Depolarization Ratio Index (DRI) as alternative method in identifying oil-filled transformer internal faults
Dissolved gas analysis is the most common and well-established method used in identifying internal faults of power transformers. However, an onsite analysis using this method accuracy still need to be improved while the laboratories analysis process takes longer time. New alternative onsite and onli...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Springer
2021
|
| Subjects: | |
| Online Access: | http://eprints.um.edu.my/26438/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Dissolved gas analysis is the most common and well-established method used in identifying internal faults of power transformers. However, an onsite analysis using this method accuracy still need to be improved while the laboratories analysis process takes longer time. New alternative onsite and online method are most urgently need to be studied. Power transformer faults identification based on extended application of polarization and depolarization current (PDC) measurement data on oil sample from in-service power transformer was investigated in this research. The changes in the molecular properties of the insulating oil due to different fault conditions found had influenced the initial time response of the PDC measurement (from 0 to 100 s). In this research, the unique characteristic pattern of each condition was determined by a single unique numerical dimensionless quantity known as depolarization ratio index (DRI). DRI is used to express the changes of depolarization current shape with different fault conditions. The ratio of depolarization current between 5/100 and 10/100 s was found to have a better correlation at least more than 90% on the incipient fault in power transformer. Indeed, the DRI at 5/100 also was observed has higher accuracy especially on units with normal, overheating and arcing condition. In addition, a graphical representation of DRI value at 5/100 and 10/100 s was proposed for fault recognition in power transformer. Both DRI and graphical representation are well validated, and the proposed interpretation techniques have accuracy of more than 90% and able to identify and classify the normal, overheating and arcing condition of the in-service power transformer. |
|---|