Off-campus Michigan Tech users: To download campus access theses or dissertations, please use the following button to log in with your Michigan Tech ID and password: log in to proxy server

Non-Michigan Tech users: Please talk to your librarian about requesting this thesis or dissertation through interlibrary loan.

Date of Award

2019

Document Type

Campus Access Master's Report

Degree Name

Master of Science in Mechanical Engineering (MS)

Administrative Home Department

Department of Mechanical Engineering-Engineering Mechanics

Advisor 1

Susanta Ghosh

Committee Member 1

Sajjad Bigham

Committee Member 2

Steven Ma

Abstract

Novel hybrid materials with micro-architectured arrangements find wide range of potential applications due their inherent capability of achieving diverse, variable and tunable physical response, through structural and material modifications. Given the need for a pre-desired response and the degree of control of parameters available with such materials, it becomes necessary to develop response databases which can be used to simplify the design process. In the present work, one such “High-symmetry interlocking micro-architecture” (Haldar et. al., 2017), has been investigated for optimization of stiffness and yield strength. As the micro-architecture exhibits periodicity, it is sufficient to analyze a characteristic Representative Volume Element (RVE) with the accurate boundary conditions. The analysis are carried out under quasi-static tensile loading through computational studies. The intricacies of the present micro-architecture arise from a large number of geometric variables which give rise to multiple combinations and a high dimensional engineering design space. Navigating such vast design space requires data-driven statistical methods to assist decision making for identifying optimal configurations. The responses from Finite Element (FE) simulations were studied using Analysis of Variance (ANOVA) methods to consolidate the design space in successive steps. The results from ANOVA are used to identify significant parameters which can be used to simplify further optimization and also assist development of predictive statistical models. Although ANOVA greatly reduces the number of configurations under study, the difficulty remains in simulating these multiple number of qualified members. Parametrization of the model allows automation, to generate results from multiple FE simulations. Computational FE softwares require geometry identification to be able to assign material properties and boundary conditions to the model. However, the complexity of the micro-architecture poses a major challenge to automate identification of these changing geometric features. To overcome this, a python program has been developed which takes minimal input from the user to conduct pre-processing and post-processing of the FE Models. Also, the results from FE simulations were analyzed to identify how the stress distribution and load transfer mechanisms change over various configurations in the design space. Based on these results from Finite Element Simulations and ANOVA , regression equations have also been developed which can be used to generate configurations for a desired level of stiffness and strength.

Share

COinS