Neural network fault prediction and its application

In this paper, the forecasting algorithm employs wavelet function to replace sigmoid function in the hidden layer of Back-Propagation Neural Network. And a Wavelet Neural Network prediction model is established to predict Anode Effect (the most typical fault) through forecasting the change rate of cell resistance. The authors have developed forecasting software based on platform of Visual Basic 6.0. The simulation results show that the proposed method not only has greatly improved fault prediction precision and real-time, but also improved the operation efficiency. That means we can increase energy efficiency and the safety of aluminum production process.

[1]  Christos Georgakis,et al.  Disturbance detection and isolation by dynamic principal component analysis , 1995 .

[2]  J.A. Starzyk,et al.  Artificial neural network for testing analog circuits , 1990, IEEE International Symposium on Circuits and Systems.

[3]  Fulei Chu,et al.  A Hierarchical Evolutionary Algorithm for Constructing and Training Wavelet Networks , 2002, Neural Computing & Applications.

[4]  Qinghua Zhang,et al.  Using wavelet network in nonparametric estimation , 1997, IEEE Trans. Neural Networks.

[5]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[6]  M. Kramer.  Nonlinear principal component analysis using autoassociative neural networks , 1991 .

[7]  Byeong Seok Ahn,et al.  The integrated methodology of rough set theory and artificial neural network for business failure prediction , 2000 .

[8]  Thian Yew Gan,et al.  Wavelet Analysis of Variability, Teleconnectivity, and Predictability of the September–November East African Rainfall , 2005 .

[9]  Skander Soltani,et al.  On the use of the wavelet decomposition for time series prediction , 2002, ESANN.

[10]  Toru Yamamoto,et al.  A Design Of Multiloop Predictive Self‐Tuning Pid Controllers , 2002 .

[11]  David Zimbra,et al.  A dynamic artificial neural network model for forecasting time series events , 2005 .

[12]  Valdis Bojarevics,et al.  WEAKLY COUPLED THERMO-ELECTRIC AND MHD MATHEMATICAL MODELS OF AN ALUMINIUM ELECTROLYSIS CELL , 2005 .

[13]  Vasile Palade,et al.  A novel fuzzy classification solution for fault diagnosis , 2004, J. Intell. Fuzzy Syst..

[14]  Yijia Cao,et al.  A Fault Diagnosis Method for Transformer Integrating Rough Set with Fuzzy Rules , 2006 .