Background

The discrete element method (DEM) is a powerful tool for to develop fundamental understanding of the behavior of geomaterials at the scale of the individual particles. This method can also be applied to simulate boundary value problems in geotechnics. Although its potential is broadly accepted, several issues with this method remain. One issue is the availability of standard example problems to verify DEM simulation codes and confirm that these codes are being correctly used. Many researchers and engineers find this difficult owing to the complexity of granular material behavior; analytical verifications exist only for limited, highly ideal cases. Furthermore, many different DEM implementations and codes exists, and it is important to understand how the accuracy is influenced by the implementation details. One solution is to quantify the error between the simulation results and the response observed in a real, physical simulation. There is a clear need for a benchmark simulation with accompanying physical test data. Consequently, the TC 105 Japanese domestic committee has prepared a round robin test for DEM simulations. We welcome anyone who would like to participate in this round robin activity.

Outline of round robin test:

Trial simulation examples:

Feasibility simulation by Dr. Naotaka Kikkawa (JNIOSH, Japan)

Feasibility simulation by Dr. Hidetaka Saomoto (AIST, Japan) with Yade
Yade: V. Šmilauer et al. (2015), Yade Documentation 2nd ed. The Yade Project. DOI: 10.5281/zenodo.34073

Schedule:

Experimental Material

The experimental material comprised an assembly of clumps. Each clump is made up of four spheres that have the same diameter of 6.2020mm. The four sphere centers are placed at the vertices of a tetrahedron and so the particles can be called tetrahedral particles. The clumps were generated by a 3D printer. The circumsphere that acts as an envelope to the clump is 10 mm (1 cm) in diameter. The clumps were produced by Dr. Daiki Takano (Port and Airport Res. Inst.).

Experiment Device

Plane strain type

The angle of repose (AOR) can be measured under plane strain conditions using this device. This device is made up of transparent acrylic plates and comprises an upper and a lower box, separated by a horizontal acrylic plate that can translate horizontally. The experiment material (clumps) is initially deposited in the upper acrylic box. During the experiment the granular material firstly falls downward under the action of gravity by translating the plate installed between the upper box and the lower box outwards. Once all the particles have come to rest, the upper box is removed, and the front vertical face of the lower box is pulled upwards by an electric motor at a constant velocity of 43mm/sec. 2150 particles were used in the experiment. The size of device can be obtained from the following link.

• device_size.pdf

The device was designed by Mr. Jun Yamaguchi and Mr. Hiroaki Kabuki (Tohoku Univ.)

Axial symmetric type

The angle of repose (AOR) can be measured under axi-symmetric conditions using this device. 2468 tetrahedral particles were pluviated from the hopper to (indicated as point (n) in the figure) into the cylindrical space (a). The cylinder comprises a fixed bottom (e) and a movable cylindrical side wall (d). During the experiments the movable side wall (d) is moved vertically downwards with a constant speed of 40mm/min. Then the particles form an approximately conical heap. The tests were conducted several times to measure the AOR because of variability of the heap.

Description

• Diameter of hopper (n)= 100mm
• Pluviation height = 340mm (from the hopper's outlet to the bottom of the cylinderical space (a))
• Diameter of cylindrical space (a)=160mm
• Initial height of cylindrical space (a)=90mm
• The coordinate axes should be defined as shown in the figure below ((e) bottom is set to Z=0m and coincident of X-Y plane.)

Measurement procedure

Material Characterisation Tests

We conducted several tests for to characterize the 3D printed granular material namely drop tests, frictional tests to determine the particle-particle & particle-wall coefficients of friction and single particle compression tests. Detailed information can be downloaded below. The numeric data of physical properties can be also downloaded.

• Density = 1111 kg/m³
• Drop & Frictional tests:
  • Numeric data
• Single particle compression test:
  • List of shear modulus
  • Data of typical force-displacement relation

Round Robin Test

We would like to invite interested engineers and scientists to perform discrete element method (DEM) simulations to measure the angle of repose for tetrahedral particles by simulating the experiments performed using Device I and/or Device II. While we would like to encourage participants that to simulate both experimental configurations, we welcome submissions that consider only one set-up.

We will acknowledge all participants on our webpage and in any resulting publications. The presentation of the results will ensure anonymity, i.e. no data presented or discussed will be attributable to any particular participant. Please let us know if you would prefer for your name to be excluded from the list of paricipants.

To participate in this scheme:

• Please declare us your interest in this round-robin test.
• Please select your DEM input parameters and develop your approach to run the simulations based on the data provided.
• Please carry out the simulation(s) using your own DEM code or the DEM code that you routinely use in your work (i.e. commercial and open-source codes can be used). Participants should decide on the appropriate number of simulations per each device. We would like to ask the participants to calculate more than the required number, because the phenomena would be expected to have a variability because of the sample size considered. The required number of simulations became at least 20 times for each device (Device 1 and Device 2) cases though a comparison between simulation and experimental results. By this, the quality of the submitted results can be ensured, and a simple comparison of each simulations result is possible.
• Please carry out the simulation(s) using your own DEM code or the DEM code that you routinely use in your work (i.e. commercial and open-source codes can be used).
• Please email the simulation condition and results. The data formats were prepared as the Zip file via the link below (named AOR_report_YourName.zip). There are two types of files for each device. The simulation condition is filled in the EXCEL file (named Device01_report.xlsx or Device02_report.xlsx). A data files providing the coordinates of all the spheres making up the tetrahedral particles at the reposed condition should also be submitted. For the tetrahedral clump particles considered, there are 4 spheres in each particle. So, (x, y, z (unit; m)) coordination of all spheres at the configuration is given the angle of repose. Please determine the angle of repose yourself, using an approach you judge to be appropriate. Each valid submission should comprise one EXCEL file and a number of Data files. If you conduct 20 simulations, please submit one EXCEL file and 20 Data files.
  • AOR_report_YourName.zip
    (EXCEL file and Data files for each device are included.)
• The deadline for submission is the 30th of November 2020.
  • Email address for submission: s_mori irides.tohoku.ac.jp
    Dr. Moriguchi (Tohoku Univ.), Secretary of domestic comittiee for TC105 in JGS

We aim to disclose the experimental results and the collected combined simulation results via this WebPage around March 2021. We will notify all participants when this collation is completed and ready for viewing.

Participants

We, TC105 national committee in JGS, express our appreciation to all participants as shown in the list.

List of Participants:

Result

We published the experimental results and the collected combined simulation results. Only the combined, overall outcome was presented and the anonymity of the participants was preserved.

Publication for Expreimental result: technical report

Data for Expreimental result: Click_here

Publication for Simulation result:technical report

Data for Simulation result: Click_here

Self-verification Material

You can find the information to verify your code by yourself. But it is not mandatory to do this verification before the Round Robin Test. We are providing this material just for self validation. The material was provided by Prof. O’Sullivan (Imperial College London).

• verification.pdf