Optimization of the processing parameters during internal oxidation of Cu–Al alloy powders using an artificial neural network

Internal oxidation is a commercial method for producing oxide dispersion strengthened copper (ODS Cu). In this paper, the dilute Cu–Al alloy powders containing 0.26 wt% of Al have been internally oxidized at temperatures (T) from 700 to 1000 °C, for holding times (t) up to 10 h. The alumina particle size has been observed and determined by electron microscopy using the two-stage preshadowed carbon replica method. By the use of backpropagation network, the non-linear relationship between internal oxidation process parameters (T,t) and alumina particle size has been established on the base of dealing with the experimental data. The results show that the well-trained backpropagation neural network can predict the alumina particle size during internal oxidation precisely and the prediction values have sufficiently mined the basic domain knowledge of internal oxidation process. Therefore, a new way of optimizing process parameters has been provided by the authors.