Parametric Sensitivity Analysis and Probabilistic Modeling of Thermo-Mechanical Processes

DEFORM™ is a finite element based process and material modeling system that is widely used in the design of thermo-mechanical processes including forging, extrusion, rolling, drawing, heat treatment and machining processes.  Modeling large plastic deformation processes helps to predict metal flow in the die cavity, flow defects and under fill conditions. Heat treatment process models account for diffusion mechanisms and phase transformations and the evolution of residual stresses and microstructure can be accurately predicted.

Parametric sensitivity analysis in DEFORM can account for variabilities in processing conditions, material data and boundary conditions that are typically encountered during manufacturing processes.  Statistical sampling methods are used to construct a set of design of experiments type of computer simulations.  DEFORM infrastructure enables simulation of multiple thermo-mechanical processes simultaneously and a Gaussian response surface is generated. Monte Carlo method is employed to systematically account for uncertainties associated with processing conditions, boundary conditions and material properties and to study its impact on the evolution of residual stresses and microstructure during thermo-mechanical processing.  This presentation will cover some of exciting applications of parametric sensitivity analysis and probabilistic modeling capabilities of DEFORM system.