|
Incorporating Shape in DEM Simulation
Garry Morrison, Mathew Price, Anthon Voigt and Thomas Landgrebe
Presented at the 1st EDEM User Group Meeting 2008 Abstract: The importance of properly representing particle shape in DEM simulation is increasingly being recognized. At DebTech we have developed novel technology for apturing 3D images of large numbers of particles quickly. Sphere-clump representations of real particles can be read into commercial DEM codes. We will discuss aspects of the 3D particle capture and present a number of examples where we have conducted DEM simulations with realistic particle shapes. These will include comparison of single sphere and sphere-clump representations when filling a container and a single particle ejector system. We will also present updated results from work to extract full 3D motion (linear and rotational) of particles from high-speed video sequences. Finally we will illustrate how curvature information derived from the 3D particle shapes can be used to convert single particle compression test results from failure load to failure stress and failure energy. |
Estimating 3D particle motion from high-speed video for simulation validation
Mathew Price and Garry Morrison
Submitted to Engineering Computations: Special issue on DEM 2007 Abstract: We present an image based method for estimating the 3D motion of rigid particles from high-speed video footage (HSV). The computed motion can be used as either a means to generate quantitative feedback for a process or to validate the accuracy of DEM simulation models. (extended version of papers below.)
|
Sphere clump generation and trajectory comparison for real particles
Mathew Price, Vasile Murariu and Garry Morrison
In proceedings of DEM 2007, Brisbane, 2007 | Abstract: An automatic method for generating sphere clump models of real particles is presented. This is achieved by a process that seeks to optimise the surface correspondence between the sphere clump and a triangular mesh model of the particle. We explore the effect of the number of spheres in the sphere clump model and the extent to which realistic trajectories can be reproduced with DEM simulation. Particle trajectories of actual interactions are extracted from high-speed video sequences using a novel pose estimation technique. Comparisons of sphere clump trajectories with particle trajectories demonstrates that improved particle shape representation with multiple spheres does not necessarily imply better simulation accuracy, unless the interacting particles exhibit highly irregular mass distributions. |
Validating rigid body simulation of real particle shapes using pose stimation from high-speed video
Mathew Price and Garry Morrison
In proceedings of DEM 2007, Brisbane, 2007. | Abstract: We present an image based method for computing 3D trajectories of rigid particles from experiments, where sequences are captured using a single high-speed video camera (HSV). The computed trajectories, which are representative of real world particle interaction, can then be directly compared with trajectories obtained from DEM simulations, hence aiding validation. Experiments consist of a diamond impacting a plate and are captured at 4000 frames per second. Simple image analysis is used to track the particle in each frame and to extract its 2D silhouette boundary. Using an approximate 3D model of the particle, generated from a multi-camera setup, a pose estimation scheme based on silhouette consistency is used in conjunction with a rigid body simulator to produce a registered 3D trajectory. We demonstrate how this information can be used to validate typical simulation characteristics, such as contact models, when applied to particles of arbitrary shape. Under reasonable conditions, the method can reliably estimate linear and angular particle velocities to within 1%. |
|
|
Last Updated on Wednesday, 18 March 2009 13:29 |
