An investigation into partial discharge pulse extraction methods. (December 2015)

• Record Type: Journal Article
• Authors: Bajwa, Abdullah Akram; Habib, Salman; Kamran, Muhammad
• Is Part Of: International journal of electrical power & energy systems. Volume 73(2015:Dec.)
• Page Start: 964
• Page End: 982
• Publication Date: 2015-12
• Subjects: Partial discharge -- Pulse extraction -- Wavelet transform -- Principle component analysis -- DBSCAN; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2015.06.028
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: Assessment of robustness of different partial discharge pulse extraction methods. Discrete Wavelet Transform (DWT) for noise removal. Analysis of energy distribution of DWT to identify partial discharge activity. PCA and DBSCAN for dimensionality reduction and classification. Abstract: Interest in condition monitoring based on partial discharge diagnostics has seen a very rapid increase from high voltage industry. Condition monitoring is an asset monitoring tool that can inform the user of real time health of their high voltage equipment/plant. This has now become a necessity in aging networks that still have equipment in operation that has fulfilled its rated life. This research is focused on investigating the performance of existing partial discharge pulse extraction methods. There are numerous pulse extraction methods present with documented results and extensive is research being carried out either to improve existing methods or to find new methods with better performance. One problem with every method is that their efficiency is highly affected when a signal with high noise is processed through them. This investigation was carried out by implementing partial discharge methods and testing their performance on different types of data sets that had different signal to noise ratios and were created for different purposes. Results were obtained by applying different data sets on various available techniques followed by selection of the best and most robust method for the task of Partial discharge pulse extraction. The most appropriate discharge pulse extraction method is based on comparison of data which will be highly useful for future research in this area.