KNOW-NOW

Main Page

KNOW-NOW

Version: v.0.0.2

Author(s): Celso R. C. Rego <celsorego@kit.edu>, Björn Mieller <bjoern.mieller@bam.de>

LIGGGHTS DEM
...

These SimStack Wanos can be used to perform a pressing simulation with the DEM solver LIGGGHTS.

Source:
https://github.com/KIT-Workflows/KNOW-NOW

README.md

LIGGGHTS WaNo icon DataDive WaNo icon

When publishing results obtained with LIGGGHTS WaNo, please consider citing it. DOI

LIGGGHTS and DataDive


These SimStack Wanos can be used to perform a pressing simulation with the DEM solver LIGGGHTS.

The LIGGGHTS WaNo creates a simulation input file using a preset template and the specified parameters. The DataDive Wano reads the results of LIGGGHTS and evaluates a stress-strain-curve.


LIGGGHTS WaNo

This WaNo creates an input file named workflow.inp for a LIGGGTHS pressing simulation. In the simulation, a defined number of particles is created within a rectangular pressing die. After a short time to allow the particles to settle, the upper pressing plate moves downwards and presses on the particles with 100 MPa. Two types of particles are specified, type 1 (ceramic) and type 2 (polymer). Polymer particles have a radius of 0.2 units.

Screenshot of the LIGGGHTS WaNo

Parameters

The customizable parameters are from top to bottom:

  • Title: An arbitrary title of this set of parameters.

  • velocity of pressing plate: The velocity of the upper pressing plate moving downwards.

  • total number of hard particles: The number of ceramic particles to be inserted. Should be max. 20000 to fit in the pressing die.

  • Class: The particle size distribution of the ceramic particles can be defined in this table. To approximate a real size distribution, five sizes and the numerical fraction of each size fraction can be specified.

  • timesteps: The number of timesteps for the pressing segment of the simulation. Depending on the filling height of the pressing die (cf. total number of hard particles), approx. 10000 timeseps are necessary to establish pressing contact.

  • dumprate trajectory: LIGGGHTS calculates the trajectory (position) of every particle in every timestep. The trajectory file can be used to visualize the movement of the particles, e.g. using OVITO. This parameter defines after how many time steps in the pressing segment of the simulation the position of all particles is recorded in the dump.compression file. Too small dumprates may create an unnecessarily large file.

  • dumprate vtk and stl: vtk and stl files can be used for the visualization of the pressing die and plates. This rate parameter determines after how many time steps in the pressing segment a vtk and an stl file with the corresponding positions are generated. Once the calculation is complete, all generated stl and vtk files are packed in a tar archive.

Output

  • forces_inputs.yml

DataDive WaNo

This WaNo reads a forces_inputs.yml file created by the LIGGGHTS WaNo and evaluates a stress-strain curve from the data. For the calculation of strain, the initial thickness of the powder filling is determined based on a specified contact pressure.

Screenshot of the DataDive WaNo

Parameters

  • forces: Name and location of the forces_inputs.yml

  • contact pressure: The stress in MPa at which the initial thickness of the powder filling is defined. Any data below the stress level is discarded for the stress-strain curve.

Input

  • forces_inputs.yml

Output

  • datadive_result.yml

  • stress-strain-curve.png

Auxiliary WaNos Mult-It and DB-Generator

  • see SimStack documentation

  • iterators to use in Mult-It

  • single variable: list(range(Mult-It.VarI-begin, Mult-It.VarI-end, Mult-It.Step))

  • two variables: itertools.product( list(np.linspace( Mult-It.VarF-begin , Mult-It.VarF-end , Mult-It.N-points )), list(range( Mult-It.VarI-begin, Mult-It.VarI-end, Mult-It.Step)) )

  • set name of the yml-database and GitHub Credentials for GitHub push in DB-Generator

  • Colab notebook for data visualization Open In Colab