中国荷斯坦牛泌乳曲线数学模型拟合及应用

发布时间：2018-08-17 08:49

【摘要】：[目的]本文旨在建立拟合中国荷斯坦牛泌乳曲线的数学模型,计算出一套可靠的305 d产奶量校正系数。[方法]分别使用Wood不完全伽玛模型、逆多项式模型、Ali-Schaeffer模型、六次多项式回归模型、Wilmink模型和Pollott模型对71个奶牛场43 812头中国荷斯坦牛的609 784条测定日记录进行了拟合。参数的估计使用SAS 9.2软件中的非线性回归过程迭代求解实现。[结果]6种数学模型的拟合度(R2值)为0.940~0.947。其中,六次多项式回归模型的拟合度(R2值)最大,模型估计误差标准差最小,在本次试验中拟合效果最好。[结论]用六次多项式回归模型计算出各泌乳时间的累积产奶量,继而提出了计算305 d产奶量的校正系数。结合实际累积产奶量和相应的校正系数能将所有未达到305 d产奶量的记录校正到305 d。
[Abstract]:[objective] the purpose of this paper is to establish a mathematical model to fit the lactation curve of Chinese Holstein cattle, and to calculate a set of reliable correction coefficient of milk yield for 305 days. [methods] the Wood incomplete gamma model, the inverse polynomial model Ali-Schaeffer model, the sixth order polynomial regression model Wilmink model and the Pollott model were used to fit the 609,784 daily records of 43,812 Chinese Holstein cattle in 71 dairy farms. The parameter estimation is realized by iterative solution of nonlinear regression process in SAS 9.2 software. [results] the fitting degree (R2 value) of the six mathematical models was 0.940 ~ 0.947. Among them, the fitting degree (R2 value) of the sixth degree polynomial regression model is the largest, and the standard deviation of the model estimation error is the smallest. The fitting effect is the best in this experiment. [conclusion] the cumulative milk yield of each lactation time was calculated by using the sixth order polynomial regression model, and the correction coefficient for calculating the milk yield of 305 days was put forward. Combined with the actual cumulative milk yield and the corresponding correction coefficient, all records of milk yield less than 305 days can be corrected to 305 days.
【作者单位】：四川农业大学畜禽遗传资源发掘与创新利用四川省重点实验室;
【基金】：四川省科技支撑项目(2014NZ0032) 四川省教育厅资助科研项目(14ZB0004)
【分类号】：S823

【相似文献】