仔猪配合料比热预测模型的构建

为探究调质、制粒等热处理过程中仔猪配合粉料传热的基本特性，有效控制热量供给并实现调质温度、时间等工艺参数的优化评估，该文以4种粉碎粒度（粉碎筛片孔径：Φ1.0、Φ1.5、Φ2.0和Φ2.5 mm）的仔猪配合粉料为研究对象，采用DSC (differential scanning calorimetry,差示扫描量热法)的方法测定了其在不同含水率范围（10%～18%，湿基）和不同温度范围（25～100℃）的比热，并分析了3个变量对比热的影响规律，构建了比热关于研究变量的预测模型。结果显示：仔猪配合粉料的比热受含水率、温度、粉碎粒度的显著影响，其中含水率的影响最为显著，温度及粉碎粒度次之；比热亦受含水率与粉碎粒度及其与温度的交互作用影响。比热随含水率和温度的增加在1.533～2.881 kJ/(kg·K)范围内升高，且与含水率呈线性关系，与温度呈非线性关系。粉碎过Φ2.5和Φ2.0 mm的仔猪料比热较粉碎过Φ1.5和Φ1.0 mm的大，在高温（60～100℃）和高含水率（18%）时表现尤为明显。基于试验数据建立了比热关于含水率、温度、粉碎粒度的多元回归模型（R2>0.993），可以有效预测仔猪配合料在此研究变量范围内的比热。研究结果为仔猪配合料调质过程传热方程的建立提供基础数据，并为其热处理工艺参数的优化提供理论依据。

Establishment of specific heat prediction model for weaned piglet mash feed

Weaned piglet’s pellet feeds are designed for healthy growth of weaned piglet. Conditioning is a key factor affecting pellet feed quality and pellet mill throughput, which can be defined as the process of transforming the physical state of the mixed mash with the addition of saturated steam to facilitate the compaction of the feed mash. Knowledge of the thermal properties of mash feed is required to perform the heat transfer calculations that are involved in the optimal design of conditioner and evaluation of operating parameters. Specific heat is an important thermodynamic parameter used in heat transfer and energy balance calculations. Conditioning processing involves broad temperature (25-95℃) and moisture content (12%-16.5%, wet base) ranges. The objectives of this investigation were to determine the specific heat of mash feeds for weaned piglet as affected by moisture content, temperature and grinding particle size. In this work, the specific heat of mash feeds for weaned piglet in 4 grinding particle sizes (geometric mean diameter of 346.34, 364.57, 388.93 and 404.25μm, and sieve aperture ofΦ1.0,Φ1.5,Φ2.0 andΦ2.5 mm) was measured as a function of moisture content (10%-18%) and temperature (25-100℃) by differential scanning calorimetry (DSC). The mash feeds were processed in Xinsanfeng feed mill in Miyun District, Beijing, while the determination of specific heat was conducted in China Agriculture University from April to November in 2015. As for the experiment design, the method of total experiment with full randomization was applied with 3 factors which included 4 kinds of grinding particle sizes (346.34, 364.57, 388.93 and 404.25μm), 16 levels of temperatures (25, 30, 35 , 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95 and 100℃) and 5 levels of moisture contents (10%, 12%, 14%, 16%and 18%). The influences of these variables on the specific heat were analyzed and a prediction model of specific heat concerning these variables was established. The specific heat was found to be significantly dependent on moisture content, temperature and grinding particle size, and the effect of moisture content was greater than the other 2 factors. In addition, the specific heat was proved to be significantly dependent on the interactions of moisture content and grinding particle size/temperature. Specific heat of mash feeds over 4 grinding particle sizes increased in the range of 1.572-2.747, 1.533-2.783, 1.562-2.869 and 1.548-2.881 kJ/kg·K with the increase in moisture content and temperature respectively. Specific heat had a linear positive correlation with moisture content and a logarithmic correlation with temperature. At 25℃ and 85℃, the specific heat values of mash feeds over 4 grinding particle sizes varied in the range of 1.533-2.037 and 2.289-2.758 kJ/kg·K with the increase of moisture content respectively. The specific heat of mash feeds with higher grinding particle size (404.25 and 388.93μm) was greater than that with lower grinding particle size, especially at high temperature (60-100℃) and high moisture content (18%). As the grinding particle size of mash feed is reduced, the surface area is increased geometrically. So per unit weight of mash feed with lower particle size will condense more heat and water. This will result in the mash feed with lower particle size having higher degrees of starch gelatinization and protein denaturation. The variations in state and components of mash feeds may contribute to the differences in specific heat. Regression models with high R2 values were established to predict the specific heat of mash feeds as a function of 2 dependent terms of only moisture content and temperature, and of 3 dependent terms of moisture content, temperature and grinding particle size. The inclusion of grinding particle size dependent term in the empirical equation is expected to enhance the predictive capacity of model at high moisture contents in comparison with that equation only including moisture content and temperature. The results provide basic data for the establishment of the heat-transfer equation in conditioning of feeds for weaned piglet, and offer theoretical reference for the optimization of process parameters in thermo-processing.