CN118134276A - Method for judging tobacco leaf finished product distribution storage location and application thereof - Google Patents
Method for judging tobacco leaf finished product distribution storage location and application thereof Download PDFInfo
- Publication number
- CN118134276A CN118134276A CN202410120735.1A CN202410120735A CN118134276A CN 118134276 A CN118134276 A CN 118134276A CN 202410120735 A CN202410120735 A CN 202410120735A CN 118134276 A CN118134276 A CN 118134276A
- Authority
- CN
- China
- Prior art keywords
- tobacco
- tobacco leaves
- determining
- finished
- distribution
- 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 201
- 235000002637 Nicotiana tabacum Nutrition 0.000 title claims abstract description 201
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000009826 distribution Methods 0.000 title claims abstract description 30
- 238000003860 storage Methods 0.000 title claims abstract description 24
- 230000001953 sensory effect Effects 0.000 claims abstract description 70
- 235000019504 cigarettes Nutrition 0.000 claims abstract description 55
- 238000004519 manufacturing process Methods 0.000 claims abstract description 12
- 238000004458 analytical method Methods 0.000 claims abstract description 10
- 238000011156 evaluation Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 claims description 20
- 239000000126 substance Substances 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 13
- 206010013911 Dysgeusia Diseases 0.000 claims description 10
- 239000007789 gas Substances 0.000 claims description 10
- 230000007794 irritation Effects 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 10
- 239000003205 fragrance Substances 0.000 claims description 7
- 239000002994 raw material Substances 0.000 claims description 7
- 150000001875 compounds Chemical class 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 6
- AFPLNGZPBSKHHQ-UHFFFAOYSA-N Betulaprenol 9 Natural products CC(C)=CCCC(C)=CCCC(C)=CCCC(C)=CCCC(C)=CCCC(C)=CCCC(C)=CCCC(C)=CCCC(C)=CCO AFPLNGZPBSKHHQ-UHFFFAOYSA-N 0.000 claims description 4
- NIDGCIPAMWNKOA-WOJBJXKFSA-N Neophytadiene Natural products [C@H](CCC[C@@H](CCCC(C)C)C)(CCCC(C=C)=C)C NIDGCIPAMWNKOA-WOJBJXKFSA-N 0.000 claims description 4
- 150000001450 anions Chemical class 0.000 claims description 4
- 150000001768 cations Chemical class 0.000 claims description 4
- 239000002027 dichloromethane extract Substances 0.000 claims description 4
- 235000014113 dietary fatty acids Nutrition 0.000 claims description 4
- 239000000194 fatty acid Substances 0.000 claims description 4
- 229930195729 fatty acid Natural products 0.000 claims description 4
- 150000004665 fatty acids Chemical class 0.000 claims description 4
- 239000004615 ingredient Substances 0.000 claims description 4
- NIDGCIPAMWNKOA-UHFFFAOYSA-N neophytadiene Chemical compound CC(C)CCCC(C)CCCC(C)CCCC(=C)C=C NIDGCIPAMWNKOA-UHFFFAOYSA-N 0.000 claims description 4
- 150000008442 polyphenolic compounds Chemical class 0.000 claims description 4
- 235000013824 polyphenols Nutrition 0.000 claims description 4
- AFPLNGZPBSKHHQ-MEGGAXOGSA-N solanesol Chemical compound CC(C)=CCC\C(C)=C\CC\C(C)=C\CC\C(C)=C\CC\C(C)=C\CC\C(C)=C\CC\C(C)=C\CC\C(C)=C\CC\C(C)=C\CO AFPLNGZPBSKHHQ-MEGGAXOGSA-N 0.000 claims description 4
- 235000019505 tobacco product Nutrition 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000002290 gas chromatography-mass spectrometry Methods 0.000 claims description 3
- 238000004949 mass spectrometry Methods 0.000 claims description 3
- 229930013930 alkaloid Natural products 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims description 2
- 150000002772 monosaccharides Chemical class 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 102000004169 proteins and genes Human genes 0.000 claims description 2
- 108090000623 proteins and genes Proteins 0.000 claims description 2
- 238000004445 quantitative analysis Methods 0.000 claims description 2
- 238000004497 NIR spectroscopy Methods 0.000 claims 1
- 235000013619 trace mineral Nutrition 0.000 claims 1
- 239000011573 trace mineral Substances 0.000 claims 1
- 238000013441 quality evaluation Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 239000000047 product Substances 0.000 description 31
- 238000012417 linear regression Methods 0.000 description 9
- YMWUJEATGCHHMB-UHFFFAOYSA-N Dichloromethane Chemical compound ClCCl YMWUJEATGCHHMB-UHFFFAOYSA-N 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000002347 injection Methods 0.000 description 6
- 239000007924 injection Substances 0.000 description 6
- 238000001228 spectrum Methods 0.000 description 5
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 4
- 238000010998 test method Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 150000002500 ions Chemical class 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 239000000843 powder Substances 0.000 description 3
- 239000010453 quartz Substances 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 3
- 239000012064 sodium phosphate buffer Substances 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- CSNNHWWHGAXBCP-UHFFFAOYSA-L Magnesium sulfate Chemical compound [Mg+2].[O-][S+2]([O-])([O-])[O-] CSNNHWWHGAXBCP-UHFFFAOYSA-L 0.000 description 2
- 150000008062 acetophenones Chemical class 0.000 description 2
- 238000009835 boiling Methods 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000002329 infrared spectrum Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000012528 membrane Substances 0.000 description 2
- 239000012074 organic phase Substances 0.000 description 2
- 238000007789 sealing Methods 0.000 description 2
- 238000007873 sieving Methods 0.000 description 2
- 238000002791 soaking Methods 0.000 description 2
- 239000011780 sodium chloride Substances 0.000 description 2
- 239000012086 standard solution Substances 0.000 description 2
- 239000006228 supernatant Substances 0.000 description 2
- 238000005119 centrifugation Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000227 grinding Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Landscapes
- Manufacture Of Tobacco Products (AREA)
Abstract
The invention discloses a method for judging a tobacco finished product distribution storage place and application thereof. According to the method, qualitative, quantitative and positioning analysis is carried out on the tobacco finished products, so that sensory quality evaluation prediction models of tobacco leaves applicable to different types of cigarettes such as fine cigarettes, medium cigarettes and conventional cigarettes are established, and the distribution flow direction of the tobacco finished products can be rapidly and accurately judged by combining the actual conditions of production equipment of different cigarette factories of tobacco industry enterprises. The method is used for determining the distribution flow direction location of the tobacco leaf finished product, is beneficial to improving the transportation and distribution energy efficiency, improving the transportation guarantee capability, and comprehensively planning the transportation capability and the storage configuration in advance.
Description
Technical Field
The invention belongs to the technical field of tobacco industry, and relates to a method for judging tobacco finished product distribution and storage places and application thereof.
Background
Tobacco industry companies purchase and transfer tobacco leaf (raw tobacco) raw materials each year, redrying the raw materials to obtain tobacco leaf (tobacco flakes) finished products, and transporting the tobacco leaf (raw tobacco flakes) finished products to a tobacco leaf warehouse of the industry companies for alcoholization for later use. Two or more cigarette factories are generally managed by an industrial company, and the factories are generally located in different cities, so that tobacco finished products after redrying are required to be distributed to different cities to enter a warehouse for warehouse alcoholization, and after alcoholization, the tobacco finished products begin to be produced and used in the cigarette factories. The equipment for producing cigarettes in different cigarette factories is different, the produced cigarette brands are different in specification, and the requirements on tobacco raw materials are different.
Currently, technicians perform sensory quality evaluation on tobacco finished products, or determine the flow direction of the tobacco finished products to a distribution place according to traffic factor limitation of a tobacco finished product delivery place, the sensory evaluation is limited by limitation of sample representative errors in a sampling link and instability limitation of subjective feelings of suction evaluation staff, and certain inaccuracy exists in the distribution mode. In the production link of finished cigarettes in cigarette factories, a large number of tobacco leaf finished products are required to be transported from a warehouse of one cigarette factory to a cigarette factory in another city for producing cigarettes, and a large number of manpower, material resources and financial resources are wasted, so that a simpler, more convenient and accurate method for judging the allocation and storage places of the tobacco leaf finished products is urgently needed to meet the application requirements.
Disclosure of Invention
Aiming at the defects and actual demands of the prior art, the invention provides a method for judging the distribution and storage sites of tobacco leaf finished products and application thereof.
In order to achieve the aim of the invention, based on consideration of the intrinsic full components (major components, semi-minor components and minor components) of tobacco leaves, a support vector machine regression (SVR) method is adopted to construct a linear correlation model between independent variables (full components) and dependent variables (sensory quality) in a high-dimensional data feature space, so as to predict the suitability of the tobacco leaves and determine proper distribution storage places. Specifically, the following technical scheme is adopted:
In a first aspect, the present invention provides a method for determining a location for dispensing and warehousing finished tobacco products, the method comprising:
(1) Sample selection of modeled tobacco: collecting single tobacco leaves, dividing the single tobacco leaves into a training set and a testing set, and classifying the applicability of the single tobacco leaves;
(2) Sensory evaluation: determining a sensory evaluation method according to industry standards, and performing sensory evaluation on the tobacco leaf samples classified in the step (1) respectively;
(3) And (3) measuring chemical components of tobacco leaves: respectively carrying out quantitative analysis on major components, semi-minor components and minor components of the tobacco leaves classified in the step (1);
(4) And (3) constructing a quality prediction model: carrying out Support Vector Regression (SVR) analysis by taking the training set tobacco leaf samples under different classifications as objects, taking the chemical components measured in the step (3) as independent variables and taking the sensory evaluation indexes in the step (2) as dependent variables, and calculating to form sensory quality prediction models of the tobacco leaves under different classifications;
(5) Determining the applicability category: carrying out chemical component measurement on the finished tobacco flakes to be measured according to the method of the step (3), identifying by adopting the model established in the step (4), and determining the applicability category of the finished tobacco flakes;
(6) Determining a tobacco leaf finished product distribution and storage place: and (3) determining the distribution flow direction of the finished tobacco flakes to be tested to a storage place according to the applicability category of the tobacco flakes determined in the step (5).
The method of the invention carries out qualitative, quantitative and positioning analysis on the tobacco finished product, thereby establishing a sensory quality evaluation prediction model suitable for different types of tobacco such as fine cigarettes, medium cigarettes and conventional cigarettes, and combining the actual conditions of production equipment of different cigarette factories of tobacco industry enterprises, the distribution flow direction of the tobacco finished product can be more rapidly and accurately judged. The method is used for determining the distribution flow direction location of the tobacco leaf finished product, is beneficial to improving the transportation and distribution energy efficiency, improving the transportation guarantee capability, and comprehensively planning the transportation capability and the storage configuration in advance.
Preferably, the tobacco leaves of step (1) comprise eight-spice type producing area tobacco leaves;
Preferably, the tobacco leaves comprise any one or a combination of at least two of upper tobacco leaves, middle tobacco leaves or lower tobacco leaves;
Preferably, the tobacco leaves comprise any one or a combination of at least two of superior tobacco leaves, medium tobacco leaves or inferior tobacco leaves.
Preferably, the classification of the applicability in the step (1) includes a fine cigarette applicable class, a medium cigarette applicable class or a conventional cigarette applicable class.
Preferably, the sensory evaluation in step (2) includes any one or a combination of at least two of aroma quality, aroma amount, miscellaneous gas, irritation or aftertaste.
Preferably, the sensory evaluation in step (2) includes the indexes of aroma quality, aroma quantity, miscellaneous gas, irritation and aftertaste.
Preferably, the score range of the sensory evaluation index is: ① fragrance quality: good, better 7.6-9.0, middle upper 6.1-7.5, middle 4.6-6.0, middle lower 3.1-4.5, worse, difference less than or equal to 3; ② fragrance amount: foot 7.6-9.0, foot 6.1-7.5, 4.6-6.0, less 3.1-4.5, less than or equal to 3; ③ miscellaneous gases: light weight 7.6-9.0, light weight 6.1-7.5, 4.6-6.0, heavy weight 3.1-4.5 and weight less than or equal to 3; ④ irritation: small 7.6-9.0, small 6.1-7.5, 4.6-6.0, large 3.1-4.5 and large less than or equal to 3; ⑤ aftertaste: comfortable 7.6-9.0, more comfortable 6.1-7.5, shang Shi 4.6-6.0, underfit 3.1-4.5 and tongue stagnation less than or equal to 3.
Preferably, the tobacco leaf chemical component determination of step (3) comprises major, semi-minor and minor components;
preferably, the macroingredients comprise any one or a combination of at least two of monosaccharides, proteins, alkaloids, anions and cations, polyphenols, polyacids and higher fatty acids, amadori compounds, pH, dichloromethane extract, solanesol or neophytadiene;
Preferably, the semi-minor ingredients include any one or a combination of at least two of anions and cations, polyphenols, polyacids and higher fatty acids, amadori compounds, pH, dichloromethane extract, solanesol or neophytadiene;
preferably, the micro-ingredients comprise volatile and/or semi-volatile substances.
Preferably, the major component or the semi-minor component is detected by a near infrared spectrum test method;
Preferably, the near infrared spectrum test method comprises the following steps: ① Before crushing the tobacco leaf sample, naturally airing until the moisture is between 6 and 8 percent, crushing the sample by adopting cyclone powder, sieving the crushed sample by a sieve with the size not more than 0.250mm, sealing and storing the sample, and refrigerating and storing the sample at the temperature of between 0 and 4 ℃; ② Sample testing is carried out by adopting a near infrared spectrometer, and the instrument conditions are as follows: scanning range: 4000cm -1～10000cm-1; resolution ratio: 8cm -1; the spectrum scanning times are not lower than 64 times; the light spot of the integrating sphere should fall in the range of 1/2 to 2/3 from the center point of the sampling cup, all samples should be collected 2 times, and 2 spectra should pass one-time inspection, and the spectrum similarity should be greater than 0.9999.
Preferably, the trace component is detected by a chromatographic mass spectrometry combined test method, wherein the detection comprises sample treatment and GC/MS analysis;
Preferably, the chromatography mass spectrometry combined test method comprises the following steps: (1) sample pretreatment: weighing 1g of tobacco leaf powder sample, adding 7-10 mL of sodium phosphate buffer solution, wherein the pH value of the sodium phosphate buffer solution is 3-3.5, soaking for 20-25 min, adding 50 mu L of 120 mu g/mL of deuterated acetophenone internal standard solution, swirling for 20min at the speed of 2000r/min, cooling for 30min in a refrigerator at-18 ℃, adding 1g of sodium chloride and 4g of anhydrous magnesium sulfate, rapidly and severely shaking, adding 5mL of dichloromethane, swirling for 20min at the speed of 2000r/min, centrifuging for 3min at the speed of 8000r/min, removing supernatant, filtering by an organic phase filter membrane, and waiting for machine-loading test;
(2) Instrument analysis conditions: ① For nonpolar or high-boiling compounds, use is made of: chromatographic column: DB-5MS elastic quartz capillary column 60m x 0.25mm x 0.25 μm; sample inlet temperature: 290 ℃; sample injection mode: not split; sample injection amount: 1 μl; programming temperature: the initial temperature is 40 ℃, kept for 3min, then 5 ℃/min is increased to 75 ℃,1 ℃/min is increased to 120 ℃,2 ℃/min is increased to 160 ℃,5 ℃/min is increased to 290 ℃, and the temperature is kept for 10min; transmission line temperature: 280 ℃; ionization mode: EI; ionization energy: 70eV; ion source temperature: 280 ℃; monitoring mode: dMRM;
② For polar or very low boiling compounds, a chromatographic column is used: DB-624 elastic quartz capillary column 60m x 0.25mm x 1.4 μm; sample inlet temperature: 235 ℃; sample injection mode: not split; sample injection amount: 1 μl; programming temperature: the initial temperature is 40 ℃ for 5min, then 2 ℃/min is increased to 160 ℃ for 1min, and 5 ℃/min is increased to 235 ℃ for 20min; transmission line temperature: 230 ℃; ionization mode: EI; ionization energy: 70eV; ion source temperature: 280 ℃; monitoring mode: dMRM.
Preferably, the specific steps of constructing the quality prediction model in the step (4) are as follows: carrying out Support Vector Regression (SVR) analysis by taking training set tobacco leaf raw materials with 3 conditions of suitable for the fine branch, suitable for the middle branch and suitable for the conventional use under different positioning as objects, taking chemical components measured in the step (3) as independent variables and taking sensory evaluation indexes in the step (2) as dependent variables, and respectively calculating to form sensory quality prediction models of the fine branch, the middle branch and the conventional type tobacco leaves;
preferably, the quality prediction model in the step (4) is further identified and trained after being established;
Preferably, the method for identifying and training comprises the following steps: taking the tobacco leaves of the training set as objects, training by adopting the model established in the step (4), and judging the accuracy of the model; and (3) taking the test set sample as an object, and adopting the model established in the step (4) to identify, so that the prediction capability of the test set sample is improved.
Preferably, the determining of the tobacco leaf finished product distribution and storage location in the step (6) specifically includes: and (3) determining the distribution flow direction of the finished tobacco pieces to be tested to a storage place according to the applicability category of the tobacco finished product determined in the step (5) and combining the production equipment conditions of different cigarette factories of tobacco industry enterprises, and integrating equipment factors and the classification result of the tobacco finished product.
In a second aspect the present invention provides the use of a method of determining a location for dispensing and warehousing finished tobacco leaves as described in the first aspect in dispensing and storing tobacco leaves.
Compared with the prior art, the invention has the following beneficial effects:
(1) The qualitative, quantitative and positioning analysis of the tobacco finished products is carried out, so that the maximum use value of the finished tobacco flakes can be brought into play;
(2) The tobacco sensory quality prediction model constructed based on Support Vector Regression (SVR) is adopted, so that the method is robust and reliable, and has good prediction precision;
(3) The tobacco sensory quality prediction problem based on machine learning provides a quantification method, and is the supplement and improvement of the traditional method;
(4) The method for judging the distribution and storage places of the tobacco leaf finished products has higher efficiency and accuracy.
Drawings
FIGS. 1a-1f are linear regression diagrams of sensory predictor predictors and actual values established by fine-count tobacco leaves according to embodiments of the present invention;
FIGS. 2a-2f are linear regression diagrams of sensory predictor predictors and actual values established by tobacco leaves in an embodiment of the invention;
fig. 3a-3f are linear regression diagrams of sensory predictor predictors and actual values established by conventional tobacco leaves according to an embodiment of the present invention.
Detailed Description
The technical means adopted by the invention and the effects thereof are further described below with reference to the examples and the attached drawings. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof.
The specific techniques or conditions are not identified in the examples and are described in the literature in this field or are carried out in accordance with the product specifications. The reagents or apparatus used were conventional products commercially available through regular channels, with no manufacturer noted.
Example 1
The embodiment provides a method for judging a tobacco finished product distribution storage place, which comprises the following steps:
Taking a tobacco leaf finished product C3F module obtained after threshing and redrying processing of raw tobacco purchased from Henan province by a certain cigarette industry enterprise as an object, wherein the cigarette industry enterprise manages 3 cigarette factories which are respectively positioned in an A city, a B city and a C city. Wherein all the production equipment of the urban cigarette factory A is the equipment for producing the fine cigarettes, all the production equipment of the urban cigarette factory B is the equipment for producing the conventional cigarettes, and all the production equipment of the urban cigarette factory C is the equipment for producing the medium cigarettes.
(1) Sample selection of modeled tobacco: extracting production history data of different specifications of tobacco group formulas of cigarettes of fine cigarettes, medium cigarettes and conventional cigarettes produced by tobacco industry enterprises, and selecting a certain number of tobacco grade modules from tobacco grade modules used in the fine cigarette group formulas as representative tobacco samples suitable for the fine cigarettes; selecting a certain number of tobacco grade modules from tobacco grade modules used in the middle-branch type tobacco group formula as representative tobacco samples suitable for the middle-branch type cigarettes; selecting a certain number of tobacco grade modules from tobacco grade modules used in a conventional tobacco grade group formula as representative tobacco samples suitable for conventional cigarettes; the selected fine-branch tobacco leaf samples, medium-branch tobacco leaf samples and conventional tobacco leaf samples all cover eight-spice type production areas and foreign importation, comprise upper tobacco leaves, middle tobacco leaves and lower tobacco leaves, comprise upper tobacco leaves, medium tobacco leaves or lower tobacco leaves, and divide the fine-branch tobacco leaf samples, medium-branch tobacco leaf samples and conventional tobacco leaf samples into training sets and test sets (the proportion is 8:2);
(2) Sensory quality evaluation: sensory evaluation is carried out on 3 tobacco samples of the conventional cigarettes, the medium cigarettes and the fine cigarettes respectively, and the sensory evaluation method comprises 5 indexes of aroma quality, aroma quantity, miscellaneous gases, irritation and aftertaste. The score ranges of the indexes are as follows: ① fragrance quality: good, better (7.6-9.0), middle and upper (6.1-7.5), middle (4.6-6.0), middle and lower (3.1-4.5), worse, and difference less than or equal to 3; ② fragrance amount: the feet are 7.6 to 9.0, the feet are 6.1 to 7.5, the feet are 4.6 to 6.0, the feet are less (3.1 to 4.5), and the quantity is less than or equal to 3; ③ miscellaneous gases: light (7.6-9.0), light (6.1-7.5), heavy (4.6-6.0), heavy (3.1-4.5) and weight less than or equal to 3; ④ irritation: small (7.6-9.0), small (6.1-7.5), large (3.1-4.5) and large (less than or equal to 3); ⑤ aftertaste: comfort (7.6-9.0), more comfort (6.1-7.5), shang Shi (4.6-6.0), underfit (3.1-4.5) and tongue stagnation less than or equal to 3;
(3) And (3) measuring chemical components of tobacco leaves: quantitatively analyzing major components, semi-trace components and trace components in conventional tobacco leaves, medium-branch tobacco leaves and fine-branch tobacco leaves respectively;
Macrocomponent, semimicroingredient experimental conditions: ① Before crushing the tobacco leaf sample, naturally airing until the moisture is between 6 and 8 percent, and crushing the sample by adopting cyclone grinding. Sieving with 60 mesh sieve, sealing and storing the sample, and refrigerating and storing at 0-4 deg.c; ② Sample testing is carried out by adopting a near infrared spectrometer, and the instrument conditions are as follows: scanning range (10 000cm < -1 > to 4 000cm -1); resolution (about 8cm -1); the spectrum scanning times should be not lower than 64 times; the light spot of the integrating sphere should fall within a range of 1/2 to 2/3 from the center point of the sampling cup, all samples should be collected 2 times, and 2 spectra should pass one-time inspection (spectral similarity should be greater than 0.9999).
Micro-component experimental conditions: ① Sample pretreatment: 1g of tobacco leaf powder sample is weighed, 10mL of sodium phosphate buffer solution is added, soaking is carried out for 20min, 200 mu L of 50 mu g/mL of deuterated acetophenone internal standard solution is added, vortex is carried out for 30min at the speed of 2000r/min, after cooling for 30min in a refrigerator at the temperature of minus 18 ℃, 1g of sodium chloride and 4g of anhydrous magnesium sulfate are added, shaking is carried out rapidly and vigorously, 5mL of dichloromethane is added, vortex is carried out for 30min at the speed of 2000r/min, and centrifugation is carried out for 3min at the speed of 8000 r/min. Removing supernatant, filtering with organic phase filter membrane, and testing;
② Sample testing was performed using GC-MS, instrument analysis conditions: chromatographic column: DB-5MS elastic quartz capillary column (60 m x 0.25mm x 0.25 μm); sample inlet temperature: 290 ℃; sample injection mode: not split; sample injection amount: 1 μl; programming temperature: the initial temperature is 40 ℃, kept for 3min, then 3 ℃/min is increased to 75 ℃,1 ℃/min is increased to 120 ℃,2 ℃/min is increased to 160 ℃,5 ℃/min is increased to 290 ℃, and the temperature is kept for 10min; transmission line temperature: 280 ℃; ionization mode: EI; ionization energy: 70eV; ion source temperature: 280 ℃; monitoring mode: SACN.
(4) And (3) constructing a quality prediction model: carrying out Support Vector Regression (SVR) analysis and operation on the fine-branch tobacco training set tobacco samples serving as objects, the chemical components of the fine-branch tobacco measured in the step (3) serving as independent variables, and the fine-branch tobacco smoking indexes obtained in the step (2) serving as dependent variables to form a fine-branch tobacco sensory quality prediction model;
the sensory quality prediction model of the medium-branch tobacco leaves and the sensory quality prediction model of the conventional tobacco leaves are obtained by the same method;
(5) Model identification and training: training by taking the fine-count tobacco leaf training set sample as an object and adopting the fine-count tobacco leaf sensory quality model established in the step (4), and judging accuracy; identifying by using the fine-branch tobacco leaf test set sample as an object and adopting the fine-branch tobacco leaf sensory quality model established in the step (4) to predict the applicability of the fine-branch tobacco leaf sensory quality model;
respectively identifying and training the sensory quality prediction model of the medium tobacco leaf and the sensory quality prediction model of the conventional tobacco leaf by the same method;
(6) Determining the applicability category: carrying out chemical component measurement on the finished cigarette flakes to be measured according to the method of the step (3), identifying by adopting the model established in the step (4), and determining the applicability category (suitable for fine count, suitable for middle count or suitable for routine) of the finished cigarette flakes to be measured;
(7) Determining a tobacco leaf finished product distribution and storage place: and (3) determining the distribution flow direction of the tobacco finished products to the storage location according to the applicability classification result of the step (6) and combining the actual conditions of production equipment of different cigarette factories of tobacco industry enterprises and comprehensive equipment factors and tobacco finished product classification results.
Results data
(1) Construction of applicability model
In the step (4), sensory quality prediction models are built for fine-branch tobacco leaves, medium-branch tobacco leaves and conventional tobacco leaves, wherein a sensory prediction index prediction value and actual measurement value linear regression diagram built for the fine-branch tobacco leaves is shown in fig. 1a-1f, and index prediction value and actual measurement value linear regression R 2 and residual statistics are shown in table 1.
TABLE 1
The results in table 1 show that the predicted value and the R 2 of the sensory index prediction model of the tobacco leaf suitable for the cigarettes are both above 0.99, the average value of the absolute value of single index residual errors is 0.01-0.03 minutes, the absolute value of the sensory total score residual errors is 0.25 minutes, the maximum value of the absolute value of the single index residual errors is 0.06-0.24 minutes, the maximum value of the sensory total score residual errors is 1.13 minutes, the sample residual errors of the single index basic 100% are within 0.25 minutes, and the sensory total score residual errors of 96% sample are within 1 minute, which indicates that the model has good prediction capability.
The sensory prediction index predicted value and actual measurement value linear regression diagram established by the medium-branch tobacco leaf is shown in fig. 2a-2f, and the index predicted value and actual measurement value linear regression R 2 and residual statistics are shown in table 2.
TABLE 2
As shown in Table 2, the predicted value and the R 2 of the sensory index prediction model of the middle-branch cigarette applicability tobacco are both above 0.99, the average value of the absolute value of single index residual errors is 0.02-0.07 min, the absolute value of sensory total score residual errors is 0.62 min, the maximum value of the absolute value of single index residual errors is 0.22-0.55 min, the maximum value of the sensory total score residual errors is 3.81 min, the sample residual errors of single index basic 100% are within 0.5 min, and the sensory total score residual errors of 80% sample are within 1 min, which indicates that the model has good prediction capability.
The sensory prediction index predicted value and actual measurement value linear regression diagram established by the conventional tobacco leaf is shown in fig. 3a-3f, and the index predicted value and actual measurement value linear regression R 2 and residual statistics are shown in table 3.
TABLE 3 Table 3
As shown in Table 3, R 2 of the predicted value and the measured value of the conventional tobacco sensory index prediction model for cigarette applicability are above 0.99, the average value of the absolute value of single index residual errors is 0.01-0.03 minutes, the absolute value of sensory total score residual errors is 0.24 minutes, the maximum value of the absolute value of single index residual errors is 0.06-0.25 minutes, the maximum value of sensory total score residual errors is 1.15 minutes, the residual errors of samples with single index of 100% are within 0.25 minutes, and the residual errors of sensory total score of samples with 99% are within 1 minute, which indicates that the model has good prediction capability.
(2) Model identification and training
The fine-branch tobacco leaf training set sample is taken as an object, training is carried out by adopting the fine-branch tobacco leaf sensory quality model established in the step (4), and accuracy is judged; taking a fine-branch tobacco leaf test set sample as an object, identifying by adopting the fine-branch tobacco leaf sensory quality model established in the step (4), and predicting the applicability of the fine-branch tobacco leaf sensory quality model
Respectively identifying and training the sensory quality prediction model of the medium tobacco leaf and the sensory quality prediction model of the conventional tobacco leaf by the same method; the results are shown in Table 4.
TABLE 4 Table 4
As can be seen from Table 4, the accuracy of the test set samples of the present invention is 100%, and the recognition rate is as high as 100%, which indicates that the accuracy of the model is higher.
(3) Model identification prediction
Taking 15 tobacco raw materials which are respectively suitable for the conventional tobacco, the medium branch tobacco and the fine branch tobacco as objects, completing the sensory quality evaluation of tobacco samples according to the step (2), completing the detection of chemical components according to the step (3), and predicting the intrinsic sensory quality according to the model established in the step (4), wherein the actual sensory quality results and the predicted results of the conventional tobacco, the medium branch tobacco and the fine branch tobacco are shown in the table 5.
TABLE 5
As can be seen from Table 5, the actual sensory quality of the tested tobacco leaves is not much different from the predicted sensory quality, and the errors are less than 5%. It follows that it is feasible to implement tobacco suitability classification by performing sensory quality model prediction on unknown finished tobacco.
(4) Determining applicability categories
And (6) taking the finished product of the C3F module tobacco obtained after threshing and redrying of the raw tobacco purchased in Henan province as an object, measuring chemical components (major component, semi-minor component and minor component) of the tobacco module according to the step (3), and substituting the chemical component values measured in the step (3) into the fine branch model, the middle branch model and the conventional model established in the step (4) respectively to obtain the perceived quality score of the tobacco module, wherein the result is shown in Table 6.
TABLE 6
| Fragrant quality | Fragrance amount | Miscellaneous gas | Irritation (irritation) | Aftertaste of | Total score | |
| Thin-support model | 6.0 | 6.5 | 6.0 | 6.0 | 6.0 | 68.3 |
| Middle support model | 5.8 | 6.0 | 5.5 | 5.7 | 5.8 | 64.7 |
| Conventional model | 5.5 | 5.8 | 5.0 | 5.5 | 5.5 | 61.7 |
Note that: total = (aroma quality x 0.3+ aroma amount x 0.3+ miscellaneous gas x 0.08+ stimulus x 0.15+ aftertaste x 0.17) x 11.11
From the data in table 6, it can be seen that the sensory evaluation of the finished product of the C3F module tobacco leaves substituted into the fine cigarette model is more excellent, so that the application category is determined, and the application category is determined to be suitable for the fine cigarette by identifying the Henan C3F module tobacco leaves through the model.
The applicant states that the detailed method of the present invention is illustrated by the above examples, but the present invention is not limited to the detailed method described above, i.e. it does not mean that the present invention must be practiced in dependence upon the detailed method described above. It should be apparent to those skilled in the art that any modification of the present invention, equivalent substitution of raw materials for the product of the present invention, addition of auxiliary components, selection of specific modes, etc., falls within the scope of the present invention and the scope of disclosure.
Claims (10)
1. A method for determining a location for dispensing and warehousing finished tobacco leaves, the method comprising:
(1) Sample selection of modeled tobacco: collecting single tobacco leaves, dividing the single tobacco leaves into a training set and a testing set, and classifying the applicability of the single tobacco leaves;
(2) Sensory evaluation: determining a sensory evaluation method according to industry standards, and performing sensory evaluation on the tobacco leaf samples classified in the step (1) respectively;
(3) And (3) measuring chemical components of tobacco leaves: respectively carrying out quantitative analysis on major components, semi-minor components and minor components of the tobacco leaves classified in the step (1);
(4) And (3) constructing a quality prediction model: carrying out Support Vector Regression (SVR) analysis by taking the training set tobacco leaf samples under different classifications as objects, taking the chemical components measured in the step (3) as independent variables and taking the sensory evaluation indexes in the step (2) as dependent variables, and calculating to form sensory quality prediction models of the tobacco leaves under different classifications;
(5) Determining the applicability category: carrying out chemical component measurement on the finished tobacco flakes to be measured according to the method of the step (3), identifying by adopting the model established in the step (4), and determining the applicability category of the finished tobacco flakes;
(6) Determining a tobacco leaf finished product distribution and storage place: and (3) determining the distribution flow direction of the finished tobacco flakes to be tested to a storage place according to the applicability category of the tobacco flakes determined in the step (5).
2. The method of claim 1, wherein the tobacco leaves in step (1) comprise octagon-type production zone tobacco leaves;
Preferably, the tobacco leaves comprise any one or a combination of at least two of upper tobacco leaves, middle tobacco leaves or lower tobacco leaves;
Preferably, the tobacco leaves comprise any one or a combination of at least two of superior tobacco leaves, medium tobacco leaves or inferior tobacco leaves.
3. The method of claim 1 or 2, wherein the classification of the suitability of step (1) comprises a fine cigarette suitability class, a medium cigarette suitability class and a regular cigarette suitability class;
Preferably, the sensory evaluation index in the step (2) includes any one or a combination of at least two of aroma quality, aroma amount, miscellaneous gas, irritation and aftertaste;
preferably, the sensory evaluation in step (2) includes the indexes of aroma quality, aroma quantity, miscellaneous gas, irritation and aftertaste.
4. A method of determining a location for dispensing and warehousing finished tobacco products according to claim 3, wherein the sensory evaluation method of step (2) is performed according to tobacco industry standard YC/T415-2011;
Preferably, the score range of the sensory evaluation index is: ① fragrance quality: good, better 7.6-9.0, middle upper 6.1-7.5, middle 4.6-6.0, middle lower 3.1-4.5, worse, difference less than or equal to 3; ② fragrance amount: foot 7.6-9.0, foot 6.1-7.5, 4.6-6.0, less 3.1-4.5, less than or equal to 3; ③ miscellaneous gases: light weight 7.6-9.0, light weight 6.1-7.5, 4.6-6.0, heavy weight 3.1-4.5 and weight less than or equal to 3; ④ irritation: small 7.6-9.0, small 6.1-7.5, 4.6-6.0, large 3.1-4.5 and large less than or equal to 3; ⑤ aftertaste: comfortable 7.6-9.0, more comfortable 6.1-7.5, shang Shi 4.6-6.0, underfit 3.1-4.5 and tongue stagnation less than or equal to 3.
5. The method of determining a point of distribution and warehousing of finished tobacco leaves according to any one of claims 1 to 4, wherein the tobacco leaf chemical component measurement of step (3) includes major, semi-minor and minor components;
preferably, the macroingredients comprise any one or a combination of at least two of monosaccharides, proteins, alkaloids, anions and cations, polyphenols, polyacids and higher fatty acids, amadori compounds, pH, dichloromethane extract, solanesol or neophytadiene;
Preferably, the semi-minor ingredients include any one or a combination of at least two of anions and cations, polyphenols, polyacids and higher fatty acids, amadori compounds, pH, dichloromethane extract, solanesol or neophytadiene;
preferably, the micro-ingredients comprise volatile and/or semi-volatile substances.
6. The method of claim 5, wherein the major or semi-minor components are detected by near infrared spectroscopy.
7. The method of claim 5 or 6, wherein the trace elements are detected by chromatographic mass spectrometry, the detection including sample processing and GC/MS analysis.
8. The method for determining a location of distribution and storage of finished tobacco leaves according to any one of claims 1 to 7, wherein the quality prediction model construction in step (4) specifically comprises the following steps: carrying out Support Vector Regression (SVR) analysis by taking training set tobacco leaf raw materials with 3 conditions of suitable for the fine branch, suitable for the middle branch and suitable for the conventional use under different positioning as objects, taking chemical components measured in the step (3) as independent variables and taking sensory evaluation indexes in the step (2) as dependent variables, and respectively calculating to form sensory quality prediction models of the fine branch, the middle branch and the conventional type tobacco leaves;
preferably, the model identification and training is further performed after the quality prediction model establishment in the step (4).
9. The method according to any one of claims 1 to 8, wherein the determining of the tobacco product distribution and storage location in step (6) specifically includes: and (3) determining the distribution flow direction of the finished tobacco flakes to be tested to a storage place according to the applicability category of the tobacco products determined in the step (5) and combining the production equipment conditions of different cigarette factories of tobacco industry enterprises.
10. Use of the method of determining a location of distribution and storage of finished tobacco leaves according to any one of claims 1 to 9 in the distribution and storage of tobacco leaves.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410120735.1A CN118134276A (en) | 2024-01-29 | 2024-01-29 | Method for judging tobacco leaf finished product distribution storage location and application thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410120735.1A CN118134276A (en) | 2024-01-29 | 2024-01-29 | Method for judging tobacco leaf finished product distribution storage location and application thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118134276A true CN118134276A (en) | 2024-06-04 |
Family
ID=91239356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410120735.1A Pending CN118134276A (en) | 2024-01-29 | 2024-01-29 | Method for judging tobacco leaf finished product distribution storage location and application thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118134276A (en) |
-
2024
- 2024-01-29 CN CN202410120735.1A patent/CN118134276A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103134850B (en) | A kind of tea leaf quality method for quick based on characteristic perfume | |
| CN105445421B (en) | A kind of method of aesthetic quality during predictive slice nicotinic alcohol by appearance index | |
| CN106680241A (en) | Novel spectrum multi-analysis classification and identification method and application thereof | |
| CN106018600B (en) | A kind of metabolism group method for distinguishing false positive mass spectrum peak-to-peak signal and quantitative correction mass spectrum peak area | |
| CN102778442A (en) | Method for rapidly identifying type of balsam material liquid for cigarette | |
| CN103674884A (en) | Random forest classification method for tobacco leaf style characteristics based on near infrared spectral information | |
| CN105469224A (en) | Odor pollution source key odor causing substance recognition method | |
| CN102967578A (en) | Method for obtaining near-infrared spectrum of beef sample online and application thereof in evaluating beef quality | |
| CN105203498A (en) | Near infrared spectrum variable selection method based on LASSO | |
| CN113340874B (en) | Quantitative analysis method based on combination ridge regression and recursive feature elimination | |
| CN105823752A (en) | Method for fast identifying variety of edible oil through near-infrared spectroscopy method | |
| CN103645144A (en) | Method for quantitatively analyzing components in white spirit | |
| CN105717066A (en) | Near-infrared spectrum recognition model based on weighting association coefficients | |
| WO2020248961A1 (en) | Method for selecting spectral wavenumber without reference value | |
| CN101339150A (en) | Method for determining octane number based on dielectric spectra technology | |
| CN111272686B (en) | Hyperspectral detection method for iron grade of iron ore concentrate powder | |
| CN109324016A (en) | A kind of determination method of redried odor type style | |
| CN109374530A (en) | The optoacoustic spectroscopy on-line monitoring method of sulfur hexafluoride gas decomposition product | |
| CN114970675A (en) | Artificial nose refrigerator food freshness detection system and method based on feature selection | |
| CN108645746B (en) | Electrical heating non-combustible cigarette sensory quality analysis method based on furnace volatile matter | |
| CN105954228A (en) | Method for measuring content of sodium metal in oil sand based on near infrared spectrum | |
| CN103344598A (en) | Method for determination of compatibleness of cut stems and tobacco leaf group | |
| CN118134276A (en) | Method for judging tobacco leaf finished product distribution storage location and application thereof | |
| CN111257452B (en) | Method for detecting addition of synthetic acetic acid into Zhenjiang aromatic vinegar | |
| CN103134764B (en) | The method of prediction true boiling point curve of crude oil is composed by transmitted infrared light |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |