基于频响函数和改进PSO算法的结构损伤识别方法  

METHOD OF STRUCTURAL DAMAGE IDENTIFICATION BASED ON FREQUENCY RESPONSE FUNCTION AND IMPROVED PSO ALGORITHM

作　　者：邹万杰[1] 马媛[1] 章云霞 郭昭君[1] 寇金鑫 

机构地区：[1]广西科技大学土木建筑工程学院,广西柳州545006

出　　处：《钢结构》2016年第7期30-33,16,共5页Steel Construction

基　　金：广西自然科学基金项目(2011GXNSFA018032)

摘　　要：为解决在一定噪声水平影响下的结构损伤识别问题,提出一种基于频响函数和改进的粒子群算法的结构损伤识别方法。以单元刚度折减因子为优化变量,采用实测频响函数和计算频响函数的相关系数来构造粒子群算法(PSO)的优化目标函数和适应度函数;考虑到简单PSO算法在寻优过程易"早熟"的问题,采用增大粒子后期位置改变量的改进策略,最后通过该算法对IASC-ASCE SHM Benchmark结构进行损伤识别。结果表明:改进后的算法是有效的,且较简单PSO算法结果更精确,收敛速度更快。In order to solve the problem in a certain noise level under the influence of structural damage identification,a combined method for structural damage identification based on frequency response function（ FRF） and improved particle swarm optimization was presented. The reducing factors of element stiffness were used as the optimization variables,and the correlation coefficients of the measured FRF and the analytical FRF were used to construct the optimization object function and the fitness function of PSO. Considering the simple PSO algorithm was prone to ＂premature ＂ in the optimization process,so the strategy to increase the position changing quantity of the particles in the later period was used,and finally IASC-ASCE SHM Benchmark structure was used to identify the damage under the noise influence by the improved PSO. The results showed that the improved algorithm was effective and more accurate,and the convergence speed was faster than the simple PSO.

关 键 词：噪声 损伤识别 频响函数 改进粒子群算法 相关系数 

分 类 号：TU317[建筑科学—结构工程]