研究目的
To develop a method for diagnosing the most probable causes of cracks on existing masonry structures using photogrammetry and distinct element modeling.
研究成果
The method successfully diagnosed the most probable causes of cracks in masonry structures by combining photogrammetry and DEM, with the real settlement scenario consistently showing the closest match to experimental crack patterns. It was applied to a real-world case study, demonstrating potential for practical use in historic preservation. Future work should focus on automation and interpolation between loading scenarios to enhance robustness and efficiency.
研究不足
The method is limited by the loading scenarios input; if a specific type of settlement is not simulated, it cannot be identified as a cause. The process is time-intensive without automation, and subsampling may not capture all joint behaviors. Assumptions in DEM, such as rigid blocks and specific joint properties, may not fully represent real-world conditions.
1:Experimental Design and Method Selection:
The study involved constructing and cracking five masonry walls in a laboratory setting, using settlement as the cracking mechanism. Photogrammetry was used to capture the state of the walls before and after cracking, and distinct element modeling (DEM) was employed to simulate various loading conditions, including the actual experimental load and other scenarios (wide, deep, different). A combined qualitative and quantitative method was used to compare crack patterns from simulations and experiments.
2:Sample Selection and Data Sources:
Five single-leaf masonry test walls were built using bricks without mortar, with specific dimensions and density. Settlement was applied using a manual jack to induce cracking, with different settlement types (local and global) to examine various crack patterns.
3:List of Experimental Equipment and Materials:
Equipment included a manual jack (OrionMotorTech Dilated Scissor Lift jack), Nikon D90 camera for photogrammetry, bricks with specified dimensions and density, cinder blocks as a fixed base, and software such as AgiSoft Photoscan, Rhino, and 3DEC for modeling and simulation.
4:Experimental Procedures and Operational Workflow:
Walls were constructed, settled to induce cracking, and documented with photogrammetry before and after damage. Images were taken with high overlap, and 3D models were generated. Computational models were created in Rhino and simulated in 3DEC under different loading conditions. Crack widths were calculated from simulation data using a custom method, and comparisons were made using subsampling and metrics like fractional distance.
5:Data Analysis Methods:
Data analysis involved calculating crack widths from photogrammetry and simulation outputs, using linear regression (slope, intercept, R2) and fractional distance to compare experimental and numerical results. A sensitivity analysis was performed on input parameters like density, stiffness, and friction angle.
独家科研数据包,助您复现前沿成果,加速创新突破
获取完整内容