Tobacco chemical value quantifying method based on near-infrared spectrum wave number K-means clustering

Abstract

The invention discloses a tobacco chemical value quantifying method based on near-infrared spectrum wave number K-means clustering, comprising the following steps: establishing a training set and a test set, and collecting near-infrared spectrum and target component content of all tobacco samples in the training set; clustering the wave number of the near-infrared spectrum of each tobacco sample in the training set through K-means clustering; after each clustering, using PLS to establish a relationship model between each subclass spectral band and the target component content, and calculating the root mean square error for cross validation of each relationship model; taking the number of clustering with the minimum sum of the root mean square error for cross validation corresponding to the relationship models as the optimal clustering number, and performing weighted summation on the relationship models corresponding to the optimal clustering number to obtain a full-spectrum model; and collecting near-infrared spectrum of each tobacco sample in the test set, and obtaining the target component content of each tobacco sample in the test set on the basis of the full-spectrum model. Compared with the existing PLS method, the method of the invention can significantly reduce the prediction error of a model.

Description

technical field [0001] The invention relates to the technical field of physical and chemical detection of tobacco, in particular to a method for quantifying the chemical value of tobacco based on near-infrared spectral wavenumber K-mean clustering. Background technique [0002] The main chemical components in tobacco, such as total sugar, nicotine, reducing sugar, and total nitrogen, have an important impact on the quality of tobacco leaves, and are the main factors that determine the strength, alcohol and degree of smoke. In the tobacco industry, the analysis and determination of conventional chemical components is of great significance to the quality control of finished cigarettes. [0003] Near-infrared spectroscopy can characterize the information of various hydrogen-containing groups in the analyte. It has the advantages of convenient sampling, no damage, no pollution, and online detection. It is very suitable for the detection of various complex mixtures. Near-infrare...

AI Technical Summary

Problems solved by technology

[0008] The existing near-infrared-based tobacco chemical value modeling method is a single PLS algorithm. During the execution of this algorithm, the local information of the spectrum is not screened or processed, resulting in some high-noise variables entering the modeling process at the same time. The spectral bands with strong correlation of the chemical values ​​to be measured have not been properly enhanced, resulting in the suboptimal prediction accuracy and interpretation ability of the model

[0009] Since the existing near-infrared-based tobacco chemical value modeling method is a single PLS algorithm, which uniformly processes each band in the near-infrared spectrum, the ability to suppress spectral noise is not strong, and the ability to mine effective information in the spectrum is insufficient. Shortcomings

Images