Prediction of flow stress of Ti-15-3 alloy with artificial neural network

来源期刊：中国有色金属学报(英文版)2001年第1期

论文作者：李萍 单德彬 薛克敏 吕炎 许沂

文章页码：95 - 97

Key words：artificial neural network; Ti-15-3 alloy; flow stress

Abstract: Hot compression experiments were conducted on Ti-15-3 alloy specimens using Gleeble-1500 Thermal Simulator．These tests were focused to obtain the flow stress data under various conditions of strain，strain rate and temperature. On thebasis of these data，the predicting model for the nonlinear relation between flow stress and deformation strain，strain rate and temperature for Ti-15-3 alloy was developed with a back-propagation artificial neural network method. Results show that the neural network can reproduce the flow stress in the sampled data and predict the nonsampled data well. Thus the neural network method has been verified to be used to tackle hot deformation problems of Ti-15-3 alloy.