基于矿物质元素指纹分析技术的浦城大米产地溯源研究

Study on Origin Traceability of Pucheng Rice Using Mineral Element Fingerprinting Analysis Technology

摘要:

通过分析浦城大米、江西地区大米、湖北地区大米、湖南地区大米和福建地区（除浦城以外）大米中25种矿物质元素质量分数特征，筛选出14种差异显著的矿物质元素，并比较主成分分析（PCA）和正交偏最小二乘判别分析（OPLS-DA）对矿物质元素的降维效果，发现经过PCA降维后的线性判别分析（LDA）结果优于OPLS-DA，进而建立了“差异分析+主成分分析+判别分析”的浦城大米产地溯源模型，筛选出钒（V）、钴（Co）、砷（As）、铷（Rb）、银（Ag）、铯（Cs）等6种特征矿物质元素，判别模型回代检验的整体正确判别率为98.1%，交叉检验的整体正确判别率为97.1%。该模型能准确识别浦城大米，为浦城大米的产地溯源和质量控制提供了参考。

Abstract:

By analyzing the mass fraction characteristics of 25 mineral elements in Pucheng rice, rice from Jiangxi region, rice from Hubei region, rice from Hunan region, and rice from other Fujian region, 14 mineral elements with significant differences were screened out. After dimension reduction of mineral elements by principal component analysis (PCA) and orthogonal partial least-squares discriminant analysis (OPLS-DA), the results of linear discriminant analysis (LDA) after dimension reduction by PCA were found to be superior to OPLS-DA. Consequently, a Pucheng rice origin traceability model of “difference analysis + principal component analysis + discriminant analysis” was established, and six characteristic mineral elements, i.e., vanadium(V), cobalt (Co), arsenic(As), rubidium(Rb), silver(Ag), and cesium(Cs), were selected. The overall correct discrimination accuracy of the back-substitution test for the discriminant model was 98.1%, and the overall correct discrimination accuracy of the cross-validation was 97.1%. It can accurately identify Pucheng rice, providing a reference for the origin traceability and quality control of Pucheng rice.