Genetic Algorithm (GA) is a highly parallel, random and adaptive searching probabilisticmethod based on the mechanics of natural selection and genetic. The scholars of domestic andforeign pay much emphasis on the research of genetic algorithm’s theory and application, andhave made an amazing progress. The achievement of Genetic Algorithm has permeated tomany fields. The structural optimization design which can determine the best schemes forengineering design by combining optimization theory with computer technique, is to makesome evaluation indicators (weight, stiffness, cost, etc) of a designed structure which meet thevarious specific requirements orconstraints achieve to the best. Genetic Algorithm as one ofthe most important algorithms of the new smart optimization theory, has a wide range ofapplications in structural optimization. But the theory and method of Genetic Algorithm arenot mature. Some insufficiencies of algorithm also wait for further improvement andconsummation.This dissertation stuty the operation mechanism of Genetic Algorithms based on theGenetic Algorithm theory. It also analysis the elements of the Genetic Algorithms on theperformance of the algorithm and the steps of the operation and propose the improvementmenthods. The new Adaptive Genetic Algorithm is proposed for the weak local search abilityand the uncertainty of the advanced computing results of the exiating Adaptive GeneticAlgorithm.And achieve the Genetic Algorithm for structural optimization problems in theMATLAB environment. The main research work in this paper is as follows:1. Analyzed the basic theory of Genetic Algorithm deeply to indicate the implementationmethod of Genetic Algorithm. To propose the corresponding improved method by analyzedthe basic properties of fitness function, selection operator, crossover operator, mutationoperator and the effect of Genetic Algorithm performance according to the disadvantages ofgenetic algorithm. 2. In-depth analysis of the basic idea of Adaptive Genetic Algorithm and an improvedAdaptive Genetic Algorithm, as well as their advantages and shortcomings. Proposed theadaptive crossover operator and mutmion operator for the deficiencies of the existingalgorithm, which can change dynamically with the evolution of population. This improvementcan enhance convergence and certainty and improve the efficiency of optimization. The newmethod proposed the elitist strategy in order to overcome the defects that the best individualof every generation failed to protect in GA.The optimization results of typical test functionsshow that the improved Adaptive Genetic Algorithm is effective.3.Through the depth analysis of the structural optimization theory and applicationrequirements for the characteristics of the truss structure and Genetic Algorithm to choose thepenalty function method to bind the handle to get the fitness function. The improved newAdaptive Genetic Algorithm presented in this paper is programmed using MATLAB. Finally,optimize the structure of three bar truss and ten bar truss based on the idea of new algorithmand the optimal results and courses of the new algorithm are compared with that of SimpleGenetic Algorithm show that the improved Adaptive Genetic Algorithm is feasible andeffective. |