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 model for incorporating robustness, material choice, and memory in the design of mechanical components
Abstract
Dominic III, the last in a series of computer programs, numerically sizes mechanical components to satisfy performance requirements. It solves the same types of problems as optimization and Taguchi Methods, but it is more flexible and natural in its approach, especially in expressing trade-offs between performance requirements. Dominic designs by explicitly considering these trade-offs rather than minimizing or maximizing a utility function. Dominic has successfully designed in such domains as V-belt drives, hydraulic cylinders, I-beams, solar heating systems, and the partial design of an automobile suspension system. This research extends the work embodied in Dominic II (Orelup 90) through the development of an algorithm to change several design variables at once and a richer model of redesign on which to base a new meta-control formulation. This research also demonstrates how to incorporate the following within a methodology for parametric design: (1) include robustness as well as performance criteria to incorporate the Taguchi philosophy of quality (i.e., minimize variability in performance caused by inherent variability in manufacturing and environmental conditions) by modelling robustness as a performance parameter; (2) handle non-numeric design variables (such as choice of materials) through meta-control and an algorithm based on attempting local improvement; and (3) utilize a memory of past solutions by applying knowledge gained in a similar, previously solved problem to the current problem. Evaluation of Dominic III shows (1) the dramatic effect changing several design variables simultaneously has on convergence to the extent of eliminating the need for meta-control; (2) a similarity between successful redesign strategies across domains; (3) effective production of robust designs plus responsiveness to trade-offs between achieving robustness and good performance; (4) designing with multi-dimensional design variables produces designs as good as exhaustively designing all possible choices; and (5) a memory of twenty problems will improve convergence and quality of solution, but probably not as well as a good heuristic method for an initial design.
Subject Area
Mechanical engineering
Recommended Citation
Orelup, Mark Ferrall, "A model for incorporating robustness, material choice, and memory in the design of mechanical components" (1992). Doctoral Dissertations Available from Proquest. AAI9233125.
https://scholarworks.umass.edu/dissertations/AAI9233125