Newswise – Structural displacement measurement is critical to the safety and integrity of civil structures. Traditional methods often face challenges such as limited measurement points and sensitivity to environmental conditions. Vision-based methods, while promising, have struggled with accuracy and robustness. Due to these issues, a new approach is necessary to advance structural health monitoring that provides more reliable and comprehensive displacement data with higher spatial resolution.
Researchers at Harbin Institute of Technology have developed a new deep-learning model, Nodes2STRNet, for structural dense dislocation detection. Study (DOI: 10.1002/msd2.12083)published in International Journal of Mechanical System Dynamics In 2023, it shows how this model outperforms traditional methods. Nodes2STRNet uses a combination of a control node estimation subnetwork (NodesEstimate) and a pose parameter recognition subnetwork (Nodes2PoseNet) to provide accurate and robust dense structural displacement measurements.
The main innovation of Nodes2STRNet lies in its ability to accurately identify structural displacements using a deformable 3D mesh model and dense optical flow. The NodesEstimate subnetwork calculates the 2D positions of control nodes from video frames, while Nodes2PoseNet maps these coordinates onto anatomical pose parameters. This method overcomes the limitations of sparse point measurements by providing a dense displacement field. The self-supervised learning strategy further enhances its efficiency, eliminating the need for extensive manual annotation. Experimental validation through seismic vibration table tests on a four-story building model confirmed the model’s high accuracy and robustness to lighting condition variations. The results showed that Nodes2STRNet consistently outperformed existing methods in recognizing displacements under various extreme ground accelerations and lighting conditions. This improvement is especially important for applications requiring high precision and reliability in dynamic environments.
Dr. Yang Xu of Harbin Institute of Technology said, “Nodes2STRNet represents a significant advance for structural displacement detection in the field of structural health monitoring. Its ability to provide accurate and dense displacement data even under challenging conditions significantly enhances Will happen.” Our ability to monitor and maintain the condition and integrity of critical structures.”
The development of Nodes2STRNet has important implications for structural health monitoring and disaster prevention. Its robust and accurate displacement measurement can be applied to a variety of structures from buildings to bridges, ensuring their safety and reliability. Additionally, this technology can aid in early detection of structural health problems, prevent potentially catastrophic failures, and improve the overall resiliency of civil infrastructure.
,
Reference
DOI
original source url
https://doi.org/10.1002/msd2.12083
Funding Information
National Natural Science Foundation of China, grant/award numbers: 52192661, 51921006, 52008138; China Postdoctoral Science Foundation, grant/award number: BX20190102, 2019M661286; Heilongjiang Provincial Natural Science Foundation, grant/award number: LH2022E070; Heilongjiang Province Postdoctoral Science Foundation, grant/award numbers: LBH TZ2016, LBH‐Z19064.
About this International Journal of Mechanical System Dynamics
International Journal of Mechanical System Dynamics (IJMSD) is an open-access journal that aims to systematically reveal the important impact of mechanical system dynamics on the entire lifecycle of modern industrial equipment. Mechanical systems can vary in scale and are integrated with electronic, electrical, optical, thermal, magnetic, acoustic, aero, fluidic systems, etc. The journal welcomes research and review articles on dynamics related to advanced theory, modelling, calculations, analysis. Software, design, control, manufacturing, testing, and evaluation of general mechanical systems.
(TagstoTranslate)Newswise(T)Cutting Edge Technologies(T)All Journal News(T)Technology(T)Top Hit Stories(T)Chinese Academy of Sciences
