Numerical simulation of soil–cone penetrometer interaction using discrete element method

Abstract

One of the most common methods to measure soil strength in-situ is cone penetrometers. In this paper the development of a three dimensional (3D) discrete element model (DEM) for the simulation of the soil-cone penetrometer interaction in a slightly cohesive loamy sand soil is presented. The aim was to investigate the effects of the soil model's geometrical (e.g., soil model cross section shape and size and model's height) changes on variations in the soil penetration resistance. The model area ratio and height ratio values were adopted to analyse the effects of the cross section size and the model's height, respectively. The results of penetration resistance of the DEM simulations were compared with the in-situ measurement with a cone penetrometer of the same geometry. This comparison allowed the derivation of the contact properties between the elements. To simulate the soil material the so-called Parallel Bond and Linear Models were used in the 3D version of the Particle Flow Code (PFC) software. Finally the mechanical properties of the soil, namely the cohesion and internal friction angle were estimated by DEM simulation of direct shear box. Results showed that the penetration process can be simulated very well using the DEM. The model's calculated penetration resistance and the corresponding in-situ measurement were in good agreement, with mean error of 14.74 %. The best performing models were a rectangular model with an area ratio of 72 and a height ratio of 1.33 and a circular model with an area ratio of 32 and a height ratio of 2. The simulation output of soil material properties with direct shear box resulted in representative values of real loamy sand soils, with cohesion values range of 6.61-8.66 kPa and internal friction angle values range of 41.34-41.60°. It can be concluded that the DEM can

Filter for penetration resistance correction in cultivated and compacted Histosols

“…According to ASABE (2019), in very hard soil where it is not possible to achieve a rate of 3 cm s −1 , slower rates do not result in significant errors. In a similar investigation, Kotrocz, Mouazen, and Kerényi (2016) used a penetration rate of 1 cm s −1 and Kuhwald et al (2016) a penetration rate of 2 cm s −1 . Finally, the cone and shaft were thoroughly washed and wiped between each penetration test in accordance with recommendations of ASABE (2019).…”

Analysation of the Effect of the Contact Properties on Soil’s Penetration Resistance in Discrete Element Simulations

“…2/a). The size of the model was chosen according to our earlier research [16] where it is proved, that the boundary of the model is far enough from the cone penetrometer, thus it has negligible effect on simulation results. The particles were created using the Simple Sequential Inhibition (SSI) technique [17], which places elements with random diameter to random locations.…”

An Evaluation of Vertical Dynamic Stress Attenuation for Compacted Coarse-Grained Soils

“…In practice, however, it is difficult to study its mesoscopic mechanical properties directly due to various restrictions of different conditions. Hence, using DEM, typically the program known as the particle flow code (PFC), is convenient to simulate the mechanical behavior of CGS fillers in terms of different particle interactions [20].…”