Title :
Fault Diagnosis against Oil-Immersed Transformer Based on PNN and GM (1,1)
Author :
Jia, Honghong ; Dai, Wenzhan
Author_Institution :
Zhejiang Sci-Tech Univ., Hangzhou
Abstract :
This paper introduces GM(1,1) and PNN technology into online fault diagnosis for oil-immersed transformer. Firstly, the seven fault types of oil-immersed transformer are characterized by the thickness of H2CH4,C2H6,C2H2 and C2H4. Secondly characteristic vector of fault status is forecasted by GM(1,1) model. Thirdly the forecasting results are chosen as the input of PNN and trained. At last, the method is applied to fault diagnose of oil-immersed transformer. The results show the approach effectively.
Keywords :
fault diagnosis; power engineering computing; power system faults; transformers; GM(1,1) model; GM(1,1) technology; PNN technology; oil-immersed transformer; online fault diagnosis; probabilistic neural network; Fault diagnosis; Gases; Hydrogen; Inspection; Oil insulation; Petroleum; Power transformer insulation; Predictive models; Sensor arrays; Temperature sensors; GM(1,1); Online Fault Diagnosis; Probabilistic Neural Network(PNN); Transformer;
Conference_Titel :
Integration Technology, 2007. ICIT '07. IEEE International Conference on
Conference_Location :
Shenzhen
Print_ISBN :
1-4244-1092-4
Electronic_ISBN :
1-4244-1092-4
DOI :
10.1109/ICITECHNOLOGY.2007.4290503
