作者: Zhanwei Yuan , Fuguo Li , Guoliang Ji , Huijuan Qiao , Jiang Li
DOI: 10.1007/S11665-013-0838-Y
关键词:
摘要: With isothermal compression tests in the Gleeble-3500 system, hot deformation behaviors of SiCp/Al composite were studied at a wide range temperatures from 623 K to 773 K, and strain rates ranging 0.001 s−1 10 s−1. Four different modeling methods such as modified Zerilli-Armstrong model, compensation Arrhenius-type double multivariate nonlinear regression (DMNR) artificial neural model (ANN) used predict flow stress. The suitability levels these models evaluated by contrasting both correlation coefficient R C average absolute relative error. results show that predictions four can adequately meet accuracy requirement according experimental data this composite. increasing numbers determined material constants complexity computing methods, predictability is enhanced. parameters selected ranges rate temperature have non-ignorable effect on predicted previous two while they slight influence DMNR ANN.