基于缩聚模态应变能与频率的结构损伤识别

STRUCTURAL DAMAGE DECTION BASED ON REDUCED MODAL STRAIN ENERGY AND FREQUENCY

在结构损伤识别中,测试振型往往是不完整的,这阻碍了结构损伤识别技术的应用效率。提出了一种采用缩聚应变能与频率相结合的损伤识别定性与定量方法。采用基于Neumann级数展开的方法对有限元模型进行缩聚,定义了单元缩聚模态应变能,并证明了缩聚模态应变能的变化对损伤的敏感性;将单元缩聚应变能变化率作为标识量来识别损伤的位置;在损伤位置初步判定的基础上,采用特征值灵敏度法求解损伤程度。以一简支钢梁为例证明了所提出的方法。结果表明:在测量模态数据有误差的情况下也能较好地给出损伤识别结果。

In the process of structural damage detection, the incompleteness of experiment mode shape is the main obstacle which hinders the efficiency of applying structural damage detection technique. A damage detection method is presented here which solved the qualitative and quantitative identification based on the reduced modal strain energy and frequency. The finite element model is reduced by Neumann series expansion. The reduced modal strain energy is defined, which is proved sensitive to damage. The variation rate of reduced modal strain energy is taken as the damage signature to identify the location of damage. With the probable damage locations determined, the eigenvalue perturbation method was employed to estimate the damage extent. The proposed method is demonstrated through the example of a simple steel beam. The results show the present method could give good damage detection results even if noise is present in test modal data.