Separation of dynamic recrystallization parameter domains from a chaotic system for Ti–6Al–4V alloy and its application in parameter loading path design

In order to acquire uniform and refined microstructures for Ti–6Al–4V alloy, the parameter loading path design plays a crucial role. Here, it is pursued to separate dynamic recrystallization parameter domains from a chaotic parameter system. A succession of isothermal compression tests in a temperature range of 1023–1323 K and strain rate range of 0.01–10 s−1 were performed to obtain the basic computation data. The processing maps were constructed, and the mapping relationships between microstructural mechanisms and power dissipation efficiency indicator (η) in such maps were identified by verification of microstructural observations. Then the inner relationships between processing parameters and microstructural mechanisms were mapped as a three-dimensional (3D) space, from which the grain refinement parameter domains with dynamic recrystallization mechanism were separated. But these domains were still chaotic, so finite element (FE) simulations were used to scatter and clear this system. The optimal parameter loading paths were determined. Five verification experiments with different varying strain rates were performed, and the grain refinement was more obvious than any constant strain rates. One of the optimization results shows that the original grains with average size of 200 μm are refined as 9.7 μm, while at a constant strain rate the original grains are refined as 22.3 μm.