The purpose of this study was to analyze the differences in the physical properties of different feed grade phosphates using principal component analysis (PCA) to streamline the physical properties. The correlation between the different physical properties of different feed grade phosphates was compared; the main components that could represent the physical properties were extracted; the comprehensive score of feed grade phosphates was preliminarily determined; and the evaluation system of feed grade phosphates was established. In this experiment five commonly used dicalcium phosphates (DCP), four calcium phosphates monobasics (MCP) and three monodicalcium phosphates (MDCP) were selected as test materials. Physical characteristics such as moisture, specific surface area, sliding friction angle, angle of repose, apparent density, tapping density, compressibility, water absorption index (WAI), water solubility index (WSI) and swelling power (SP) were measured and analyzed. The PCA method was used to obtain the correlation coefficient matrix, the variance contribution analysis table, the load matrix and the eigenvector of each principal component. The comprehensive score of each sample was calculated. The results showed that: 1) There were certain differences in the physical properties of different feed grade phosphates, namely, the WAI of feed grade phosphates were between 0.28-1.35; the SP were between 0.62-2.40; the WAI and SP of DCP were higher; the WAI and SP of No. 11 (granularity of MCP) and No. 11 (granularity of MDCP) were higher than those of powder; the WSI of DCP were much lower than those of MCP and MDCP; and the angle of repose of DCP were higher, both of which were greater than 40°. The specific surface area of DCP was higher, and the granularities of MCP and MDCP were lower than those of other groups. 2) There is a certain correlation between the physical properties of phosphates in different feed grades. The specific surface area of feed grade phosphate was significantly and positively correlated with the angle of repose (r=0.919, P<0.01). The apparent density was negatively correlated with the specific surface area, sliding friction angle and angle of repose (r=-0.816, -0.809, -0.898, P<0.05). The degree of compressibility was significantly and positively correlated with the angle of repose (r=0.926, P<0.01), the specific surface area (r=0.874, P<0.05), and the apparent density (r=-0.812, P<0.05). The WSI was negatively correlated with the specific surface area and angle of repose and WAI (r=-0.812, -0.810, -0.864, P<0.05). There was a significant positive correlation between SP and WAI (r=0.930, P<0.01), a very significant negative correlation with WSI (r=-0.987, P<0.01), and a significant positive correlation with specific surface area (r=0.801, P<0.05). 3) Through PCA, three active principal components were extracted, and the contribution rates of the first, second and third principal components were 60.07%, 18.10% and 10.10% respectively. The cumulative contribution rate of the three principal components was 88.27%. The scores of the physical characteristics of feed grade phosphates were DCP, MDCP and MCP, and the score of granular phosphate was lower. Based on the comprehensive analysis, considering the physical characteristics of feed grade phosphates, DCP could be preferred for the production of higher quality feed pellets.

The purpose of this experiment was to study the effects of different processing parameters and replacement ratio of yellow mealworm protein replacing fish meal on the quality of low starch slow-sinking aquatic extruded feed, and to optimize the optimal operating parameters. In this study, a Box-Behnken experimental design was used to optimize parameters of the three independent variables with the replacement ratio of yellow mealworm protein, the moisture content of mash feed after conditioning and the die temperature, and the variation ranges were from 0 to 50%, 26% to 30%, 100 to 140 ℃, respectively. The bulk density, sinking speed, the sinking ratio at 10 min and energy consumption of extruded pellet feed were used as the dependent variables to optimize the three independent variables. The results showed that the bulk density of low starch slow-sinking extruded feed gradually increased with the increase of the replacement ratio of yellow mealworm protein, increased with the increase of moisture content, and increased slightly with the increase of die temperature. The sinking speed gradually increased with the increase of the replacement ratio of yellow mealworm protein, and increased first and then decreased with the increase of the die temperature. When the replacement ratio of yellow mealworm protein was 0 ~ 25%, the sinking speed gradually increased with the increase of moisture content. When the replacement ratio of yellow mealworm protein was 50%, the sinking speed increased first and then decreased. The sinking ratio at 10 min gradually increased with the increase of replacement ratio of yellow mealworm protein, and increased with the increase of moisture content. The sinking ratio at 10 min increased first and then decreased with the increase of die temperature. The energy consumption gradually increased with the increase of the replacement ratio of yellow mealworm protein, decreased with the increase of moisture content after conditioning, and gradually decreased with the increase of die temperature. The results of variance analysis showed that the order of effect of each factor on the quality of low starch slow-sinking aquatic extruded pellet feed was as follows: replacement ratio of yellow mealworm protein, moisture content and die temperature. The optimal process parameters and the optimal replacement ratio for processing low starch slow-sinking aquatic extruded pellet feed were optimized: the moisture content after the conditioning was 27.6%, the die temperature was 108 °C, and the replacement ratio was 15.0%. Under this process parameters, the sinking speed was 7.72 cm/s, the sinking ratio at 10 min was 95%, and the energy consumption was 34.39 (kW·h)/t. The relative error of verification test results was less than 4%, the model optimization results were reliable, and the research results can provide a reference for the production of low starch slow-sinking aquatic extruded feed.

The purpose of this experiment was to clarify the effect of different combinations of alternative soybean meal protein on the physicochemical properties and pelleting characteristics of mash feed. A correlation model was also established between physicochemical properties, pelleting efficiency and pellet quality. Seven unconventional protein sources were selected, including clostridia ethanol protein (CAP), cottonseed protein concentrate (CPC), corn gluten meal (CGM), peanut meal (PNM), rapeseed meal (RSM), cottonseed meal (CSM) and distillers dried grains with soluble (DDGS). Six groups were prepared for the alternative soybean meal protein combinations. These feedstuffs were grounded by hammer mill with a 2-mm screen-sized plate sieve, and then mixed with the rest feed raw materials in the same proportion. The pelleting experiment was carried out on the ring die pellet mill. Specifically, the die hole diameter was 3 mm, and the length-diameter ratio of ring die was 10:1. The conditioning temperature and time were 80 ˚C and 135 s respectively. The mash and pellet feed samples were taken to record the processing parameters. The parameters were then measured, including the physicochemical properties of mash feed (apparent density, tap density, angle of repose, angle of friction, water absorption, water solubility, protein dispersibility, and viscosity), and pelleting characteristics (electricity consumption per ton, pelleting rate, pellet durability index, and hardness). Partial least squares regression was utilized to establish the correlation between physicochemical properties and pelleting characteristics. Entropy weight TOPSIS analysis was made to evaluate the pelleting characteristics of different mash feed. The results showed that: 1) There were the significant differences in the physicochemical properties, pelleting efficiency and pellet quality of different groups. The water absorption index varied from 1.79 to 2.20 g/g, the final-viscosity was from 741.50~1 665.00 mPa∙s, the electricity consumption per ton was from 7.46~9.28 kWh/t, and the hardness was from 58.75~84.15 N. 2) The physicochemical properties shared a significant effect on the pelleting efficiency and pellet quality. The compression degree was negatively correlated with the pelleting rate (r=-0.519, P<0.05), positively correlated with the pellet durability index (r=0.545, P<0.05), and significantly and positively correlated with the hardness (r=0.570, P<0.01). The water absorption index and swelling degree were positively correlated with the pellet durability index (r=0.450, 0.548, P<0.05), and significantly and positively correlated with the electricity consumption per ton (r=0.869, 0.903, P<0.01). The viscosity characteristic values were significantly and positively correlated with the electricity consumption per ton (r=0.883, 0.891, 0.860, P<0.01). 3) In terms of pelleting efficiency, the best performance was found in the group with the addition of 38% peanut meal (CGP1 group), whereas, the group with the addition of 100% soybean meal (SBM group) was the worst. In terms of pellet quality, the best quality was observed in the group with the addition of ethanol Clostridium albumenum protein (CAP group), whereas, the worst was in the group with the addition of 32% corn gluten meal (CGP2 group). The finding can provide the basic data and reference for the selection and processing application of non-soybean meal diet protein source. The reduction and substitution of soybean meal can be promoted to alleviate the excessive dependence on the soybean meal import for the national food security.

The purpose of this study was to investigate the effects of different proportions of aged paddy rice replacing corn on the physicochemical properties, pelleting performance, and product quality of corn-soybean meal mixed powder. The processing parameters were then optimized under different replacement ratios. The best replacement ratio and process parameters were obtained after optimization. The aged paddy rice, corn, and soybean meal were used as the feed raw materials and then ground by a hammer mill with a sieve diameter of 2 mm. These powders were mixed evenly, according to the given proportion. Three samples of mixed powder were randomly taken and stored in a sealed bag, in order to determine the physical and chemical properties. Box-Behnken design was used in the experiment. A total of 17 groups of pelleting groups were constructed with the aged paddy rice replacement ratio, the conditioning temperature, and the conditioning time as the independent variables, while the feed pellet durability index, pelleting rate, and power consumption per ton as the dependent variables. A pellet mill was also used in the test, where the diameter of the die hole was 3 mm and the length-diameter ratio was 10:1. The pellet feed samples were taken to qualify the process parameters of the conditioner and pellet mill. The working parameters were also recorded during production. The physicochemical properties of mixed powder were measured, such as angle of repose, angle of frictional, bulk density, tap density, compressibility, water absorption index, water solubility index, swelling power, viscosity characteristic value and pelleting characteristics, such as pellet durability index, pelleting rate and power consumption per ton. Three independent variables were then optimized. The results showed as follows: 1) In properties of the material, the angle of repose, the angle of friction, bulk density, and tap density decreased significantly (P<0.05) with the increase of the proportion of aged paddy rice replacing corn; The water absorption, peak viscosity and holding viscosity increased first and then decreased, while the pasting temperature and final viscosity increased significantly (P<0.05). 2) In process parameters, the pelleting rate first increased and then decreased, and the power consumption per ton and pellet durability index gradually increased (P<0.05). The power consumption per ton decreased first and then increased significantly (P<0.05). The pellet durability index first increased and then decreased gradually. Furthermore, the pelleting rate gradually increased with the extension of conditioning time, and the power consumption per ton decreased first and then increased. The pellet durability index first increased and then significantly decreased (P<0.05). 3) The response surface variance analysis showed that the descending order was obtained: the replacement ratio of aged paddy rice > conditioning temperature > conditioning time, representing the effects of each factor on the pelleting characteristics. 4) The optimization and prediction of process parameters showed that the adjustable range of conditioning temperature increased first and then decreased with the increase of the proportion of aged paddy rice replacing corn, whereas, the conditioning time decreased first and then increased. The aged paddy rice instead of corn in feed processing can significantly improve the product quality and the energy consumption of pelleting.