A systematic framework for DEM study of realistic gravel-sand mixture from particle recognition to macro- and micro-mechanical analysis. (May 2022)

• Record Type: Journal Article
• Authors: Li, Ruidong; Hu, Xiangyu; Chen, Fan; Wang, Xiang; Xiong, Hao; Wu, Hao
• Is Part Of: Transportation geotechnics. Volume 34(2022)
• Publication Date: 2022-05
• Subjects: Gravel sand mixture -- Particle shape -- Deep learning -- DEM simulation; Engineering geology -- Periodicals; Soil mechanics -- Periodicals; Rock mechanics -- Periodicals; Transportation -- Periodicals; 624.15105
• DOI: 10.1016/j.trgeo.2021.100693
• Languages: English
• ISSNs: 2214-3912

Abstract

Abstract: Reproducing realistic particle shape is vital for discrete modeling of granular construction materials, such as sands and gravels, which are widely used foundation soils in civil engineering. The existing algorithms used to determine particle shapes from the raw images of construction materials are very limited, especially when dealing with binary mixtures, such as sandy gravel. To address this issue, this study aims to develop a systematic framework for realistic simulation of the gravel-sand mixture based on deep-learning-enhanced discrete element method (DEM). An efficient and convenient method is proposed to quickly identify particle contour and establish particle shape libraries based on a combination of YOLOv5 and U-Net algorithms. Furthermore, DEM-based biaxial compression tests are conducted on two groups of gravel-sand mixture based on the acquired realistic coarse and fine particle shapes. The influences of coarse particle shapes and fine sand content on the macroscopic and microscopic behaviors of gravel-sand mixtures are quantitatively and comparatively studied. The proposed framework for deep-learning-enhanced particle shape acquisition and realistic DEM simulation will provide researchers with more convincing physics-based insights into granular mechanics and has the potential to be extended into 3D to benefit practical problems.