CN113804648A - Tobacco online real-time monitoring device and application thereof in tobacco quality nondestructive rapid quality control - Google Patents
Tobacco online real-time monitoring device and application thereof in tobacco quality nondestructive rapid quality control Download PDFInfo
- Publication number
- CN113804648A CN113804648A CN202111097714.5A CN202111097714A CN113804648A CN 113804648 A CN113804648 A CN 113804648A CN 202111097714 A CN202111097714 A CN 202111097714A CN 113804648 A CN113804648 A CN 113804648A
- Authority
- CN
- China
- Prior art keywords
- tobacco
- correction
- coefficient
- model
- reference value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 241000208125 Nicotiana Species 0.000 title claims abstract description 232
- 235000002637 Nicotiana tabacum Nutrition 0.000 title claims abstract description 232
- 238000012806 monitoring device Methods 0.000 title claims abstract description 36
- 238000003908 quality control method Methods 0.000 title claims abstract description 19
- 239000000463 material Substances 0.000 claims abstract description 76
- 230000003595 spectral effect Effects 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims abstract description 26
- 230000005540 biological transmission Effects 0.000 claims abstract description 17
- 238000004519 manufacturing process Methods 0.000 claims abstract description 12
- 238000010926 purge Methods 0.000 claims abstract description 9
- 238000012937 correction Methods 0.000 claims description 96
- 238000002790 cross-validation Methods 0.000 claims description 65
- 230000002452 interceptive effect Effects 0.000 claims description 58
- 238000004445 quantitative analysis Methods 0.000 claims description 57
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 claims description 28
- 238000005259 measurement Methods 0.000 claims description 28
- 241000196324 Embryophyta Species 0.000 claims description 23
- 238000001228 spectrum Methods 0.000 claims description 21
- 238000012795 verification Methods 0.000 claims description 19
- 230000003287 optical effect Effects 0.000 claims description 17
- ZAMOUSCENKQFHK-UHFFFAOYSA-N Chlorine atom Chemical compound [Cl] ZAMOUSCENKQFHK-UHFFFAOYSA-N 0.000 claims description 14
- 229910002651 NO3 Inorganic materials 0.000 claims description 14
- NHNBFGGVMKEFGY-UHFFFAOYSA-N Nitrate Chemical compound [O-][N+]([O-])=O NHNBFGGVMKEFGY-UHFFFAOYSA-N 0.000 claims description 14
- ZLMJMSJWJFRBEC-UHFFFAOYSA-N Potassium Chemical compound [K] ZLMJMSJWJFRBEC-UHFFFAOYSA-N 0.000 claims description 14
- 229930013930 alkaloid Natural products 0.000 claims description 14
- 150000003797 alkaloid derivatives Chemical class 0.000 claims description 14
- 239000000460 chlorine Substances 0.000 claims description 14
- 229910052801 chlorine Inorganic materials 0.000 claims description 14
- 229910052757 nitrogen Inorganic materials 0.000 claims description 14
- 239000011591 potassium Substances 0.000 claims description 14
- 229910052700 potassium Inorganic materials 0.000 claims description 14
- 229910052736 halogen Inorganic materials 0.000 claims description 9
- 150000002367 halogens Chemical class 0.000 claims description 9
- 239000004575 stone Substances 0.000 claims description 9
- WFKWXMTUELFFGS-UHFFFAOYSA-N tungsten Chemical compound [W] WFKWXMTUELFFGS-UHFFFAOYSA-N 0.000 claims description 9
- 229910052721 tungsten Inorganic materials 0.000 claims description 9
- 239000010937 tungsten Substances 0.000 claims description 9
- 238000007664 blowing Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 5
- 238000000034 method Methods 0.000 claims description 5
- 230000001066 destructive effect Effects 0.000 claims description 4
- 238000010238 partial least squares regression Methods 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 2
- 230000000087 stabilizing effect Effects 0.000 claims description 2
- 238000002329 infrared spectrum Methods 0.000 abstract description 24
- 238000013178 mathematical model Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 38
- 238000010200 validation analysis Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000004497 NIR spectroscopy Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000010453 quartz Substances 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 2
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005194 fractionation Methods 0.000 description 1
- 125000000524 functional group Chemical group 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- -1 polytetrafluoroethylene Polymers 0.000 description 1
- 229920001343 polytetrafluoroethylene Polymers 0.000 description 1
- 239000004810 polytetrafluoroethylene Substances 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000007430 reference method Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
- 238000001845 vibrational spectrum Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
Abstract
A tobacco online real-time monitoring device comprises a control system, a transmission system and a light path system, wherein the control system is electrically connected with the transmission system and the light path system, the transmission system is used for stably transmitting a tobacco sample and comprises a conveyor belt and a stepping motor, the light path system is used for collecting the spectral data of the tobacco sample and comprises an annular cover, a light source, a convex lens, a spectrometer and a purging component, the light path system is arranged above the conveyor belt, and the control system is arranged above the light path system; according to the tobacco online real-time monitoring device and the application thereof in tobacco quality nondestructive rapid quality control, the functions of accurately identifying foreign matters of tobacco materials, judging the height fluctuation of the tobacco materials, rapidly and quantitatively analyzing the quality of the tobacco materials without damage and accurately grading the similar tobacco materials are realized by a near infrared spectrum instrument erected on a tobacco material inclined plane conveying production line and by combining mathematical model analysis, so that the technical problem of tobacco material online quality control is solved.
Description
Technical Field
The invention belongs to the technical field of tobacco online monitoring, and particularly relates to a tobacco online real-time monitoring device and application thereof in nondestructive rapid quality control of tobacco quality.
Background
With the development of productivity and the advancement of society, the online quality control of industrial and agricultural products is an important demand to be solved urgently in the production line in recent years. Different from the traditional off-line detection technology, the on-line detection technology requires that a detection instrument device is erected on a production line, real-time quality control data of products are obtained in a very short time, and the fault-tolerant capability of uncertain factors in the product transmission process on the production line is very high.
Tobacco is an important commercial crop. For a long time, the quality control of tobacco is always an important technical means for improving the product quality of tobacco production enterprises. With the development of the tobacco industry, the demand for the online quality control of tobacco materials is increasingly prominent. The four technical requirements become technical problems to be solved urgently by current tobacco production enterprises, wherein the tobacco material foreign matter is accurately identified, the tobacco material height fluctuation is judged, the tobacco material quality is nondestructively and rapidly analyzed quantitatively, and the similar tobacco materials are accurately classified.
Disclosure of Invention
In order to solve the problems, the invention provides an online real-time tobacco monitoring device and application thereof in nondestructive rapid quality control of tobacco quality.
The technical scheme adopted by the invention is as follows:
the utility model provides an online real-time monitoring device of tobacco, includes control system, transmission system and optical path system, control system and transmission system, optical path system electricity be connected, transmission system is used for stabilizing the conveying to the tobacco sample, contains conveyer belt and step motor, optical path system is used for the collection to tobacco sample spectral data, contains annular cover, light source, convex lens, spectrum appearance and sweeps the part, the conveyer belt top sets up optical path system, optical path system's top sets up control system, is provided with computer system and touch-sensitive screen among the control system.
The control system controls the stepping motor to operate to drive the transmission belt to transmit the tobacco materials upwards, and the transmission speed of the stepping motor driving the transmission belt is 1.2 m/s-1.8 m/s, preferably 1.5 m/s.
The control system is used for controlling the opening and closing of the device, controlling the spectrometer to collect spectral data, storing and exporting the data, implanting and calling functions of the model and displaying results.
The upper side of the conveying belt is provided with an annular cover, the side face of the annular cover is provided with a halogen tungsten lamp and a blowing hole, the top face of the annular cover is provided with a convex lens, and a near-infrared spectrometer is arranged above the convex lens.
The near-infrared spectrometer is positioned at the focus above the convex lens, and the number of the blowing holes is four, and the four blowing holes are respectively positioned at the upper side and the lower side of the halogen tungsten lamp.
The vertical distance between the lower opening of the annular cover and the conveyor belt is 100mm, the convex lens is made of near-infrared quartz material, the diameter is 80mm, the focal length is 150mm, and the waveband range of the spectrometer is 908 nm-1676 nm; the integration time of the spectrometer is 0.08-0.12 second, preferably 1.00 second, the system is controlled to start a purging component to purge and clean the light path system, and the pressure of the purging gas flow is 0.3-1.0 MPa, preferably 0.4-0.7 MPa.
The tobacco online real-time monitoring device is used for realizing the application of accurate identification of tobacco foreign matters, the tobacco online real-time monitoring device is used for acquiring spectral data of common foreign matters in tobacco and tobacco production, the common foreign matters comprise withered branches, sponge, hay, plant leaves, paper, hair and broken stones, the tobacco and the spectral data of various foreign matters are used as independent variables, the tobacco, the withered branches, the sponge, the hay, the plant leaves, the paper, the hair and the broken stones are assigned reference values of 0, 1, 2, 3, 4, 5, 6 and 7 respectively, and a tobacco foreign matter accurate identification partial least square-discriminant analysis model is established by adopting partial least square regression and a full-interactive verification algorithm;
factor number of established tobacco foreign body accurate identification partial least square-discriminant analysis modelN f = 6, correction data prediction value-reference value regression equation y = 0.9987x + 0.0051, correction correlation coefficientr C = 0.9993, correction of measurement coefficientR 2 C = 0.9987, correct root mean square errorRMSEC= 0.08, cross validation data prediction value-reference value regression equationy = 0.9983x + 0.0072, interactive proof correlation coefficientr CV = 0.9992, interactive proof determination coefficientR 2 CV = 0.9985, cross validation root mean square errorRMSECV= 0.09, and the accuracy rates for tobacco, dead branches, sponges, hay, plant leaves, paper, hair, and crushed stones were 100%, and 100%, respectively, using 0.7, 1.5, 2.5, 3.5, 4.5, 5.5, and 6.5 as thresholds, respectively.
The tobacco online real-time monitoring device is used for judging the height fluctuation of tobacco materials, the tobacco online real-time monitoring device collects tobacco material spectrum data which are higher than the tobacco material reference height by 0mm, 5.6 mm, 11.2 mm, 21.5 mm, 32.6 mm, 43.7 mm and 53.9 mm, the tobacco material spectrum data with different heights are used as independent variables, the tobacco material reference height data which are higher than the tobacco material are used as dependent variables, and a partial least square regression and full interactive verification algorithm are adopted to establish a partial least square-discriminant analysis model for judging the height fluctuation of the tobacco materials;
factor number of partial least square-discriminant analysis model for judging height fluctuation of tobacco materialN f = 2, correction data prediction value-reference value regression equationy = 0.9982x+ 0.0428, correction of correlation coefficientr C = 0.9991, correction of measurement coefficientR 2 C = 0.9982, correct root mean square errorRMSEC= 0.79, regression equation of predicted value-reference value of cross validation datay = 0.9983x+ 0.0392, cross-validation correlation coefficientr CV = 0.9991, interactive proof determination coefficientR 2 CV = 0.9982, cross validation root mean square errorRMSECVAnd = 0.80, which are respectively 100%, 45% of the accuracy of the fluctuation in height of the tobacco material, which is higher than the reference height of the tobacco material by 0mm, 5.6 mm, 11.2 mm, 21.5 mm, 32.6 mm, 43.7 mm, 53.9 mm, using 5, 8, 15, 25, 36, 45 as a threshold value.
The tobacco online real-time monitoring device is used for acquiring online spectral data of tobacco materials and synchronously sampling online, reference values of total plant alkaloid, total sugar, reducing sugar, total nitrogen, nitrate, total potassium and chlorine in the tobacco samples are respectively obtained by the same method, the online spectral data of the tobacco materials are used as independent variables, the reference values of the total plant alkaloid, the total sugar, the reducing sugar, the total nitrogen, the nitrate, the total potassium and the chlorine in the tobacco samples are respectively sampled synchronously online, and a tobacco quality lossless rapid quantitative analysis model is established by combining partial least squares regression with a full interactive verification algorithm;
factor number of the total plant alkaloid quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.8885x+ 0.1146, correction of correlation coefficientsr C = 0.9426, correction of measurement coefficientR 2 C = 0.8885, correct root mean square errorRMSEC= 0.028; cross validation data prediction value-reference value regression equationy = 0.8768x+ 0.1266, interactive proof correlation coefficientr CV = 0.9306, interactive proof determination coefficientR 2 CV = 0.8668, cross validation root mean square errorRMSECV= 0.031; model (model)RPD = 2.71。
Factor number of total sugar quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.7483x+ 2.4445, correction of correlation coefficientsr C = 0.8650, correction of measurement coefficientR 2 C = 0.7483, correct root mean square errorRMSEC = 0.20; cross validation data prediction value-reference value regression equationy = 0.7229x+ 2.6916, interactive proof correlation coefficientr CV = 0.8399, interactive proof determination coefficientR 2 CV = 0.7070, cross validation root mean square errorRMSECV = 0.22; model (model)RPD = 2.37。
Factor number of established reducing sugar quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.7874x+ 1.7250, correction of correlation coefficientsr C = 0.8874, correction of measurement coefficientR 2 C = 0.7874, correct root mean square errorRMSEC = 0.15；Cross validation data prediction value-reference value regression equationy = 0.7633x+ 1.9204, interactive proof correlation coefficientr CV = 0.8650, interactive proof determination coefficientR 2 CV = 0.7497, cross validation root mean square errorRMSECV= 0.16; model (model)RPD = 2.72。
Factor number of total nitrogen quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.8374x+ 0.3002, correction of correlation coefficientsr C = 0.9151, correction of measurement coefficientR 2 C = 0.8374, correct root mean square errorRMSEC= 0.042; cross validation data prediction value-reference value regression equationy = 0.8141x+ 0.3436, cross validation correlation coefficientr CV = 0.8962, interactive proof determination coefficientR 2 CV = 0.8046, cross validation root mean square errorRMSECV = 0.046; model (model)RPD = 2.37。
Factor number of established nitrate quantitative analysis modelN f = 7; correction data prediction value-reference value regression equationy = 0.7118x+ 0.0961, correction of correlation coefficientr C = 0.8437, correction of measurement coefficientR 2 C = 0.7118, correct root mean square errorRMSEC= 0.012; cross validation data prediction value-reference value regression equationy = 0.6964x+ 0.1012, interactive proof correlation coefficientr CV = 0.8280, interactive proof determination coefficientR 2 CV = 0.6876, cross validation root mean square errorRMSECV = 0.012; model (model)RPD = 2.28。
Factor number of total potassium quantitative analysis modelN f = 13; correction data prediction value-reference value regression equationy = 0.8106x+ 0.4173, correction of correlation coefficientsr C = 0.9003, correction of measurement coefficientR 2 C = 0.8106, correct root mean square errorRMSEC = 0.038; interactive verification data predictionValue-reference value regression equationy = 0.7774x+ 0.4903, interactive proof correlation coefficientr CV = 0.8699, interactive proof determination coefficientR 2 CV = 0.7599, cross validation root mean square errorRMSECV= 0.044; model (model)RPD = 2.93。
Number of factors of the established chlorine quantitative analysis modelN f = 15; correction data prediction value-reference value regression equationy = 0.8683x+ 0.0832, correction of correlation coefficientsr C = 0.9318, correction of measurement coefficientR 2 C = 0.8683, correct root mean square errorRMSEC= 0.015; cross validation data prediction value-reference value regression equationy = 0.8367x+ 0.1032, interactive proof correlation coefficientr CV = 0.8996, interactive proof determination coefficientR 2 CV = 0.8096, cross validation root mean square errorRMSECV= 0.018; model (model)RPD = 3.05。
The online real-time tobacco monitoring device realizes the application of the same-class tobacco precise grading function, a spectrum matching value calculation method is adopted to realize the same-class tobacco precise grading, 11 samples of the same-class tobacco samples are subjected to 10 times of parallel acquisition of spectral data by the online real-time tobacco monitoring device respectively, the 10 times of parallel spectral data of the same-class tobacco samples are subjected to average data calculation, Standard Normal Variate (SNV) and first-order 5-point derivative data preprocessing are carried out on the average data in sequence, and then the spectrum matching value is calculated by taking the average spectrum subjected to the data preprocessing as a reference; the range of a discrimination Threshold (Threshold) is set to be 0.99977-0.99985, preferably 0.99979-0.99982, further preferably 0.99980, and on the basis of taking the column as a reference, the spectral matching values of other tobacco samples are smaller than the Threshold except that the diagonal SMV value is 1.0000, namely the tobacco sample of the same class has a self matching value of 1.0000, so that the other samples have differences, namely the tobacco samples of the same class are identified as the tobacco samples of the same class but different classes, and accurate classification of the tobacco of the same class is realized.
The application of any one of the above in the nondestructive rapid quality control of tobacco quality.
The near infrared spectrum belongs to molecular vibration spectrum and is mainly derived from frequency combination and frequency doubling absorption of hydrogen-containing functional groups in substance molecules. The modern near infrared spectroscopy analysis technology integrates three scientific and technological pillars of a near infrared spectroscopy theory, an instrument precise micro-electro-mechanical manufacturing technology and a chemometrics algorithm, not only has the capability of online quality control of products in theory, but also can technically realize nondestructive, rapid and efficient quality control of the products.
The invention relates to a tobacco online real-time monitoring device and application thereof in nondestructive rapid quality control of tobacco quality, wherein the device is erected on a tobacco material inclined plane conveying production line based on a near infrared spectrum analysis technology, and near infrared spectrum data of tobacco materials on an inclined plane conveying belt are collected through a near infrared spectrometer; through mathematical model analysis, accurate identification of foreign matters in tobacco materials, height fluctuation judgment of the tobacco materials, lossless rapid quantitative analysis of the tobacco materials and accurate classification of similar tobacco materials are realized in a very short time, so that the requirement of online quality control of the tobacco materials is met.
The invention has the beneficial effects that: according to the tobacco online real-time monitoring device and the application thereof in the tobacco quality nondestructive rapid quality control, the functions of accurately identifying foreign matters of tobacco materials, judging the height fluctuation of the tobacco materials, rapidly and quantitatively analyzing the quality of the tobacco materials without damage and accurately grading similar tobacco materials are realized by a near infrared spectrum instrument erected on a tobacco material inclined plane conveying production line and by combining mathematical model analysis, so that the technical problem of tobacco material online quality control is solved.
Drawings
FIG. 1 is a schematic side view of an online real-time tobacco monitoring device according to the present invention.
FIG. 2 is a schematic side view of an optical path system of the tobacco on-line real-time monitoring device according to the present invention.
FIG. 3 is a scatter diagram of No. 2 and No. 3 principal component score of near infrared spectrum for non-destructive rapid identification of tobacco foreign matter.
FIG. 4 is a correlation diagram of the predicted value-reference value of the correction data of the near infrared spectrum model for the nondestructive rapid identification of tobacco foreign matters.
FIG. 5 is a diagram of the relationship between the predicted value and the reference value of the full-interactive verification data of the nondestructive rapid identification near-infrared spectrum model of the tobacco foreign matter.
FIG. 6 is a scatter diagram of the 1 st and 2 nd principal components of the near infrared spectrum for rapid highly nondestructive identification of tobacco materials according to the present invention.
FIG. 7 is a correlation diagram of the predicted value-reference value of the correction data of the tobacco material high-nondestructive rapid identification near infrared spectrum model.
FIG. 8 is a correlation diagram of the predicted value-reference value of the full-interactive verification data of the tobacco material high-nondestructive rapid identification near infrared spectrum model.
FIG. 9 is a correlation diagram of the tobacco quality (total plant alkaloid) near infrared spectrum quantitative analysis model correction data prediction value-reference value.
FIG. 10 is a correlation diagram of the tobacco quality (total plant alkaloid) near infrared spectrum quantitative analysis model full-interactive verification data prediction value-reference value.
FIG. 11 is a correlation diagram of the predicted value-reference value of the correction data of the tobacco quality (total sugar) near infrared spectrum quantitative analysis model.
FIG. 12 is a correlation diagram of the predicted value-reference value of the full-interactive verification data of the tobacco quality (total sugar) near infrared spectrum quantitative analysis model.
FIG. 13 is a correlation diagram of the tobacco quality (reducing sugar) near infrared spectrum quantitative analysis model correction data prediction value-reference value.
FIG. 14 is a correlation diagram of a predicted value-reference value of full-interactive verification data of a tobacco quality (reducing sugar) near infrared spectrum quantitative analysis model.
FIG. 15 is a correlation diagram of the tobacco quality (total nitrogen) NIR quantitative analysis model correction data prediction value-reference value according to the invention.
FIG. 16 is a correlation diagram of the tobacco quality (total nitrogen) near infrared spectrum quantitative analysis model full-interactive verification data prediction value-reference value.
FIG. 17 is a correlation diagram of the tobacco quality (nitrate) near infrared spectrum quantitative analysis model correction data prediction value-reference value.
FIG. 18 is a correlation diagram of the tobacco quality (nitrate) near infrared spectrum quantitative analysis model full-interactive verification data prediction value-reference value.
FIG. 19 is a correlation diagram of the tobacco quality (total potassium) near infrared spectrum quantitative analysis model correction data prediction value-reference value.
FIG. 20 is a correlation diagram of the predicted value-reference value of the full-interactive verification data of the tobacco quality (total potassium) near infrared spectrum quantitative analysis model.
FIG. 21 is a correlation diagram of the predicted value-reference value of the correction data of the tobacco quality (chlorine) near infrared spectrum quantitative analysis model.
FIG. 22 is a correlation diagram of the tobacco quality (chlorine) near infrared spectrum quantitative analysis model full-interactive verification data prediction value-reference value.
Wherein: the device comprises a conveyor belt 1, a control system 2, a touch screen 3, an optical path system 4, an annular cover 5, halogen tungsten lamps 6-1 and 6-2, a convex lens 7, a near-infrared spectrometer 8 and blowing holes 9-1, 9-2, 9-3 and 9-4; the LS light source emits light, the LR sample reflects the light, the LF convex lens condenses the light, and the D transmission direction is realized; h1 annular shroud to conveyor distance, H2 convex lens focal length; wherein H1 = 100mm, H2 = 150mm, and the diameter phi = 80mm of the convex lens (7).
Detailed Description
The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention. Unless otherwise specified, the technical means used in the examples are conventional means well known to those skilled in the art, and the raw materials used are commercially available products.
The embodiments described below are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, unless otherwise specified, the terms "top," "bottom," "upper," "lower," and the like refer to orientations or positional relationships illustrated in the drawings, which are used for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the referenced system or element must have a particular orientation, be constructed in a particular orientation, and be operated, and thus should not be construed as limiting the present invention.
It is to be understood that, unless otherwise expressly stated or limited, the term "coupled" is used in a generic sense as defined herein, e.g., fixedly attached or removably attached or integrally attached; may be directly connected or indirectly connected through an intermediate. The specific meaning of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The embodiment 1 "tobacco on-line real-time monitoring device" is described with reference to fig. 1 and 2.
The tobacco on-line real-time monitoring device and the application thereof in the tobacco quality nondestructive rapid quality control comprise the following structures and can realize the following functions:
(1) the structure is as follows: comprises a control system 2, a transmission system and an optical path system 4;
(2) the functions are as follows: the functions of accurately identifying foreign matters in the tobacco, judging the height fluctuation of tobacco materials, quickly and quantitatively analyzing the tobacco quality without damage and accurately grading the similar tobacco are realized.
The control system 2 is used for controlling the device to be opened and closed, controlling the spectrometer to collect spectrum data, storing and exporting data, implanting and calling functions of a model and displaying results, comprises a computer system and a touch screen 3, and controls the device through the touch screen 3; the control system 2 is electrically connected with the transmission system and the optical path system 4.
The transmission system is used for stably conveying tobacco samples and comprises a conveyor belt 1 and a stepping motor, the control system 2 controls the stepping motor to operate to drive the conveyor belt 1 to convey tobacco materials upwards, and the conveying speed of the stepping motor driving the conveyor belt 1 is 1.2 m/s-1.8 m/s, preferably 1.5 m/s.
An optical path system 4 is arranged above the conveyor belt 1, a control system 2 is arranged above the optical path system 4, and a computer system and a touch screen 3 are arranged in the control system 2.
The optical path system 4 is used for collecting the spectral data of the tobacco sample and comprises an annular cover 5, a light source, a convex lens 7, a spectrometer and a purging component. An annular cover 5 is arranged above the conveyor belt 1, halogen tungsten lamps 6-1 and 6-2 and blowing holes 9-1, 9-2, 9-3 and 9-4 are arranged on the side face of the annular cover 5, a convex lens 7 is arranged on the top face of the annular cover 5, and a near-infrared spectrometer 8 is arranged above the convex lens 7. The blowing component is an air compressor which is communicated with the blowing hole through a pipeline.
The vertical distance between the lower opening of the annular cover 5 and the conveyor belt 1 is 100 mm; the light source is preferably a halogen tungsten lamp; the convex lens is made of near-infrared quartz material, and has the diameter of 80mm and the focal length of 150 mm; the spectrometer is positioned at a focus above the convex lens 7, and preferably a near-infrared spectrometer; the upper side and the lower side of the halogen tungsten lamp 6-1 and 6-2 of the sweeping component are respectively provided with 2 air inlet holes 9-1, 9-2, 9-3 and 9-4.
The tobacco material is conveyed upwards through the conveyor belt 1 and is irradiated by the light source when passing through the light path system 4, reflected light is converged to the spectrograph positioned at the focus of the convex lens through the convex lens 7, the spectrograph receives the reflected converged light and converts the reflected converged light into spectral data, and the spectral data is recorded and exported through the control system 2. The spectrometer is a near infrared spectrometer 8; the spectrometer adopts a polytetrafluoroethylene white board as a spectrum reference; the waveband range of the spectrometer is 908 nm-1676 nm; the integration time of the spectrometer is 0.08-0.12 seconds, preferably 1.00 second; and after the spectrum is collected, the control system 2 starts a purging component to purge and clean the light path system 4, and the pressure of the purging gas flow is 0.3-1.0 MPa, preferably 0.4-0.7 MPa.
Example 2 accurate identification of tobacco foreign bodies
The description of "tobacco foreign matter accurate identification" in example 2 is made with reference to fig. 3, 4 and 5.
The tobacco online real-time monitoring device provided by the invention is adopted to realize accurate identification of tobacco foreign matters. The device is used for collecting the spectral data of common foreign matters in tobacco and tobacco production respectively. The common foreign matters comprise deadwood, sponge, hay, plant leaves, paper, hair and broken stones. The tobacco and different object spectral data are used as independent variables, reference values of 0, 1, 2, 3, 4, 5, 6 and 7 are assigned to tobacco, deadwood, sponge, hay, plant leaves, paper, hair and broken stones respectively, and a partial least square regression and full-interactive verification algorithm are adopted to establish a tobacco foreign object accurate identification partial least square-discriminant analysis model.
The 2 nd and 3 rd principal component score scatter diagrams of the established tobacco foreign matter accurate identification partial least square-discriminant analysis model are shown in the attached figure 3. As can be seen from the attached figure 3, the 2 nd and 3 rd principal component score scatter diagrams can clearly distinguish 7 kinds of foreign matters from tobacco, and the established partial least square-discriminant analysis model for accurately identifying the foreign matters in the tobacco has good accurate identification capability on the foreign matters in the tobacco.
Factor number of established tobacco foreign body accurate identification partial least square-discriminant analysis modelN f And (6). The correlation diagram of the predicted value and the reference value of the correction data of the established tobacco foreign body accurate identification partial least square-discriminant analysis model is shown in figure 4, and the regression equation of the predicted value and the reference value of the correction datay = 0.9987x+ 0.0051, correction of correlation coefficientr C = 0.9993, correction of measurement coefficientR 2 C = 0.9987, correct root mean square errorRMSECAnd = 0.08. The correlation diagram of the predicted value and the reference value of the interactive validation data of the established tobacco foreign body accurate identification partial least square-discriminant analysis model is shown in figure 5, and the regression equation of the predicted value and the reference value of the interactive validation datay = 0.9983x+ 0.0072, interactive proof correlation coefficientr CV = 0.9992, interactive proof determination coefficientR 2 CV = 0.9985, cross validation root mean square errorRMSECV = 0.09。
The accuracy rates of tobacco, deadwood, sponge, hay, plant leaves, paper, hair and broken stone are respectively 100%, 100% and 100%, respectively, with 0.7, 1.5, 2.5, 3.5, 4.5, 5.5 and 6.5 as threshold values.
Example 3 tobacco material height fluctuation determination
The "tobacco material height fluctuation judgment" in example 3 will be described with reference to FIG. 6, FIG. 7 and FIG. 8.
The tobacco online real-time monitoring device provided by the invention is adopted to realize the judgment of the height fluctuation of tobacco materials. The device is used for collecting tobacco material spectrum data which are 0mm, 5.6 mm, 11.2 mm, 21.5 mm, 32.6 mm, 43.7 mm and 53.9 mm higher than the reference height of the tobacco material respectively. And establishing a partial least square-discriminant analysis model for judging the height fluctuation of the tobacco material by adopting partial least square regression and combining a full-interactive verification algorithm by taking the spectral data of the tobacco materials with different heights as independent variables and the data higher than the reference height of the tobacco materials as dependent variables.
The scatter diagram of the No. 1 and No. 2 principal component score of the built tobacco material height fluctuation judgment partial least square-discriminant analysis model is shown in the attached figure 6. As can be seen from the attached figure 6, the 1 st and 2 nd principal component scatter diagram can clearly distinguish tobacco materials with 7 heights, and the partial least square-discriminant analysis model for judging the height fluctuation of the tobacco materials has good judgment capability on the height fluctuation of the tobacco materials.
Factor number of partial least square-discriminant analysis model for judging height fluctuation of tobacco materialN f And (2). The correlation diagram of the predicted value and the reference value of the correction data of the built partial least square-discriminant analysis model for determining the height fluctuation of the tobacco material is shown in figure 7, and the regression equation of the predicted value and the reference value of the correction datay = 0.9982x+ 0.0428, correction of correlation coefficientr C = 0.9991, correction of measurement coefficientR 2 C = 0.9982, correct root mean square errorRMSEC= 0.79. The correlation diagram of the predicted value-reference value of the interactive validation data of the built tobacco material height fluctuation judgment partial least square-discriminant analysis model is shown in figure 8, and the regression equation of the predicted value-reference value of the interactive validation data is shown in figurey = 0.9983x+ 0.0392, cross-validation correlation coefficientr CV = 0.9991, interactive proof determination coefficientR 2 CV = 0.9982, cross validation root mean square errorRMSECV = 0.80。
The accuracy of the fluctuation of the height of the tobacco material with the height of 0mm, 5.6 mm, 11.2 mm, 21.5 mm, 32.6 mm, 43.7 mm and 53.9 mm higher than the reference height of the tobacco material is respectively 100%, 100% and 100% by taking 5, 8, 15, 25, 36 and 45 as threshold values.
Example 4 non-destructive Rapid quantitative analysis of tobacco quality
Example 4, "non-destructive rapid quantitative analysis of tobacco quality" is described with reference to fig. 9 to 22.
The tobacco online real-time monitoring device provided by the invention is adopted to realize nondestructive rapid quantitative analysis of tobacco quality. The device of the invention is used for collecting the online spectral data of tobacco materials and synchronously sampling online, and reference values of total plant alkaloid, total sugar, reducing sugar, total nitrogen, nitrate, total potassium and chlorine in the tobacco samples are respectively obtained based on a reference method. On-line spectral data of tobacco materials are used as independent variables, reference values of total plant alkaloid, total sugar, reducing sugar, total nitrogen, nitrate, total potassium and chlorine in on-line synchronously sampled tobacco samples are used as dependent variables, and a partial least squares regression and full-interactive verification algorithm are combined to establish a tobacco quality lossless rapid quantitative analysis model.
Factor number of the total plant alkaloid quantitative analysis modelN f = 10. The correlation diagram of the predicted value of correction data and the reference value of the total plant alkaloid quantitative analysis model is shown in FIG. 9, and the regression equation of the predicted value of correction data and the reference valuey = 0.8885x+ 0.1146, correction of correlation coefficientsr C = 0.9426, correction of measurement coefficientR 2 C = 0.8885, correct root mean square errorRMSEC= 0.028. The correlation diagram of the predicted value-reference value of the cross validation data of the total plant alkaloid quantitative analysis model is shown in FIG. 10, and the regression equation of the predicted value-reference value of the cross validation data is shown in FIG. 10y = 0.8768x+ 0.1266, interactive proof correlation coefficientr CV = 0.9306, interactive proof determination coefficientR 2 CV = 0.8668, cross validation root mean square errorRMSECV= 0.031; model (model)RPD = 2.71。
Factor number of total sugar quantitative analysis modelN f = 10. The correlation graph of correction data prediction value-reference value of the total sugar quantitative analysis model is shown in FIG. 11, and the regression equation of correction data prediction value-reference valuey = 0.7483x + 2.4445, correction of correlation coefficientsr C = 0.8650, correction of measurement coefficientR 2 C = 0.7483, correct root mean square errorRMSEC= 0.20. The correlation diagram of the predicted value-reference value of the cross validation data of the total sugar quantitative analysis model is shown in FIG. 12, and the regression equation of the predicted value-reference value of the cross validation data is shown in FIG. 12y = 0.7229x+ 2.6916, interactive proof correlation coefficientr CV = 0.8399, interactive proof determination coefficientR 2 CV = 0.7070, cross validation root mean square errorRMSECV= 0.22; model (model)RPD = 2.37。
Factor number of established reducing sugar quantitative analysis modelN f = 10. The correlation diagram of correction data predicted value-reference value of the established reducing sugar quantitative analysis model is shown in figure 13, and the regression equation of correction data predicted value-reference valuey = 0.7874x+ 1.7250, correction of correlation coefficientsr C = 0.8874, correction of measurement coefficientR 2 C = 0.7874, correct root mean square errorRMSEC= 0.15. the correlation diagram of the predicted value and the reference value of the interactive validation data of the established quantitative reducing sugar analysis model is shown in the attached figure 14, and the regression equation of the predicted value and the reference value of the interactive validation data is shown in the figurey = 0.7633x+ 1.9204, interactive proof correlation coefficientr CV = 0.8650, interactive proof determination coefficientR 2 CV = 0.7497, cross validation root mean square errorRMSECV = 0.16; model (model)RPD = 2.72。
Factor number of total nitrogen quantitative analysis modelN f = 10. The correlation graph of the predicted value of correction data and the reference value of the total nitrogen quantitative analysis model is shown in FIG. 15, and the regression equation of the predicted value of correction data and the reference valuey = 0.8374x+ 0.3002, correction of correlation coefficientsr C = 0.9151, correction of measurement coefficientR 2 C = 0.8374, correct root mean square errorRMSEC = 0.042. The correlation diagram of the predicted value-reference value of the cross validation data of the total nitrogen quantitative analysis model is shown in FIG. 16, and the regression equation of the predicted value-reference value of the cross validation data is shown in FIG. 16y = 0.8141x+ 0.3436, cross validation correlation coefficientr CV = 0.8962, interactive proof determination coefficientR 2 CV = 0.8046, cross validation root mean square errorRMSECV= 0.046; model (model)RPD= 2.37。
Factor number of established nitrate quantitative analysis modelN f And = 7. The correlation graph of the corrected data predicted value-reference value of the established nitrate quantitative analysis model is shown in FIG. 17, and the regression equation of the corrected data predicted value-reference valuey = 0.7118x+ 0.0961, correction of correlation coefficientr C = 0.8437, correction of measurement coefficientR 2 C = 0.7118, correct root mean square errorRMSEC= 0.012. The correlation diagram of the predicted value-reference value of the cross validation data of the established nitrate quantitative analysis model is shown in FIG. 18, and the regression equation of the predicted value-reference value of the cross validation datay = 0.6964x+ 0.1012, interactive proof correlation coefficientr CV = 0.8280, interactive proof determination coefficientR 2 CV = 0.6876, cross validation root mean square errorRMSECV= 0.012; model (model)RPD= 2.28。
Factor number of total potassium quantitative analysis modelN f = 13. The correlation diagram of the predicted value of correction data and the reference value of the total potassium quantitative analysis model is shown in FIG. 19, and the regression equation of the predicted value of correction data and the reference valuey = 0.8106x+ 0.4173, correction of correlation coefficientsr C = 0.9003, correction of measurement coefficientR 2 C = 0.8106, correct root mean square errorRMSEC= 0.038. The correlation diagram of the predicted value-reference value of the cross validation data of the established total potassium quantitative analysis model is shown in FIG. 20, and the regression equation of the predicted value-reference value of the cross validation data is shown in the figurey = 0.7774x+ 0.4903, interactive proof correlation coefficientr CV = 0.8699, interactive proof determination coefficientR 2 CV = 0.7599, cross validation root mean square errorRMSECV= 0.044; model (model)RPD= 2.93。
Number of factors of the established chlorine quantitative analysis modelN f And (5) = 15. The correlation graph of the correction data prediction value and the reference value of the chlorine quantitative analysis model is shown in FIG. 21, and the regression equation of the correction data prediction value and the reference valuey = 0.8683x+ 0.0832, correction of correlation coefficientsr C = 0.9318, correction of measurement coefficientR 2 C = 0.8683, correct root mean square errorRMSEC= 0.015. The correlation diagram of the predicted value-reference value of the cross validation data of the chlorine quantitative analysis model is shown in FIG. 22, and the regression equation of the predicted value-reference value of the cross validation data is shown in FIG. 22y = 0.8367x+ 0.1032, interactive proof correlation coefficientr CV = 0.8996, interactive proof determination coefficientR 2 CV = 0.8096, cross validation root mean square errorRMSECV= 0.018; model (model)RPD = 3.05。
The parameters of each model built include: factor number (Nf), Calibration Coefficient of Regression (R2C), Calibration Root Mean Square Error (RMSEC), Calibration Regression Equation (REC), Calibration Regression Coefficient (rC), Cross Validation measurement Coefficient (R2 CV), Cross Validation Root Mean Square Error (RMSECV), Cross Validation Regression Equation (RECV), Cross Validation Coefficient of Regression (rCV), and Relative Performance (RPD) as shown in Table 1. As can be seen from Table 1, the RPD values of the models are all not less than 2.30, which indicates that the models can be used for nondestructive rapid quantitative analysis of the quality of total plant alkaloid, total sugar, reducing sugar, total nitrogen, nitrate, total potassium and chlorine in tobacco.
Example 5 precision fractionation of the same tobacco
The on-line real-time tobacco monitoring device provided by the invention realizes the function of 'same type tobacco precise grading'. And (3) realizing the accurate classification of the similar tobaccos by adopting a Spectrum Matching Value (SMV) calculation method. The method comprises the steps of respectively carrying out 10-time parallel acquisition on spectral data on 11 samples of the same grade of tobacco samples by using the tobacco online real-time monitoring device, calculating average data on the 10-time parallel spectral data of the same grade of tobacco samples, carrying out Standard Normal Variate (SNV) and first-order 5-point derivative data preprocessing on the average data in sequence, and calculating a spectral matching value by taking an average spectrum subjected to data preprocessing as a reference, wherein the average spectrum is shown in a table 2.
The discrimination Threshold (Threshold) range is set to 0.99977 to 0.99985, preferably 0.99979 to 0.99982, and more preferably 0.99980. As shown in table 2, based on the column, the spectrum matching values of other tobacco samples are all smaller than the threshold except that the diagonal SMV value is 1.0000, i.e., the matching value of the same-class tobacco sample and the matching value of the same-class tobacco sample are 1.0000, which indicates that the other tobacco samples are different, i.e., the other similar tobacco samples are identified as the same-class tobacco samples but different-class tobacco samples, thereby realizing the accurate classification of the same-class tobacco.
Although the invention has been described in detail hereinabove with respect to a general description and specific embodiments thereof, it will be apparent to those skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (8)
1. The utility model provides a tobacco on-line real-time supervision device which characterized in that: including control system (2), transmission system and optical path system (4), control system (2) be connected with transmission system, optical path system (4) electricity, transmission system is used for stabilizing the conveying to the tobacco sample, contain conveyer belt (1) and step motor, optical path system (4) are used for the collection to tobacco sample spectral data, contain annular cover (5), the halogen tungsten lamp, convex lens (7), near-infrared spectrometer (2) and sweep the part, conveyer belt (1) top sets up optical path system (4), the top of optical path system (4) sets up control system (2), be provided with computer system and touch-sensitive screen (3) in control system (2).
2. The tobacco online real-time monitoring device according to claim 1, characterized in that: the upper portion of conveyer belt (1) set up annular cover (5), the side of annular cover (5) is provided with halogen tungsten lamp and purge hole, the top surface of annular cover (5) sets up convex lens (7), the top of convex lens (7) sets up near-infrared spectrometer (8).
3. The tobacco online real-time monitoring device according to claim 2, characterized in that: the near-infrared spectrometer (8) is positioned at the focus above the convex lens (7), and four blowing holes are respectively positioned at the upper side and the lower side of the halogen tungsten lamp.
4. The application of the tobacco online real-time monitoring device to realize accurate recognition of foreign matters in tobacco according to claim 1, is characterized in that: the tobacco online real-time monitoring device collects tobacco and spectral data of common foreign matters in tobacco production, the common foreign matters comprise withered branches, sponges, hay, plant leaves, paper, hair and broken stones, the tobacco, the withered branches, the sponges, the hay, the plant leaves, the paper, the hair and the broken stones are assigned reference values of 0, 1, 2, 3, 4, 5, 6 and 7 respectively by taking the tobacco and the spectral data of the various foreign matters as independent variables, and a tobacco foreign matter accurate identification partial least square-discriminant analysis model is established by adopting partial least square regression and a full-interactive verification algorithm;
factor number of established tobacco foreign body accurate identification partial least square-discriminant analysis modelN f = 6, correction data prediction value-reference value regression equation y = 0.9987x + 0.0051, correction correlation coefficientr C = 0.9993, correction of measurement coefficientR 2 C = 0.9987, correct root mean square errorRMSEC= 0.08, cross validation data prediction value-reference value regression equationy = 0.9983x + 0.0072, interactive proof correlation coefficientr CV = 0.9992, interactive proof determination coefficientR 2 CV = 0.9985, cross validation root mean square errorRMSECV= 0.09, and the accuracy rates for tobacco, dead branches, sponges, hay, plant leaves, paper, hair, and crushed stones were 100%, and 100%, respectively, using 0.7, 1.5, 2.5, 3.5, 4.5, 5.5, and 6.5 as thresholds, respectively.
5. The application of the tobacco online real-time monitoring device to the judgment of the height fluctuation of tobacco materials according to claim 1 is characterized in that: the tobacco online real-time monitoring device collects tobacco material spectral data which are 0mm, 5.6 mm, 11.2 mm, 21.5 mm, 32.6 mm, 43.7 mm and 53.9 mm higher than the tobacco material reference height, and establishes a tobacco material height fluctuation judgment partial least square-discriminant analysis model by adopting partial least square regression and a full-interactive verification algorithm by taking the tobacco material spectral data with different heights as independent variables and the tobacco material reference height data as dependent variables;
factor number of partial least square-discriminant analysis model for judging height fluctuation of tobacco materialN f = 2, correction data prediction value-reference value regression equationy = 0.9982x+ 0.0428, correctionCorrelation coefficientr C = 0.9991, correction of measurement coefficientR 2 C = 0.9982, correct root mean square errorRMSEC= 0.79, regression equation of predicted value-reference value of cross validation datay = 0.9983x+ 0.0392, cross-validation correlation coefficientr CV = 0.9991, interactive proof determination coefficientR 2 CV = 0.9982, cross validation root mean square errorRMSECVAnd = 0.80, which are respectively 100%, 45% of the accuracy of the fluctuation in height of the tobacco material, which is higher than the reference height of the tobacco material by 0mm, 5.6 mm, 11.2 mm, 21.5 mm, 32.6 mm, 43.7 mm, 53.9 mm, using 5, 8, 15, 25, 36, 45 as a threshold value.
6. The application of the tobacco online real-time monitoring device to realize nondestructive rapid quantitative analysis of tobacco quality according to claim 1, characterized in that: the tobacco online real-time monitoring device collects tobacco material online spectrum data and synchronously samples the tobacco material online, reference values of total plant alkaloid, total sugar, reducing sugar, total nitrogen, nitrate, total potassium and chlorine in a tobacco sample are respectively obtained by the same method, the tobacco material online spectrum data are used as independent variables, the reference values of total plant alkaloid, total sugar, reducing sugar, total nitrogen, nitrate, total potassium and chlorine in the tobacco sample sampled online synchronously are used as dependent variables, and a partial least squares regression and full-interactive verification algorithm are combined to establish a tobacco quality lossless rapid quantitative analysis model;
factor number of the total plant alkaloid quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.8885x+ 0.1146, correction of correlation coefficientsr C = 0.9426, correction of measurement coefficientR 2 C = 0.8885, correct root mean square errorRMSEC= 0.028; cross validation data prediction value-reference value regression equationy = 0.8768x+ 0.1266, interactive proof correlation coefficientr CV = 0.9306, interactive proof determination coefficientR 2 CV = 0.8668, interactive authentication allRoot error of squareRMSECV= 0.031; model (model)RPD = 2.71；
Factor number of total sugar quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.7483x+ 2.4445, correction of correlation coefficientsr C = 0.8650, correction of measurement coefficientR 2 C = 0.7483, correct root mean square errorRMSEC = 0.20; cross validation data prediction value-reference value regression equationy = 0.7229x+ 2.6916, interactive proof correlation coefficientr CV = 0.8399, interactive proof determination coefficientR 2 CV = 0.7070, cross validation root mean square errorRMSECV = 0.22; model (model)RPD = 2.37；
Factor number of established reducing sugar quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.7874x+ 1.7250, correction of correlation coefficientsr C = 0.8874, correction of measurement coefficientR 2 C = 0.7874, correct root mean square errorRMSEC = 0.15; cross validation data prediction value-reference value regression equationy = 0.7633x+ 1.9204, interactive proof correlation coefficientr CV = 0.8650, interactive proof determination coefficientR 2 CV = 0.7497, cross validation root mean square errorRMSECV= 0.16; model (model)RPD = 2.72；
Factor number of total nitrogen quantitative analysis modelN f = 10; correction data prediction value-reference value regression equationy = 0.8374x+ 0.3002, correction of correlation coefficientsr C = 0.9151, correction of measurement coefficientR 2 C = 0.8374, correct root mean square errorRMSEC= 0.042; cross validation data prediction value-reference value regression equationy = 0.8141x+ 0.3436, cross validation correlation coefficientr CV = 0.8962, interactive proof determination coefficientR 2 CV = 0.8046, cross validation root mean square errorRMSECV = 0.046(ii) a Model (model)RPD = 2.37；
Factor number of established nitrate quantitative analysis modelN f = 7; correction data prediction value-reference value regression equationy = 0.7118x+ 0.0961, correction of correlation coefficientr C = 0.8437, correction of measurement coefficientR 2 C = 0.7118, correct root mean square errorRMSEC= 0.012; cross validation data prediction value-reference value regression equationy = 0.6964x+ 0.1012, interactive proof correlation coefficientr CV = 0.8280, interactive proof determination coefficientR 2 CV = 0.6876, cross validation root mean square errorRMSECV = 0.012; model (model)RPD = 2.28；
Factor number of total potassium quantitative analysis modelN f = 13; correction data prediction value-reference value regression equationy = 0.8106x+ 0.4173, correction of correlation coefficientsr C = 0.9003, correction of measurement coefficientR 2 C = 0.8106, correct root mean square errorRMSEC = 0.038; cross validation data prediction value-reference value regression equationy = 0.7774x+ 0.4903, interactive proof correlation coefficientr CV = 0.8699, interactive proof determination coefficientR 2 CV = 0.7599, cross validation root mean square errorRMSECV= 0.044; model (model)RPD = 2.93；
Number of factors of the established chlorine quantitative analysis modelN f = 15; correction data prediction value-reference value regression equationy = 0.8683x+ 0.0832, correction of correlation coefficientsr C = 0.9318, correction of measurement coefficientR 2 C = 0.8683, correct root mean square errorRMSEC= 0.015; cross validation data prediction value-reference value regression equationy = 0.8367x+ 0.1032, interactive proof correlation coefficientr CV = 0.8996, interactive proof determination coefficientR 2 CV = 0.8096, cross validation root mean square errorRMSECV= 0.018; model (model)RPD = 3.05。
7. The application of the online real-time tobacco monitoring device according to claim 1 to the realization of the accurate classification function of similar tobaccos is characterized in that: the method comprises the steps of realizing accurate classification of the same type of tobacco by adopting a spectrum matching value calculating method, parallelly collecting spectrum data for 10 times by using the tobacco online real-time monitoring device for 11 samples in different levels of the same type of tobacco samples, calculating average data of the parallel spectrum data for 10 times of the same type of tobacco samples in the same level, sequentially performing Standard Normal Variate (SNV) and first-order 5-point derivative data preprocessing on the average data, and calculating a spectrum matching value by using an average spectrum subjected to data preprocessing as a reference; the range of a discrimination Threshold (Threshold) is set to be 0.99977-0.99985, preferably 0.99979-0.99982, further preferably 0.99980, and on the basis of taking the column as a reference, the spectral matching values of other tobacco samples are smaller than the Threshold except that the diagonal SMV value is 1.0000, namely the tobacco sample of the same class has a self matching value of 1.0000, so that the other samples have differences, namely the tobacco samples of the same class are identified as the tobacco samples of the same class but different classes, and accurate classification of the tobacco of the same class is realized.
8. Use of any one of claims 4 to 7 for the non-destructive rapid quality control of tobacco quality.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111097714.5A CN113804648A (en) | 2021-09-18 | 2021-09-18 | Tobacco online real-time monitoring device and application thereof in tobacco quality nondestructive rapid quality control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111097714.5A CN113804648A (en) | 2021-09-18 | 2021-09-18 | Tobacco online real-time monitoring device and application thereof in tobacco quality nondestructive rapid quality control |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113804648A true CN113804648A (en) | 2021-12-17 |
Family
ID=78939527
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111097714.5A Pending CN113804648A (en) | 2021-09-18 | 2021-09-18 | Tobacco online real-time monitoring device and application thereof in tobacco quality nondestructive rapid quality control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113804648A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4971077A (en) * | 1989-08-02 | 1990-11-20 | R. J. Reynolds Tobacco Company | On-line tobacco evaluation system and method |
| CN101933650A (en) * | 2009-06-30 | 2011-01-05 | 福建中烟工业公司 | Near infrared spectroscopy introduced tobacco foreign material removing device and removing method thereof |
| CN102680427A (en) * | 2012-05-27 | 2012-09-19 | 红云红河烟草(集团)有限责任公司 | Method for identifying surface fragrance quality of cigarettes by near infrared spectral analysis technology |
| CN103090802A (en) * | 2013-02-02 | 2013-05-08 | 中国农业科学院烟草研究所 | Method for detecting thickness of tobacco leaf |
| CN106053383A (en) * | 2016-06-27 | 2016-10-26 | 四川威斯派克科技有限公司 | Near-infrared online detection method for tobacco processing process |
| CN113003253A (en) * | 2021-02-23 | 2021-06-22 | 中国烟草总公司郑州烟草研究院 | Tobacco material storage and distribution uniformity detection control equipment and method based on laser scanning |
| CN113092409A (en) * | 2021-05-14 | 2021-07-09 | 深圳市智程机电技术有限公司 | Tobacco online near infrared spectrum detection device and method based on linear optical filter |
-
2021
- 2021-09-18 CN CN202111097714.5A patent/CN113804648A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4971077A (en) * | 1989-08-02 | 1990-11-20 | R. J. Reynolds Tobacco Company | On-line tobacco evaluation system and method |
| CN101933650A (en) * | 2009-06-30 | 2011-01-05 | 福建中烟工业公司 | Near infrared spectroscopy introduced tobacco foreign material removing device and removing method thereof |
| CN102680427A (en) * | 2012-05-27 | 2012-09-19 | 红云红河烟草(集团)有限责任公司 | Method for identifying surface fragrance quality of cigarettes by near infrared spectral analysis technology |
| CN103090802A (en) * | 2013-02-02 | 2013-05-08 | 中国农业科学院烟草研究所 | Method for detecting thickness of tobacco leaf |
| CN106053383A (en) * | 2016-06-27 | 2016-10-26 | 四川威斯派克科技有限公司 | Near-infrared online detection method for tobacco processing process |
| CN113003253A (en) * | 2021-02-23 | 2021-06-22 | 中国烟草总公司郑州烟草研究院 | Tobacco material storage and distribution uniformity detection control equipment and method based on laser scanning |
| CN113092409A (en) * | 2021-05-14 | 2021-07-09 | 深圳市智程机电技术有限公司 | Tobacco online near infrared spectrum detection device and method based on linear optical filter |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106198448A (en) | A kind of automatic high speed lossless sorting live body male and female Pupa bombycis or the technique of live body male and female Bombyx bombycis | |
| CN113533236B (en) | Fruit spectrum feature extraction method, model training method and detection method | |
| CN110542658A (en) | tobacco non-smoke substance classification method based on hyperspectral imaging technology | |
| CN111257277B (en) | Tobacco leaf similarity judgment method based on near infrared spectrum technology | |
| CN104568822B (en) | A kind of capsule of weeping forsythia medicinal material multi objective while quick determination method | |
| CN106053383A (en) | Near-infrared online detection method for tobacco processing process | |
| CN110132880B (en) | Tobacco leaf overall sensory quality evaluation method based on near infrared spectrum | |
| CN105044022A (en) | Method for rapidly nondestructively detecting wheat hardness based on near infrared spectrum technology and application | |
| CN111537469A (en) | Apple quality rapid nondestructive testing method based on near-infrared technology | |
| CN105606552A (en) | Method for analyzing quality trends of cigarette tobaccos based on whole spectral range molecular spectrum | |
| CN108318442A (en) | A kind of detection method suitable for Chinese medicine suspending system | |
| CN108548794A (en) | A kind of biological products method for transferring near infrared model | |
| CN113804648A (en) | Tobacco online real-time monitoring device and application thereof in tobacco quality nondestructive rapid quality control | |
| CN114088661A (en) | Online prediction method for chemical components in tobacco leaf curing process based on transfer learning and near infrared spectrum | |
| CN113655027A (en) | Method for rapidly detecting tannin content in plant by near infrared | |
| CN109283153A (en) | A kind of method for building up of soy sauce Quantitative Analysis Model | |
| CN110907369B (en) | Wuyi rock tea production place identification method fusing different detection method characteristic variables | |
| Masithoh et al. | NIR reflectance spectroscopy and SIMCA for classification of crops flour | |
| CN115824996A (en) | Tobacco conventional chemical component general model modeling method based on near infrared spectrum | |
| Wang et al. | Fast discrimination of bamboo species using VIS/NIR spectroscopy | |
| CN109540837A (en) | The method that near-infrared quickly detects Boehmeria nivea leaves wood fibre cellulose content | |
| CN112782116A (en) | Method for detecting moisture content of large traditional Chinese medicine honeyed pill by utilizing near infrared spectrum and application | |
| CN113916824A (en) | Alpha in milks1Intermediate infrared rapid batch detection method of casein | |
| Andre et al. | On-line monitoring of the buffer capacity of particleboard furnish by near-infrared spectroscopy | |
| CN109406420B (en) | Method for predicting content of scopoletin in fructus lycii based on hyperspectral imaging technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |