Classification of Multiple Partial Discharge Sources Using Time-Frequency Analysis and Deep Learning
Source abstract
Abstract
Partial discharge (PD) analysis is critical for diagnosing insulation degradation in high-voltage equipment. While conventional methods struggle with multi-source PD classification due to signal overlap and noise, this study proposes a hybrid approach combining five time–frequency analysis (TFA) techniques with deep learning (GoogLeNet for simulation, ResNet50 for experiments). PD data are generated through Finite Element Method (FEM) simulations and validated via laboratory experiments. The Scatter Wavelet Transform (SWT) achieves 96.67% accuracy (F1-score: 0.967) in simulation and perfect 100% accuracy (F1-score: 1.000) in experiments, outperforming other TFAs like HHT (70.00% experimental accuracy). The Wigner–Ville Distribution (WVD) also shows strong experimental performance (94.74% accuracy, AUC: 0.947), though its computational complexity limits real-time use. These results demonstrate the SWT’s superiority in handling real-world noise and multi-source PD signals, providing a robust framework for insulation diagnostics in power systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: High voltage insulation and dielectric phenomena · Electrostatic Discharge in Electronics · Power Transformer Diagnostics and Insulation
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Faculty of Electrical Engineering, Czech Technical University in Prague, Technická 2, 166 27 Prague, Czech Republic
Funding
Czech Office for Standards, Metrology and Testing, under the Metrology Development Program 2025 · Czech Technical University in Prague
References (44)
Show source-provided references
1. Pattern recognition techniques and their applications for automatic classification of artificial partial discharge sources10.1109/TDEI.2013.6508749 ↗
2. Ali, N.H.N., Rapisarda, P., and Lewin, P.L. (2016, January 25–28). Separation of multiple partial discharge sources within a high voltage transformer winding using time frequency sparsity roughness mapping. Proceedings of the 2016 International Conference on Condition Monitoring and Diagnosis (CMD), Xi’an, China.10.1109/CMD.2016.7757802 ↗
3. Partial discharge source discrimination using a support vector machine10.1109/TDEI.2010.5412017 ↗
4. A Comprehensive Review of Signal Processing and Machine Learning Technologies for UHF PD Detection and Diagnosis (I): Preprocessing and Localization Approaches10.1109/ACCESS.2021.3077483 ↗
5. Unsupervised Deep Learning for Detecting Number of Partial Discharge Sources in Stator Bars10.1109/TDEI.2023.3306324 ↗
6. A new methodology for the identification of PD in electrical apparatus: Properties and applications10.1109/TDEI.2005.1430391 ↗
7. Classification of partial discharges10.1109/14.249365 ↗
8. Condition Monitoring Based on Partial Discharge Diagnostics Using Machine Learning Methods: A Comprehensive State-of-the-Art Review10.1109/TDEI.2020.009070 ↗
9. Carvalho, I.F., da Costa, E.G., Nobrega, L.A.M.M., and Silva, A.D.d.C. (2024). Identification of Partial Discharge Sources by Feature Extraction from a Signal Conditioning System. Sensors, 24.10.3390/s24072226 ↗
10. Albarracín-Sánchez, R., Álvarez Gómez, F., Vera-Romero, C., and Rodríguez-Serna, J. (2020). Separation of Partial Discharge Sources Measured in the High-Frequency Range with HFCT Sensors Using PRPD-teff Patterns. Sensors, 20.10.3390/s20020382 ↗
11. Separation and Classification of Concurrent Partial Discharge Signals Using Statistical-Based Feature Analysis10.1109/TDEI.2020.009043 ↗
12. A Partial Discharge Signal Separation Method Applicable for Various Sensors Based on Time-Frequency Feature Extraction of t-SNE
13. Partial Discharge Pattern Recognition Based on a Multifrequency F-P Sensing Array, AOK Time-Frequency Representation, and Deep Learning10.1109/TDEI.2022.3199189 ↗
14. Deep learning and data augmentation for partial discharge detection in electrical machines10.1016/j.engappai.2024.108074 ↗
15. Partial Discharge Identification Based on Unsupervised Representation Learning Under Repetitive Impulse Excitation With Ultra-Fast Slew-Rate10.1109/TPWRD.2023.3338209 ↗
16. Pattern Recognition of Partial Discharge in High-Voltage Cables Using TFMT Model10.1109/TPWRD.2024.3465660 ↗
17. Yang, R., Liu, Y., Geng, J., Wang, Y., Liu, Y., and Zheng, L. (2025). Detection of Partial Discharge Acoustic Signals in Converter Valve Halls Using Multivariate Time-Frequency Slices. IEEE Sens. J., early access.
18. Zheng, J., Chen, Z., Wang, Q., Qiang, H., and Xu, W. (2022). GIS Partial Discharge Pattern Recognition Based on Time-Frequency Features and Improved Convolutional Neural Network. Energies, 15.10.3390/en15197372 ↗
19. Chen, B., Hu, Y., and Wu, L. (2025). Deep Learning-Based Multi-Source Partial Discharge Pattern Recognition Integrated with Auxiliary Prior Localization Information. IEEE Trans. Dielectr. Electr. Insul., 1–9. in press.10.1109/TDEI.2025.3561094 ↗
20. Comparative analysis of machine learning and deep learning techniques on classification of artificially created partial discharge signal10.1016/j.measurement.2024.114947 ↗
21. Improving RF-Based Partial Discharge Localization via Machine Learning Ensemble Method10.1109/TPWRD.2019.2907154 ↗
22. Deep Learning Approach for Cable Partial Discharge Pattern Identification10.1007/s00202-024-02571-w ↗
23. Antenna contactless partial discharges detection in covered conductors using ensemble stacking neural networks10.1016/j.eswa.2022.118910 ↗
24. A Novel Application of Deep Belief Networks in Learning Partial Discharge Patterns for Classifying Corona, Surface and Internal Discharges10.1109/TIE.2019.2908580 ↗
25. A Novel Deep-Learning Framework to Identify and Locate Single and Multiple Partial Discharge Events10.1109/TDEI.2023.3289482 ↗
26. Florkowski, M. (2020). Classification of Partial Discharge Images Using Deep Convolutional Neural Networks. Energies, 13.10.3390/en13205496 ↗
27. Pattern Recognition of Partial Discharge Based on VMD-CWD Spectrum and Optimized CNN with Cross-Layer Feature Fusion10.1109/ACCESS.2020.3017047 ↗
28. Zhou, X., Wu, X., Ding, P., Li, X., He, N., Zhang, G., and Zhang, X. (2020). Research on Transformer Partial Discharge UHF Pattern Recognition Based on CNN-LSTM. Energies, 13.10.3390/en13010061 ↗
29. Xi, Y., Zhou, F., and Zhang, W. (2023). Partial Discharge Detection and Recognition in Insulated Overhead Conductor Based on Bi-LSTM with Attention Mechanism. Electronics, 12.10.3390/electronics12112373 ↗
30. Partial discharge detection on aerial covered conductors using time-series decomposition and long short-term memory network10.1016/j.epsr.2020.106318 ↗
31. Partial discharge detection of insulated conductors based on CNN-LSTM of attention mechanisms10.1007/s43236-021-00239-3 ↗
32. Nguyen, M.T., Nguyen, V.H., Yun, S.J., and Kim, Y.H. (2018). Recurrent Neural Network for Partial Discharge Diagnosis in Gas-Insulated Switchgear. Energies, 11.10.3390/en11051202 ↗
33. Gao, A., Zhu, Y., Cai, W., and Zhang, Y. (2018). Partial Discharge Recognition with a Multi-Resolution Convolutional Neural Network. Sensors, 18.10.3390/s18103512 ↗
34. Study on Deep-Learning Model for Phase Resolved Partial Discharge Pattern Classification Based on Convolutional Neural Network Algorithm10.1007/s42835-024-01967-9 ↗
35. Fault detection and classification on insulated overhead conductors based on MCNN-LSTM10.1049/rpg2.12380 ↗
36. Recognition of single and multiple partial discharge sources in transformers based on ultra-high frequency signals10.1049/iet-gtd.2013.0131 ↗
37. Separation Techniques of Partial Discharges and Electrical Noise Sources: A Review of Recent Progress10.1109/ACCESS.2020.3035249 ↗
38. Aging and related phenomena in modern electric power systems10.1109/14.237744 ↗
39. Automated recognition of partial discharges10.1109/94.469976 ↗
40. Krivda, A. (1995, January 22–25). Recognition of discharge patterns during ageing. Proceedings of the 1995 Conference on Electrical Insulation and Dielectric Phenomena, Virginia Beach, VA, USA.
41. A Convolutional Neural Network-Based Deep Learning Methodology for Recognition of Partial Discharge Patterns from High-Voltage Cables10.1109/TPWRD.2019.2906086 ↗
42. Importance of unraveling memory propagation effects in interpreting data on partial discharge statistics10.1109/14.249364 ↗
43. A generalized approach to partial discharge modeling10.1109/94.407017 ↗
44. Forssén, C. (2008). Modelling of Cavity Partial Discharges at Variable Applied Frequency. [Ph.D. Thesis, KTH].