论文部分内容阅读
群搜索(GSO)算法是一种新的群智能优化算法,适用于结构优化设计.本文通过对GSO算法进行改进,简化了算法的计算过程,提高了优化性能.对算法的改进主要有二个方面:一是采用随机搜索,放弃了按角度搜索的方式;二是在生成个体新位置时,增加了一个随迭代次数递减的控制变量——分量变异概率,用于限制允许变异的维的数量.通过对经典桁架算例的优化以及与标准GSO算法的计算结果比较,可以看出改进后的群搜索优化算法(SGSO)具有更好的收敛速度和收敛精度,SGSO算法的结构比GSO算法更简单、易于实现并且计算用时更少.
The group search (GSO) algorithm is a new swarm intelligence optimization algorithm, which is suitable for structural optimization design.This paper improves the GSO algorithm by simplifying the calculation process of the algorithm and improves the optimization performance. There are two main improvements to the algorithm Aspects: First, the use of random search, to give up the way to search by angle; the second is to generate a new location in the individual, with a decreasing number of iterations of control variables - component mutation probability, used to limit the number of allowable variation of the number of By comparing with the classical GSO algorithm and comparing with the standard GSO algorithm, it can be seen that the improved SGSO algorithm has better convergence speed and convergence precision. The structure of SGSO algorithm is more than that of GSO algorithm Simple, easy to implement and less time consuming.