Thus, PD detection is an effective way to evaluate the degradation state of cable insulation. In this study, PD data are extracted from two different environments : a low voltage and high voltage cable using two different experimental devices. Original feature extraction, ranking and selection methods are implemented to design optimized Support Vector Machines classifiers (SVMs) that attribute an ageing state to the cable insulation for both cable types. Given the great interest of transferring a cable diagnosis model from one cable to any other cable type a domain adaptation technique is implemented. Recognition rate exceeding 99% are achieved when a classifier built in one environment is used to predict the ageing state of a new cable type based on PD data acquired in different conditions. The methodology presented in this work could be applied in realistic settings, for the monitoring of different cable types used in power grid system based on PD measurements.

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

[2]  Javier Ortego,et al.  Signal analysis and feature generation for pattern identification of partial discharges in high-voltage equipment , 2013 .

[3]  Sheng Chen,et al.  Orthogonal least squares methods and their application to non-linear system identification , 1989 .

[4]  A. Cavallini,et al.  Identification of Partial Discharge Phenomena in HVDC Apparatus , 2008, Conference Record of the 2008 IEEE International Symposium on Electrical Insulation.

[5]  Gian Carlo Montanari,et al.  AC and DC partial discharge measurements on defective cables , 2016, 2016 IEEE Electrical Insulation Conference (EIC).

[6]  T. Ditchi,et al.  Characterization of partial discharges in solid insulators under DC voltage using physical cavity properties , 2017, 2017 International Symposium on Electrical Insulating Materials (ISEIM).

[7]  Olga Fink,et al.  Domain Adaptive Transfer Learning for Fault Diagnosis , 2019, 2019 Prognostics and System Health Management Conference (PHM-Paris).

[8]  Giovanni Mazzanti,et al.  Life and reliability models for high voltage DC extruded cables , 2017, IEEE Electrical Insulation Magazine.

[9]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[10]  Paul Lewin,et al.  A lower dimensional feature vector for identification of partial discharges of different origin using time measurements , 2012 .

[11]  Gian Antonio Susto,et al.  Machine Learning for Predictive Maintenance: A Multiple Classifier Approach , 2015, IEEE Transactions on Industrial Informatics.

[12]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[13]  Hazlee Azil Illias,et al.  High noise tolerance feature extraction for partial discharge classification in XLPE cable joints , 2017, IEEE Transactions on Dielectrics and Electrical Insulation.

[14]  G. B. Kumbhar,et al.  Detection, Measurement, and Classification of Partial Discharge in a Power Transformer: Methods, Trends, and Future Research , 2018 .

[15]  Xiaoxin Chen,et al.  Partial discharge feature extraction through contourlet transform for XLPE cable defect models classification , 2016, 2016 International Conference on Condition Monitoring and Diagnosis (CMD).

[16]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[17]  Yoshua Bengio,et al.  Domain Adaptation for Large-Scale Sentiment Classification: A Deep Learning Approach , 2011, ICML.

[19]  Chao Wang,et al.  Discrete Wavelet Transform Decomposition Level Determination Exploiting Sparseness Measurement , 2013 .

[20]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .