Improved Ant Colony Algorithm And Its Application In Truss Structure Optimization

To make some of evaluation indicators of the designed structure getting the best isthe purpose of structure optimization, for example the cost, the stiffness, the weight, etc.And must ensure that the norms and specific requirements restrictions can be met.Trussstructure is a very popular structure.Truss structure can be repeatedly used, it has a goodassembly,a higher material utilization, light quality, and its carrying capacity is verystrong. Therefore, the truss structure is widely used in engineering structures, so,trussstructure optimization design becomes more important.In practical engineering, the problem becomes more complex, traditionaloptimization methods have been unable to resolve, so computational intelligencemethods become the key issues of contemporary research. Ant colony algorithmcombines the greedy type search, positive feedback mechanism and distributed parallelcomputing, so that it can search global optimal solution. However, the ant colonyalgorithm requires a longer search time, stagnation occurs very easily. In this paper, forthe shortcomings of the ant colony algorithm, proposed a series of improvements, a newant colony algorithm is formed. It can effectively avoid local optimization problem, butalso accelerate the convergence rate. And it is successfully applied to optimize thedesign of the truss structure, improve the optimization of efficiency and cost savingsprojects.This paper studies the basic ant colony algorithm and refers to the predecessorsvarious improvements of the ant colony algorithm. According to the advantages anddisadvantages, proposed an improved ant colony algorithm. And designed a specificprocess of the improved ant colony algorithm. Formed a complete optimization process.Use the Improved ant colony algorithm in truss optimization problems, the mainproblem in this paper is the cross-sectional area of discrete optimization of trussstructures. Use Matlab7.0in structural optimization procedures for the preparation. Byanalyzing some typical truss structure, and comparing with other methods, prove thispaper improved ant colony algorithm is feasible and effective.In the section of the truss structure optimization, the improved ant colonyalgorithm this paper achieved a good application, greatly reducing the project cost,saving material. So, putting forward a feasible optimization method, and can beingeffective in practical engineering applications, but also can bring considerable economic benefits is the significance of this issue lies.