本平台为互联网非涉密平台,严禁处理、传输国家秘密或工作秘密

基于多因素辅材参数的短支卷烟烟气有害成分预测模型

Multifactor prediction models for harmful components in mainstream smoke of short cigarettes based on auxiliary material parameters

  • 摘要: 为考察卷烟辅助材料参数对短支卷烟7种有害成分(CO、HCN、NNK、氨、BaP、苯酚和巴豆醛)释放量及H值(危害性指数)的影响,选取卷烟纸定量、卷烟纸透气度、卷烟纸助燃剂质量分数、卷烟纸助燃剂钾钠比、滤棒压降和接装纸透气度共6个因素,采用中心组合结合正交设计制备50个短支卷烟样品,使用样品的检测结果建立多元线性模型、二次多项式模型、多因子及互作项模型,并筛选出各指标的最优模型。通过调整卷烟辅材的参数和烟丝配方对预测模型进行了验证。结果表明:短支卷烟的7种有害成分及H值预测模型预测精度良好,8个模型的验证数据中97.5%的相对误差在±15%以内,平均绝对误差在3.4%~6.5%之间。该预测模型对短支卷烟具有良好的适用性。

     

    Abstract: To investigate the effects of auxiliary material parameters on the release of seven harmful components (CO, HCN, NNK, NH3, BaP, phenol, and crotonaldehyde) in mainstream smoke and the H index value (the hazard index) of short cigarettes, 50 short cigarette samples were prepared according to a central composite design combined with an orthogonal design, and the selected six factors were cigarette paper grammage, cigarette paper air permeability, combustion improver content in cigarette paper, K/Na ratio in the combustion improver, filter rod pressure drop, and cigarette tipping paper air permeability. Multivariate linear model, quadratic polynomial model, and multifactor and interaction model were developed using the test results, and the optimal model for each index was determined. The prediction models were verified by adjusting the parameters of the cigarette auxiliary materials and tobacco blends. The results showed that the eight models for the seven harmful components and the H index value of short cigarette had good prediction accuracies, with the relative errors within ±15% and the mean absolute errors between 3.4% and 6.5% in 97.5% of the validation data of the eight models. The applicability of the prediction models for short cigarette is thus considered satisfactory.

     

/

返回文章
返回