Thermography and artificial intelligence in transformer fault detection
| Autor: | Ronaldo R. B. de Aquino, Milde M. S. Lira, Gustavo Maciel dos Santos |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Engineering
Transformer oil business.industry 020209 energy Applied Mathematics Dissolved gas analysis 020208 electrical & electronic engineering Intelligent decision support system 02 engineering and technology Temperature measurement Fuzzy logic Automotive engineering Fault detection and isolation law.invention law Thermography 0202 electrical engineering electronic engineering information engineering Electronic engineering Electrical and Electronic Engineering business Transformer |
| Zdroj: | Electrical Engineering. 100:1317-1325 |
| ISSN: | 1432-0487 0948-7921 |
| DOI: | 10.1007/s00202-017-0595-2 |
| Popis: | Dissolved gas analysis of insulating oil in refrigerated power transformer oil is a widespread technique for detecting incipient faults. However, this technique involves safety procedures for the collection of oil samples, laboratory response times and, in some cases, removing the transformer from operation. Removing a transformer from operation in certain situations can become very costly as there are production environments that must run uninterrupted so that stoppages such as this represent financial loss. Infrared thermography is a non-destructive temperature measurement technique commonly used to detect anomalies and predict possible faults without disrupting system operation. This paper presents studies based on the use of infrared temperature measurement to detect incipient faults in transformers through dissolved gas analysis of the insulating oil. This study’s methodology uses intelligent systems to analyse transformer face temperatures and detect incipient faults. The results obtained in this work present 86 and 83% of classification correctness using artificial neural networks and fuzzy logic, respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink