3D finite element simulation of a single-tip horizontal penetrometer–soil interaction. Part I: Development of the model and evaluation of the model parameters

A fundamental step to employing a horizontal penetrometer for on-the-go measurement of soil compaction is to characterize the horizontal penetrometer resistance (PR) as affected by soil mechanical/physical properties and design/operational parameters of the penetrometer. Such complicated soil–tool interactions are preferably investigated using numerical analyses, e.g. finite element (FE) method. A symmetric 3D FE model for a single-tip horizontal penetrometer–soil interaction was developed and evaluated in ABAQUS/Explicit. In the first paper of this study, we focus on the development of the model and the evaluation of some affecting parameters, namely (i) boundary effects, (ii) element size (mesh density) around the conical tip of the penetrometer, (iii) sensitivity of the predicted PR with respect to (a) soil mechanical properties (b) tip extension (i.e. the distance between cone base and shin of the carrying tine), (c) working depth, and (iv) soil layering and the effect of working depth in relation to the tine critical depth. The elastic–plastic parameters of the soil constitutive model (Drucker–Prager) were determined from samples taken within the working depth of the penetrometer in a soil bin. The results showed that for the tested penetrometer dimensions and the given soil specifications (i) the simulated soil box should be at least 0.8 m long and 0.4 m wide to avoid interference from the boundaries on predicted PR; (ii) the element size around the cone should be fine enough (9 mm) for a stable steady-state simulated PR; and (iii) the tip extension must be >4 cm to avoid the effect of tine on PR. PR was found to be most sensitive to soil internal angle of friction and compressive yield stress, whilst Poissonʹs ratio had no significant effect, and Youngʹs modulus of elasticity and soil-metal coefficient of friction showed only minor impacts on PR. Due to soil surcharge, PR increased by 30% when increasing the working depth of the penetrometer from 5 to 40 cm in a homogenous soil profile. An effective radius of 6 cm was found for the soil failure around the cone. The simulations showed that when the cone was located above the tine critical depth (i.e. inside the crescent failure zone), the PR was lower and the PR signal was unstable compared with when the cone worked out of crescent failure zone at the same depth. The model developed here can be used to characterize PR with respect to the state of soil strength/compaction across soil types.