Partial Discharge

Partial discharge (PD) refers to localized electrical discharges that occur in insulating materials under high voltage stress, serving as early warning signs of impending equipment failure in high-voltage systems. Current research focuses on improving the automated classification of PD signals from various sources (e.g., metallic protrusions, particles) using advanced machine learning techniques, such as neural networks and ensemble methods like Random Forests and Support Vector Machines, to enhance diagnostic accuracy and speed. Accurate PD detection and classification are crucial for preventative maintenance in power systems, preventing costly and potentially catastrophic failures in equipment like transformers and gas-insulated systems.

Papers

July 17, 2023

Generalizable Classification of UHF Partial Discharge Signals in Gas-Insulated HVDC Systems Using Neural Networks
Steffen Seitz, Thomas Götz, Christopher Lindenberg, Ronald Tetzlaff, Stephan Schlegel
Neural Network Abnormal Signal Learning Generalizable Biomedical Signal Partial Discharge

October 21, 2022

Feature Engineering and Classification Models for Partial Discharge in Power Transformers
Jonathan Wang, Kesheng Wu, Alex Sim, Seongwook Hwangbo
Classification Model Feature Engineering Feature Set Power Transformer Partial Discharge

Partial Discharge

Papers

Generalizable Classification of UHF Partial Discharge Signals in Gas-Insulated HVDC Systems Using Neural Networks

Feature Engineering and Classification Models for Partial Discharge in Power Transformers