Towards Identification of Transformer Oil Faults via Novel Combined DGA Approach | ||||
| Delta University Scientific Journal | ||||
| Volume 7, Issue 3, November 2024, Page 218-227 PDF (748.09 K) | ||||
| Document Type: Original research papers | ||||
| DOI: 10.21608/dusj.2024.433464 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Shimaa A. Ibrahim  1; Sayed A. Ward1, 2; Adel EL-Faraskoury3; Diaa-Eldin A. Mansour4, 5
| ||||
| 1Department of Electrical Engineering, Faculty of Engineering at Shoubra, Benha University | ||||
| 2Department of Electrical Engineering, Faculty of Engineering Delta University for Science and Technology 11629 Cairo, Egypt | ||||
| 3Extra High Voltage Research Centre, Egyptian Electricity Holding Company, Egypt | ||||
| 4Electrical Power Engineering Department, School of Electronics, Communication and Computer Engineering, Egypt-Japan University of Science and Technology | ||||
| 5Electrical Power and Machines Engineering Department, Faculty of Engineering, Tanta University Tanta, Egypt | ||||
| Abstract | ||||
| Transformers are expensive but essential pieces of machinery in electric power systems. As a result, the electrical network suffers large financial losses when transformers fail. Therefore, early identification of likely transformer problems has a favorable impact on the dependability and continued functioning of electrical grids. One common chemical test for identifying potential problems with power transformers is the dissolved gas analysis. Conventional DGA approaches include the Rogers ratio, key gas ratio, Dornenburg ratio technique, Duval, and IEC gas ratio procedures. Unconventional DGA techniques include, among others, conditional probability, clustering, Rogers Refined, and IEC Refined. This work offers a novel method for accurately interpreting transformer faults in this regard. The proposed methodology is built around a blend of traditional and non-traditional methods with innovative processes. The Egyptian Electricity Transmission Company (EETC) provided 386 datasets with known defects, which are used to demonstrate the beneficial aspects of the suggested approach. In order to verify the enhanced accuracy of the proposed DGA technique, it is then compared with more contemporary DGA techniques utilizing the same concentration of sample gases. The developed strategy came to the conclusion that the overall accuracy of fault detection is improved by integrating several DGA techniques with varying fault accuracy. Furthermore, a case study on aged transformers with a rating of 66/11Kv has been provided to examine how aging affects dissolved gases. | ||||
| Keywords | ||||
| (Power transformers; Condition monitoring; Dissolved gas analysis (DGA); ANN; Conditional probability technique | ||||
|
Statistics Article View: 73 PDF Download: 42 |
||||
