Genetic Algorithm And Beam Structure Topology Optimization Design Study

Genetic algorithms is a novel optimum method based on the biological evolution theory, which attracts the many researchers' attention in the fields of computer science and the optimum design in recent years. Comparing to the conventional optimum algorithm, genetic algorithms has two remarkable merits including the hiding concurrency and the global convergence. Moreover with the continuous development of its capacity, this algorithms is becoming more suitable for engineering application. Meanwhile, the topology optimization is applied widely in many engineering structures. And the research about it is also very important.In this thesis, the basic theory of genetic algorithm is first analyzed. And the program system of this algorithms is realized. Then the improvement of genetic algorithms is completed. By virtue of two-order fitting technique, a new method for fast determining genetic operator is discussed. This method is very useful to promote the efficiency of genetic algorithms and precision of results. Finally the correctness and feasibility of this method are verified by several numerical examples.In present the optimum design of beam structure is a very meaningful and valuable topic in engineering fields. In order to solve the problem of topology optimization of beam structure, the mathematic model of beam is investigated. With this model, the topology optimization is completed by employing the software ANSYS. It is believed that the work presented in this thesis should provide good theoretical background for the design of the beam structure.