Research on fault diagnosis method based on Modified Elman Neural Network
暂无分享,去创建一个
This paper presents fault diagnosis method in aluminum electrolysis based on Modified Elman Neural Network. According to the mechanism of fault occurring in aluminum electrolysis, time and type of fault is determined by neural network. Simulation results show that the fault diagnosis method based on Modified Elman Neural Network can predict fault during aluminum electrolysis rapidly and accurately. And it has a certain value in engineering applications.
[1] Valdis Bojarevics,et al. WEAKLY COUPLED THERMO-ELECTRIC AND MHD MATHEMATICAL MODELS OF AN ALUMINIUM ELECTROLYSIS CELL , 2005 .
[2] Geoffrey E. Hinton,et al. Reducing the Dimensionality of Data with Neural Networks , 2006, Science.