噪声影响下基于改进损伤识别因子和遗传算法的结构损伤识别  

Structural damage identification based on improved damage identification factor and genetic algorithm under noise

作　　者：黄民水[1,2] 吴玏[1] 朱宏平[2] 

机构地区：[1]武汉工程大学环境与城市建设学院交通研究中心,武汉430073 [2]华中科技大学土木工程与力学学院控制结构湖北省重点实验室,武汉430074

出　　处：《振动与冲击》2012年第21期168-174,共7页Journal of Vibration and Shock

基　　金：国家杰出青年基金(50925828);湖北省自然科学基金(2009CDB080);湖北省教育厅科学技术研究计划重点项目(D20111512)

摘　　要：提出一种考虑噪声影响基于改进损伤识别因子和遗传算法的结构损伤识别方法。介绍结构损伤识别的参数化思想,并提出一种考虑不同模态信息和权重系数的损伤识别因子。在选择算子、交叉算子、变异算子以及自适应交叉和变异等方面对遗传算法进行改进,并在无噪声情况下,对一简支梁和连续梁进行了损伤识别,识别效果非常好,验证了方法的正确性和有效性。分别考虑随机噪声和高斯白噪声对简支梁进行损伤识别,发现该方法在噪声水平不很高情况下,识别效果很好。最后,在一钢框架基准有限元模型基础上,考虑四种损伤工况进行损伤识别,识别结果较好,该方法能够很好地应用于结构的损伤识别和健康监测。A Structural damage identification method was presented here, based on improved damage identification factor and genetic algorithm under noise. Firstly, parametrization for damage identification was introduced, and a new damage identification factor was presented considering different modal messages and weighting coefficients. Second｜y, Genetic algorithm was improved in several aspects of selecting operator, crossover operator and mutation crossover; at the same time, adaptive crossover and adaptive mutation were investigated. And damage identification was performed for a simply supported beam and a continuous beam without noise, the effect was very good and the effectiveness of the method was verified. Then, two types of noise, random noise and white Gaussian noise, were considered for the simply supported beam, it was shown that identification performance is good when the noise level is not very high. In the end, based on a baseline finite element model of a steel frame, damage identification was conducted in four damaged cases, it was shown that the performance is better, the method can be used in damage identification and health monitoring.

关 键 词：结构损伤识别 改进损伤识别因子 遗传算法 随机噪声 高斯白噪声 基准有限元模型 

分 类 号：TU375[建筑科学—结构工程]