Power transformer diagnostic prediction research based on quantum neural networks and evidence theory
Qiang Song1, Ai-min Wang2
1School of Mechanical Engineering, Anyang Institute of Technology, Henan, Anyang City 455000, China
2School of Computer and Information Engineering, Anyang Normal University, Henan, Anyang City 455000, China
Aiming at the fault of power transformer fault information diversity and uncertainty, a large amount of data and no regularity characteristics, a new fault diagnosis method of quantum neural network based on information fusion. In order to accurately and effectively identify transformer fault model, combining the quantum neural network and evidence theory combination of transformer fault diagnosis. A quantum neural networks to collect data on the macroscopic, microscopic quantum corrections in the interval of fuzzy intersection data according to a certain proportion of the rational allocation of the associated mode, so as to improve the accuracy of pattern recognition; use of the evidence theory can improve the convergence speed of quantum neural networks. The results were compared with the diagnosis and BP neural network input, that this method has a higher accuracy in transformer fault pattern recognition.