HighlightsThe relationship between the angle of repose and compound fertilizer particle size distribution and sphericity was revealed.The range of the rolling friction coefficient obtained by different modeling was calibrated by the golden section method.Scanning modeling required 80% of the rolling friction coefficient of spherical modeling to reflect actual motion.The relative error of scanning modeling (7.91%) was better than that of spherical modeling (30.84%).Abstract. The discrete element method (DEM) is widely used to simulate the behavior of granular materials. The accurate establishment of the particle model has a great influence on the accuracy of the simulation. Most of the existing discrete element simulations of granular compound fertilizers use spherical modeling without considering the shape or particle size distribution. To compare the simulation accuracy of spherical modeling (spherical particles with equivalent diameters) and scanning modeling (a three-dimensional model of fertilizers was obtained by a scanner and then automatically filled by EDEM2020 software) on compound fertilizer discharge, three kinds of granular compound fertilizers (Changqingshu, Xinshidai, and Munvhong) were selected as the research objects. The characterization parameters of fertilizer shape and fluidity were measured. Principal component analysis and Pearson correlation analysis were used to analyze the influence of the compound fertilizer characterization parameters on the angle of repose. The results showed that the particle shape and particle size distribution of compound fertilizer were the main factors affecting its angle of repose. Through the golden section method and a single-factor experiment, the rolling friction coefficients of the three granular fertilizers under the two granular modeling methods for spherical modeling and scanning modeling were determined. The results showed that the rolling friction coefficient when using scanning modeling to approximate the actual angle of repose was approximately 80% of that when using the spherical modeling method. The influence of different modeling methods on the total rotational kinetic energy and the formation of the angle of repose was discussed. For the same fertilizer, the total rotational kinetic energy was greater in spherical modeling than in scanning modeling, and the formation time of the angle of repose was longer in scanning modeling than in spherical modeling. The results of an external tank wheel fertilization simulation and bench test showed that considering the scanning modeling of fertilizer shape and particle size distribution, the fertilization variation coefficient obtained from the simulation test was closer to that of the bench test. This research helps to better understand the influence of particle shape and particle size distribution on DEM simulations and provides references for discrete element modeling of other granular fertilizers. Keywords: Compound fertilizer, Modeling method, Shape, Simulated flow characteristics, Size distribution.
Effect of soil particle-size distribution (PSD) on soil-subsoiler interactions in the discrete element model
