Establishing flow stress behaviour of Ti-6Al-4V alloy and development of constitutive models using Johnson-Cook method and Artificial Neural Network for quasi-static and dynamic loading

Ti-6Al-4V alloy is one of the most widely used material in both research as well as in commercial industries at present due to its high strength-to-weight ratio, low density and excellent high-temperature properties. Understanding the behaviour of this alloy at various deformation conditions (strain, strain rate and temperature) is crucial to fit the material in demanding applications. In our present work, the mechanical behaviour of α + β dual phase Ti-6Al-4V alloy is studied in the temperature ranging from 25 ℃ to 200 ℃ and rate of deformation ranging from 10-3 s−1 to 104 s−1. Tensile tests were done using uniaxial tensile tester for quasi-static range of deformation and Split-Hopkinson bar machine for dynamic range of deformation. Post deformation fracture surface analysis were carried out to understand the effects of strain rate and temperature variations on the deformation behaviour, and to establish a structure–property correlation. For the studied range of deformation parameters, constitutive models are developed. As Johnson-Cook model is one of the widely used model in various numerical analysis software, a modified Johnson-Cook model has been developed for the alloy. Moreover, artificial intelligence (AI) is found to be an efficient tool these days to solve complex problems in various fields of engineering. An attempt is also made in this study to develop an artificial neural network (ANN) framework for the prediction of flow stress at various deformation condition for Ti-6Al-4V alloy. Our study revealed that the AI based ANN technique is more efficient in calculating the flow stress as compared with the traditional Johnson-Cook model.