摘要以不同来源的菊花为研究载体，建立菊花中总黄酮的常规含量测定方法——比较不同溶剂及浓度、提取时间、料液比对提取效果的影响。采集菊花样品近红外光谱（NIRS）并对其光谱预处理方法等进行考察，借助化学计量学方法将常规分析方法与NIRS进行关联，建立菊花中总黄酮的NIR快速质量评价模型。73035

实验结果表明：不同体积分数的乙醇对菊花中总黄酮的提取效率相对较高，因此选择不同体积分数乙醇作为正交试验的考察因素考察60%、70%、80%体积分数的乙醇发现，70%体积分数的乙醇提取效果最好。随着提取时间的增加，菊花中黄酮提取率随之升高。可见反应时间越长，提取越充分，但当提取时间超过 60min 时，提取率趋于稳定，增长较小，因此选择以提取时间30 min、45 min、60 min 作为正交试验的三个水平。随着溶剂的增加，菊花中黄酮提取率随之逐渐增大， 当料液比为 1:50 时达到菊花黄酮提取的最大值，之后提取溶剂再增加，提取率又逐渐降低。说明随着溶剂用量的增加，可能降低了提取物中有效成分的浓度。 因此，选定料液比为 1:40、 1:50、 1:60 作为正交试验的三个水平。根据单因素试验的结果，选取影响菊花总黄酮提取效果各因素中有意义的水平做正交试验，对结果进行极差分析，以确定最佳的提取条件。采用 L9(34)正交表,以 提取时间(A)、乙醇体积分数(B)、料液比 (C)、作为 3个考察因素， 选取 3 个水平进行试验。由表2的极差R分析结果可以看出:RC>RB>RA，影响提取质量的主要因素为:B、C，根据各因素组合试验结果，确定较佳工艺条件C3B2A3，即料液比:1:50、乙醇体积分数:70%、提取时间:60min。

采用NIRS法建立菊花中总黄酮快速含量测定方法研究中，对模型建立方法、光谱预处理方法、模型建立中最佳潜变量数等进行筛选，确定以PLS法进行模型的建立，以SNV法结合1D 、ND法对菊花NIR原始光谱进行预处理，并通过比较预测残差平方和（PRESS）、交叉验证均方根误差（RMSECV）等参数确定了模型建立时最佳潜变量数为8。采用上述方法所建立的校正模型，其相关系数为0。8329，校正误差均方根（RMSEC）为0。0661，预测误差均方根（RMSEP）为0。1000，预测值和真实值之间相关性较好，说明所建立的PLS模型能够有效地预测未知含量的菊花样品中总黄酮的含量。该方法的建立将对菊花的快速质量评价体系的研究提供一定的科学依据。

Abstract With different sources of chrysanthemum as the research carrier, the establishment of the total flavonoids in chrysanthemum was determined by the conventional content measurement method -- comparison of different solvents and concentration, extraction time, the effect of the ratio of material to extract。 Chrysanthemum samples collected near infrared spectroscopy (NIRS) and the spectral preprocessing methods were investigated。 Using chemometric methods will be routine analysis method and NIRS Association, total flavonoids in Chrysanthemum NIR rapid quality evaluation model is established。

Experimental results show that the extraction efficiency of total flavonoids from Chrysanthemum with different volume fraction of ethanol is relatively high, so the choice of different volume fraction of ethanol as the study of orthogonal test factors study of 60%, 70%, 80%, volume fraction of ethanol found that 70% volume fraction of ethanol extraction effect best。 With the increase of extraction time, the extraction rate of flavonoids in Chrysanthemum increased。 Visible longer reaction time, extracting more fully, but when the extraction time more than 60 minutes, extraction rate tends to be stable, the increase was smaller, so the choice of the extraction time of 30 min, 45 min, 60 min as the three levels of the orthogonal test。 With the increase of solvent, the extraction rate of flavonoids in Chrysanthemum gradually increased, reached the maximum when the chrysanthemum flavones extracted when the solid-liquid ratio was 1:50, after extracting solvent and increase the extraction rate was gradually decreased。 It showed that the concentration of effective components in the extract could be decreased with the increase of the amount of solvent。 Therefore, the selected material liquid ratio of 1:40, 1:50, 1:60 as the three levels of orthogonal test。 According to the results of single factor experiment, the orthogonal experiment was conducted to determine the effect of the factors on the extraction of total flavonoids from chrysanthemum, and the results were analyzed to determine the best extraction conditions。 L9 (34) orthogonal table was used to extract the time (A), ethanol volume fraction (B), the ratio of material to liquid (C), as the 3 factors, and 3 levels were selected。 From table 2 the range R analysis results can be seen: RC>RB>RA, influence of the extraction quality of the main factors: B, C, according to the various factors combination test results and determine optimum process conditions C3B2A3, i。e。 feed liquor ratio: 1:50, the volume fraction of ethanol: 70%, extraction time 60min。