Off-campus UMass Amherst users: To download dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users, please click the view more button below to purchase a copy of this dissertation from Proquest.

(Some titles may also be available free of charge in our Open Access Dissertation Collection, so please check there first.)

A feature-based representation to support the design process and the manufacturability evaluation of mechanical components

David William Rosen, University of Massachusetts Amherst

Abstract

To support the needs of intelligent computer-aided design (CAD) tools for mechanical design, manufacturing, and concurrent engineering, new computer-based artifact representations are needed. The overall goal of this project was to develop a representation of mechanical components that supports design at multiple design stages and supports manufacturability and other life-cycle evaluations. The concept for constructing intelligent CAD tools in this dissertation revolves around a central product database that captures the designer's view of the design. Since the designer's view of the product may not be suitable for subsequent downstream (manufacturing, assembly) or performance evaluations, a method of converting the design representation into suitable secondary representations is proposed. The development of a general conversion methodology is shown to enable the successful implementation of these concepts. A Feature-Based Design-For-Manufacturability (FBDFM) system for thin-walled components has been developed that embodies these concepts. Design activity and manufacturability evaluations are supported at both the configuration (qualitative information) and parametric (quantitative information) design stages. Manufacturability evaluations for tooling cost and cycle time have been developed and integrated into FBDFM for injection molding and die casting processes. FBDFM has been applied to the design, manufacturability evaluation, and redesign of thin-walled components taken from actual products. Features are meaningful abstractions of the geometry and intended behavior of parts of components that are relevant to one or more life-cycle activities. Design-with features have relevance to design activity. Their use as design primitives facilitates not only the design process, but also facilitates conversions to secondary representations, composed of features relevant to those secondary activities. Thus, the central artifact representation, called the primary representation, consists of design-with features and includes their connectivity, geometry, size parameters, and tolerances. A principle result of this project was the formalization of features, feature-based representations, and representation conversions utilizing mathematical models. The first model investigated feature-based representation conversion as a mathematical function. It was shown that conversion functions can have discontinuities. The second math model was an application of formal language theory. Graph grammars were developed that described the generation of thin-walled components in FBDFM and that described the conversion of resulting components into tooling cost manufacturing features. The design grammar was a formal approach to the development of a suitable means of communication between designers and CAD systems where the sentences in the language could be used as artifact representations. The use of a feature-based design representation was shown to enable design activity and downstream evaluations at multiple design stages and promises to be a suitable foundation for future intelligent CAD systems. Theoretical models support the intuitive concepts and experimental results and guided the development of computational models, algorithms, and the FBDFM system.

Subject Area

Mechanical engineering

Recommended Citation

Rosen, David William, "A feature-based representation to support the design process and the manufacturability evaluation of mechanical components" (1992). Doctoral Dissertations Available from Proquest. AAI9233152.
https://scholarworks.umass.edu/dissertations/AAI9233152

Share

COinS