基于SNV和MSC结合遗传算法对羊肉葡萄糖含量可见-近红外光谱建模的效果  

Modeling effect of visible-near infrared spectrum on mutton glucose content based on SNV and MSC combined with genetic algorithm

作　　者：尹成诚 康景 刘建新 年芳[1] 唐德富 YIN Chengcheng;KANG Jing;LIU Jianxin;NIAN Fang;TANG Defu(College of Science,Gansu Agricultural University,Lanzhou 730070,Gansu,China)

机构地区：[1]甘肃农业大学理学院,甘肃兰州730070

出　　处：《草业科学》2024年第10期2427-2434,共8页Pratacultural Science

基　　金：国家自然科学基金项目(31860655)。

摘　　要：为提高羊肉中营养成分可见-近红外光谱预测模型的稳定性和准确性,本研究以葡萄糖(GLU)指标为例,采用遗传算法(GA)提取特征波长后,结合标准正态变换(SNV)和多元散射校正(MSC)两种预处理方式进行偏最小二乘法建立预测模型,对比SNV、MSC预处理直接进行偏最小二乘的建模效果。结果显示:在标准正态变换下遗传偏最小二乘模型(GA-SNV-PLS)优于直接在标准正态变换下偏最小二乘模型(FS-SNV-PLS);经交叉验证后,该模型的均方根误差(RMSE)为0.122,决定系数(R^(2))为0.930,相对分析误差(RPD)为2.295;相较于全光谱偏最小二乘模型(FSPLS)、全波段多元散射校正FS-MSC-PLS和多元散射下GA-MSC-PLS,其R^(2)和RPD分别提高了95.80%、50.21%、85.05%和62.65%、37.08%、52.54%。结果表明,由SNV结合遗传算法建立的偏最小二乘模型能够提高模型的预测能力。To improve the stability and prediction ability of the visible-near infrared spectral model for mutton nutrients,taking glucose(GLU)as an example,the characteristic wavelength was extracted by genetic algorithm(GA)and a prediction model was established.Two preprocessing methods,standard normal transformation(SNV)and multivariate scattering correction(MSC),were used to directly model the partial least squares regression and the results were compared.Genetic partial least squares model under SNV(GA-SNV-PLS)was better than the direct partial least squares model under SNV(FSSNV-PLS).After cross-validation,the root mean square error(RMSE)of the model was 0.122,determinant coefficient R^(2)was 0.930,and relative analysis error(RPD)was 2.295.Compared with the full spectrum,the R^(2)and RPD for MSC and genetic partial least square model under MSC increased by 95.80%,50.21%,85.05%;62.65%,37.08%,and 52.54%,respectively.

关 键 词：近红外光谱 羊肉 葡萄糖 标准正态变换 多元散射校正 遗传算法 

分 类 号：O657.33[理学—分析化学] TP18[理学—化学] TS251.53[自动化与计算机技术—控制理论与控制工程]