Cornell University Graduate School >
Cornell Theses and Dissertations >
Please use this identifier to cite or link to this item:
|Title: ||Multi-scale Computational Techniques For Design Of Polycrystalline Materials|
|Authors: ||Sundararaghavan, Veeraraghavan|
|Keywords: ||Computational mechanics|
|Issue Date: ||20-Jul-2007|
|Abstract: ||Microstructures play an important role in controlling distribution of properties in engineering materials. It is possible to develop components with tailored distribution of properties such as
strength and stiffness by controlling microstructure evolution during the manufacturing process. When forming metallic components by imposing large deformations, mechanisms such as slip and lattice rotation drive formation of texture in the underlying polycrystalline microstructure. Such microstructural changes affect the final distribution of material properties in the component. By carefully designing the imposed deformation, one
could potentially tailor the microstructure and obtain desired property distributions. This thesis focuses on development of
novel computational strategies for designing deformation processes to realize materials with desired properties. The techniques presented are an interplay of several new tools developed
recently, such as reduced order modeling, graphical cross-plots, statistical learning, microstructure homogenization and
multi-scale sensitivity analysis. The primary outcomes of this thesis are listed below:
1. Development of reduced-order representations and graphical methodologies for representing process-property-texture relationships.
2. Development of adaptive reduced-order optimization techniques for identification of processing paths that lead to desirable microstructure-sensitive properties.
3. Development of homogenization techniques for predicting microstructure evolution in large deformation processes.
4. Development of multi-scale sensitivity analysis of poly-crystalline material deformation for optimizing microstructure-sensitive properties during industrial forming processes.
The framework for design of polycrystalline microstructures leads
to increased product yield in industrial forming processes and simultaneously allows control distribution of properties such as stiffness and strength in forged products. Multi-scale design problems leading to billions of unknowns have been solved using parallel computing techniques. The computational framework can be readily used for selecting optimal processing paths for achieving
desired properties. The methodology developed is a fundamental effort at providing detailed deformation process design solutions needed for controlling properties of performance-critical hardware components in automotive, structural and aerospace applications.|
|Description: ||Your submission was rejected by Pattie Place:I am rejecting because on page 95 there is too much white space.
But Page 95 is at the end of a chapter?|
|Appears in Collections:||Cornell Theses and Dissertations|
Items in eCommons are protected by copyright, with all rights reserved, unless otherwise indicated.