【作者】 张立钢；

【导师】 孙文志；

【作者基本信息】 东北农业大学 ， 动物营养与饲料科学， 2000， 硕士

【摘要】 本研究采用系统仿真技术，在广泛查阅文献的基础上建立了适温条件下猪生长的动态营养模型。在该营养模型中，主要研究采食养分与动物生产性能之间的关系。模型中考虑了性别差异对动物生长可能带来的影响。 在营养模型中主要围绕体蛋白和体脂肪来展开。研究当输入养分在数量和质量上发生变化时，其转化为体蛋白和体脂肪的日沉积速率。模型中仅考虑了实际输入养分流中的两种：即能量流与蛋白质流。并假设除能量和蛋白质以外的输入养分始终都处于最适水平，对生长没有任何限制作用。在体组成之间数量关系的描述上，主要采用了异速生长公式。在自由采食量的预测方面，用累计采食能量对时间进行多项式拟合。并可应用到对模型的调试和验证中去。在对文献进行综合分析及模型验证的基础上，本文对猪的维持能耗值采用Em=110*LBW^0.75kcal/day，将蛋白质的最大沉积速率作为活体重的函数处理，根据试验结果将蛋白质的日沉积速率定为100.53（母猪）和102.77（阉公猪），变化趋势按照NRC第十版猪的营养需要（1998）提供的缺省曲线（各种变化趋势的折衷）处理。 本研究进行现场试验，针对市售三元杂交商品猪（长白×大白×杜洛克），建立阉公猪和母猪的动态营养仿真模型。结合饲养试验、消化试验、屠宰试验，运用动态拟合技术，得到猪的自由消化能采食量与生长时间的多项式方程。通过屠宰试验，得到该品种猪的胴体体成分指标，用异速生长方程拟合体蛋白质与与空体重、活重与消化能采食量。用Gompertz方程拟合时间与体重的关系。 本研究通过利用自行饲养试验动物进行模型验证，可以证明，虽然模型中的许多参数的取值参考了国外研究的试验数据，但具有一定通用性。验证结果营养模型的基本框架和具体参数取值具有较好的合理性与准确性。尽管在建模过程中为把握模型主体而对问题进行了一定简化，模型析因过程在理论上也还需要深入探讨，但在一定范围内具有意义。随着生产实际中得到数据的丰富，模型中参数取值可以通过与试验数据拟合得以精确化。 在本研究中总结了对某一品种猪进行模型仿真的一般步骤。 在现有营养模型的基础上逐步做深入工作，在模型中考虑的对营养利用影响因子中加入环境因素的影响、以及生产过程的投入产出和成本核算等等，模型实用性将提高。当建立起更为有效的生长仿真模型之后，便可以在遗传、营养及环境等各方面因素都发生变化的条件下对动物的生长表现准确地做出预测。这有助于在生产实践中对饲养策略硕士学位论文 猪生长的营养仿翼筷型的建立 东北农业大学一的制订和优化；并在生产经营决策方面起到辅助支持作用。生长仿真模型还可以对某一饲粮配方结果的生产性能和经济效益做出预测和评价；如果将配方优化技术与模型仿真技术相结合，将有助于以最大效益为优化目标的新一代饲料配方软件的实现。更多还原

【Abstract】 In this study, a nutritional model of pig growth in thermoneutral conditions was built up by technique of systematic simulation. In this model, the relationship between ingested nutrients and animal performance was studied chiefly. The influence of sex was considered. A frame of model was developed primarily, which suitable for pig growing. Some important parameters in the model were estimated from experimental data. The animal growth processes are simplified as the changes of amount of body compositions. Two important compositions are body protein and body lipid, and the key of quality or quantity of ingested nutrients. Only two flows of nutrients, energy and protein, are considered in the model; other nutrients besides arc supposed obtaining in sufficient levels and are not limited to animal growth. Ailoinetric equations are used to describe the relationship among body compositions. In the prediction of voluntazy feed intake, a method of fitting cumulative energy intake with a polynomial equation to time is introduced. Protein plays a central role in the simulation model. For instance, body protein is deposited in prior to lipid: the compositions excluding body lipid are determined by allometric equations with body protein; digestible energy intake (DEl) is divided into protein energy and protein-free digestible energy; the turnover rate of body protein influences the rationality of classical factorial equation for energy, and so on. The maximal retention rate of body protein (Pdmax) is a key variable in the model, its value depends on strain and sex of animals. In the model Pdmax was considered to be a function of live body weight.The daily average protein deposition rate was lOO.53g for gilt and 102.77g for castrate according the result of experiment .The trend of Pdmax was considered to be the same as the recommended curve of NRC (1998) A great disagreement exists concerning how to express the maintenance energy requirement (Em) and the unitazy energy cost of protein retention (bp) . It is unreasonable in statistics when using a classical factorial equation to determine Em and bp simultaneously. A constant Em of 110* LBW拁? kcal /day and umtaiy energy cost of lipid retention (bp) is 53kJ/g. On the basis of this study, it is feasible that using the foilowing equation to estimate the energy .3 - requirement of pig growing: Using the data from the experiments, fitting the equation of corpus and corpus protein, time and feed intake and so on. Instead of the normal iterative algorithm, a new deductive solution was developed to determine the daily retention rate of body protein( Pd )and lipid( Ld). This algorithm simplifies the calculation process, saves rtmning time of model, and clarifies the inner structure of model. The effect of minimal ratio of retention rate of lipid to protein (M) on predicted animal performance is not significant when pig fed ad libitum. since the energy intake of animal in normal condition id sufficient to fit the minimal physiological requirement of lipid retention. The model was tested with some practical performance data of pig growth. A good fitness exists between the real values and predicted values made by the growth model, so the model is a valid one. It is concluded that the basic frame of the model and the taking value of parameters are quite reasonable and accuracy Although the model makes many simplifications and contains some更多还原