利用CNCPS V5.0预测我国杂种肉牛干物质采食量的效果评价  

Evaluation on Dry Matter Intake Predictions by the Cornell Net Carbohydrate and Protein System with Crossbred Beef in China

作　　者：杜晋平[1] 杨晓林[1] 殷裕斌[1] 解祥学[2] 孟庆翔[2] 

机构地区：[1]长江大学动物科学学院,湖北荆州434025 [2]中国农业大学动物科技学院肉牛研究中心,北京100193

出　　处：《中国草食动物》2010年第2期16-20,共5页China Herbivores

摘　　要：通过两个动物试验(子试验1和子试验2)比较了CNCPS模型预测的我国杂种肉牛干物质采食量(DMI)与实际观察值间的吻合程度,目的在于评估该模型是否适合于预测我国杂种肉牛生产性能。结果表明:①子试验1中3个处理组分别有93%、80%和73%的点落在-0.4至0.4kg/d的范围内,子试验2的4个处理组分别有87%、73%、73%和80%的点落在-0.4至0.4kg/d的范围内,说明CNCPS较好地预测了我国杂种肉牛的DMI。②经线性回归分析,两个子试验中观察和预测的DMI间相关系数分别为0.83和0.79,具有较高的可信度;所有处理组的DMI误差均方根(RMSE)都很小。以上结果说明,CNCPS对中国杂种肉牛DMI具有较好的预测能力。Two separate animal trials （sub-trial 1 and sub-trial 2） were conducted to evaluate the coincidence of dry matter intake （DMI） values predicted by CNCPS V 5.0 and observed actually.The results showed that three treatments in sub-trial 1 had 93%,80%,and 73% of predicted points fell within the range from-0.4 to 0.4 kg/d.Similarly,four treatments in sub-trial 2,about 87%,73%,73%,and 80% predicted points fell within these ranges,respectively,suggesting that the CNCPS model could predict accurately DMI in these trials.By regression analysis,acceptable values of R^2 （R^2=0.83 and 0.79 for trial 1 and trial 2,respectively） and small root mean square error （RMSE） values of all treatments were obtained.Results indicated that the CNCPS model could be used precisely to predict DMI of China＇s crossbred beef.

关 键 词：康奈尔净碳水化合物和蛋白质体系模型 杂种肉牛 干物质采食量 预测评估 

分 类 号：S823.4[农业科学—畜牧学]