English  /  中文
ISSN 2097-0498e-ISSN 2773-0077CN 61-1520/U
Guan Chenhui, Li Zhe, Liu Xiaoyan, Liu Lulu, Li Wen, Cai Guojun. 2026: Settlement characteristics and prediction of typical loess-filled subgrade considering constraint conditions. Journal of Road Engineering, 6(2): 266-286. DOI: 10.1016/j.jreng.2025.06.007
Citation: Guan Chenhui, Li Zhe, Liu Xiaoyan, Liu Lulu, Li Wen, Cai Guojun. 2026: Settlement characteristics and prediction of typical loess-filled subgrade considering constraint conditions. Journal of Road Engineering, 6(2): 266-286. DOI: 10.1016/j.jreng.2025.06.007

Settlement characteristics and prediction of typical loess-filled subgrade considering constraint conditions

  • Post-construction settlement of loess subgrade is a critical factor influencing road stability. This study, based on the Wubu–Suide section of China National Highway G307, investigates the settlement behavior of loess-filled subgrades under various constraint conditions. Field tests conducted at typical sections provide data on the deformation of different subgrades. Focusing on lateral deformation of soil units, the study explores the mechanism behind differential settlement in filled subgrades. A predictive model for loess-filled subgrade settlement, incorporating constraint conditions, is also proposed. The results indicate that constraint conditions exert significant control over settlement distribution. Under one-, two-, and three-sided constraints, the subgrades' normalized differential settlements are 1.03, 0.56, and 0.18, respectively, confirming that multi-sided constraints progressively reduce differential settlement while enhancing uniformity in filled subgrades, thus facilitate settlement control. For subgrades with different constraints, the further away from the constraints, the greater the settlement. When constraints are symmetrically distributed, maximum settlement occurs at the constraint center. Based on the exponential model with a fitting correlation coefficient value of the settlement-time curve reaching 0.99, a settlement prediction model consisting of three stages: initial, deceleration, and stability was established, and the parameter determination was described in detail. As constraint dimensionality increases, the core parameter A1/A2 of the model is 0.755, 1.06, and 1.54, respectively, which is positively correlated with the number of constraints. The prediction model, based on the testing data, aligns well with actual engineering observations and can be applied to practical projects, particularly for predicting subgrade settlement following continuous filling.
  • loading

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return