Modeling and experiments of chewing mechanical properties of pellet feed using discrete element method

In order to explore the mechanical properties and breaking behavior of pellet feed during chewing, the experiments of texture mechanics, followed by the modeling and simulation of pellet feed based on the discrete element method were carried out in this research. Five wet basis moisture contents (8%, 10%, 12%, 14% and 16%, respectively), two kinds of loading directions (L and D direction, respectively) of pig pellet feed were selected as variables. First, mechanical parameters including hardness, elasticity, tackiness and chewiness were measured by a texture analyzer. The results of compression tests showed that the hardness of pig pellet feed was 5.44-32.43 N, the elastic index was 0.04-0.94 mm, the tackiness was 0.07- 6.63 N, the chewiness was 5.52-27.39 mJ. Moreover, the hardness, tackiness and chewiness of pig pellet feed decreased significantly with the increase of moisture content but the elasticity showed an adverse varying trend. The hardness, elasticity, tackiness and chewiness along D direction outweighed that of L direction in numerical data at the same moisture content. Then, the chewing and breaking processes of pellet feed were simulated based on the discrete element method (DEM) combined with bonding particle model, in which the whole pellet feed were considered as agglomerations of micro-particles and broke when the stress between micro-particles had exceeded the maximum limit. Multi-parameter optimization experiments were carried out using quadratic orthogonal rotation design, in which stiffness coefficient (X1), critical stress (X2) and bonding radius (X3) were the influencing factors, hardness (Y1), elasticity (Y2), tackiness (Y3) and chewiness (Y4) were evaluating indicators. Based on the regression analysis of the Design-Expert 8.0.6 software and response surface analysis method, the relationship between the three influencing factors and evaluating indicators was established. The similarity between experimental and simulated results in feed morphology and mechanical index proved that the modeling method for pellet feed based on DEM was effective and accurate. This work can provide a reference for the feed forming process and the optimization design of the related feed machinery. Meanwhile, the DEM model provided a new method for evaluating the texture and palatability of pellet feed.