Modeling of thermal annealing of Zno:Ga thin films for transparent conductive oxide using neural networks

In this paper, we present the nonlinear thermal annealing modeling for the electrical properties of ZnO:Ga thin films on annealing temperature and film thickness using neural network based on error backpropagation (BPNN) algorithm and multilayer perceptron (MLP). The thermal annealing process of ZnO:Ga thin films were characterized by general factorial experimental design. To model the nonlinear annealing process, 6 experiments are trained by BPNN which has 2-4-1 structures. The output response models on carrier concentrations, mobility and resistivity of ZnO:Ga thin films trained by BPNN are represented by surface plot of response surface model. The predicted models by training experiments using BPNN were verified by 2 additional experiments not included to the training experiments, and the performance of models is measured by root mean square error (RMSE) and R-square value. Based on the modeling results, neural network can provide sufficient correspondence between the predicted output values and the measured. The annealing process is nonlinear and complex but the output response can be predicted by the neural network model.