摘要
以柔性曲拱为研究对象,提出一种双尺度数字图像相关方法。在整数像素尺度,通过傅里叶变换快速找到变形参数的像素级初始估计;在亚像素尺度,结合像素级初始估计和逆合成高斯-牛顿迭代,得到亚像素精度的振动测量。在获得结构振动响应后,使用频域分解方法识别结构的固有频率、振型。最后,开展柔性曲拱的视觉测量实验,并辅以激光测振仪测量进行对比分析,验证了所提方法在柔性结构振动测量和模态辨识中的有效性。
A two-scale digital image correlation method based on flexible curved arch is proposed in this paper.In integer pixel scale,the initial estimation of deformation parameters at pixel level is quickly found by Fourier transform.In sub-pixel scale,combined with pixel level initial estimation and inverse synthesis Gaussian-Newton iteration,the deformation measurement results with sub-pixel precision is obtained.After the vibration response of the structure is obtained,the natural frequency and mode of the structure are identified by frequency domain decomposition method.Finally,the visual measurement experiment was conducted on a flexible curved arch,supplemented by comparative analysis of laser vibration measurement,to verify the effectiveness of the proposed method in vibration measurement and modal identification of flexible structures.
作者
李婉培
吕中荣
谢培悦
焦柯
汪利
LI Wanpei;LÜZhongrong;XIE Peiyue;JIAO Ke;WANG Li(School of Aeronautics and Astronautics,Sun Yat-sen University,Shenzhen 518107,China;Guangdong Architectural Design&Research Institute Company Limited,Guangzhou 510010,China)
出处
《中山大学学报（自然科学版）（中英文）》
CAS
CSCD
北大核心
2024年第2期108-114,共7页
Acta Scientiarum Naturalium Universitatis Sunyatseni
基金
广东省重点领域研发计划(2022B0101080001)
国家自然科学基金(11972336)。
关键词
数字图像相关方法
双尺度位移恢复
柔性大变形结构
模态辨识
digital image correlation
two-scale displacement recovery
flexible and large deformation structure
modal identification
