NITS近红外光谱技术检测溶液中克螨特含量研究
本论文的主要研究对象是浓度范围在0。006～0。12 mg/kg的克螨特水溶液样品。通过10种不同的光谱预处理方法对克螨特水溶液的近红外光谱进行预处理，通过对光谱数据的预处理结果进行比较，得出了二阶导数15点平滑与多元散射校正(MSC)和Mean Center相结合的光谱预处理方法，能够取得较好的预处理结果。再通过挑选合理的有效波长宽度、挑选出最佳的主因子数目，建设了利用偏最小二乘回归算法的近红外光谱的数学模型，其中建立的校正集中的样本的预计值与实际值之间的相关系数为0。9899，外部交叉验证均方根RMSECV为0。0049，预测集中的样本预测值与实际值之间的相关系数为0。9948，预测均方根RMSEP为0。0025，证明了研究获得的模型具有较好的预测能力和较高的预测精准度。
毕业论文关键词：近红外光谱法；克螨特；偏最小二乘回归法
Abstract The purpose of the experiment is ptoposing a rapid method to evaluate the amount of propargite in solution based on near-infrared transmittance spectroscopy(NITS)。In this experiment, we make use of near-infrared transmittance spectroscopy。 The scope of monitoring is 900~1700 nm, the resolution is 8cm-1, the number of scan is 32。A total of 39 propargite solution samples were used for modeling by partial least squares (PLS) 。 The mass ratio of these samples contained from0。006 to 0。12 mg/kg。 The 39 samples were pided into three groups, two of them is the validation sets, one is the prediction set。
The main subject of the present paper is the mixed sample concentration in the range of 0。006 ~ 0。12mg / kg of an aqueous solution of propargite。 We design ten different with each other spectral pretreatment methods of the Near-infrared Spectroscopy of these mixed solution。 At the last, the experiment showed us 15 points the second derivative smoothing, the multiplicative scatter correction (MSC) and Mean Center way of pretreatment can make the best result。 And then we selected the appropriate effective wavelength width, pick out the best of the main factors the number of Principal Component Analysis using partial least squares regression analysis, the predicted value and the actual value of the correction to establish which set of samples the correlation coefficient of 0。9899 between the external cross-validation RMSECV as 0。0049, the correlation coefficient value prediction sample set and the actual value of 0。9948, 0。0025 root mean square prediction(RMSEP)。 The experiment proved that the construction of the model has good predictive ability and high prediction accuracy。
Key words: Near-Infrared Spectroscopy；propargite ；partial least squares；
目 录
第一章 绪 论 1
1。1 引言 1
1。2 近红外光谱技术的历史 1
1。2。1 近红外光谱技术的发展历程 1
1。3 近红外光谱技术的基本原理 2
1。3。1 建立光谱数据的初级模型 6
1。3。2 初级模型的扩充 7
1。3。3 模型的数量 7
1。4 近红外光谱技术的特点 8
1。4。1 近红外光谱技术的优点 8
1。4。2近红外光谱技术的不足 9
1。5 化学计量学在近红外光谱技术中的利用 9
1。5。1 光谱的预处理 10