Genetic algorithm based optimization in engineering design using fuzzy constraints and fitness functions

Due to the character of the original source materials and the nature of batch digitization, quality control issues may be present in this document. Please report any quality issues you encounter to digital@library.tamu.edu, referencing the URI of the item.Includes bibliographical references (leaves 60-63).Issued also on microfiche from Lange Micrographics.The motivation for this work has been the use of tools, such as genetic algorithms and fuzzy sets, to address the various issues that are involved in an engineering design optimization problem. In order to address the variety, generality and uncertainty in the design problems, a genetic algorithm based approach was developed that could readily handle these difficulties. The focus of this approach has been on the selection process, which forms the heart of any genetic algorithm. In particular, a selection process that could quickly evolve possible solutions that would span the entire design domain was developed and implemented. The selection process was also tailored in such a way that it could handle the uncertainties and imprecision in the design specifications and constraints. This approach, in its basic form, was developed, tested and modified repeatedly using increasingly difficult problems. It was subsequently modified to solve a general non-linear design optimization problem. In order to handle the large design space in certain problems, the concept of alpha cuts was used to constrain the design space to a smaller and more desirable region. Design problems that are used as examples were selected from current research papers and solved. The entire algorithm was coded in MATLAB as several independent modules so that it can be used similar to any other toolbox