Assessment of computational intelligence and conventional dissolved gas analysis methods for transformer fault diagnosis
| Autor: | Milad Soleimani, Jawad Faiz |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
010302 applied physics
Downtime Computer science 020209 energy Dissolved gas analysis Computational intelligence 02 engineering and technology 01 natural sciences Fault detection and isolation Reliability engineering law.invention Electric power system Electricity generation Hazardous waste law 0103 physical sciences 0202 electrical engineering electronic engineering information engineering Electrical and Electronic Engineering Transformer |
| Zdroj: | IEEE Transactions on Dielectrics and Electrical Insulation. 25:1798-1806 |
| ISSN: | 1558-4135 1070-9878 |
| Popis: | Transformers are vital components of power systems as they are situated between energy generation and consumers and their failure disrupts the use of electrical energy. Therefore, diagnosing an incipient fault is essential in avoiding hazardous operating conditions and minimizes downtime cost. In transformers, faults take place due to electrical or thermal stresses that cause insulation decomposition in transformers. In oil-filled transformers, insulations are cellulose and oil, and the products of the insulation decomposition are gases which can be dissolved in the oil. Therefore, dissolved gas analysis (DGA) can be used for fault diagnosis in oil filled transformers. In this paper, DGA interpretation methods, conventional and intelligence, are investigated and compared. For evaluating consistency and accuracy of the methods, "No Result" cases are not considered. It can help the newcomers to this field to have access to a comprehensive comparison about the application of computational intelligence and conventional methods in transformer fault detection using DGA. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|