The following article has been retracted due to special reason of the author. This paper published in Vol.5 No. 2, 2013, has been removed from this site.
Electrochemical techniques and fractal theory were employed to study the corrosion behaviors and pits distribution characteristics on the corroded surfaces of 304 stainless steel exposed in FeCl3 solution. Fractal fea...Electrochemical techniques and fractal theory were employed to study the corrosion behaviors and pits distribution characteristics on the corroded surfaces of 304 stainless steel exposed in FeCl3 solution. Fractal features of pits distribution over the corroded surfaces were observed and described by the fractal dimension. A 5-8-2 back-propagation (BP) artificial neural network model for the diagnoses of the pitting corrosion rate and pits deepness of 304 stainless steel under various conditions was developed by considering the fractal dimension as a key parameter for describing the pitting corrosion characteristics. The predicted results are well in agreement with the experimental data of pitting corrosion rate and pit deepness. The max relative errors between their experimental and simulation data are 6.69% and 4.62%, respectively.展开更多
文摘The following article has been retracted due to special reason of the author. This paper published in Vol.5 No. 2, 2013, has been removed from this site.
基金the Natural Science Foundation of Liaoning Province (No.972210)
文摘Electrochemical techniques and fractal theory were employed to study the corrosion behaviors and pits distribution characteristics on the corroded surfaces of 304 stainless steel exposed in FeCl3 solution. Fractal features of pits distribution over the corroded surfaces were observed and described by the fractal dimension. A 5-8-2 back-propagation (BP) artificial neural network model for the diagnoses of the pitting corrosion rate and pits deepness of 304 stainless steel under various conditions was developed by considering the fractal dimension as a key parameter for describing the pitting corrosion characteristics. The predicted results are well in agreement with the experimental data of pitting corrosion rate and pit deepness. The max relative errors between their experimental and simulation data are 6.69% and 4.62%, respectively.