Transformers are one of the important and at the same time expensive components of power systems. On timely diagnosis of fault in such systems is still among the researchers interest. Fault diagnosis of transformers based on the dissolved gas analysis is a new technique in the field of fault diagnosis of power transformers. IEC, Roger’s and Dornenburg techniques are the mostly used techniques. However, each have a limited range for diagnosis and may fail a correct diagnosis when the amount of dissolve gases are close to the predefined margins. This problem can be solved by using intelligent techniques such as Fuzzy logics or neural networks. In this paper, a combined technique is proposed for the fault diagnosis of power transformers. It employs a combination of fuzzy Roger’s, fuzzy IEC, and fuzzified dissolved gas thresholds for the fault diagnosis. The proposed technique is applied on some transformers with a sum of over 100 available data on their conditions and showed an accuracy of over 98%.
Rights and permissions | |
This Journal is an open access Journal Licensed under the Creative Commons Attribution-NonCommercial 4.0 International License. (CC BY NC 4.0) |