A GA-optimized Johnson-Cook model of flow behavior for Ti-6554 alloy in cross-phase temperature range. (21st August 2019)

• Record Type: Journal Article
• Authors: Long, Shuai; Zhou, Jie; Qiu, Ze-Lin; Zhou, Yu-Ting; Li, Shi-Shan
• Is Part Of: Materials research express. Volume 6:Number 10(2019)
• Publication Date: 2019-08-21
• Subjects: Ti-6Cr-5Mo-5V-4Al alloy -- hot compression tests -- microstructure -- temperature separated type Johnson-Cook model -- genetic algorithm; Materials science -- Research -- Periodicals; Materials science -- Periodicals; 620.11
• DOI: 10.1088/2053-1591/ab3885
• Languages: English
• ISSNs: 2053-1591

Abstract

Abstract: Hot compression tests of Ti-6Cr-5Mo-5V-4Al (Ti-6554) alloy were conducted on a Gleeble-3500 isothermal simulator at the temperature range of 963 ∼ 1113 K and strain rate range of 0.001 ∼ 1 s −1 . The effects of temperature, strain rate and strain on flow behavior were analyzed according to the flow curves and microstructure observations. At low temperature range (963 ∼ 1053 K), massive globular α forms barriers to dislocation movement and enhances peak stress. Subsequently, the intensive incidence of dynamic globularization of α leads to strong softening with the increasing strain. At high temperature range (1053 ∼ 1113 K), the volume fraction reduction of α promotes dislocation movement and lowers peak stress. Dynamic crystallization of β operates at low strain rate condition which results in relatively evident decreasing of flow stress. At higher strain rate, squashed α transforms into flocculent structure which can be deemed as the result of dynamic recovery. With the increasing strain, the size of globular α enlarged and phase transform occurs due to the generation of deformation heat. To describe the hot deformation behavior, a modified Johnson-Cook (m-JC) model considering the coupled effects of strain rate and temperature was constructed and a temperature separated type Johnson-Cook model (ts-JC) was proposed to optimize the describing accuracy. Meanwhile, for the further optimization of prediction capability, genetic algorithm (GA) was introduced to optimize the constants of ts-JC model (GA-JC). The prediction capability evaluators ln(MSE)-values and Adj. R 2 -values of m-JC, ts-JC and GA-JC were 4.94, 4.59, 4.29 and 0.990 39, 0.987 85, 0.983 08, respectively, which reveal good optimization results of GA-JC.