Discrete element method (DEM) has presented promising advantages in simulation of soil-machine/tool interaction. To develop reliable DEM simulations, selection and parametrization of a suitable contact model is crucial. The plate sinkage test is widely applied to characterize the soil behavior during compaction. This study aimed at setting up a DEM model for plate sinkage test using the hysteretic spring-linear cohesion contact model to (i) find the key parameters of the model which need to be calibrated and (ii) develop a calibration method for DEM parameters of the model for varying soil water content and bulk density of a cohesive soil. It was tested whether confined compression tests in combination with direct shear tests could be used to calibrate DEM parameters of the contact model. Soil cohesion and internal coefficient of friction were directly obtained and used from experimental direct shear test whereas particle yield strength was estimated by matching the stress-sinkage relationship of experimental and DEM-simulated confined compression tests. A clay loam soil at two water contents (11 and 16 %) and two dry bulk densities (1000 and 1150 kg m⁻³) was used to test the calibration method. Simulations using the parameterization obtained from the combination of confined compression and direct shear tests resulted in remarkable underestimation of plate sinkage. A satisfactory agreement with measurements was only found when the simulated cohesion and internal coefficient of friction were several times higher than measured. The factors were indicated to be constant with changes in soil water content and bulk density. A strong correlation was found between the particle yield strength and precompression stress estimated from confined compression tests. Using the factors obtained for increasing the cohesion and internal coefficient of friction and estimation of particle yield strength from precompression stress, a viable method for calibration of plate sinkage test could be proposed.

离散元方法（DEM）在土壤-机器/工具相互作用的模拟中显示了有希望的优势。为了开发可靠的DEM模拟，选择合适的接触模型并对其进行参数化至关重要。板下沉测试被广泛应用于表征压实过程中的土壤行为。这项研究旨在使用滞后弹簧-线性内聚接触模型建立用于板沉测试的DEM模型，以（i）查找模型中需要校准的关键参数，以及（ii）开发用于DEM参数的校准方法改变粘性土的土壤水分和堆积密度的模型。测试了是否可以使用约束压缩试验与直接剪切试验相结合来校准接触模型的DEM参数。直接从实验直接剪切试验中获得并使用了土壤内聚力和内部摩擦系数，而通过匹配实验和DEM模拟的密闭压缩试验的应力-沉降关系来估算颗粒的屈服强度。两种水含量（11％和16％）和两种干容重（1000和1150 kg·m的粘土壤土）⁻³）用于测试校准方法。使用从密闭压缩试验和直接剪切试验的组合中获得的参数化进行的模拟导致钢板沉陷的明显低估。仅当模拟的内聚力和内部摩擦系数比测量值高出几倍时，才能找到与测量值令人满意的一致性。表明这些因素随土壤含水量和堆积密度的变化是恒定的。发现在有限的压缩试验中估计的颗粒屈服强度与预压缩应力之间存在很强的相关性。利用获得的增加内聚力和内摩擦系数的因素以及通过预压缩应力估算颗粒屈服强度的方法，可以提出一种可行的方法来校准板沉试验。