Title :
The study of variant DGA feature neural network multilayer diagnostic model
Author :
Hu, Qing ; Chen, Weigen ; Du, Lin ; Li, Nan ; Sun, Caixin
Author_Institution :
Key Lab. of High Voltage Eng. & Electr. New Technol. of MOE, Chongqing Univ., Chongqing
Abstract :
Selecting appropriate features has vital effect on the effectiveness of fault diagnosis, and DGA is widely used in transformer fault diagnosis, so this paper, using ANN as method and 5 gas concentrations as available features, studies the the key feature gases according to fault types, and their roles in fault diagnosis. Based on this, this paper provides the variant feature neural network multilayer diagnosis model.
Keywords :
chemical analysis; fault diagnosis; neural nets; power engineering computing; power transformers; dissolved gas analysis; gas concentration; transformer fault diagnosis; variant DGA feature neural network multilayer diagnostic model; Appropriate technology; Artificial neural networks; Automation; Dissolved gas analysis; Fault diagnosis; Intelligent control; Multi-layer neural network; Neural networks; Power transformers; Sun; DGA; Fault Diagnosis; Neural Network; Transformer;
Conference_Titel :
Intelligent Control and Automation, 2008. WCICA 2008. 7th World Congress on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4244-2113-8
Electronic_ISBN :
978-1-4244-2114-5
DOI :
10.1109/WCICA.2008.4594268
