Zhongru Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhongru Yu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Dongzhe Ren
School of Intelligent Information Engineering, Chongqing Aerospace Polytechnic, Chongqing 400021, China
1 shared publication - Gang Liu
School of Civil Engineering Chongqing University Chongqing 400045 China cqu.edu.cn
1 shared publication - Guojun Deng
School of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, China
1 shared publication - Shuai Shao
School of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, China
1 shared publication - Zhixiang Zhou
College of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, China
1 shared publication
Latest Publications
- Open Access
Full-Field Dynamic Parameters and Tension Identification of Stayed Cables Using a Novel Holographic Vision-Based Method
Due to the slender geometry and low-amplitude vibrations of stayed cables, existing vision-based methods often fail to accurately identify their full-field dynamic parameters, especially the higher-order modes. This paper proposes a novel holographic vision-based method to accurately identify the high-order full-field dynamic parameters and estimate the tension of the stayed cables. Particularly, a full-field optical flow tracking algorithm is proposed to obtain the full-field dynamic displacement information of the stayed cable by tracking the changes in the optical flow field of the continuous motion signal spectral components of holographic feature points. Frequency-domain analysis is applied to extract the natural frequencies and damping ratios, and the vibration frequency method is used to estimate the tension. Additionally, an Eulerian-based amplification algorithm—holographic feature point video magnification (HFPVM)—is proposed for enhancing weak visual motion signals of the stayed cables, so that the morphological motion information of the stayed cables can be visualized. The effectiveness of the proposed method has been validated through experiments on the stayed cable models. Compared with the results obtained using contact sensors, the proposed holographic vision-based method can accurately identify the first five natural frequencies with overall errors below 5% and a maximum deviation of 6.86% in cable tension estimation. The first three normalized holographic mode shapes and dynamic displacement vectors are successfully identified, with the MAC value reaching up to 99.51%. This entirely non-contact vision-based method offers a convenient and low-cost approach for cable tension estimation, and this is also the first study to propose a comprehensive, visual, and quantifiable strategy for periodic or long-term monitoring of cable-supported structures, highlighting its strong potential in practical applications.
Open AccessFull-Field Dynamic Parameters and Tension Identification of Stayed Cables Using a Novel Holographic Vision-Based Method
Due to the slender geometry and low-amplitude vibrations of stayed cables, existing vision-based methods often fail to accurately identify their full-field dynamic parameters, especially the higher-order modes. This paper proposes a novel holographic vision-based method to accurately identify the high-order full-field dynamic parameters and estimate the tension of the stayed cables. Particularly, a full-field optical flow tracking algorithm is proposed to obtain the full-field dynamic displacement information of the stayed cable by tracking the changes in the optical flow field of the continuous motion signal spectral components of holographic feature points. Frequency-domain analysis is applied to extract the natural frequencies and damping ratios, and the vibration frequency method is used to estimate the tension. Additionally, an Eulerian-based amplification algorithm—holographic feature point video magnification (HFPVM)—is proposed for enhancing weak visual motion signals of the stayed cables, so that the morphological motion information of the stayed cables can be visualized. The effectiveness of the proposed method has been validated through experiments on the stayed cable models. Compared with the results obtained using contact sensors, the proposed holographic vision-based method can accurately identify the first five natural frequencies with overall errors below 5% and a maximum deviation of 6.86% in cable tension estimation. The first three normalized holographic mode shapes and dynamic displacement vectors are successfully identified, with the MAC value reaching up to 99.51%. This entirely non-contact vision-based method offers a convenient and low-cost approach for cable tension estimation, and this is also the first study to propose a comprehensive, visual, and quantifiable strategy for periodic or long-term monitoring of cable-supported structures, highlighting its strong potential in practical applications.