【摘要】 采用多元校正技术偏最小二乘法结合具有代表性的烟气总粒相物萃取液的傅里叶变换近红外透射光谱220个,分别建立了焦油、烟碱和一氧化碳的校正模型,并对光谱数据预处理方法和异常值诊断进行了讨论。所建的焦油、烟碱和一氧化碳模型相关系数分别为0.9922、0.9828和0.9734;均方预测残差(RMSECV)分别为0.3392、0.0329和0.3478;预测范围分别为预测范围分别为11.0～15.0mg、1.15～1.45mg和13.0～17.5mg。通过对模型进行t检验,在显著性水平大于0.05的条件下,其测定结果与标准方法的测定结果对比,两者无显著性差异。该分析方法应用于同时测定卷烟焦油、烟碱和一氧化碳的释放量,结果令人满意。更多还原

【Abstract】 The contents of tar, nicotine and carbon monoxide in smoking is the standard of quality for coming into the market. Determination of these contents plays an important role in their quality control. The calibration models were established for determination of tar,nicotine and carbon monoxide by partial least squares and Fourier transform near infrared transmittance spectrometry of 220 representative extractions from total particulate matter of smoking. Spectral data preprocessing and outliers diagnostics were also discussed. Correlation coefficients of the models were 0.9922, 0.9828 and 0.9734, and the root mean square error of cross validation (RMSECV) were 0.3392, 0.0329 and 0.3478 in the predicted range of 11.0～15.0 mg, 1.15～1.45 mg and 13.0～17.5 mg for tar, nicotine and carbon monoxide, respectively. By statistical significance test, the results of determination were compared with those of standard methods with no significant difference at 0.05 level. The method has been applied to simultaneous determination of tar, nicotine and carbon monoxide in smoking with satisfactory results.更多还原