Research On Optimization Method And Application Of Continuous Bridges Based On Genetic Algorithm

Structural optimization can not only reduce the weight and construction cost of structures, but also improve the performances of structures effciently. Since 1960's, with the development of FEM and computer, structural optimization has developed very quickly. Traditional mathematical programming is still adopted by structural optimization now. However,the traditional optimization methods are unable to solve the global optimization problem. In bridge structural optimization problem, there are a number of complex and non-linear restriction qualification, which increases the difficulty of solving such problems. So new optimization methods should be taken into account.Recently, genetic algorithm as a new kind of intelligent optimization method which is set up based on the evolutionary theory of biology has been applied in many fields, structural optimization is an important aspect of its application. Such weakpoints of GA exist, like much computing time, slow velocity to convergence, and easy to fall into the local optimal solution. It is necessary to improve SGA to increase its realibility.The main research contents and innovative achievement go as follows:1. Based on literature and study of the relevant fields, the basic concept and characteristic is studied, and the common structural optimization methods and their application in bridge engineering are summarized.2. The basic theory and characteristic of GA is introduced. Particular attentions are paid on studys of the mathematical model, restriction qualification disposal, the original population and the choice of the controlling genetic parameter of the GA applied into structural optimization. The feasibility and rationality of GA to solve engineering structural optimization are tested by practical optimization example.3. The shortcomings of SGA applied in bridge structural optimization are analysed. The quasi-exactness penalization function is adopted to solve the complex restriction qualification in continuous bridge optimization problem. And the hybrid method of Complex methods and GA is presented to improve its searching ability. The feasibility and rationality of improved GA to solve prestressed concrete bridge structural optimization are tested by practical optimization example.4. On the basis of current research and engineering experiences, the key design parameter of prestressed concrete bridge are summarize when optimization. More attentions are paid on to solve the optimization problem of the superstructure of the prestressed concrete bridge, and he mathematical model is built,. The objective function, the design variables and the restriction qualification are explained in detail. Also the flow chart of the optimization procedure is built. The feasibility is argued.5. Prestressed concrete continuous bridge optimization design is investigated by using the improved genetic algorithms which is introduced in this paper. The reliability and effectiveness o fthe algorithms are verified by all of the results of this actual engineering example. And the calculated results show that the algorithm is successful and simple to solve the actual engineering structural optimal design problem. And that based on the results of computation, some relative conclusions are presented.