The Application Research Of Genetic Algorithms In Structure Optimization Design Based On Reliability

The structure optimization design based on relizaility considers syntherically the capacity ,economic agent , reliability and safty of structures .It's the development trend of structure optimization design .However . The structure optimization design based on relizaility is difficult to some extent .Genetic Algorithms(GA),based on Darwin's biological evolution and Mendel's theory of heredity,It's a adaptive, illuminating, probabilitic and iterative algorithm .It doesn't need any auxiliary konwledges such as differential, and only need objective functions which influence the search direction and the corresponding fitness function .Thus ,it's fit for complex problems. Since the appearance of GA ,some improved GA have been raised . The convergence of the aggorithms is one of the important research aspects . In this paper ,a improved GA was used to carry out structure optimization design based on reliability ,and the convergence was improved . Main points of this paper is as following :1) Reappeared the characteristic of GA using Matlab GA toolbox .The results provided the basis of improvements of subsequent improved GA.2) Discussed the iteration of AF0M(2), improved it has better iteration efficiency and convergence.3) Put forward a improved GA , i.e. OMFGA, with decimal codeing and optimum maintaining.4) Proved the convergence of OMFGA, using two general precondition of the convergence of FGA.