摘要
【目的】解决烤鸭传统挂炉烤制过程中中心温度难以在线精确监测的问题。【方法】通过测定烤鸭的品质指标,利用多元线性回归、偏最小二乘回归、支持向量回归、人工神经网络等方法对北京烤鸭中心温度进行在线客观预测。【结果】烤鸭胸肉的L*、a*、b*、脱氧肌红蛋白、氧合肌红蛋白、高铁肌红蛋白、水分含量、脂肪含量、蛋白二级结构等指标均可用于有效识别北京烤鸭的中心温度;线性模型多元线性回归和偏最小二乘回归的预测集决定系数RC2分别为0.9543和0.9384,均方根误差SEC分别为5.8205℃和6.7634℃,MLR模型预测效果优于偏最小二乘回归模型;非线性模型支持向量回归优于人工神经网络模型,其预测集决定系数RC2和交叉验证决定系数RCV2分别为0.9837和0.9496,均方根误差SEC和交叉验证均方根误差SECV分别为3.5215℃和6.1236℃,北京烤鸭中心温度预测模型构建以支持向量回归模型效果最好;支持向量回归验证集的决定系数RV2较高,达到0.9748,均方根误差SEV为5.5204℃,结合建模结果得出支持向量回归模型预测挂炉烤制北京烤鸭的中心温度效果最佳。【结论】北京烤鸭胸肉的色度、肌红蛋白、水分含量、脂肪含量、蛋白二级结构等可有效识别北京烤鸭的中心温度;基于品质指标的SVR模型可准确预测烤鸭的中心温度。
【Objective】 The aim of this study was to solve the problem in detecting center temperature of Beijing roast duck during traditional Gua-lu roasting accurately and timely. 【Method】 The prediction models of center temperature were established by using multiple linear regression, partial least-squares regression, support vector regression and artificial neural network according to the quality indicators. 【Result】 The results showed that the models were effective to identify center temperature of Beijing roast duck by L*, a*, b*, deoxymyoglobin, oxymyoglobin, metmyoglobin, moisture and fat content, as well as protein secondary structure of duck breast. The RC2 of multiple linear regression and partial least-squares regression were 0.9543 and 0.9384, and SEC of 5.8205℃ and 6.7634℃, respectively. The prediction effect of multiple linear regression was better than partial least-squares regression, while the prediction model of support vector regression was superior to artificial neural network. RC2 and RCV2 of support vector regression were 0.9837 and 0.9496, SEC and SECV were 3.5215℃ and 6.1236℃, respectively, so the support vector regression was the best prediction model of center temperature. The RV2 of the verified models of support vector regression was 0.9748, and the SEV was 5.5204℃. The model obtained by support vector regression together with the modeling results could accurately predict the center temperature of Beijing roast duck. 【Conclusion】 The color, myoglobin, water content, fat content and protein secondary structure of the breast of Beijing roast duck could effectively identify the central temperature. The SVR model was the most accurate prediction model for the center temperature.
作者
柳艳霞
王振宇
郑晓春
朱瑶迪
陈丽
张德权
LIU YanXia;WANG ZhenYu;ZHENG XiaoChun;ZHU YaoDi;CHEN Li;ZHANG DeQuan(Institute of Food Science and Technology,Chinese Academy of Agricultural Sciences/Key Laboratory of Agro-Products Processing,Ministry of Agriculture and Rural Affairs,Beijing 100193;College of Food science and Technology,Henan Agricultural University/Henan Key Laboratory of Meat Processing and Quality Safety Control,Zhengzhou 450002)
出处
《中国农业科学》
CAS
CSCD
北大核心
2020年第8期1655-1663,共9页
Scientia Agricultura Sinica
基金
国家重点研发计划(2016YFD0401505)。
关键词
北京烤鸭
中心温度
品质
预测模型
Beijing roast duck
center temperature
quality
prediction model
