Prediction of the properties of semiconductor Zn$_x$Mg$_y$O sol-gel layers using artificial neural networks

Using artificial neural networks, the properties of semiconductor sol-gel layers of Zn$_x$Mg$_y$O were predicted. To form a training data set and a data set for testing neural networks by the sol-gel method, layers were formed based on ZnO : Mg films. The measurement of the photoelectric characteristics of the sol-gel coatings was carried out on an automated basic laser testing complex in accordance with National Standart-17772-88. The experiments were performed for 150 input parameters, 135 of which were used to train neural networks. In this work, we studied the influence of the architecture of neural networks on the accuracy of predicting the properties of Zn$_x$Mg$_y$O semiconductor sol-gel layers.