Optimization Frameworks for Discrete Composite Laminate Stacking Sequences

(ABSTRACT) Composite panel structure optimization is commonly decomposed into panel optimization subproblems, with specified local loads, resulting in manufacturing incompatibilities between adjacent panel designs. Using genetic algorithms to optimize local panel stacking sequences allows panel populations of stacking sequences to evolve in parallel and send migrants to adjacent panels, so as to blend the local panel designs globally. The blending process is accomplished using the edit distance between individuals of a population and the set of migrants from adjacent panels. The objective function evaluating the fitness of designs is modified according to the severity of mismatches detected between neighboring populations. This lays the ground work for natural evolution to a blended global solution without leaving the paradigm of genetic algorithms. An additional method applied here for constructing globally blended panel designs uses a parallel decomposition antithetical to that of earlier work. Rather than performing concurrent panel genetic optimizations, a single genetic optimization is conducted for the entire structure with the parallelism solely within the fitness evaluations. A guide based genetic algorithm approach is introduced to exclusively generate and evaluate valid globally blended designs, utilizing a simple master-slave parallel implementation, implicitly reducing the size of the problem design space and increasing the quality of discovered local optima. ACKNOWLEDGEMENTS I would like to thank Prof. Layne T. Watson for serving as my research advisor and committee chairperson. His broad perspective and vision coupled with his tireless work ethic create in him the paragon of what a research advisor should be. He's seen me through to the end of an era and helped place me on the road to the next. I can not thank him enough for his encouraging words and understanding stance during these last few months of my life. I wish to also thank Zafer Gürdal and Omprakash Seresta for providing me with the engineering reviews required for the work. I am indebted to the Terascale Computing Facility at Virginia Tech for a generous grant of computer time on System X as well as time on the Anantham cluster. To my friend D. Slotta who in times of great need always seemed to be there to lend a helping hand and remind me to keep things in perspective, I will not forget. To P. Haggerty who graciously let me live in his home during the completion of this work while I was homeless, I am indebted forever. Finally, to Elissa, …