Two artificial neural network models are developed to predict the effect of corrosion behavior of different series of aluminum alloys when exposed to two corrosive environments. Given the corrosion environment and time of exposure the first neural network predicts the ASTM G34 corrosion rating and the resulting material loss. The trained and limited test results predicted from this network are in good comparison to the experimental data. The effects of corrosion environment and material type from neural network simulation are presented to illustrate the trends. The second network predicts the cycles for final fatigue failure and the residual static strength of a particular type of material, given the amount of material loss due to corrosion. Based on the preliminary results, the neural network approach to corrosion and fatigue predictions is encouraging and can be used for a variety of materials and environments if more data is available. It is intended that the approach developed here will assist in the structural integrity evaluation of corrosion in aging aircraft.
Corrosion prediction in aging aircraft materials using neural networks
2000
10 Seiten, 13 Quellen
Aufsatz (Konferenz)
Englisch
Corrosion Prediction in Aging Aircraft Materials using Neural Networks
British Library Conference Proceedings | 2000
|British Library Online Contents | 1996
|Online Contents | 1996
|