Date of Award
2018
Document Type
Open Access Dissertation
Degree Name
Doctor of Philosophy in Mechanical Engineering-Engineering Mechanics (PhD)
Administrative Home Department
Department of Mechanical Engineering-Engineering Mechanics
Advisor 1
Ossama Abdelkhalik
Committee Member 1
Nina Mahmoudian
Committee Member 2
Mo Rastgaar
Committee Member 3
Nilufer Onder
Abstract
In this dissertation, the concept of hidden genes genetic algorithms is developed. In system architecture optimization problems, the topology of the solution is unknown and hence, the number of design variables is variable. Hidden genes genetic algorithms are genetic algorithm based methods that are developed to handle such problems by hiding some genes in the chromosomes. The genes in the hidden genes genetic algorithms evolve through selection, mutation, and crossover operations. To determine if a gene is hidden or not, binary tags are assigned to them. The value of the tags determine the status of the genes. Different mechanisms are proposed for the evolution of the tags. Some mechanisms utilize stochastic operations while others are based on deterministic operations. All the proposed mechanisms are tested on mathematical and space trajectory optimization problems. Moreover, Markov chain models of the mechanisms are derived and their convergence is investigated analytically. The results show that the proposed concept are capable to search for the optimal solution by autonomously enabling the algorithms to assign the hidden genes.
Recommended Citation
Ahmadi Darani, Shadi, "System Architecture Optimization Using Hidden Genes Genetic Algorithms with Applications in Space Trajectory Optimization", Open Access Dissertation, Michigan Technological University, 2018.