Development of a Robust Computational Design Simulator for Industrial Deformation Processes

As part of this continuing DMII-funded NSF project, computationally rigorous gradient-based optimization methodologies are being addressed for a virtual materials process design that is based on quantified product quality and accounts for process targets and constraints including economic aspects. Computational design techniques will be developed that can be used to select the necessary sequence of forming and intermediate thermal-stage processes, select appropriate dies and preforms and control/design the various process parameters such that, for a given raw material with a given initial geometry, one can obtain a final product with desired microstructure and shape under various process constraints and with minimal utilization rates and overall cost. A framework for preform as well as process parameter optimization for singleand multi-stage metal forming processes is considered. The design of each individual process will be performed using gradientoptimization techniques that are based on a rigorous continuum sensitivity analysis. Multi-stage sensitivity algorithms are proposed that allow the sensitivity fields of individual processes to be used in multi-stage process design. Optimal microstructure evolution paths, ideal forming techniques and knowledge based expert systems will be used to select the required sequence of processes and to develop feasible initial designs. The reliability of the design process will be quantified with respect to uncertainties in the physical and mathematical model. The combination of ideal forming and sensitivity analysis provides a powerful virtual design environment for deformation processes. To focus the proposed work towards the design of industrial deformation processes, we will emphasize the use of polycrystalline constitutive models that account for the induced microstructure changes during processing, realistic modeling and representation of frictional and contact conditions, complicated die and preform geometries and practical process constraints and objectives. Materials and aerospace manufacturing industries will also collaborate with us during this investigation thus allowing industrial use of the proposed techniques and providing the Cornell team with valuable technical information. The use of ideal forming to obtain an initial design and a continuum sensitivity analysis for computing the optimal process design constitutes a mathematically and physically rigorous and computationally effective methodology for process design of metallic components in advanced manufacturing applications. These developments will lead to a virtual process laboratory that will assist industry in reducing lead time for process and product development, in trimming the cost of an extensive experimental trial-anderror process development effort, in developing processes for tailored material properties and in increasing volume/time yield.