Optimum design of steel frame with genetic algorithm

SUMMARY Optimum Design of Steel Frame with Genetic Algorithm Genetic algorithm (GA) is an optimization method that obtains an ideal result among available choices that is solution of problem or system by ensuring desired criteria. According to the method, a set of possible solutions for problem is simulated real populations struggling for survival in nature. Difficulties occuring during life provide the survival of the fittest while the others are eliminated from these populations. If all of the populations represent a set of available solutions for problem, GA tries to find the best individual among these populations by starting with any of them. In this study, the minimum weight design of steel framework is done by using GA with different encoding, intelligent mutation and flexible crossover operators. For this purpose, a program is coded in FORTRAN. In the program, stress, stability, slope and geometrical conditions are included as constraint. In the analysis of the frame systems, Matrix Analysis of Structures is used. This study consists of four chapters. In the first chapter, some information about optimization technique and the purpose of the study are given. In the second chapter, some steel framework examples taken from literature are optimized with GA, and the design results obtained in this study are compared with the design results taken from literature. In the third chapter, the conclusions drawn from this study are presented and some recommendations are given. In the forth chapter, the references are presented. In conclusion, it appears that varieties of different encoding with intelligent mutation and flexible crossover operators increase performance of GA. Keywords: Minimum Weight Design, Genetic Algorithm, Variety of Different Encoding, Flexible Crossover, Intelligent Mutation, Steel Framework VI