基于岭回归的加热不燃烧卷烟水分检测模型构建

Study on Moisture Detection Model for Heat-Not-Burn Cigarette Based on Ridge Regression

为了提高微波法在HNB卷烟水分检测模型的稳定性和准确性,本文基于岭回归法和线性回归法对HNB卷烟水分检测模型进行对比和建模分析。对295个HNB卷烟样品进行微波数据采集,使用气相色谱法作为水分化学值检测方法,在使用SPXY方法按照7∶3的比例划分样品校正样品和预测集样品后,分别采用岭回归和线性回归对HNB卷烟水分建模分析。结果表明:采用3变量、岭参数λ=0.06建立岭回归模型,其校正相关系数R_(c)、预测相关系数R p较线性回归模型均有所提高,分别从0.8686、0.9110提高到0.9471、0.9222;标准偏差从0.8961%、0.7207%,分别减小到0.7777%、0.6454%,提高了模型的稳定性,对于后期模型泛化研究提供了一定的思路。

In order to improve the stability and accuracy of microwave moisture detection model for HNB cigarette,the comparison and modeling analysis of HNB cigarette moisture detection model based on ridge regression and linear regression were carried out.The microwave data of 295 HNB cigarette samples were collected,the chemical value of water was detected by gas chromatography.Using SPXY method to divide the calibration and prediction sets samples according to the ratio of 7∶3,ridge regression and linear regression were used to analyze the moisture content of HNB cigarettes.The results showed that the ridge regression model was established with three variables and the ridge parameter was 0.06,the corrected correlation coefficient RC and predicted correlation coefficient RP were increased from 0.8686 and 0.9110 to 0.9471 and 0.9222,and the standard deviation decreased from 0.8961%and 0.7207%to 0.7777%and 0.6454%,respectively.The stability of the model was improved and some ideas were provided for the later research of model generalization.