CN116343944A - Cigarette auxiliary material parameter and physical index and main stream smoke component influence prediction method - Google Patents
Cigarette auxiliary material parameter and physical index and main stream smoke component influence prediction method Download PDFInfo
- Publication number
- CN116343944A CN116343944A CN202310281731.7A CN202310281731A CN116343944A CN 116343944 A CN116343944 A CN 116343944A CN 202310281731 A CN202310281731 A CN 202310281731A CN 116343944 A CN116343944 A CN 116343944A
- Authority
- CN
- China
- Prior art keywords
- cigarette
- main stream
- stream smoke
- parameters
- cigarettes
- 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
- 235000019504 cigarettes Nutrition 0.000 title claims abstract description 282
- 239000000779 smoke Substances 0.000 title claims abstract description 92
- 238000000034 method Methods 0.000 title claims abstract description 69
- 239000000463 material Substances 0.000 title claims abstract description 32
- 238000007637 random forest analysis Methods 0.000 claims abstract description 63
- 238000009423 ventilation Methods 0.000 claims description 40
- 238000012549 training Methods 0.000 claims description 13
- 229910019142 PO4 Inorganic materials 0.000 claims description 12
- 239000010452 phosphate Substances 0.000 claims description 12
- 230000000391 smoking effect Effects 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 11
- NBIIXXVUZAFLBC-UHFFFAOYSA-K phosphate Chemical compound [O-]P([O-])([O-])=O NBIIXXVUZAFLBC-UHFFFAOYSA-K 0.000 claims description 10
- KRKNYBCHXYNGOX-UHFFFAOYSA-K Citrate Chemical compound [O-]C(=O)CC(O)(CC([O-])=O)C([O-])=O KRKNYBCHXYNGOX-UHFFFAOYSA-K 0.000 claims description 9
- NPYPAHLBTDXSSS-UHFFFAOYSA-N Potassium ion Chemical compound [K+] NPYPAHLBTDXSSS-UHFFFAOYSA-N 0.000 claims description 9
- FKNQFGJONOIPTF-UHFFFAOYSA-N Sodium cation Chemical compound [Na+] FKNQFGJONOIPTF-UHFFFAOYSA-N 0.000 claims description 9
- 229910001414 potassium ion Inorganic materials 0.000 claims description 9
- 229910001415 sodium ion Inorganic materials 0.000 claims description 9
- MUBZPKHOEPUJKR-UHFFFAOYSA-N Oxalic acid Chemical compound OC(=O)C(O)=O MUBZPKHOEPUJKR-UHFFFAOYSA-N 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims 1
- 238000013461 design Methods 0.000 abstract description 8
- 230000008569 process Effects 0.000 abstract description 5
- 238000003908 quality control method Methods 0.000 abstract description 3
- 238000012827 research and development Methods 0.000 abstract description 3
- SNICXCGAKADSCV-JTQLQIEISA-N (-)-Nicotine Chemical compound CN1CCC[C@H]1C1=CC=CN=C1 SNICXCGAKADSCV-JTQLQIEISA-N 0.000 description 25
- 229960002715 nicotine Drugs 0.000 description 25
- SNICXCGAKADSCV-UHFFFAOYSA-N nicotine Natural products CN1CCCC1C1=CC=CN=C1 SNICXCGAKADSCV-UHFFFAOYSA-N 0.000 description 25
- 230000035699 permeability Effects 0.000 description 14
- 238000011002 quantification Methods 0.000 description 6
- 238000002485 combustion reaction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000012417 linear regression Methods 0.000 description 3
- 238000000611 regression analysis Methods 0.000 description 3
- 241000208125 Nicotiana Species 0.000 description 2
- 235000002637 Nicotiana tabacum Nutrition 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- BITYAPCSNKJESK-UHFFFAOYSA-N potassiosodium Chemical compound [Na].[K] BITYAPCSNKJESK-UHFFFAOYSA-N 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000012569 chemometric method Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Databases & Information Systems (AREA)
- Mathematical Optimization (AREA)
- Chemical & Material Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Algebra (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing Of Cigar And Cigarette Tobacco (AREA)
Abstract
The invention discloses a method for predicting the influence of auxiliary parameters, physical indexes and main stream smoke components of cigarettes, which comprises the steps of obtaining the auxiliary parameters, the physical indexes and the main stream smoke data of known cigarette samples in advance, obtaining the influence degree of the auxiliary parameters, the physical indexes and the main stream smoke data of the cigarette samples in specific component content regression of the main stream smoke of the cigarettes through random forest analysis, obtaining a corresponding random forest model regression result, and predicting the specific component data of the main stream smoke of unknown cigarette samples by using the established random forest model. The method utilizes the characteristics that the random forest method is not easy to fall into overfitting, has remarkable noise resistance and can process higher dimensional data, effectively reveals the correlation between the parameters and physical indexes of the auxiliary materials of the cigarettes and preset components in main stream smoke of the cigarettes, accurately predicts the component data in the main stream smoke of unknown cigarette samples, and is sufficient to provide reliable reference for design, research and development of cigarette products and quality control.
Description
Technical Field
The invention relates to the field of tobacco detection, in particular to a method for predicting influences of auxiliary parameters, physical indexes and main stream smoke components of cigarettes.
Background
The release amount of conventional components (tar, nicotine and CO) of the smoke is an important index for the design of the cigarette, and the release amount of the tar, the nicotine and the CO in the main stream smoke of the cigarette can be accurately predicted, so that the method has important significance for guiding the digital design of cigarette products. Related work has been carried out on the prediction of tar, nicotine and CO emissions in the mainstream smoke of cigarettes: if the quantitative relation between the gram weight of the cigarette paper, the air permeability of the forming paper, the air permeability of the tipping paper and the suction resistance of the filter stick and the release amount of tar, nicotine and CO in the main stream smoke is inspected by using a chemometric method, a multi-element linear prediction model is established; for another example, a mathematical model for predicting the release amount of tar, nicotine and CO in main stream smoke of cigarettes is established by multiple linear regression analysis of the quantitative of the cigarette paper, the air permeability of the forming paper, the air permeability of the tipping paper and the suction resistance of the filter stick; for another example, a multi-element linear regression equation prediction model of filter ventilation, filter rod pressure drop, cigarette paper quantification, cigarette paper ventilation, cigarette paper combustion improver mass fraction and potassium-sodium ratio in the cigarette paper combustion improver and release amounts of tar, nicotine and CO in main stream smoke of cigarettes is established; in addition, for example, a linear regression method and a stepwise regression method are adopted to construct a multi-element linear prediction model of the release amount of tar, nicotine and CO of the medium cigarette based on parameters such as the air permeability, the quantification, the combustion improver consumption, the combustion improver potassium-sodium ratio, the filter rod pressure drop, the tow specification, the air permeability of the forming paper, the air permeability of the tipping paper and the like of the cigarette paper.
However, most of existing algorithms for predicting the release amounts of tar, nicotine and CO in main stream smoke of cigarettes are based on linear fitting models, and the linear fitting models have the defects of low accuracy, easiness in overfitting, large influence of noise and the like.
Disclosure of Invention
In view of the above, the present invention aims to provide a method for predicting the influence of auxiliary parameters and physical indexes of cigarettes and main stream smoke components, so as to solve the defects in the current prediction process of the release amount of specific components of main stream smoke of cigarettes based on the auxiliary parameters and physical indexes of cigarettes.
The technical scheme adopted by the invention is as follows:
the invention provides a method for predicting the influence of auxiliary parameters, physical indexes and main stream smoke components of cigarettes, which comprises the following steps:
inputting cigarette auxiliary material parameters, physical indexes and preset component data in main stream smoke of a known cigarette sample in advance;
determining a random forest method from a system, and setting corresponding algorithm parameters;
obtaining importance ordering results of cigarette auxiliary material parameters and physical indexes of known cigarette samples in preset component regression of cigarette main stream smoke and corresponding random forest model fitting results by utilizing the random forest method;
based on a random forest model fitting result, the cigarette auxiliary material parameters and physical indexes of the unknown cigarette sample are combined, and the preset components in the main stream smoke of the unknown cigarette sample are predicted.
In at least one possible implementation manner, the obtaining the importance ranking result of the cigarette auxiliary parameters and the physical indexes of the known cigarette sample in the preset component regression of the main stream smoke of the cigarette includes:
sampling according to the set training sample size from the data of the known cigarette samples in a random and repeatable mode to form a training set;
combining the optimal segmentation modes of the given auxiliary parameters and physical indexes of the cigarettes in the training set to obtain a known cigarette sample which is not extracted;
when the preset components of the main stream smoke of the cigarettes are regressed, the known cigarette samples which are not extracted are taken as test samples, and the importance of the cigarette auxiliary material parameters and the physical indexes of each cigarette sample in the preset components is evaluated by adopting a random sampling mode.
In at least one possible implementation manner, the importance of the cigarette auxiliary material parameters and physical indexes is represented by adopting standardized data with poor inter-tree prediction precision.
In at least one possible implementation manner, the algorithm parameters include the number of models to be constructed, a sample size, a maximum node number, a maximum tree depth and a minimum child node number.
In at least one possible implementation manner, the cigarette auxiliary material parameters and physical indexes comprise a plurality of combinations of the following various parameter indexes: the cigarette paper ventilation degree, the cigarette paper ration, the cigarette paper citrate content, the cigarette paper phosphate content, the cigarette paper oxalate content, the cigarette paper potassium ion content, the cigarette paper sodium ion content, the tipping paper ventilation degree, the cigarette weight, the standard smoking resistance, the closed smoking resistance, the filter rod ventilation rate, the cigarette paper ventilation rate and the total ventilation rate.
In at least one possible implementation manner, the preset component data in the main stream smoke of the cigarette are obtained by actual measurement according to the established standard in the industry.
In at least one possible implementation thereof, the random forest method includes a random forest classification method or a random forest regression method.
Compared with the prior art, the main design concept of the invention is that the cigarette auxiliary material parameters and the physical indexes of the known cigarette samples and the data of the main stream smoke and the preset components are detected in advance, and then the influence degree of the cigarette auxiliary material parameters and the physical indexes of the cigarette samples in the regression of the content of the specific components in the main stream smoke of the cigarette and the regression result of the corresponding random forest model are obtained after random forest analysis, and the prediction result is carried out on the data of the specific components in the main stream smoke of the unknown cigarette samples by using the established random forest model. The method utilizes the characteristics that the random forest method is not easy to fall into overfitting, has remarkable noise resistance and can process higher dimensional data, effectively reveals the correlation strength between the parameters and physical indexes of auxiliary materials of cigarettes and preset components in main stream smoke of cigarettes, accurately predicts the component data in main stream smoke of unknown cigarette samples, and is sufficient to provide reliable reference for design, research and development of cigarette products and quality control.
Drawings
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described with reference to the accompanying drawings, in which:
FIG. 1 is a flow chart of a method for predicting the influence of parameters and physical indexes of auxiliary materials and main stream smoke components of cigarettes provided by an embodiment of the invention;
FIG. 2 is a schematic diagram of linear fitting of predicted and actual values of tar release in mainstream smoke of 8 unknown cigarette samples provided by an embodiment of the present invention;
FIG. 3 is a schematic diagram of linear fitting of predicted and actual values of nicotine release in mainstream smoke of 12 unknown cigarette samples provided by an embodiment of the present invention;
fig. 4 is a schematic linear fitting diagram of predicted values and actual values of CO release amounts in mainstream smoke of 12 unknown cigarette samples provided in an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
The invention provides an embodiment of a method for predicting the influence of auxiliary parameters, physical indexes and main stream smoke components of cigarettes, specifically, as shown in fig. 1, the method comprises the following steps:
step S1, inputting cigarette auxiliary material parameters of known cigarette samples, physical indexes and preset component data in main stream smoke of cigarettes in advance;
step S2, determining a random forest method (comprising a random forest classification method or a random forest regression method, which can be selected as the random forest regression method) from a system, and setting corresponding algorithm parameters, wherein the algorithm parameters can comprise the number of models to be constructed, the sample size, the number of maximum nodes, the maximum tree depth and the minimum number of child nodes; for example, the number of models is set to 100, the sample size is set to 1.0, and other construction parameters can be preset default values.
S3, obtaining an importance ordering result of the cigarette auxiliary material parameters and the physical indexes of the known cigarette samples in the regression of the preset components of the main stream smoke of the cigarette and a corresponding random forest model fitting result by utilizing the random forest method;
and S4, based on a random forest model fitting result, predicting preset components in main stream smoke of the unknown cigarette sample by combining the cigarette auxiliary parameters and physical indexes of the unknown cigarette sample.
It should be noted here that the present invention is primarily directed to these three main stream smoke constituent objects, particularly with respect to tar, nicotine and CO emissions, but the present invention is not limited to these three elements, either necessarily or only.
In some preferred embodiments, the obtaining the importance ranking result of the cigarette accessory parameters and the physical indexes of the known cigarette sample in the preset component regression of the main stream smoke of the cigarette includes:
step S31, sampling is carried out in a random and repeatable mode according to the set training sample size from the data of the known cigarette samples to form a training set;
step S32, combining the optimal segmentation modes of the given auxiliary parameters and physical indexes of the cigarettes in the training set to obtain a known cigarette sample which is not extracted;
and step S33, when the preset components of the main stream smoke of the cigarettes are regressed, taking the known cigarette samples which are not extracted as test samples, and evaluating the importance of the cigarette auxiliary material parameters and physical indexes of each cigarette sample in the preset components by adopting a random sampling mode.
The cigarette auxiliary material parameters and physical indexes specifically can comprise a plurality of combinations of the following various parameter indexes: the cigarette paper ventilation degree, the cigarette paper ration, the cigarette paper citrate content, the cigarette paper phosphate content, the cigarette paper oxalate content, the cigarette paper potassium ion content, the cigarette paper sodium ion content, the tipping paper ventilation degree, the cigarette weight, the standard smoking resistance, the closed smoking resistance, the filter rod ventilation rate, the cigarette paper ventilation rate and the total ventilation rate. And the data of preset components (such as the release amounts of tar, nicotine and CO) in the main stream smoke of the cigarettes are obtained by actual measurement according to the established standard in the industry.
Here, taking an analysis process of a random forest method of the release amount of CO in main stream smoke of cigarettes as an example: the total number of known cigarette samples is N, the size of the training sample is y, and the number of constructed models is N tress The number of cigarette auxiliary parameters and physical index variables of the cigarette samples randomly extracted at the nodes is m; then, sampling is preferably performed by adopting a bootstrap sampling method, namely sampling is performed by adopting a random and repeatable sampling mode according to the size of a set training sample from N pieces of data of known cigarette samples to form a group of training sets; then, carrying out regression on the known cigarette samples which are not taken by using an optimal segmentation mode of m variables in the training set, wherein each time the samples which are not taken are N× (1-y), and the samples which are not taken can be collectively called as out-of-bag data; when the release amounts of tar, nicotine and CO are regressed, the out-of-bag data can be used as a test sample, and a random sampling method is adopted to evaluate the importance of the material parameters of each cigarette sample in the prediction of the release amounts of tar, nicotine and CO. In the concrete calculation, the importance of the cigarette auxiliary parameters and physical indexes of the cigarette sample is preferably represented by standardized data with poor inter-tree prediction precision.
The invention is further specifically described with reference to the following examples, wherein: the method is characterized in that 14 auxiliary parameters and physical indexes of a cigarette sample with known tar release amount are used as variables, a random forest method is adopted to analyze the importance of the cigarette auxiliary parameters and physical indexes of the cigarette sample in the regression of the tar release amount in the main stream smoke of the cigarette, and the tar release amount in the main stream smoke of an unknown cigarette sample is predicted.
The tar release amounts of the main stream smoke of the produced 55 cigarette samples were subjected to fitting analysis (table 1, table 2) by taking 14 cigarette auxiliary parameters and physical indexes of the ventilation degree of the cigarette paper, the quantification of the cigarette paper, the citrate content of the cigarette paper, the phosphate content of the cigarette paper, the oxalate content of the cigarette paper, the potassium ion content of the cigarette paper, the sodium ion content of the cigarette paper, the ventilation degree of the tipping paper, the weight of the cigarette, the standard smoking resistance, the closed smoking resistance, the ventilation rate of the filter stick, the ventilation rate of the cigarette paper and the total ventilation rate as variables. The results show that the tar release amount in the main stream smoke of the cigarette samples can be accurately fitted by preferably but not exclusively adopting a random forest regression method configured in an IBM SPSS model 18.0 data processing system according to the cigarette auxiliary parameters and physical indexes of each cigarette sample.
TABLE 1 random forest regression analysis results of tar release from 55 cigarette samples
TABLE 2 analysis of Tar Release random forest method fitting results for 55 cigarette samples
Next, the importance of each cigarette accessory parameter and physical index is shown in Table 3. As can be seen from table 3, the importance ranking among the 14 detected cigarette auxiliary parameters and physical indexes is: cigarette paper potassium ion content & gt filter stick ventilation rate & gt standard smoke resistance & gt total ventilation rate & gt tipping paper air permeability & gt cigarette paper ventilation rate & gt cigarette paper citrate content & gt cigarette paper quantitative & gt cigarette paper sodium ion content, cigarette paper phosphate content, cigarette paper oxalate content, cigarette weight and closed smoke resistance, because the minimum field variation is less than 0.05, the cigarette paper phosphate is automatically removed by a random forest regression method in a data processing system, and the cigarette paper phosphate is not listed in importance ranking.
TABLE 3 importance ordering of 14 auxiliary parameters and physical indicators in tar release fitting
Then, random forest predictions were performed on 8 unknown cigarette samples according to the fitting results obtained by the foregoing method (see table 4). The result shows that according to the cigarette auxiliary material parameters and physical indexes of the cigarette samples and the established random forest model of the tar release amount in the main stream smoke, the tar release amount in the main stream smoke of unknown cigarette samples can be accurately predicted by adopting a random forest regression method (see table 5 and figure 2), and R2 is 0.959.
Tab 4 8 unknown cigarette sample main stream smoke tar release prediction results
Table 5 8 random forest method prediction result analysis of tar release amount of unknown cigarette samples
In addition to the above examples, the method further uses 14 auxiliary parameters and physical indexes of the cigarette sample with known nicotine release amount as variables, adopts a random forest method to analyze the importance of the cigarette auxiliary parameters and physical indexes of the cigarette sample in the regression of the nicotine release amount in the main stream smoke of the cigarette, and predicts the nicotine release amount in the main stream smoke of the unknown cigarette sample.
The tobacco release amounts of the produced main stream smoke of 63 cigarette samples were subjected to fitting analysis (table 6, table 7) by taking 14 cigarette auxiliary parameters and physical indexes of the ventilation degree of the cigarette paper, the quantification of the cigarette paper, the citrate content of the cigarette paper, the phosphate content of the cigarette paper, the oxalate content of the cigarette paper, the potassium ion content of the cigarette paper, the sodium ion content of the cigarette paper, the ventilation degree of the tipping paper, the weight of the cigarette, the standard smoking resistance, the closed smoking resistance, the ventilation rate of the filter stick, the ventilation rate of the cigarette paper and the total ventilation rate as variables. The result shows that the release amount of nicotine in main stream smoke of the cigarette samples can be accurately fitted by adopting a random forest regression method according to the cigarette auxiliary material parameters and physical indexes of each cigarette sample.
TABLE 6 random forest regression analysis of nicotine Release from 63 cigarette samples
TABLE 7 analysis of random forest method fitting results for nicotine Release from 63 cigarette samples
Next, the importance of 14 auxiliary parameters and physical indexes of the cigarette samples on the release amount of nicotine in the main stream smoke is analyzed, and the importance of each auxiliary parameter and physical index of each cigarette is shown in table 8. As can be seen from table 8, the importance ranking of the detected 14 cigarette auxiliary parameters and physical indexes is: the cigarette paper potassium ion content is larger than the cigarette paper ventilation rate is larger than the cigarette paper quantification is larger than the cigarette paper ventilation rate is larger than the filter stick ventilation rate is larger than the tipping paper air permeability is larger than the cigarette paper citrate content is larger than the standard smoke resistance is larger than the total ventilation rate is larger than the cigarette paper air permeability is larger than the cigarette paper sodium ion content, and the cigarette paper phosphate radical content, the cigarette paper oxalate radical content, the cigarette weight and the closed smoke resistance are automatically removed by a random forest regression method in the system because the minimum field variation is smaller than 0.05, so that the cigarette paper phosphate radical content, the cigarette paper oxalate radical content, the cigarette weight and the closed smoke resistance are not listed in the importance ranking.
TABLE 8 importance ranking of 14 auxiliary parameters and physical indicators in Nicotine Release amount fitting
Then, random forest predictions were made for 12 unknown cigarette samples according to the random forest model that had been established (table 9). The results show that the random forest regression method can accurately predict the release amount of nicotine in the main stream smoke of an unknown cigarette sample according to the cigarette auxiliary material parameters and physical indexes of the cigarette sample and the established random forest model of the release amount of nicotine in the main stream smoke (see table 10 and figure 3), and R2 is 0.965.
TABLE 9 prediction results of nicotine Release in mainstream Smoke from 12 unknown cigarette samples
TABLE 10 analysis of prediction results of nicotine Release from 12 unknown cigarette samples by random forest method
Finally, the method also takes 14 auxiliary parameters and physical indexes of the cigarette sample with known CO release as variables, adopts a random forest method to analyze the importance of the cigarette auxiliary parameters and physical indexes of the cigarette sample in the regression of the CO release in the main stream smoke of the cigarette and predicts the CO release in the main stream smoke of the unknown cigarette sample.
The CO release amounts in the main stream smoke of the produced 63 cigarette samples were subjected to fitting analysis (table 11, table 12) by taking 14 cigarette auxiliary parameters and physical indexes of the air permeability of the cigarette paper, the quantification of the cigarette paper, the citrate content of the cigarette paper, the phosphate content of the cigarette paper, the oxalate content of the cigarette paper, the potassium ion content of the cigarette paper, the sodium ion content of the cigarette paper, the air permeability of the tipping paper, the weight of the cigarette, the standard smoking resistance, the closed smoking resistance, the ventilation rate of the filter stick, the ventilation rate of the cigarette paper and the total ventilation rate as variables. The result shows that the CO release amount in the main stream smoke of the cigarette samples can be accurately fitted by adopting a random forest regression method according to the cigarette auxiliary material parameters and physical indexes of each cigarette sample.
TABLE 11 random forest regression analysis results of CO Release from 63 cigarette samples
TABLE 12 random forest method fitting result analysis of CO Release amount for 63 cigarette samples
Next, the importance of the 14 auxiliary parameters and physical indexes of the cigarette sample on the release amount of CO in the main stream smoke is analyzed, and the importance of each auxiliary parameter and physical index of the cigarette is shown in table 13. As can be seen from table 13, the importance ranking of the detected 14 cigarette auxiliary parameters and physical indexes is: cigarette paper potassium ion content & gt, filter stick ventilation rate & gt, standard smoke resistance & gt, cigarette paper ventilation rate & gt, cigarette paper quantitative & gt, cigarette paper air permeability & gt, total ventilation rate & gt, cigarette paper citrate content & gt, tipping paper air permeability & gt, cigarette paper sodium ion content, and cigarette paper phosphate content, cigarette paper oxalate content, cigarette weight and closed smoke resistance are automatically removed by a random forest regression method in a system due to the fact that the minimum field variation is smaller than 0.05, so that importance ranking is not included.
Table 13 importance ranking of 14 auxiliary parameters and physical indicators in CO release fitting
Then, random forest predictions were performed on 12 unknown cigarette samples according to the random forest model method that was already established (table 14). The result shows that according to the cigarette auxiliary material parameters and physical indexes of the cigarette samples and the established random forest model of the CO release amount in the main stream smoke, the CO release amount in the main stream smoke of the unknown cigarette samples can be accurately predicted by adopting a random forest regression method (see table 15 and figure 4), and R2 is 0.950.
TABLE 14 prediction results of CO Release in mainstream smoke of 12 unknown cigarette samples
TABLE 15 analysis of CO Release random forest prediction results for 12 unknown cigarette samples
In summary, the main design concept of the invention is to detect and obtain the cigarette auxiliary parameters and physical indexes of the known cigarette samples and the data of the main stream smoke and the preset components in advance, then obtain the influence degree of the cigarette auxiliary parameters and physical indexes of the cigarette samples in the regression of the specific component content in the main stream smoke of the cigarette and the regression result of the corresponding random forest model after random forest analysis, and use the established random forest model to predict the specific component data in the main stream smoke of the unknown cigarette samples. The method utilizes the characteristics that the random forest method is not easy to fall into overfitting, has remarkable noise resistance and can process higher dimensional data, effectively reveals the correlation strength between the parameters and physical indexes of auxiliary materials of cigarettes and preset components in main stream smoke of cigarettes, accurately predicts the component data in main stream smoke of unknown cigarette samples, and is sufficient to provide reliable reference for design, research and development of cigarette products and quality control.
In the embodiments of the present invention, "at least one" means one or more, and "a plurality" means two or more. "and/or", describes an association relation of association objects, and indicates that there may be three kinds of relations, for example, a and/or B, and may indicate that a alone exists, a and B together, and B alone exists. Wherein A, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of the following" and the like means any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c may represent: a, b, c, a and b, a and c, b and c or a and b and c, wherein a, b and c can be single or multiple.
The construction, features and effects of the present invention are described in detail according to the embodiments shown in the drawings, but the above is only a preferred embodiment of the present invention, and it should be understood that the technical features of the above embodiment and the preferred mode thereof can be reasonably combined and matched into various equivalent schemes by those skilled in the art without departing from or changing the design concept and technical effects of the present invention; therefore, the invention is not limited to the embodiments shown in the drawings, but is intended to be within the scope of the invention as long as changes made in the concept of the invention or modifications to the equivalent embodiments do not depart from the spirit of the invention as covered by the specification and drawings.
Claims (7)
1. A method for predicting the influence of auxiliary parameters, physical indexes and main stream smoke components of cigarettes is characterized by comprising the following steps:
inputting cigarette auxiliary material parameters, physical indexes and preset component data in main stream smoke of a known cigarette sample in advance;
determining a random forest method from a system, and setting corresponding algorithm parameters;
obtaining importance ordering results of cigarette auxiliary material parameters and physical indexes of known cigarette samples in preset component regression of cigarette main stream smoke and corresponding random forest model fitting results by utilizing the random forest method;
based on a random forest model fitting result, the cigarette auxiliary material parameters and physical indexes of the unknown cigarette sample are combined, and the preset components in the main stream smoke of the unknown cigarette sample are predicted.
2. The method for predicting the influence of parameters and physical indexes of auxiliary materials on main stream smoke components of cigarettes according to claim 1, wherein the obtaining the importance sequencing result of the parameters and physical indexes of auxiliary materials of known cigarettes in the regression of the preset components of main stream smoke of cigarettes comprises:
sampling according to the set training sample size from the data of the known cigarette samples in a random and repeatable mode to form a training set;
combining the optimal segmentation modes of the given auxiliary parameters and physical indexes of the cigarettes in the training set to obtain a known cigarette sample which is not extracted;
when the preset components of the main stream smoke of the cigarettes are regressed, the known cigarette samples which are not extracted are taken as test samples, and the importance of the cigarette auxiliary material parameters and the physical indexes of each cigarette sample in the preset components is evaluated by adopting a random sampling mode.
3. The method for predicting the influence of the parameters and the physical indexes of the auxiliary materials of the cigarettes and the components of the main stream smoke according to claim 2, wherein the importance of the parameters and the physical indexes of the auxiliary materials of the cigarettes is represented by using standardized data with poor inter-tree prediction precision.
4. The method for predicting the effects of auxiliary parameters, physical indexes and main stream smoke components of a cigarette according to claim 1, wherein the algorithm parameters comprise the number of models to be constructed, the sample size, the maximum node number, the maximum tree depth and the minimum child node number.
5. The method for predicting the effect of cigarette accessory parameters and physical indicators and main stream smoke components according to claim 1, wherein the cigarette accessory parameters and physical indicators comprise a plurality of combinations of the following parameter indicators: the cigarette paper ventilation degree, the cigarette paper ration, the cigarette paper citrate content, the cigarette paper phosphate content, the cigarette paper oxalate content, the cigarette paper potassium ion content, the cigarette paper sodium ion content, the tipping paper ventilation degree, the cigarette weight, the standard smoking resistance, the closed smoking resistance, the filter rod ventilation rate, the cigarette paper ventilation rate and the total ventilation rate.
6. The method for predicting the influence of auxiliary parameters, physical indexes and main stream smoke components of cigarettes according to claim 1, wherein the preset component data in the main stream smoke of the cigarettes are obtained by actual measurement according to established standards in the industry.
7. The method for predicting the influence of cigarette auxiliary parameters, physical indexes and main stream smoke components according to any one of claims 1 to 6, wherein the random forest method comprises a random forest classification method or a random forest regression method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310281731.7A CN116343944A (en) | 2023-03-17 | 2023-03-17 | Cigarette auxiliary material parameter and physical index and main stream smoke component influence prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310281731.7A CN116343944A (en) | 2023-03-17 | 2023-03-17 | Cigarette auxiliary material parameter and physical index and main stream smoke component influence prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116343944A true CN116343944A (en) | 2023-06-27 |
Family
ID=86889002
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310281731.7A Pending CN116343944A (en) | 2023-03-17 | 2023-03-17 | Cigarette auxiliary material parameter and physical index and main stream smoke component influence prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116343944A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117493945A (en) * | 2023-12-25 | 2024-02-02 | 山东产研青创信息科技有限公司 | Flue gas analysis early warning method and system based on computer vision |
-
2023
- 2023-03-17 CN CN202310281731.7A patent/CN116343944A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117493945A (en) * | 2023-12-25 | 2024-02-02 | 山东产研青创信息科技有限公司 | Flue gas analysis early warning method and system based on computer vision |
| CN117493945B (en) * | 2023-12-25 | 2024-03-29 | 山东产研青创信息科技有限公司 | Flue gas analysis early warning method and system based on computer vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1860961B1 (en) | Smoking machine | |
| CN107796782B (en) | Redrying quality stability evaluation method based on tobacco leaf characteristic spectrum consistency measurement | |
| CN116343944A (en) | Cigarette auxiliary material parameter and physical index and main stream smoke component influence prediction method | |
| CN105445421B (en) | A kind of method of aesthetic quality during predictive slice nicotinic alcohol by appearance index | |
| CN108732127B (en) | Method for detecting mixing proportion of each component in cut tobacco | |
| CN101949862A (en) | Method for measuring wrapped ash state after burning of cigarette | |
| CN103344572A (en) | Method for evaluating blending homogeneity of cut rolled stems and regenerated cut tobaccos in cigarettes | |
| CN113190967B (en) | Multi-objective screening-based fine cigarette material combination method | |
| CN110274848A (en) | Cigarette shreds based on density of tobacco rod are distributed characterizing method | |
| CN104899424B (en) | A kind of cured tobacco leaf maturity and blade construction objective measure based on microscopic feature | |
| CN105628646B (en) | A kind of online tar prediction of cigarette and method for early warning | |
| CN101147623A (en) | Cigarette supplementary lean combined assembling method | |
| CN103892435A (en) | Method for preparing tobacco powder raw materials and stabilizing quality of reconstituted tobacco | |
| CN105223140A (en) | The method for quickly identifying of homology material | |
| CN108398520B (en) | Method for predicting cigarette combustion intensity based on detection of flow field distribution | |
| CN104090082B (en) | The Forecasting Methodology of the full smoke pH of a kind of cigarette | |
| CN114563425A (en) | Tobacco leaf potassium content determination method based on tobacco leaf microstructure | |
| CN110333177B (en) | Cigarette segmented suction resistance nondestructive rapid detection evaluation method based on cigarette paper air permeability stability | |
| CN113092689B (en) | Quantitative determination method for tobacco stem penetration degree | |
| CN101701885B (en) | Gas collecting bottle for connecting smoking machine with electronic nose and use method thereof | |
| CN117272239A (en) | Cigarette ash packing performance prediction method based on random forest algorithm | |
| CN104598752B (en) | A kind of CO burst size forecast model construction methods in cigarette smoke based on combustion adjuvant | |
| CN113222268A (en) | Multi-mode reasoning-based tobacco baking quality prediction model establishing method | |
| CN102823943B (en) | Cigarette design adjusting method based on NNK release amount in deep smoking case | |
| CN112526009A (en) | Method for measuring water content of heated cigarette core material based on water activity |
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 |