Details

Design Optimization of Mechanical Components

by DESHMUKH, DINAR VIVEK

Abstract (Summary)
The need for high performance, low cost designs in the engineering industry is what drives designers to search for a utopian design. Design optimization demands greater importance in product development than any other aspect involved. Usually for a design problem under consideration, there are multiple parameters that need to be optimized and often these are conflicting in nature. The designer needs to arrive at a suitable design that fulfills all the requirements but also meets all the optimization parameters to the greatest extent possible.With this goal, an integrated approach to optimization is presented here with the scope of designing mechanical components in mind. The method presented incorporates a goal programming approach called the Compromise Decision Support Problem Technique combined with a Branch and Bound Algorithm to solve multi-objective, constrained non-linear problems with mixed variables. The approach is initially presented qualitatively, and later illustrated with practical design problems.The approach is well suited to solving real life engineering problems that are highly constrained, have multiple objectives and the designs generated need to adhere to industry standards. We have applied the approach to designing a helical compression spring and a three-stage gear train; components that find universal application in mechanical engineering. The nature of Pareto Optimal solutions for multiple objective problems is explored and a trade-off analysis is proposed. The results obtained are compared with existing solutions in literature, and comparisons are drawn.
Bibliographical Information:

Advisor:

School:University of Cincinnati

School Location:USA - Ohio

Source Type:Master's Thesis

Keywords:multiple objective optimization pareto optimality mechanical design compromise decision support problem branch and bound algorithm

ISBN:

Date of Publication:01/01/2002

© 2009 OpenThesis.org. All Rights Reserved.