The isothermal constant strain rate compression experiment was carried out on titanium alloy TC4 ELI at the two phase region temperature of 750

9

50 ℃ and the strain rate range of 0.001

7

0 s

-1

by using Gleeble-3500 thermal simulation machine.The hot deformation behavior of the alloy was analyzed

and the constitutive model of the alloy was established by the Arrhenius equation and BP artificial neural network model.The results reveal that the effect of strain rate and deformation temperature on the flow stress of titanium alloy TC4 ELI exhibits remarkable

and the flow stress is decreased with increasing in deformation temperature and decreasing in strain rate.In the deformation of the two-phase region

theαphase in the original microstructure has a certain degree of spheroidization/dynamic recrystallization

and the low strain rate can promote occurrence of the spheroidization/dynamic recrystallization.The average errors reach 17.51% and 1.36% for establishment of Arrhenius equation and BP artificial neural network model for the constitutive equations

and BP artificial neural network model has a higher precision

which is more suitable for predicting the flow stress of titanium alloy TC4 ELI.