Constitutive modelling and microstructural analysis of 92W-5Co-3Ni alloy subjected to high strain rate testing at elevated temperatures

• Texture variation at a higher strain rate can activate the formation of Adiabatic Shear Bands. • Adiabatic Shear Bands have been noticed at high strain rates above 2700 s −1 . • Arrhenius’s model captured accurate dynamic flow-softening behaviour. • Random Forest model displayed the best flow stress prediction behaviour. The mechanical response of 92 W-5Co-3Ni alloy was analysed at dynamic strain rates ranging from 1800 s −1 to 4200 s −1 and temperatures from 323 K to 873 K. The marginal strain hardening, followed by significant strain softening, was seen in flow stress behaviour. The rise in the yield strength at room temperature and higher strain rates was mainly due to the increased fraction of low-angle grain boundaries and kernel average misorientation. The severe elongation of tungsten grains was noticed for all strain rates. The Adiabatic Shear Bands (ASBs) were noticed above strain rates of 2700 s −1 with an average shear band’s width of 45 µm. Moreover, the average size of the ASB shear band’s width was reduced to 30 µm for higher strain rates (4200 s −1 ), which exhibited the enhancement of the self-sharpening capacity of the kinetic energy penetrator. Subsequently, the Arrhenius constitutive model and machine-learning-based Random Forest model (RF) were developed to predict flow stress. The Arrhenius model accurately predicted the flow stress behaviour with a correlation coefficient (R) of 0.9766 and an average absolute error (AARE) of 2.033 %. However, the RF model demonstrated the best prediction of flow stress behaviour with an R of 0.9912 and an AARE of 1.145 %.