Answers are grounded in this paper and cite their sources.
Yellow sand substrate in solar greenhouses in southern Xinjiang exhibits stratified moisture distribution, which affects the accuracy of soil–tool interaction simulations. This study proposed a discrete element method (DEM) parameter calibration approach that accounts for moisture variation. Nonbonded contact parameters at moisture contents of 1–18% were calibrated with the Hertz–Mindlin (no slip) model; the mean relative error of the simulated angle of repose was 2.49%. At moisture contents of 9%, 10.5%, and 12%, uniaxial compression and cone penetration slopes were used as dual responses. Twenty-four Latin hypercube sampling (LHS) combinations were generated, and Bonding V2 model parameters were inversely identified using Gaussian process regression (GPR) and differential evolution (DE). The cross-validated R2 values were 0.924 and 0.826, respectively, and critical tangential stress (Z4) contributed most to both responses. A stratified DEM model was then established using field moisture contents of 8.1%, 9.5%, and 10.7%. The mean field soil fragmentation rate was 91.3%. Quantitative layer-wise validation used cone tip resistance, with a mean relative error of 8.5%. The proposed method can characterize contact and bonding properties under varying moisture conditions and support the rapid construction of DEM models for greenhouse tillage.
No takes yet. Share an insight, caveat, or question.
Lu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: