Modelling approach for soil displacement in tillage using discrete element method

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review


The majority of studies dedicated to the dynamic behaviour of agricultural soils are basically focused on prediction
of physical conditions within the soil-tool interface whereas the dynamic soil response, such as soil
displacement, has not attracted sufficient attention. Considering its importance for engineering applications, the
objective of the presented study was to develop a modelling approach to study the soil displacement due to soil
interaction with operating tool during sweep cultivation. The developed approach employs the similarity criteria
to account for the effect of scaling-up soil particle sizes on model quality and combines (i) an empirical study of
soil displacement during a sweep operation and (ii) related numerical simulations using a discrete element
model. Soil displacement for three different soil types in soil bins was measured using a tracer methodology. The
discrete element model was formulated for cohesive soils with parallel bond contacts between aggregates. The
model allows simulation of mechanical tests (used for calibration purposes) and simulation of soil interaction
with operating tool during sweep cultivation. This work shows that due to the necessary change in particle sizes
of model soil, the conditions of physical equivalence between the model and natural soil can be difficult to
satisfy. Such condition can cause significant dissimilarity between measured and simulated soil displacement.
However, this can be overcome if the model scaling does not exceed a certain limit which is determined by soil
structure and the potential contact density on a tool’s contact surface during the soil-tool interaction. By using
these findings, our simulations demonstrate that prediction of soil displacement is possible within the suggested
approach with sufficient accuracy
TidsskriftSoil & Tillage Research
Sider (fra-til)60-71
Antal sider12
StatusUdgivet - 2018

Se relationer på Aarhus Universitet Citationsformater

ID: 128118804