Prediction of reduced glass transition temperature of metallic alloys based on a neural network

Abstract The reduced glass transition temperature Trg is an important forming ability parameter. describes the formation in materials and behaviour of at between solid liquid states parameter for analysis, development, production process. This article process results research on development a system prediction metallic alloys based recurrent neural network algorithms. developed can predict analysis its chemical formula with high accuracy. accuracy was evaluated using 3 metrics: MSE, RMSE, MAE. Obtained values are: MSE value 0.000678, RMSE 0.0260, MAE 0.01835.