热裂解-气相色谱-质谱法结合偏最小二乘法鉴别地理标志产品镇江香醋

Identification of geographically indicated Zhenjiang aromatic vinegar by pyrolysis-gas chromatography-mass spectrometry combined with partial least squares

中文关键词:  镇江香醋  热裂解  气相色谱质谱联用  偏最小二乘法

英文关键词:Zhenjiang aromatic vinegar  pyrolysis  gas chromatography-mass spectrometry  partial least squares

中文摘要:

      目的 建立热裂解-气相色谱-质谱法(pyrolysis-gas chromatography-mass spectrometry, Py-GC-MS)结合偏最小二乘法(partial least squares, PLS)鉴别地理标志产品镇江香醋的方法。方法 利用Py-GC/MS采集106个食醋样本的总离子流图, 挑选21个共有峰, 并通过美国国家标准与技术研究院质谱数据库(National Institute of Standards and Technology Mass Spectral Database, NIST MS)对其进行定性分析。21个共有峰的峰面积组成样品的基本矢量数据, 经标准化处理后, 通过“去N法”筛选模型特征变量, 并利用经典的PLS算法构建地理标志产品镇江香醋鉴别模型。结果 模型验证结果显示: 建模集食醋样品(71个)的正确识别率为94.4%, 验证集食醋样品(106个样品)的正确识别率为96.2%, 表现出较高的准确性和稳定性。结论 本研究提供了一种简便、可靠的地理标志产品镇江香醋鉴别方法, 具有较高的应用价值。

英文摘要:

      Objective To establish a method for identification of geographically indicated Zhenjiang aromatic vinegar by pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS) combined with partial least squares (PLS). Methods The total ion flow pattern of 106 vinegar samples was collected by Py-GC/MS, and 21 common peaks were selected. It was qualitatively analyzed by the National Institute of Standards and Technology Mass Spectral Database (NIST MS). The peak areas of 21 common peaks constituted the basic vector data of the sample. After standardized processing, the feature variables of the model were screened by the “leave-N-out”, and the identification model of the geographical indicated Zhenjiang aromatic vinegar was constructed by the classical PLS algorithm. Results The model verification results showed that the correct recognition rate of the model set vinegar samples (71 samples) was 94.4%, and the correct recognition rate of the verification set vinegar samples (106 samples) was 96.2%, showing high accuracy and stability. Conclusion This study provides a simple and reliable method for identifying the geographically indicated product Zhenjiang aromatic vinegar, demonstrating significant practical value.