•Canonical correlation analysis (CCA) used for quantifying parameter sensitivities in a constitutive model.•Conventionally used local sensitivity techniques were compared to global methods.•CCA, a multivariate global sensitivity technique, facilitated a better understanding of parameter interrelationships.•Independent validation tests were conducted based on both experimental observations and literature sources•Both validation exercises showed that CCA predictions match the trends in optimized parameters for titanium alloys.Successful modeling of a material's deformation behavior is dependent on the development of realistic constitutive models that can mimic the actual response of the material of interest. In-depth knowledge about the parameters in a constitutive model will lead to a better understanding of the relationship between flow stress and deformation conditions and will eventually aid in better design of the deformation process. Curve-fitting is generally employed to calibrate constitutive models using experimental data. However, the relative impact of constitutive model parameters on the mechanical response for different models has not been extensively explored. In this study, both local and global sensitivity analysis methods are used to understand and quantify the contribution of the parameters. Canonical correlation analysis (CCA) has been used to understand the effect of constitutive model parameters on the flow stress behavior of titanium alloys. This analysis has been performed for the Mechanical Threshold Stress (MTS) model, which is a constitutive description based on the physics of dislocation motion. The limitations of local sensitivity methods have been highlighted and it has been shown that CCA provides a measure of both an individual variable's contribution and the effectiveness of the parameter set as a whole.Download high-res image (239KB)Download full-size image

•Constitutive modeling of uniaxial compression tests for Ti-5553 in both the single-phase and two-phase regimes.•Voce and Mechanical Threshold Stress (MTS) models implemented in the viscoplastic self-consistent (VPSC) framework.•Voce model shown to fit the individual deformation responses accurately; however, no consistent trends observed.•Dislocation-based MTS model shown to fit for a wide range of deformation conditions accurately using the same parameters.•Validation of the single-phase MTS model, showing that same parameter set is applicable to the set of related alloys.Titanium alloy Ti-5Al-5Mo-5V-3Cr (Ti-5553) is a near beta alloy used in structural aircraft components because of its excellent mechanical properties. Simulating the mechanical response of this material using constitutive models is an important step in understanding the relationship between its microstructure and properties. Uniaxial compression tests were conducted at temperatures both below and above the β transus and at different loading rates. A viscoplastic self-consistent (VPSC) model was used to match the stress-strain response of the Ti-5553 alloy based on uniaxial compression tests across a range of temperatures and strain rates. Sets of parameter values were determined for two different hardening models, namely the modified Voce model, which is empirical, and the Mechanical Threshold Stress (MTS) model, which is based on dislocation theory. No consistent trends in the Voce parameter values were found as a function of temperature or strain rate. By contrast, the physically-based MTS model, which is explicitly designed to cover wide ranges of deformation conditions, was able to fit a wide range of temperatures and strain rates. It was further validated by comparison with experimental measurements with other β titanium alloys that had similar compositions.Download high-res image (263KB)Download full-size image

•A full-field thermoelastic spectral method is presented.•Spherical and cylindrical, homogeneous and inhomogeneous inclusion configurations are studied.•The spectral method predictions show agreement with analytical solutions for different geometries.•The effect of voxelization on local fields near interfaces is also assessed.A numerical method based on Fast Fourier Transforms to compute the thermoelastic response of heterogeneous materials is presented and validated by comparison with analytical solutions of the Eshelby inclusion problem. Spherical and cylindrical, homogeneous and inhomogeneous inclusion configurations are used to validate the results of the proposed spectral method. Dependencies of the numerical solutions on homogeneity, geometry and resolution are also explored, and the differences with respect to known analytical solutions are quantified and discussed. In the case of homogeneous inclusions, the proposed numerical method is direct, i.e. does not require iteration. Using enough resolution, the micromechanical fields predicted for these simple geometries are shown to be in good agreement with the analytical results. The specific way in which inclusions are voxelized is also explored, and its effect on local fields near interfaces is assessed.

A brief summary of simulation techniques for recrystallization is given. The limitations of the Potts model and the cellular automaton model as used in their standard forms for grain growth and recrystallization are noted. A new approach based on a hybrid of the Potts model (MC) and the cellular automaton (CA) model is proposed in order to obtain the desired limiting behavior for both curvature-driven and stored energy-driven grain boundary migration.