Jiang Shengqiang， Hu Liu， Chen Yihao， Li Xu
Engineering Research Center of Complex Tracks Processing Technology and Equipment of Ministry of Education， School of Mechanical Engineering and Mechanics， Xiangtan University， Xiangtan 411105， China
Abstract
Objective This study aims to calibrate the discrete element method （DEM） contact parameters of lithium nickel cobalt manganese oxide （NCM） powder. Considering the small particle size and strong adhesion characteristics of the powder， the calibration of discrete element parameters aims to improve the prediction accuracy of DEM simulations during powder conveying， mixing， and packing processes.
Methods Physical tests were first conducted to obtain the intrinsic properties of the NCM powder. The particle size distribution was measured using a laser diffraction analyzer， yielding d10=1.2 μm， d50=2.5 μm， and d90=4.0 μm. Optical microscopy was used to characterize the particle morphology， which revealed irregular particle shapes and significant particle agglomeration. The bulk density and tapped density were measured according to GB/T 1479.1-2011， with values of 0.90 g/cm³ and 2.36 g/cm³， respectively. The static angle of repose was measured according to GB/T 31057.3-2018. The measurement was repeated 20 times， and the mean value was calculated at a confidence level of 0.90， yielding a reference angle of repose of 58.3°. During DEM model construction， a particle scaling strategy based on similarity principles was applied to reduce computational cost， and the equivalent particle diameter was set to 0.625 mm. Considering the adhesion effect between powder particles， the Johnson-Kendall-Roberts （JKR） contact model was adopted to describe the interparticle adhesion force caused by surface energy. A geometric model consisting of a funnel and a receiving plate was established according to the national standard. The angle of repose of the simulated particle pile was obtained using a MATLAB-based image processing method. The pile images were converted into grayscale and binarized images， the boundary contour was extracted， and the left and right boundaries were fitted separately using the least squares method. The average value of the two fitted angles was used as the response variable. During the parameter calibration stage， a Plackett-Burman （PB） design was used to screen nine contact parameters， including static friction， dynamic friction， restitution coefficient， surface energy， and rolling damping， to identify significant factors. Subsequently， a steepest ascent test with six gradient levels was performed to approach the optimal parameter range of the significant factors， while the remaining insignificant parameters were fixed at their intermediate levels. Finally， a central composite design （CCD） with 13 runs was carried out to establish a quadratic regression model between the angle of repose and the significant factors.
Results and Discussion The PB test results showed that interparticle surface energy （P<0.01） and rolling damping coefficient （P<0.01） were significant factors influencing the angle of repose， while the other parameters showed no significant effects （P> 0.05）. The steepest ascent test showed that the relative error reached its minimum value in the fourth group， indicating that this parameter range was close to the optimal region. The regression model established from the CCD showed a high degree of fit， with R²=0.969 0， adjusted R²=0.946 8， and predicted R²=0.916 1. The model was highly significant （P<0.000 1） and passed the lack-of-fit test （P>0.05）. Based on this model， the optimal parameter combination was determined. A verification simulation was conducted， yielding an angle of repose of 58.8°， with a relative error of only 0.94% compared with the experimental value of 58.3°.
Conclusion A combined physical–simulation approach， incorporating MATLAB image processing and statistical test design， is adopted for calibration. A JKR contact model incorporating particle scaling is established to simulate cohesive powder， and image processing of the pile boundary provides a stable response variable for calibration. Surface energy and rolling damping are identified as dominant factors， and a regression model is subsequently developed. The optimal parameters are determined and validated， demonstrating excellent agreement between the simulation and physical tests. The calibrated parameters are reliable and can provide a solid basis for DEM simulations of NCM powder conveying， mixing， and packing processes.
Keywords： lithium nickel cobalt manganese oxide powder； discrete element method； angle of repose； contact parameter calibration； response surface methodology
Get Citation: Jiang Shengqiang， Hu Liu， Chen Yihao， et al. Discrete element parameter calibration of lithium nickel cobalt manganese oxide powder based on static angle of repose［J］. China Powder Science and Technology， 2027， 33（1）： 1-12.
Rceived：2026-05-10， Revised： 2026-08-23，Online： 2026-09-14。
Funding： The research was supported by the General Program of the National Natural Science Foundation of China (Grant No. 52575540) and the Young Scientists Fund of the National Natural Science Foundation of China (Grant No. 12302513).
CLC No.：TB44
Type Code：A
Serial No.：1008-5548（2027）01-0001-12
