Title :
The module fault diagnosis of power transformer based on GA-BP algorithm
Author :
Sun, Hui-qin ; Sun, Li-hua ; Liang, Yong-Chun ; Guo, Ying-Jun
Author_Institution :
Hebei Univ. of Sci. & Technol., Shijiazhuang, China
Abstract :
According to the parameters of voltage and current of power transformer, the faults of power transformer are divided into interior and exterior modules. Genetic algorithm is adopted to optimize the initial value in neural network. BP (back propagation) algorithm is utilized to search in local part and fast gets the matrix of the weight value and the threshold. Then it realizes the fault diagnosis of power transformer. The result proves that the convergence rate of neural network based on genetic algorithm is faster than BP neural network, and improves the speed of fault diagnosis of power transformer.
Keywords :
backpropagation; fault diagnosis; genetic algorithms; neural nets; power transformer protection; GA-BP algorithm; back propagation algorithm; genetic algorithm; module fault diagnosis; neural networks; power transformer current parameters; power transformer fault diagnosis; power transformer voltage parameters; Biological cells; Constraint optimization; Convergence; Fault diagnosis; Genetic algorithms; Intelligent networks; Neural networks; Power transformers; Protective relaying; Sun; BP algorithm; Genetic algorithm; fault diagnosis; module; power transformer;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527199
