基于品种、年份和产地对绿豆矿物元素产地溯源的指标筛选

Indicator Selection of Source Tracing of Production Area of Mineral Elements of Mung Bean Based on Variety,Year and Production Area

通过分析品种、年份与产地及其交互作用对绿豆矿物元素含量造成的差异,筛选有效特征指标,结合统计分析进行产地判别。以连续2年(2018—2019年)来自吉林白城、黑龙江泰来、黑龙江杜尔伯特蒙古、山东泗水地区的绿豆样本为研究对象,采用电感耦合等离子体质谱仪测定样品的52种元素。结果表明:Na、Mg、P、K、Mn、Cu、As和Se元素受产地影响较大;Na、As、Pd、B、Mn、Co、Cu、Pt、Pb和Lu元素受年份影响较大;P、Mn、Y、Cd、Cs、Ba、Lu、Ir和Pb元素受品种影响较大。实验对筛选到与产地直接相关的21种元素进行PCA分析和Fisher判别分析。7个主成分累计贡献率81.98%。建立的判别模型对4个产地的判别正确率为96.7%,交叉验证判别率为93.3%。说明由这些元素组成的模型可以对样本实现正确判别。

By analyzing the differentiation caused by variety, year and production area and interaction function towards the mineral elements content of the mung bean, effective feature indicators were selected, and statistics analysis was combined to conduct production area judgement. The mung bean samples from the cities of Baicheng, Jilin, Heilongjiang Tailai, Heilongjiang Dulbert Mongolia, Shandong Sishui area in successive two years(2018—2019) were used as the research materials, and inductively coupled plasma-mass spectrum(ICP-MS) was adopted to measure the fifty-two kinds of elements in the samples. The results indicated that: the elements of Na, Mg, P, K, Mn, Cu, As and Se were greatly influenced by the production area;the elements of Na, As, Pd, B, Mn, Co, Cu, Pt, Pb and Lu were greatly influenced by the year;and the elements of P, Mn, Y, Cd, Cs, Ba, Lu, Ir and Pb were greatly influenced by the variety. The experiment conducted the main components analysis and judgement analysis towards the selected twenty-one kinds of elements which were directly related to the production area. And the accumulated contribution rate of the seven main components was 81.98%. The correctness rate of the judgement of the established judgement model towards the four production areas was 96.7%, crossing verification judgement rate was 93.3%, which illustrated that the models formed by these elements can realize correct judgement towards the samples.