A Prediction Method for the California Bearing Ratio of Soil-Rock Mixture Based on the Discrete Element Method and CT Scanning. (1st October 2020)

• Record Type: Journal Article
• Authors: Ji, Xiaoping; Li, Jia; Cui, Zhifei; Li, Shouwei; Xiong, Yue; Hu, Jianming; Jiang, Yingjun
• Other Names: Zhang Chunshun Academic Editor.
• Is Part Of: Advances in civil engineering. Volume 2020(2020)
• Publication Date: 2020-10-01
• Subjects: Civil engineering -- Periodicals; Civil engineering; Periodicals; 624
• DOI: 10.1155/2020/9794756
• Languages: English
• ISSNs: 1687-8086

Abstract

Abstract : Because of the large amount of gravel with particle sizes over 40 mm in the soil-rock mixture (SRM), it is impossible to determine its California Bearing Ratio (CBR) via the indoor test method, which is a key parameter for designing the backfill in underground mined cavities or the road subgrade constructed with SRM. In this paper, X-ray computed tomography (CT) scanning and 3D reconstruction technology were used to construct the 3D structure of SRM particles with a particle size greater than 5 mm. Based on the vertical vibration test method (VVTM) and PFC 3D, the numerical simulation method (NSM-CBR) of SRM was established. The CBR of the SRM with a maximum particle size over 40 mm (SRM-G) was studied by NSM-CBR, and the effects of factors such as maximum particle size, soil content, and large-size particle content ( d ≥ 40 mm) on the CBR were investigated via NSM-CBR. Based on the laboratory tests and NSM-CBR, the prediction model and the determining method of CBR of SRM-G were established and verified. The results show that the maximum deviation between the CBR obtained from NSM-CBR and laboratory tests was 7.4%. The CBR of SRM-G decreases linearly with the increase in soil content and increases with the increase in maximum particle size and large-size particle content. The practical project shows that the maximum deviation between the predictive and measured values of the CBR of SRM-G was less than 1.5%, indicating that the prediction model and the method established in this paper have high reliability.