基于核极限学习机奶牛日粮能量消化预测方法

为准确评价饲料营养价值和提升饲喂管理水平,精准预测奶牛日粮能量消化指标具有重要意义。传统上主要基于线性回归(LR)方法预测奶牛日粮能量消化指标,但受参数模型假设限制,预测结果精度低,甚至偏离实际。文章首次将核极限学习机(KELM)方法应用于奶牛日粮消化能(DE)和能量消化率(ED)预测,KELM作为一种典型非参数机器学习模型,无需提前对预测模型作任何假设,仅通过学习训练样本数据,便可拟合出最接近实际的函数,特别适用于奶牛日粮能量消化等复杂系统预测问题。与传统LR方法和其他非参数模型RBF-ANN、SVM及标准ELM对比验证与讨论,结果表明,基于KELM预测方法在MAE、MAPE、RMSE及RT等多数指标上优于其他方法,特别是与传统LR方法相比,KELM方法预测精度更高,可作为对奶牛日粮DE和ED预测新型参考方法,为人工智能与机器学习在预测和评价动物饲粮营养价值应用研究提供借鉴。

Predicting the diet energy digestion of dairy cows using kernel extreme learning machine

In order to effectively evaluate the diet nutritional value and improve feeding management, it is essential to accurately predict the indicators of diet energy digestion of dairy cows.The traditional mathematical models used to predict the indicators of diet energy digestion of dairy cows are usually based on linear regression(LR) method. However, the LR method may be limited by regression function assumption, its prediction results are not accurate enough, and sometimes deviate from the actual. In this study, a kernel extreme learning machine(KELM) technique was applied to predict the indicators of digestible energy(DE) and energy digestibility(ED) of dairy cows for the first time. KELM is a typical non-parametric machine learning model that does not require any specific assumptions about the regression function in advance and only by learning the training sample data,then can fit the function closest to the actual in most cases. Therefore, this method is especially suitable for complex system prediction problems such as diets energy digestion of dairy cows. To evaluate prediction accuracy effectively, the KELM technique is compared with the traditional LR method and other non-parametric model such as radial basis function artificial neural network(RBFANN), support vector regression(SVR) and standard ELM methods. The prediction results indicate that the proposed KELM-based prediction technique is superior to other methods in most metrics such as MAE, MAPE, RMSE, and RT. In particular, it has higher prediction accuracy than traditional LR method, and can be used as a new reference method for predicting DE and ED indicators of dairy cows. At the same time, the proposed KELM-based prediction technique provides a reference for the application of artificial intelligence and machine learning in predicting and evaluating the nutritional value of animal diets.