scholarly journals A GPU Accelerated Continuous-Based Discrete Element Method for Elastodynamics Analysis

2011 ◽  
Vol 320 ◽  
pp. 329-334 ◽  
Author(s):  
Zhao Song Ma ◽  
Chun Feng ◽  
Tian Ping Liu ◽  
Shi Hai Li

This paper presents a GPU computing algorithm, used to accelerate the Continuous-based Discrete Element Method (CDEM). Using a NVIDIA GTX VGA card, the computing speed achieved an average 650 times speedup ratio vs. Intel Core-Dual 2.66 GHz CPU. To parallelize the CDEM algorithm, the clone node force refreshing process is separated from the elemental calculation, and is replaced by a “Node Group” force assignment process, which ensures the data independence in parallel execution.

2021 ◽  
Author(s):  
Christoph Rettinger ◽  
Sebastian Eibl ◽  
Ulrich Rüde ◽  
Bernhard Vowinckel

<p>With the increasing computational power of today's supercomputers, geometrically fully resolved simulations of particle-laden flows are becoming a viable alternative to laboratory experiments. Such simulations enable detailed investigations of transport phenomena in various multiphysics scenarios, such as the coupled interaction of sediment beds with a shearing fluid flow. There, the majority of available simulations as well as experimental studies focuses on setups of monodisperse particles. In reality, however, polydisperse configurations are much more common and feature unique effects like vertical size segregation.</p><p>In this talk, we will present numerical studies of mobile polydisperse sediment beds in a laminar shear flow, with a ratio of maximum to minimum diameter up to 10. The lattice Boltzmann method is applied to represent the fluid dynamics through and above the sediment bed efficiently. We model particle interactions by a discrete element method and explicitly account for lubrication forces. The fluid-particle coupling mechanism is based on the geometrically fully resolved momentum transfer between the fluid and the particulate phase. We will highlight algorithmic aspects and communication schemes essential for massively parallel execution.</p><p>Utilizing these capabilities allows us to achieve large-scale simulations with more than 26.000 densely-packed polydisperse particles interacting with the fluid. With this, we are able to reproduce effects like size segregation and to study the rheological behavior of such systems in great detail. We will evaluate and discuss the influence of polydispersity on these processes. These insights will be used to improve and extend existing macroscopic models.</p>


Particuology ◽  
2011 ◽  
Vol 9 (4) ◽  
pp. 446-450 ◽  
Author(s):  
Ji Xu ◽  
Huabiao Qi ◽  
Xiaojian Fang ◽  
Liqiang Lu ◽  
Wei Ge ◽  
...  

TAPPI Journal ◽  
2019 ◽  
Vol 18 (2) ◽  
pp. 101-108
Author(s):  
Daniel Varney ◽  
Douglas Bousfield

Cracking at the fold is a serious issue for many grades of coated paper and coated board. Some recent work has suggested methods to minimize this problem by using two or more coating layers of different properties. A discrete element method (DEM) has been used to model deformation events for single layer coating systems such as in-plain and out-of-plain tension, three-point bending, and a novel moving force picking simulation, but nothing has been reported related to multiple coating layers. In this paper, a DEM model has been expanded to predict the three-point bending response of a two-layer system. The main factors evaluated include the use of different binder systems in each layer and the ratio of the bottom and top layer weights. As in the past, the properties of the binder and the binder concentration are input parameters. The model can predict crack formation that is a function of these two sets of factors. In addition, the model can predict the flexural modulus, the maximum flexural stress, and the strain-at-failure. The predictions are qualitatively compared with experimental results reported in the literature.


2021 ◽  
Vol 910 ◽  
Author(s):  
Yiyang Jiang ◽  
Yu Guo ◽  
Zhaosheng Yu ◽  
Xia Hua ◽  
Jianzhong Lin ◽  
...  

Abstract


2021 ◽  
pp. 014459872110135
Author(s):  
Zhen Tian ◽  
Shuangxi Jing ◽  
Lijuan Zhao ◽  
Wei Liu ◽  
Shan Gao

The drum is the working mechanism of the coal shearer, and the coal loading performance of the drum is very important for the efficient and safe production of coal mine. In order to study the coal loading performance of the shearer drum, a discrete element model of coupling the drum and coal wall was established by combining the results of the coal property determination and the discrete element method. The movement of coal particles and the mass distribution in different areas were obtained, and the coal particle velocity and coal loading rate were analyzed under the conditions of different helix angles, rotation speeds, traction speeds and cutting depths. The results show that with the increase of helix angle, the coal loading first increases and then decreases; with the increase of cutting depth and traction speed, the coal loading rate decreases; the increase of rotation speed can improve the coal loading performance of drum to a certain extent. The research results show that the discrete element numerical simulation can accurately reflect the coal loading process of the shearer drum, which provides a more convenient, fast and low-cost method for the structural design of shearer drum and the improvement of coal loading performance.


Sign in / Sign up

Export Citation Format

Share Document