Artificial neural network algorithm for predict the photocatalytic activity of the Mn co-doped MgAl2O4: Ce composite photocatalyst

A gamma-ray irradiation assisted polyacrylamide gel method have been used to synthesize the Mn co-doped MgAl
₂
O
₄
: Ce composite photocatalyst. The effects of catalyst load, dye concentration, pH value and irradiation time on the photocatalytic activity of the Mn co-doped MgAl
₂
O
₄
: Ce composite photocatalyst were investigated by the degradation of the rhodamine B dye under simulated sunlight irradiation. An artificial neural network model was well-established by using variables including catalyst load, dye concentration, pH value and irradiation time as input layer and degradation rate as output layer. The appropriate neural network model was established to achieve better prediction for the degradation rate and then the response input parameters were used to screen the output data for optimizing the experimental parameters of photocatalytic reaction. The application of presented artificial neural network algorithm can be used as an important tool for predict the photocatalytic activity of the Mn co-doped MgAl
₂
O
₄
: Ce composite photocatalyst. This algorithm shows potential applications in material performance prediction, space radiation environment prediction and geological disaster prediction in three gorges reservoir area.