LI Mingyuea， QIAN Fupingb， MA Xiaob， LI Yifeia， YIN Xinrua， ZHENG Zhiminb
a. School of Civil Engineering and Architecture， b. School of Energy and Environment， Anhui University of Technology， Maanshan 243032， China
Abstract Objective In steel production technology， accurately simulating the behaviour of steel slag particles under specific operating con⁃ ditions remains a significant challenge. To address this issue， this study establishes reliable simulation parameters for steel slag particles， providing a robust parameter basis for constructing discrete element method （DEM） simulation models using EDEM software. Through systematic calibration of key contact parameters， such as the collision restitution coefficient， rolling friction coefficient， and static friction coefficient， a complete set of parameter values is derived to approximate the real-world behaviour of steel slag particles. In addition， the results offer a theoretical reference for DEM simulations of steel slag particle transport in pipelines， where flow characteristics are affected by factors including pipeline geometry， flow velocity， and particle-wall interactions. By using the optimal parameter combination，this work enhances the modeling accuracy for these complex interactions and improves the prediction capability for steel slag particle behaviour in pipelines.
Methods A combination of experimental and simulation methods was used to determine the basic parameters of steel slag particles， such as particle size distribution， density， particle shape， and stacking angle. To accurately calibrate the particleparticle contact parameters in discrete element simulations， an orthogonal experimental design method was used. The parameter levels for collision restitution coefficient， rolling friction coefficient， and static friction coefficient were selected based on the existing literature and pretest results to ensure scientific rigor. To minimize experimental errors and enhance result reliability， an empty column was set up for analyzing the random errors during the experimental process. Through rigorous planning of test combinations， multiple sets of comparative tests were carried out to systematically investigate the influence of each parameter on steel slag particle behavior. For precise measurement of the stacking angle， the Open CV library in Python was used for image processing steps， including grayscale conversion， binarization， and edge contour extraction. This methodology ultimately yielded an optimal parameter combination among the coefficients.
Results and Discussion Through rigorous orthogonal experimental design and several rounds of simulation validation， the optimal particle-particle contact parameters were determined as follows： rolling friction coefficient of 0.1， static friction coefficient of 0. 35， and collision restitution coefficient of 0.1. Experimental results showed that the measured natural stacking angle of steel slag particles was 26. 39°， and the discrete element simulation using the optimal parameter combination yielded a stacking angle of 25. 92°. The relative error between the two was only 1. 78%， validating the parameter combination and the reliability of the simulation model. Through an in-depth analysis of the experimental data， it was found that among the parameters affecting the stacking angle of steel slag particles， the collision restitution coefficient had the most prominent influence. It significantly affected postcollision particle trajectories， thereby substantially impacting the stacking behavior. The rolling friction coefficient was the second most important factor， and the static friction coefficient had a relatively minimal influence but played a role in particle stabilization during the particle accumulation stage.
Conclusion To obtain the contact parameters for steel slag particle simulations， a combination of experimental and simulation methods was used， and the Generic EDEM Material Model （GEMM） database was employed to establish a range of contact parameters. During the stacking angle measurement stage， the Open CV library in Python was utilized for image reading and processing， enabling precise extraction of the outer contour line of the pile. A systematic analysis was conducted to determine the effect of different contact parameters on stacking behavior. Polar analysis and analysis of variance（ ANOVA） revealed that with a rolling friction coefficient of 0. 1， static friction coefficient of 0. 35， and coefficient of restitution of 0. 1， the simulated stacking angle was 25. 92°， demonstrating exceptional agreement with the experimentally measured stacking angle， with a mere relative error of 1. 78%. These results confirm the validity of this parameter combination as the optimal configuration for steel slag particle simulation in DEM software.
Keywords： steel slag particle； stacking angle； image processing； contact parameter calibration； orthogonal test； numerical simulation
Get Citation: LI Mingyue， QIAN Fuping， MA Xiao， et al. Calibration of particle contact parameters based on steel slag particle stacking angle using EDEM［J］. China Powder Science and Technology， 2026， 32（3）： 1−10.
Received: 2025-03-26 .Revised: 2025-06-10,Online: 2025-07-21.
Funding Project:国家自然科学基金项目，编号：52176148；安徽省重点研究与开发计划项目，编号202104i07020016。
First Author: 李明月（1999—），女，硕士生，研究方向为工业通风与空气净化。E-mail：18956973307@163. com。
Corresponding Author:钱付平（1974—），男，教授，博士生导师，入选“安徽省优秀青年人才支持计划”，安徽省战略性新兴产业技术领军人才。 研究方向为通风除尘系统及设备优化研究。E-mail：fpingqian@ahut. edu. cn
DOI：10.13732/j.issn.1008-5548.2026.03.017
CLC No: TP391.9； TB301； TB4 Type Code: A
Serial No:1008-5548（2026）03-0001-10
