Title :
Application of CP Modular Neural Networks on DGA Based Power Transformer Fault Diagnosis
Author :
Xiao-ming, Wang ; Xiaobo, Liu ; Yongping, Lu
Author_Institution :
Ultra High Voltage Troansmission Subcompany, Jiangxi Electr. Power Corp., Nangchang, China
Abstract :
Counter propagation arithmetic is make up of this paper presents a fault diagnosis model for power transformer based on counter propagation network. The compound neural networks model is build first and parameters of CP Networks are confirmed by comparing the results in different situations. The diagnostic examples indicate the validity of the proposed method. The diagnosis correctness of the new method has been much enhanced than the ordinary BP neural network and the improved three rations method.
Keywords :
backpropagation; fault diagnosis; neural nets; power transformer testing; CP modular neural networks; backpropagation; counter propagation arithmetic; diagnosis correctness; dissolved gases analysis; power transformer fault diagnosis; Artificial neural networks; Counting circuits; Dissolved gas analysis; Fault diagnosis; Neural networks; Neurons; Oil insulation; Power system security; Power transformer insulation; Power transformers; CP compound neural networks; Dissolved Gases Analysis (DGA); Fault diagnosis; Transformer;
Conference_Titel :
High Voltage Engineering and Application, 2008. ICHVE 2008. International Conference on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4244-3823-5
Electronic_ISBN :
978-1-4244-2810-6
DOI :
10.1109/ICHVE.2008.4774000
