Show simple item record

dc.contributor.authorPatel, Jayankumar
dc.date.accessioned2018-05-23T20:18:43Z
dc.date.available2018-05-23T20:18:43Z
dc.date.issued2017
dc.identifier.isbn9780355310696
dc.identifier.other1983446471
dc.identifier.urihttp://hdl.handle.net/10477/77347
dc.description.abstractSolid freeform (SFF) fabrication of functionally graded material (FGM) objects has garnered much research interest since last decade. To move from research sample and prototypes to commercially viable functional FGM parts, it is necessary to develop an integrated approach for modeling, optimization, and process planning for SFF fabrication process. While solid modeling of FGM objects has been studied in detail, the build orientation optimization and process planning of FGM objects remain largely unsolved. In this paper, we introduce a novel approach for build orientation optimization (BOO) for additive manufacturing of FGM Objects. The build orientation cost function is implemented using material error and geometric error as primary factors. To create a solid model of the FGM object, we first create a solid geometric model and then map material function on it using distance field computation. Geometric error considers volumetric stair-case error and material error considers discretization of material composition as the primary factor. Discretization of material composition and multi-scale random error computation algorithm are discussed in detail. Since the build cost function cannot be calculated analytically and an expansive parametric sweep is too computationally expensive to implement, we treat cost function as a black box and use surrogate model based optimization to find the optimum build orientation. The algorithm first conducts initial few build cost computations to create a minimal design space based on Latin Hypercube Sampling (LHS) method. A surrogate model is then fitted to approximate the build cost function. Using the surrogate model, a new set of sample points is generated and used to progressively improve the surrogate model. This process is iterated until optimal orientation is achieved. We finally test our optimization algorithm on various test objects to illustrate the overall methodology.
dc.languageEnglish
dc.sourceDissertations & Theses @ SUNY Buffalo,ProQuest Dissertations & Theses Global
dc.subjectApplied sciences
dc.subjectAdditive manufacturing
dc.subjectBuild orientation optimization
dc.subjectFunctionally graded materials
dc.subjectSurrogate model based optimization
dc.titleBuild Orientation Optimization for Additive Manufacturing of Functionally Graded Materials
dc.typeDissertation/Thesis


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record