Partial Discharge Detection Based on Ultrasound Using Optimized Deep Learning Approach

Abstract

Electrical equipment is prone to different types of Partial Discharge (PD) failures that are varying between minor and severe level. In this paper, Three developed models for Convolution Neural Network (CNN) are proposed to detect and classify four different partial discharge types which are arcing, corona discharge, tracking, looseness as well as healthy equipment situation. Notably, the resulting models exhibited an impressive overall accuracy of more than 94%, which is particularly significant considering the inherent presence of noise in the real-world samples obtained as representative field failures. These findings underscore the robustness and effectiveness of the CNN models in accurately identifying PDs, despite the intricate challenges associated with real-world data.

Development of an Industrial Partial Discharge Calibrator and Its Performance Tests

“…Traditional techniques, which are electrical, comply with the IEC-60270 standard, serving as the benchmark for PD measurement. Conversely, unconventional methods capture high-frequency PD signals via alternative approaches, including acoustic detection, optical monitoring, Ultra High Frequency (UHF) analysis, and the use of High-Frequency Current Transformers (HFCT), expanding the scope of PD detection beyond the limitations of standard electrical methods [16], [17]. Where appropriate techniques are used, it is also possible to observe the effects of the discharge in the test sample later.…”