(Dr. Shreyes Melkote, advisor)
"Fixture-Workpiece Contact Modeling for a Compliant Workpiece"
A fixture consists of a set of unilateral frictional contacts that locate, constrain and support a workpiece during machining. The problem of fixture design optimization involves determination of the number of fixture elements (locators and clamps), their layout, and the magnitude of the clamping forces that will minimize the error due to workpiece deformation arising from clamping and machining forces. Central to this problem is the need for an accurate model of the fixture-workpiece system to base the design optimization on. This is especially critical for fixturing of compliant workpieces, which can undergo large elastic deformations induced by the clamping and machining forces.
Finite element models take into account all the compliances in the system and therefore can predict the workpiece deformation accurately. However, accuracy of this modeling approach depends largely on how the fixture-workpiece contact interface is modeled. Commonly used boundary conditions like nodal degree of freedom (d.o.f) constraints result in erroneous contact deformation predictions. Non-linear contact elements offer a potential solution to this problem, but there exist no detailed studies to develop guidelines for their use in modeling fixture-workpiece contact problems.
In this thesis, detailed analysis of the different ways of modeling the fixture-workpiece contact interface using the finite element method, focusing on model accuracy and computational cost, was conducted. Specifically, different ways of modeling locator/clamp-workpiece contacts using point/distributed force boundary conditions, displacement boundary conditions, and non-linear contact elements were examined. Surface-to-surface contact elements were found to be the most suitable method for modeling the fixture-workpiece contact. Guidelines were generated for modeling the fixture-workpiece contact interface for two common contact geometries: planar-planar and spherical-planar. Experimental validation for single and multiple-contact cases was performed.
Clamping force optimization was then performed for several example fixture-workpiece
systems modeled using the guidelines generated in the single contact study.
Response surface method was employed to develop an analytic objective function
for the optimization to avoid several solutions of the computationally
intensive finite element model. Good agreement was found between prediction