Detection and Separation of multiple Partial Discharges sources on Power Transformers — Case Study
Advanced Partial Discharge (PD) measurement techniques, including electrical and UHF methods, have been developed to detect and localize the insulation defects that could occur on Power Transformers. However, challenges like environmental noise and multiple PD sources can complicate the interpretation of PD measurement results. The advancements in PD source separation techniques, such as 3PARD, have significantly enhanced the accuracy of diagnosing insulation defects in power transformers. The application of these methods, coupled with UHF and electrical PD tests, allows for a detailed analysis of measurement results. The correlation between high PD values, suspicious PRPD patterns, and actual physical damage observed upon inspection underscores the reliability of these diagnostic tests. Such precise matching of PD patterns with visual damages confirms the effectiveness of PD diagnostics in identifying insulation system failures. This paper describes, through a case study, the most important aspects of Electrical (and UHF) PD tests on power transformers. The Results obtained show high PD's values on the transformer and reveal many suspicious PRPD patterns. With the using of 3PARD tool, many insulation defaults were detected, and their origin were identified. The transformer was opened on the workshop, and with the visual inspection of different parts, we discovered many serious damages on the insulation system. These findings exactly matched with the PRPD patterns obtained during the different PD measurements stages, and it became clear that our assumptions based on the PD measurement results were all correct at 100%.
Authors
- Sofiane Bakkay
- Mükremin Yanan
- Îlker Çalişkan
Institutions
- Omron (Japan) (JP)
Publication Details
- Published
- 2024-10-07
- DOI
- https://doi.org/10.1109/powerafrica61624.2024.10759367
- Citations
- 1
- Primary Topic
- High voltage insulation and dielectric phenomena
- Type
- article
- Field-Weighted Citation Impact
- 0.12