基于改进灰狼优化算法的结构损伤识别  

Structural damage identification based on improved gray wolf optimization algorithm

作　　者：谢少鹏 吴柏生 赵秀婷 邹敏清[3] 林丹锐 XIE Shao-peng;WU Bai-sheng;ZHAO Xiu-ting;ZOU Min-qing;LIN Dan-rui(School of Advanced Manufacturing,Guangdong University of Technology,Jieyang 522000,China;School of Electro-Mechanical Engineering,Guangdong University of Technology,Guangzhou 510006,China;Experimental Teaching Department,Guangdong University of Technology,Guangzhou 51006,China)

机构地区：[1]广东工业大学先进制造学院,揭阳522000 [2]广东工业大学机电工程学院,广州510006 [3]广东工业大学实验教学部,广州510006

出　　处：《计算力学学报》2024年第2期256-262,共7页Chinese Journal of Computational Mechanics

基　　金：国家自然科学基金(11672118);广工揭阳理工(筹)紧缺型人才科研启动基金资助项目.

摘　　要：结合模态柔度矩阵、广义模态柔度矩阵和振型三个识别精度较好的指标,构造新的目标函数求解损伤识别问题。通过Nelson方法求解得到的频率与振型的导数,得到对结构刚度发生变化时更具敏感性的位置,然后在这些位置布置传感器以提取结构信息。针对原有的灰狼算法虽然全局搜索能力强,但是存在局部搜索精度差的问题,本文从初始种群和收敛因子等方面着手,改善灰狼算法的局部搜索能力及收敛速度。最后利用提出的方法,通过识别梁模型及桁架模型中的损伤单元说明本文方法的有效性。In this paper,a new objective function is constructed to solve the damage identification problem by combining three indicators with good identification accuracy:modal flexibility matrix,generalized modal flexibility matrix and mode shape.The derivative of a frequency and a mode shape obtained by Nelson method is used to obtain the positions that are more sensitive to the change of structural stiffness,and then sensors are arranged at these positions to extract structural information.Although the original gray wolf algorithm has strong global search ability,it has the problem of poor local search accuracy.This paper improves the local search ability and convergence speed of gray wolf algorithm from the aspects of initial population,convergence factor,etc.Finally,the effectiveness of the proposed method is demonstrated by identifying the damage elements in a beam model and a truss model.

关 键 词：结构损伤识别 灰狼算法 灵敏度定位法 动力特性参数 Nelson方法 

分 类 号：TU311[建筑科学—结构工程] O346.5[理学—固体力学]