CN112016046B - Construction method and device of tobacco shred structure prediction model - Google Patents
Construction method and device of tobacco shred structure prediction model Download PDFInfo
- Publication number
- CN112016046B CN112016046B CN202010710238.9A CN202010710238A CN112016046B CN 112016046 B CN112016046 B CN 112016046B CN 202010710238 A CN202010710238 A CN 202010710238A CN 112016046 B CN112016046 B CN 112016046B
- Authority
- CN
- China
- Prior art keywords
- tobacco
- tobacco shred
- leaf
- score
- rate
- 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.)
- Active
Links
- 241000208125 Nicotiana Species 0.000 title claims abstract description 477
- 235000002637 Nicotiana tabacum Nutrition 0.000 title claims abstract description 477
- 238000010276 construction Methods 0.000 title claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 61
- 238000000611 regression analysis Methods 0.000 claims abstract description 51
- 238000000034 method Methods 0.000 claims abstract description 46
- 239000011159 matrix material Substances 0.000 claims abstract description 39
- 238000000513 principal component analysis Methods 0.000 claims abstract description 23
- 238000001514 detection method Methods 0.000 claims description 42
- 238000004458 analytical method Methods 0.000 claims description 30
- 238000010219 correlation analysis Methods 0.000 claims description 23
- 239000012634 fragment Substances 0.000 claims description 23
- 238000005520 cutting process Methods 0.000 claims description 9
- 230000000717 retained effect Effects 0.000 claims description 6
- 241000037488 Coccoloba pubescens Species 0.000 claims description 5
- 230000000694 effects Effects 0.000 description 21
- 238000000540 analysis of variance Methods 0.000 description 17
- 230000001186 cumulative effect Effects 0.000 description 11
- 238000013467 fragmentation Methods 0.000 description 7
- 238000006062 fragmentation reaction Methods 0.000 description 7
- 238000012847 principal component analysis method Methods 0.000 description 7
- 238000012417 linear regression Methods 0.000 description 6
- 238000000547 structure data Methods 0.000 description 6
- 238000012795 verification Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 5
- 238000012216 screening Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 239000000779 smoke Substances 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 235000019504 cigarettes Nutrition 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 239000000945 filler Substances 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
- 238000002360 preparation method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
Classifications
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
Abstract
The application discloses a construction method and a construction device of a tobacco shred structure prediction model, which are used for detecting leaf structures of obtained tobacco flake samples and cut tobacco structures after shredding; carrying out principal component analysis on the leaf structure of each tobacco flake sample, selecting preset number of principal components with accumulated variance contribution rate more than or equal to 95% as target principal components, and outputting a score coefficient matrix of the target principal components; constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculating to obtain the score of each target principal component; stepwise multiple regression analysis is carried out on the scores of the target main components and the tobacco shred structures, and a regression model of the scores of the target main components and the tobacco shred structures is output; the tobacco shred structure prediction model of the leaf structure is obtained by combining the score calculation model and the regression model, so that the technical problems that the tobacco shred structure of the predicted leaf structure and the actual tobacco shred structure of the leaf in the existing tobacco shred structure prediction method have large difference and the prediction precision is not high are solved.
Description
Technical Field
The application relates to the technical field of tobacco processing, in particular to a method and a device for constructing a tobacco shred structure prediction model.
Background
The tobacco shred structure, that is, the distribution of tobacco shred size, is the core index affecting the physical characteristics of single weight, density, suction resistance, total ventilation rate, hardness and the like of the rolled product, combustibility and smoke release behavior. The leaf structure determines the distribution of the sizes of cut tobacco after shredding, and the leaf structure has important significance for guiding the leaf structure after threshing and redrying, so that the cigarette rolling quality can be improved by establishing a determined relation between the cut tobacco structure and the leaf structure. The tobacco shred structure of the predicted blade structure of the existing tobacco shred structure prediction method has a larger difference with the actual tobacco shred structure of the blade, and the problem of low prediction precision exists.
Disclosure of Invention
The application provides a construction method and a construction device of a tobacco shred structure prediction model, which are used for solving the technical problems that the tobacco shred structure of a prediction blade structure of the existing tobacco shred structure prediction method has a larger difference from the actual tobacco shred structure of the blade, and the prediction precision is not high.
In view of the above, the first aspect of the present application provides a method for constructing a tobacco shred structure prediction model, including:
obtaining a plurality of tobacco flake samples, and detecting the leaf structure of each tobacco flake sample and the cut tobacco structure after shredding;
Carrying out principal component analysis on the leaf structures of the tobacco flake samples, selecting a preset number of principal components with accumulated variance contribution rate greater than or equal to 95% as target principal components, and outputting a scoring coefficient matrix of the target principal components;
constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculating the score of each target principal component based on the score calculation model;
step-by-step multiple regression analysis is carried out on the score of each target principal component and the tobacco shred structure, and a regression model of the score of each target principal component and the tobacco shred structure is output;
and combining the score calculation model and the regression model to obtain the tobacco shred structure prediction model of the leaf structure.
Optionally, detecting a leaf structure of each of the tobacco leaf samples includes:
and detecting the numerical value of each leaf structural index in each tobacco flake sample to obtain the leaf structure of each tobacco flake sample, wherein the leaf structural indexes comprise a large leaf rate, a medium leaf rate, a small leaf rate, a fragment rate and a small fragment rate.
Optionally, detecting a cut tobacco structure of each cut tobacco sample after shredding, including:
Detecting the proportion of each tobacco shred structural index in the tobacco shred samples obtained based on the tobacco shred sample shredding, and obtaining the tobacco shred structure of each tobacco shred sample.
Optionally, the obtaining a plurality of tobacco flake samples, detecting a leaf structure of each tobacco flake sample and a shredded tobacco shred structure, and then further includes:
performing correlation analysis on the tobacco shred structural indexes and the leaf structural indexes in each tobacco flake sample based on a correlation analysis method of binary distance variables to obtain correlation coefficients of the tobacco shred structural indexes and the leaf structural indexes;
when the correlation coefficient of the leaf structural index and any tobacco shred structural index is in a first preset range, reserving the tobacco shred structural index and the leaf structural index in the tobacco flake sample;
and when the correlation coefficients of the leaf structural indexes and all the tobacco shred structural indexes are not in the first preset range, removing the leaf structural indexes.
Optionally, when the correlation coefficients of the leaf structural index and all the tobacco shred structural indexes are not in the first preset range, removing the leaf structural index, and then further including:
When the detection index and the characteristic index exist in the reserved blade structure index at the same time, preferentially reserving the detection index;
the detection indexes comprise a large piece rate, a medium piece rate, a small piece rate, a piece rate and a piece breaking rate, and the characteristic indexes comprise a large piece rate, a medium piece rate and a small piece breaking rate.
Optionally, when it is detected that the detection index and the feature index exist in the retained blade structural index at the same time, the detection index is preferentially retained, and then further includes:
carrying out correlation analysis on the reserved blade structural indexes based on a correlation analysis method of binary distance variables to obtain correlation coefficients between every two blade structural indexes;
and when the correlation coefficient between every two blade structural indexes is not in a second preset range, removing the tobacco flake sample corresponding to the correlation coefficient which is not in the second preset range.
Optionally, the combining the score calculation model and the regression model to obtain a tobacco shred structure prediction model of the leaf structure further includes:
and detecting leaf structures of the obtained tobacco leaves to be detected, inputting the leaf structures of the tobacco leaves to be detected into the tobacco shred structure prediction model to predict the tobacco shred structures, and outputting the tobacco shred structures of the tobacco leaves to be detected.
The second aspect of the present application provides a device for constructing a tobacco shred structure prediction model, comprising:
the tobacco shred cutting device comprises an acquisition unit, a cutting unit and a cutting unit, wherein the acquisition unit is used for acquiring a plurality of tobacco flake samples and detecting the leaf structure of each tobacco flake sample and the cut tobacco structure after cutting;
the first analysis unit is used for carrying out principal component analysis on the leaf structures of the tobacco flake samples, selecting a preset number of principal components with the accumulated variance contribution rate being more than or equal to 95% as target principal components, and outputting a score coefficient matrix of the target principal components;
a construction unit, configured to construct a score calculation model of each of the target principal components and the blade structure based on the score coefficient matrix, and calculate a score of each of the target principal components based on the score calculation model;
the second analysis unit is used for carrying out stepwise multiple regression analysis on the scores of the target principal components and the tobacco shred structures and outputting regression models of the scores of the target principal components and the tobacco shred structures;
and the combining unit is used for combining the score calculation model and the regression model to obtain the tobacco shred structure prediction model of the leaf structure.
Optionally, the acquiring unit specifically includes:
The acquisition subunit is used for acquiring a plurality of tobacco flake samples;
the first detection subunit is used for detecting the numerical value of each leaf structural index in each tobacco flake sample to obtain the leaf structure of each tobacco flake sample, wherein the leaf structural indexes comprise a large leaf rate, a medium leaf rate, a small leaf rate, a fragment rate and a small fragment rate;
and the second detection subunit is used for detecting the proportion of the tobacco shred structural indexes in the tobacco shred samples obtained based on the shredding of the tobacco flake samples to obtain the tobacco shred structures of the tobacco flake samples.
Optionally, the method further comprises:
the prediction unit is used for detecting leaf structures of the obtained tobacco leaves to be detected, inputting the leaf structures of the tobacco leaves to be detected into the tobacco shred structure prediction model for tobacco shred structure prediction, and outputting the tobacco shred structures of the tobacco leaves to be detected.
From the above technical scheme, the application has the following advantages:
the application provides a construction method of a tobacco shred structure prediction model, which comprises the following steps: obtaining a plurality of tobacco flake samples, and detecting the leaf structure of each tobacco flake sample and the cut tobacco structure after shredding; carrying out principal component analysis on the leaf structure of each tobacco flake sample, selecting preset number of principal components with accumulated variance contribution rate more than or equal to 95% as target principal components, and outputting a score coefficient matrix of the target principal components; constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculating the score of each target principal component based on the score calculation model; stepwise multiple regression analysis is carried out on the scores of the target main components and the tobacco shred structures, and a regression model of the scores of the target main components and the tobacco shred structures is output; and combining the score calculation model and the regression model to obtain the tobacco shred structure prediction model of the leaf structure.
According to the construction method of the tobacco shred structure prediction model, leaf structure and tobacco shred structure detection are carried out on the obtained tobacco flake samples, and in order to avoid the problem of multiple collinearity among leaf structures, main component analysis is carried out on the leaf structures; in order to ensure the accuracy of a tobacco shred structure prediction model, selecting a preset number of main components with the accumulated variance contribution rate being more than or equal to 95% as target main components, and outputting a scoring coefficient matrix of the target main components; the method comprises the steps of constructing a score calculation model of each target principal component and a leaf structure based on a score coefficient matrix, calculating to obtain the score of each target principal component, performing stepwise multiple regression analysis on the score of each target principal component and a tobacco shred structure to obtain a regression model of the score of each target principal component and the tobacco shred structure, combining the score calculation model and the regression model to obtain a tobacco shred structure prediction model of the leaf structure, accurately detecting the tobacco shred structure corresponding to the leaf structure through the tobacco shred structure prediction model, and detecting that the obtained tobacco shred structure is close to an actual tobacco shred structure, so that the technical problems that the tobacco shred structure of the predicted leaf structure and the actual tobacco shred structure of the leaf in the existing tobacco shred structure prediction method have large difference and low prediction precision are solved.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the application, and that other drawings can be obtained from these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a schematic flow chart of a method for constructing a tobacco shred structure prediction model according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a device for constructing a tobacco shred structure prediction model according to an embodiment of the present application.
Detailed Description
The application provides a construction method and a construction device of a tobacco shred structure prediction model, which are used for solving the technical problem of low prediction precision in the existing tobacco shred structure prediction method.
In order to make the present application better understood by those skilled in the art, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
For easy understanding, referring to fig. 1, an embodiment of a method for constructing a tobacco shred structure prediction model according to the present application includes:
step 101, obtaining a plurality of tobacco flake samples, and detecting the leaf structure of each tobacco flake sample and the cut tobacco structure after shredding.
Preferably, the formulated tobacco flakes of different years, different production places, different parts, different grades, different leaf structures and different combinations thereof are used as the tobacco flake samples. Then, the leaf structure of each tobacco flake sample and the cut tobacco structure after shredding are detected.
102, carrying out principal component analysis on the leaf structure of each tobacco flake sample, selecting a preset number of principal components with the cumulative variance contribution rate being greater than or equal to 95% as target principal components, and outputting a score coefficient matrix of the target principal components.
In order to avoid the problem of multiple collinearity among leaf structures, in the embodiment of the application, a principal component analysis method is adopted to analyze leaf structures of all tobacco flake samples, principal components are extracted, a preset number of principal components with a cumulative variance contribution rate greater than or equal to 95% are selected as target principal components in order to ensure the accuracy of an established tobacco shred structure prediction model, and the cumulative variance contribution rate of 3 principal components is generally greater than or equal to 95%, so that 3 principal components with a cumulative variance contribution rate greater than or equal to 95% are preferably selected as target principal components in the embodiment of the application. And 3 main components are extracted from the blade structure, so that a scoring coefficient matrix of the 3 main components is obtained.
And 103, constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculating the score of each target principal component based on the score calculation model.
And constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, thereby establishing a relation model of the score of each target principal component and the blade structure, and calculating the score of each target principal component through the score calculation model. The fractional coefficient matrix and the blade structure data can be input into SPSS software, and the score calculation model and the score of each target principal component can be obtained through calculation by the SPSS software.
Step 104, stepwise multiple regression analysis is carried out on the scores of the target principal components and the tobacco shred structures, and a regression model of the scores of the target principal components and the tobacco shred structures is output.
The scores of the 3 target main components and the tobacco shred structure data can be input into SPSS software, gradual multiple regression analysis is carried out on the scores of the 3 target main components and the tobacco shred structure through the SPSS software, the score of the target main component with obvious regression effect and the regression model of the tobacco shred structure are output, and therefore a relation model between the tobacco shred structure and the score of the target main component is established.
And 105, combining the score calculation model and the regression model to obtain a tobacco shred structure prediction model of the leaf structure.
Through the steps, a relation model (a score calculation model) between the score of the target main component and the leaf structure and a relation model (a regression model) between the cut tobacco structure and the score of the target main component can be established, so that the relation model between the cut tobacco structure and the leaf structure is indirectly established, and a cut tobacco structure prediction model of the leaf structure can be obtained by combining the two models.
According to the construction method of the tobacco shred structure prediction model, the leaf structure and the tobacco shred structure of the obtained tobacco flake sample are detected, and in order to avoid the problem of multiple collinearity among the leaf structures, the leaf structures are subjected to principal component analysis; in order to ensure the accuracy of a tobacco shred structure prediction model, selecting a preset number of main components with the accumulated variance contribution rate being more than or equal to 95% as target main components, and outputting a scoring coefficient matrix of the target main components; the method comprises the steps of constructing a score calculation model of each target principal component and a leaf structure based on a score coefficient matrix, calculating to obtain the score of each target principal component, performing stepwise multiple regression analysis on the score of each target principal component and a tobacco shred structure to obtain a regression model of the score of each target principal component and the tobacco shred structure, combining the score calculation model and the regression model to obtain a tobacco shred structure prediction model of the leaf structure, accurately detecting the tobacco shred structure corresponding to the leaf structure through the tobacco shred structure prediction model, and detecting that the obtained tobacco shred structure is close to an actual tobacco shred structure, so that the technical problems that the tobacco shred structure of the predicted leaf structure and the actual tobacco shred structure of the leaf in the existing tobacco shred structure prediction method have large difference and low prediction precision are solved.
The above is one embodiment of a method for constructing a tobacco shred structure prediction model provided by the application, and the following is another embodiment of a method for constructing a tobacco shred structure prediction model provided by the application.
The application provides another embodiment of a construction method of a tobacco shred structure prediction model, which comprises the following steps:
step 201, a plurality of tobacco flake samples are obtained, and the leaf structure of each tobacco flake sample and the cut tobacco structure after shredding are detected.
Preferably, the tobacco flakes of the formula of single-material cigarettes and different combinations thereof with different years, different producing places, different parts, different grades and different leaf structures are used as tobacco flake samples, and specific leaf sample information is shown in table 1.
Table 1 tobacco flake sample information
Detecting the numerical value of each leaf structural index in each tobacco flake sample, for example, detecting the large and medium leaf rate, the large leaf rate, the medium leaf rate and the like of each tobacco flake sample, so as to obtain the leaf structure of each tobacco flake sample, wherein the proportion of each leaf structural index in each tobacco flake sample is detected, which belongs to the prior art, and the specific process is not repeated here. The blade structure indexes comprise a large piece rate, a medium piece rate, a small piece rate, a fragment rate and a small fragment rate.
In the embodiment of the application, 3 tobacco flakes are respectively selected from 4 producing places in table 1 to serve as tobacco flake samples, and the tobacco flake samples are subjected to leaf structure detection by adopting a loose and rewet image method, and the detection results are shown in table 2.
Table 2 blade Structure test results
After the leaf structure is detected and recorded, tobacco flake samples are shredded, single-component tobacco flakes can be selected for shredding, the shredding width can be 1.0mm, the tobacco shred samples can be obtained, the proportion of each tobacco shred structural index in the tobacco shred samples is detected, the tobacco shred structures of each tobacco flake sample are obtained, the tobacco shred structural index can be the size of each layer of tobacco shred, each layer of tobacco shred is obtained through screening of sieve holes with different sizes, and then the proportion of each layer of tobacco shred in the tobacco shred is obtained through calculation, wherein the sieve holes can be specifically set according to actual conditions, and specific limitation is not made here. In the embodiment of the application, the tobacco shred structure detection is carried out on tobacco shred samples cut by 4 producing areas according to the method specified in YC/T178-2003 method for measuring the tobacco shred whole tobacco shred rate and the tobacco shred breaking rate, the detection result is shown in Table 3, 4 layers of tobacco shred structures are obtained by screening tobacco shreds by adopting 4 sieve holes, and therefore 4 tobacco shred structure indexes are provided.
TABLE 3 tobacco shred Structure detection results
And 202, carrying out correlation analysis on tobacco shred structural indexes and leaf structural indexes in each tobacco flake sample to obtain correlation coefficients of each tobacco shred structural index and each leaf structural index.
In the embodiment of the application, a correlation analysis method of binary distance variables is adopted to carry out correlation analysis on tobacco shred structural indexes and leaf structural indexes in each tobacco flake sample, so as to obtain correlation coefficients of the tobacco shred structural indexes and leaf structural indexes in each tobacco flake sample, wherein the correlation coefficients comprise Pearson correlation coefficients and significance P values (Sig values). In the embodiment of the present application, the leaf structure of table 2 and each index in the tobacco shred structure of table 3 are subjected to correlation analysis, and the obtained correlation result is referred to table 4.
TABLE 4 correlation between various indices of lamina structure and tobacco shred structure
Step 203, when the correlation coefficient between the leaf structural index and any tobacco shred structural index is within a first preset range, retaining the tobacco shred structural index and leaf structural index in the tobacco flake sample; and when the correlation coefficients of the leaf structural indexes and all tobacco shred structural indexes are not in the first preset range, removing the leaf structural indexes.
When the absolute value of the Pearson correlation coefficient of the leaf structural index and any tobacco shred structural index is more than or equal to 0.500 or the P value of the leaf structural index and any tobacco shred structural index is less than or equal to 0.01, the correlation between the tobacco shred structural index and the leaf structural index is obvious, and the tobacco shred structural index and the leaf structural index in a tobacco flake sample are reserved; when the absolute value of the Pearson correlation coefficient of the leaf structural index and all tobacco shred structural indexes is less than 0.500 or the P value of the leaf structural index and all tobacco shred structural indexes is more than 0.01, the correlation between the tobacco shred structural indexes and the leaf structural indexes is not obvious, and the leaf structural indexes corresponding to the absolute value of the Pearson correlation coefficient is less than 0.500 or the P value is more than 0.01 are removed. As can be seen from table 4, the correlation between the middle sheet rate and each index in the tobacco shred structure is not significant, so that the middle sheet rate in the leaf structure is removed, and the remaining leaf structure indexes are kept for subsequent analysis.
Step 204, when the existence of the detection index and the characteristic index in the reserved blade structure index is detected, the detection index is reserved preferentially.
Detecting whether the detection index and the characteristic index exist in the leaf structural indexes reserved in each tobacco flake sample at the same time, if so, preferentially reserving the detection index to avoid repeated operation of data, otherwise, reserving the existing leaf structural indexes without any treatment. Wherein, the detection indexes comprise a large slice rate, a medium slice rate, a small slice rate, a fragment rate and a fragment rate, and the characteristic indexes comprise a large slice rate, a medium slice rate and a small fragment rate (a small slice rate, a fragment rate and a fragment rate). Referring to table 4, there are both detection indexes (large-scale, small-scale, fragmentation rate and fragmentation rate) and feature indexes (large-medium-scale and small-scale), the small-scale, fragmentation rate and fragmentation rate in the detection indexes are preferentially reserved, the small-scale is removed, and the large-medium-scale (large-scale + medium-scale) in the feature indexes includes detection index information (large-scale) but the detection indexes do not include feature index information (large-medium-scale), so that the integrity of the data is considered to a certain extent, at this time, the large-medium-scale in the feature indexes is reserved, the large-scale in the detection indexes is removed, and therefore, the large-medium-scale, the small-scale, the fragmentation rate and the fragmentation rate are finally reserved for subsequent analysis.
And 205, carrying out correlation analysis on the leaf structure indexes reserved in each tobacco flake sample to obtain correlation coefficients between every two leaf structure indexes.
The leaf structure indexes have strong correlation, and the leaf structure indexes are directly adopted to construct the relationship with the tobacco shred structure, so that the problem of multiple collinearity exists. The precondition of constructing the tobacco shred structure prediction model by adopting the principal component analysis method is that the leaf structures have strong correlation, and if the principal component of the leaf structures is directly extracted by adopting the principal component analysis method to construct the tobacco shred structure prediction model, the principal component construction among leaf structures with weak correlation is likely to be extracted to obtain the tobacco shred structure prediction model with low accuracy.
Therefore, in the embodiment of the application, the suitability of the blade structure is checked before the main component is extracted. Specifically, a correlation analysis method of binary distance variables is adopted to conduct correlation analysis on leaf structure indexes reserved in each tobacco flake sample, and correlation coefficients between every two leaf structure indexes are obtained, wherein the correlation coefficients comprise Pearson correlation coefficients and significance P values.
And 206, removing the tobacco flake samples corresponding to the correlation coefficients which are not in the second preset range when the correlation coefficients between every two blade structure indexes are not in the second preset range.
When the absolute value of the Pearson correlation coefficient between every two leaf structure indexes in the tobacco flake sample is more than or equal to 0.500 or the P value is less than or equal to 0.01, the correlation relationship between leaf structures of the tobacco flake sample is stronger, and the tobacco flake sample is suitable for extracting main components by adopting a main component analysis method; when the absolute value of the Pearson correlation coefficient between every two leaf structural indexes in the tobacco flake sample is less than 0.500 or the P value is more than 0.01, the tobacco flake sample is not suitable for the principal component analysis method, and the tobacco flake sample is removed. In the embodiment of the application, the leaf structural index reserved for a certain tobacco leaf sample is subjected to correlation analysis, and the analysis result is shown in table 5.
TABLE 5 correlation between blade structural indicators
As can be seen from Table 5, there is a very significant correlation between the blade structure indices, so that the method is suitable for extracting the principal component by using the principal component analysis method, so as to avoid the problem of multiple collinearity between the blade structures.
Step 207, performing principal component analysis on the leaf structure reserved by each tobacco flake sample, selecting a preset number of principal components with the cumulative variance contribution rate being greater than or equal to 95% as target principal components, and outputting a score coefficient matrix of the target principal components.
In order to avoid the problem of multiple collinearity among leaf structures, in the embodiment of the application, a principal component analysis method is adopted to analyze leaf structures of all tobacco flake samples, principal components are extracted, and in order to ensure the accuracy of an established tobacco shred structure prediction model, a preset number of principal components with the cumulative variance contribution rate of more than or equal to 95% are selected as target principal components. In the embodiment of the application, principal component analysis is performed on the blade structure, principal components are extracted, the variance contribution rate of each principal component is shown in table 6, and the cumulative variance contribution rate of the extracted 3 principal components is more than 95%, so that 3 principal components with the cumulative variance contribution rate greater than or equal to 95% are selected as target principal components in the embodiment of the application, namely, 3 new variable substitution blade structures are constructed. Tables 7 and 8 show the component matrices of the extracted 3 main components.
TABLE 6 variance contribution ratio of principal components
TABLE 7 component matrix
TABLE 8 component matrix after rotation
The principal component extraction is performed by SPSS software to obtain a score coefficient matrix of 3 target principal components by extracting principal components from the leaf structure, and the score coefficient matrix of the principal components is output, please refer to table 9.
TABLE 9 scoring coefficient matrix for principal components
And step 208, constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculating the score of each target principal component based on the score calculation model.
And constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, thereby establishing a relation model of the score of each target principal component and the blade structure, and calculating the score of each target principal component through the score calculation model. The fractional coefficient matrix and the blade structure data can be input into SPSS software, and the score calculation model and the score of each target principal component can be obtained through calculation by the SPSS software. Specifically, the score calculation model of each target principal component is:
wherein Y is j Calculating a model for the score of the jth target principal component, j=1, 2,3; alpha ij The scoring coefficient of the jth target principal component of the blade structure index i, n is the number of the reserved blade structure indexes and X i As the structural index i of the blade,for the mean value of all leaf structural indexes i in all tobacco leaf samples, +.>Standard deviation for all leaf structure index i in all tobacco leaf samples.
Using the previous example, 12 large and medium sheet rates X are calculated from the 12 sheet samples of the 4 production places 1 Mean of (2)77.57, standard deviation->6.37; chip rate X 2 Mean>16.78, standard deviation->3.92; fraction X 3 Mean>5.62, standard deviation->2.67; rate of shatter of X 4 Mean>0.025, standard deviation->0.012.
Based on the above parameters, and in combination with the score coefficients of table 9, the score calculation model of the first target principal component can be obtained as follows:
Y 1 ＝-0.213*(X 1 -77.57)/6.37-0.668*(X 2 -16.78)/3.92+1.495*(X 3 -5.62)/2.67-0.295*(X 4 -0.025)/0.012；
the score calculation model of the second target principal component is:
Y 2 ＝-0.338*(X 1 -77.57)/6.37-0.365*(X 2 -16.78)/3.92-0.275*(X 3 -5.62)/2.67+1.727*(X 4 -0.025)/0.012；
the score calculation model of the third target principal component is:
Y 3 ＝-0.978*(X 1 -77.57)/6.37+2.695*(X 2 -16.78)/3.92-1.630*(X 3 -5.62)/2.67-1.887*(X 4 -0.025)/0.012。
and correspondingly inputting the large and medium piece rate, the small piece rate, the piece rate and the piece rate in the leaf structure indexes in the 12 tobacco leaf samples of the 4 producing places into each score calculation model to correspondingly obtain the score of each target main component, and referring to Table 10.
TABLE 10 score of target principal component
Step 209, performing stepwise multiple regression analysis on the score of each target principal component and the tobacco shred structure, and outputting a regression model of the score of the target principal component and the tobacco shred structure.
The scores of the 3 target principal components and the tobacco shred structure data can be input into SPSS software, gradual multiple regression analysis is carried out on the scores of the 3 target principal components and the tobacco shred structure through the SPSS software, the scores of the target principal components with obvious regression effects and a regression model of the tobacco shred structure are output, and therefore a relation model between the tobacco shred structure and the scores of the target principal components is established, and the regression model is a linear regression model composed of coefficients and constant coefficients of the target principal components of the model with the best regression fitting effect.
According to the scores of the target principal components in table 10, stepwise multiple regression analysis is performed on the scores of the 3 target principal components and the layers of the tobacco shred structure, and the regression analysis results are shown in tables 11 to 22.
(1) Regression model of score of first layer tobacco shred structure and target principal component
As can be seen from table 11, the correlation coefficient of the score of the first target principal component and the first layer tobacco shred structure is 0.918, the determination coefficient is 0.828, which indicates that the analysis result of the score of the first target principal component and the first layer tobacco shred structure regression analysis (model 1) is remarkable; the correlation coefficient of the score of the second target principal component and the first layer tobacco shred structure is 0.972, the determination coefficient is 0.944, and the regression effect of the score of the second target principal component and the regression analysis (model 2) of the first layer tobacco shred structure is better than that of the model 1. Since the scores of the first, second, and third target principal components are not significant as the analysis result of the cut tobacco structure regression analysis (model 3), i.e., the P value is not significant, i.e., model 3 is the same as model 2, model 3 is excluded. In model 2 and model 3, the fitting effect of model 2 is better, and therefore, the coefficients of model 2 are used to build the final regression model.
From the analysis of variance results in table 12, it is clear that the P value of the linear regression reaches an extremely significant level, which indicates that the score of the first target principal component and the score of the second target principal component have good fitting effect with the first layer tobacco shred structure. As can be seen from the parameters of the regression model of Table 13, the first target principal component score of model 2 had a coefficient of-7.733, the second target principal component score had a coefficient of-2.677, and the constant coefficient had a coefficient of 59.526. Therefore, the regression model of the score of the first layer tobacco shred structure and the target principal component is:
Z 1 ＝-7.733Y 1 -2.677Y 2 +59.526；
TABLE 11 score of target principal component and regression analysis result of first layer cut tobacco structure
TABLE 12 score of target principal component and analysis of variance of regression analysis of first layer tobacco cut structure
TABLE 13 score of target principal component and first layer tobacco shred structural regression model parameters
/>
(2) Regression model of score of second-layer tobacco shred structure and target principal component
Stepwise regression analysis is performed on the score of the target main component and the second-layer tobacco shred structure, the analysis result is shown in table 14, the correlation coefficient of the model 2 is 0.963, the decision coefficient is 0.926, and the regression effect is better than that of the model 1, so that the regression model is constructed by adopting the correlation coefficient of the model 2.
From the analysis of variance results in table 15, it is clear that the P value of the linear regression reaches an extremely significant level, indicating that the score of the first target principal component, the score of the second target principal component and the second layer tobacco shred structure fit effect are good. As can be seen from the parameters of the regression model in table 16, the linear fitting effect of the score of the first target principal component and the score of the second target principal component is good, and therefore, the regression model of the score of the second layer tobacco shred structure and the target principal component is:
Z 2 ＝4.706*Y 1 +1.200*Y 2 +21.918
TABLE 14 score of target principal component and regression analysis result of second layer tobacco shred structure
| Model | R | R square | R side after adjustment | Error of standard estimation |
| 1 | 0.933 a | 0.870 | 0.857 | 1.90861 |
| 2 | 0.963 b | 0.926 | 0.910 | 1.51216 |
TABLE 15 score of target principal component and analysis of variance of regression analysis of second-layer tobacco shred structure
TABLE 16 score of target principal component and second-layer tobacco shred structural regression model parameters
(3) Regression model of score of third layer tobacco shred structure and target principal component
Stepwise regression analysis was performed on the score of the target principal component and the third-layer tobacco cut filler structure, and the analysis results are shown in table 17, wherein the correlation coefficient of model 2 was 0.961, the determination coefficient was 0.924, and the regression effect was better than that of model 1. From the analysis of variance results in table 18, it is clear that the P value of the linear regression reaches an extremely significant level, which indicates that the score of the first target principal component, the score of the second target principal component and the third layer tobacco shred structure fit effect is good. As can be seen from the parameters of the regression model in table 19, the linear fitting effect of the score of the first target principal component and the score of the second target principal component is good, and therefore, the regression model of the score of the third layer tobacco shred structure and the target principal component is:
Z 3 ＝2.867*Y 1 +1.368*Y 2 +17.716；
table 17 score of target principal component and regression analysis result of third layer cut tobacco structure
| Model | R | R square | R side after adjustment | Error of standard estimation |
| 1 | 0.868 a | 0.753 | 0.728 | 1.72388 |
| 2 | 0.961 b | 0.924 | 0.907 | 1.00783 |
TABLE 18 score of target principal component and analysis of variance of regression analysis of third layer tobacco cut structure
TABLE 19 score of target principal component and third layer tobacco shred structural regression model parameters
(4) Regression model of score of fourth-layer tobacco shred structure and target principal component
Stepwise regression analysis is performed on the score of the target main component and the fourth layer tobacco shred structure, the analysis result is shown in table 20, the correlation coefficient of the model 1 is 0.702, the decision coefficient is 0.493, and the regression effect is good. From the analysis of variance results in table 21, the P value of the linear regression was close to the very significant level, indicating that the score of the first target principal component was better fitted to the fourth layer tobacco structure. As can be seen from the parameters of the regression model in Table 22, the score line fitting effect of the first target principal component is good, and therefore, the regression model of the score of the fourth layer tobacco shred structure and the target principal component is:
Z 4 ＝0.085*Y 1 +0.768；
table 20 score of target principal component and regression analysis result of fourth layer tobacco shred structure
| Model | R | R square | R side after adjustment | Error of standard estimation |
| 1 | 0.702 a | 0.493 | 0.442 | 0.08991 |
TABLE 21 score of target principal component and analysis of variance of regression analysis of fourth layer tobacco shred structure
Table 22 shows score of principal component and fourth layer tobacco shred structure regression model
And 210, combining the score calculation model and the regression model to obtain a tobacco shred structure prediction model of the leaf structure.
Through the steps, a relation model (a score calculation model) between the score of the target main component and the leaf structure and a relation model (a regression model) between the tobacco shred structure and the score of the target main component can be established, so that the relation model between the tobacco shred structure and the leaf structure is indirectly established, and a tobacco shred structure prediction model of the leaf structure can be obtained by combining the two models, namely, the score calculation model is substituted into the regression model, and the tobacco shred structure prediction model of the leaf structure is obtained.
In the embodiment of the application, the tobacco shred structure prediction model of the obtained leaf structure is also verified, two groups of tobacco sheets of Sichuan and Hunan are respectively selected as verification samples, the leaf structure of the verification samples is shown in a table 23, and the actual value of the tobacco shred structure and the predicted value predicted by the tobacco shred structure prediction model are shown in a table 24, and the absolute deviation range of the predicted value and the actual value is 0.19% -1.184% from the table 24.
Table 23 leaf structure of Sichuan and Hunan tobacco flakes
Table 24 verification results of tobacco shred Structure prediction model based on lamina Structure
Step 211, carrying out leaf structure detection on the obtained tobacco leaves to be detected, inputting the leaf structure of the tobacco leaves to be detected into a tobacco shred structure prediction model to carry out tobacco shred structure prediction, and outputting the tobacco shred structure of the tobacco leaves to be detected.
And obtaining a tobacco sheet needing to be detected in the tobacco shred structure as a leaf sample to be detected, carrying out leaf structure detection on the tobacco sheet, obtaining leaf structure data, inputting the leaf structure data into a tobacco shred structure prediction model to carry out tobacco shred structure prediction, and outputting the tobacco shred structure of the tobacco sheet to be detected.
The above is an embodiment of a method for constructing a tobacco shred structure prediction model provided by the present application, and the following is a specific application example of the method for constructing a tobacco shred structure prediction model provided by the present application.
1. Sample preparation
The tobacco flake samples are selected as middle upper waiting tobacco modules after threshing and redrying in different producing areas in 2017, and specific information is shown in table 25.
Table 25 tobacco flake sample information
2. Detecting blade structure
Blade structure detection is carried out by adopting a loose and rewet image method, and the detection results are shown in table 26.
Blade structure of watch 26
3. Cut tobacco detecting structure
Tobacco flakes of single component are cut into shreds with a shredding width of 1.0mm, and after shredding, the tobacco shred structures are detected by using a vibrating screen Retsch AS400 (Germany), and the results are shown in table 27, wherein the sizes of the screening holes are 8, and the number of corresponding tobacco shred structure indexes is 8 in the embodiment of the application.
Surface 27 cut tobacco structure
4. Screening blade structural index
(1) The leaf structure indexes of the 4 producing areas and the tobacco shred structure indexes are subjected to correlation analysis, the analysis results are shown in table 28, and according to the table 28, the correlation between the middle cut rate and the ratio of the tobacco shred structures of each layer is not obvious, so that the middle cut rate is removed, and the other indexes are reserved and enter the subsequent analysis.
Table 28 correlation between leaf Structure index and tobacco shred Structure index
/>
(2) Among the reserved blade structural indexes, detection indexes (large-piece rate, small-piece rate, fragment rate and fragment rate) and the characteristic indexes constructed by the detection indexes (large-piece rate, medium-piece rate, small-piece rate (small-piece rate, fragment rate and fragment rate)) are simultaneously present, so that repeated operation of data is avoided, the integrity of the data is considered, the large-piece rate and the small-piece rate are removed, and the large-piece rate, the medium-piece rate, the small-piece rate, the fragment rate and the fragment rate are reserved to participate in subsequent analysis.
5. Checking and judging suitability of principal component analysis
The retained blade structure is subjected to correlation analysis, the analysis results are shown in tables 29 and 30, and the blade structure indexes shown in the results have extremely obvious correlation relation, so that the method is suitable for constructing new variables through a principal component analysis method, and the problem of multiple collinearity among the blade structures is avoided.
TABLE 29 correlation between blade structural indicators
/>
TABLE 30 correlation between blade structural indicators
/>
6. Construction of new variables using principal component analysis
The results of KMO and butralite sphericity tests on the leaf structure are shown in table 31, and the results show very significant significance, indicating that the leaf structure is suitable for principal component analysis.
Table 31 KMO and Bartlite sphericity test results
Three principal components are extracted from the blade structure, and the cumulative variance contribution rate (table 32) of the three principal components reaches 100% (> 95.00%), so that modeling accuracy can be ensured, and therefore, the three principal components are extracted from the blade structure, namely, three new variables are constructed to replace the blade structure index. Tables 33 and 34 are component matrices for extracting three main components.
Table 32 variance contribution ratio of principal component
TABLE 33 component matrix
/>
Table 34 rotated component matrix
7. Calculating the score of the target principal component
The principal component extraction is performed by SPSS software to obtain a score coefficient matrix of 3 target principal components by extracting principal components from the leaf structure, and the score coefficient matrix of the principal components is output, please refer to table 35.
Table 35 scoring coefficient matrix of principal component
The score calculation model for calculating the first target principal component is as follows:
Y 1 ＝-0.213*(X 1 -77.57)/6.37-0.668*(X 2 -16.78)/3.92+1.495*(X 3 -5.62)/2.67-0.295*(X 4 -0.025)/0.012；
The score calculation model of the second target principal component is:
Y 2 ＝-0.338*(X 1 -77.57)/6.37-0.365*(X 2 -16.78)/3.92-0.275*(X 3 -5.62)/2.67+1.727*(X 4 -0.025)/0.012；
the score calculation model of the third target principal component is:
Y 3 ＝-0.978*(X 1 -77.57)/6.37+2.695*(X 2 -16.78)/3.92-1.630*(X 3 -5.62)/2.67-1.887*(X 4 -0.025)/0.012。
wherein X is 1 -large and medium sheet rate, unit: the%;
X 2 -platelet rate, unit: the%;
X 3 -fragmentation rate, unit: the%;
X 4 -fines ratio, unit: the%;
and correspondingly inputting the large and medium piece rate, the small piece rate, the piece rate and the piece rate in the blade structure indexes in the 12 tobacco flake samples of the 4 producing places into each score calculation model to correspondingly obtain the score of each target main component, and referring to a table 36.
TABLE 36 score of target principal component
| Numbering device | First target principal component | Second target principal component | Third target principal component |
| 1 | 1.52 | 1.08 | -1.29 |
| 2 | 1.65 | -0.45 | 0.73 |
| 3 | 1.69 | -0.43 | 0.58 |
| 4 | -0.39 | 1.30 | 1.31 |
| 5 | -0.81 | 1.40 | 1.66 |
| 6 | -0.82 | 1.22 | -1.49 |
| 7 | -0.76 | -1.16 | 0.24 |
| 8 | -0.52 | -1.29 | 0.33 |
| 9 | -0.74 | -1.05 | -0.23 |
| 10 | -0.27 | -0.13 | -0.92 |
| 11 | -0.37 | -0.10 | -0.86 |
| 12 | -0.19 | -0.38 | -0.06 |
8. Establishing regression model of score of tobacco shred structure and target principal component
(1) Regression model of score of first layer tobacco shred structure and target principal component
The scores of the 3 target principal components were subjected to stepwise multiple regression analysis with the tobacco shred structural index, and as shown in table 37, the correlation coefficient between the score of the first target principal component and the tobacco shred structure of the first layer was 0.842, and the determination coefficient was 0.710, which indicates that the score of the first target principal component and the analysis result of the tobacco shred structure regression analysis (model 1) of the first layer were extremely remarkable. From the analysis of variance results in table 38, it can be seen that the P value of the linear regression reaches a very significant level, indicating that the score of the first target principal component has a good fit to the first layer tobacco structure. According to the parameters of the regression model in table 39, the score of the first target principal component and the coefficient of the constant are all tested to reach the very significant level, so the regression model of the score of the first layer tobacco shred structure and the first target principal component is:
H 1 ＝-8.941*Y 1 +18.860；
Table 37 regression analysis results of the target principal component and first layer tobacco shred Structure
Table 38 analysis of variance results of regression analysis of the target principal component and the first layer tobacco shred structure
Table 39 main objective component and first layer tobacco shred structure regression model parameters
(2) Regression model of scores of second and third-layer tobacco shred structures and target principal components
And (3) carrying out stepwise multiple regression analysis on the scores of the 3 target main components and the second layer and the third layer of tobacco shred structures respectively, wherein the regression analysis results cannot meet the condition of an accurate regression model because the correlation between the second layer of tobacco shred structures and the third layer of tobacco shred structures and the leaf structures is weak, so that the regression model is not established for the second layer of tobacco shred structures and the third layer of tobacco shred structures. When the tobacco shred structural index and the leaf structural index are subjected to correlation analysis to screen leaf structural indexes, the tobacco shred structural index can be screened, when the absolute value of the Person correlation coefficient of a certain tobacco shred structural index and all leaf structural indexes is less than 0.5 or the P value is more than 0.1, the tobacco shred structural index is removed from participating in subsequent modeling, meanwhile, the tobacco shred structural index classification method can be used for checking whether the tobacco shred structural index classification is reasonable or not, and when the correlation of the tobacco shred structural index and all leaf structural indexes is not high, the tobacco shred structural index classification can be considered again.
(3) Regression model of score of fourth-layer tobacco shred structure and target principal component
Stepwise multiple regression analysis is performed on the scores of the 3 target principal components and the fourth-layer tobacco shred structure respectively, and according to tables 40-42, the correlation coefficient of the score of the first target principal component and the fourth-layer tobacco shred structure is 0.768, the determination coefficient is 0.590, which indicates that the fitting effect of the score of the first target principal component and the fourth-layer tobacco shred structure is good. The regression model of the score of the fourth layer tobacco shred structure and the first target main component is as follows:
H 4 ＝0.480*Y 1 +8.063；
table 40 score of target principal component and regression analysis result of fourth layer tobacco shred structure
Table 41 score of target principal component and analysis of variance of regression analysis of fourth layer tobacco shred structure
Table 42 score of target principal component and fourth layer tobacco shred structural regression model parameters
(4) Regression model of score of fifth-layer tobacco shred structure and target principal component
Stepwise multiple regression analysis is performed on the scores of the 3 target principal components and the fifth-layer tobacco shred structure respectively, and according to tables 43-45, the correlation coefficient of the score of the first target principal component and the fifth-layer tobacco shred structure is 0.841, and the determination coefficient is 0.707, which indicates that the score of the first target principal component and the fifth-layer tobacco shred structure have good fitting effect. The regression model of the score of the fifth tobacco shred structure and the first target main component is as follows:
H 5 ＝1.501*Y 1 +14.323；
Table 43 score of target principal component and regression analysis result of fifth layer tobacco shred structure
Table 44 analysis of variance results of regression analysis of the score of the target principal component and the fifth-layer tobacco shred structure
Table 45 score of target principal component and fifth layer tobacco shred structural regression model parameters
(5) Regression model of score of sixth-layer tobacco shred structure and target principal component
The scores of the 3 target main components are respectively subjected to stepwise multiple regression analysis with the sixth-layer tobacco shred structure, and according to tables 46-48, the results of regression models established by the scores of the first target main component and the second target main component and the sixth-layer tobacco shred structure are good, the correlation coefficient of the models is 0.945, the decision coefficient is 0.894, and the variance analysis result shows that the regression effects are very remarkable. The regression model of the score of the sixth layer tobacco shred structure and the first target principal component and the score of the second target principal component is as follows:
H 6 ＝5.244*Y 1 +1.602*Y 2 +23.646；
table 46 score of target principal component and regression analysis result of sixth layer tobacco shred structure
Table 47 score of target principal component and analysis of variance of regression analysis of sixth layer tobacco shred structure
Table 48 score of target principal component and sixth layer tobacco shred structural regression model parameters
(6) Regression model of score of seventh-layer tobacco shred structure and target principal component
The scores of the 3 target principal components are respectively subjected to stepwise multiple regression analysis with the sixth-layer tobacco shred structure, and according to tables 49-51, the score of the first target principal component and the regression model established by the seventh-layer tobacco shred structure have good effect, the correlation coefficient of the model is 0.891, the decision coefficient is 0.794, and the analysis of variance results show that the regression effect is very remarkable. The regression model of the score of the seventh layer tobacco shred structure and the first target main component is as follows:
H 7 ＝1.656*Y 1 +4.884；
Table 49 score of target principal component and seventh layer tobacco shred structural regression analysis result
Table 50 score of target principal component and analysis of variance of regression analysis of seventh layer tobacco shred structure
Table 51 score of target principal component and seventh layer tobacco shred structural regression model parameters
(6) Regression model of score of eighth layer tobacco shred structure and target principal component
The scores of the 3 target main components are respectively subjected to stepwise multiple regression analysis with the sixth-layer tobacco shred structure, and according to tables 52-54, the score of the first target main component and the regression model established by the eighth-layer tobacco shred structure have good effect, the correlation coefficient of the model is 0.750, the decision coefficient is 0.563, and the analysis of variance result shows that the regression effect is very remarkable. The regression model of the score of the eighth layer tobacco shred structure and the first target main component is as follows:
H 8 ＝0.045*Y 1 +0.151；
table 52 score of target principal component and regression analysis result of eighth layer tobacco shred structure
Table 53 score of target principal component and analysis of variance of regression analysis of eighth layer tobacco shred structure
Table 54 score of target principal component and eighth layer tobacco shred structural regression model parameters
9. Tobacco shred structure prediction model for establishing leaf structure
(1) Score calculation model
The score calculation model of the first target principal component is:
Y 1 ＝-0.213*(X 1 -77.57)/6.37-0.668*(X 2 -16.78)/3.92+1.495*(X 3 -5.62)/2.67-0.295*(X 4 -0.025)/0.012；
The score calculation model of the second target principal component is:
Y 2 ＝-0.338*(X 1 -77.57)/6.37-0.365*(X 2 -16.78)/3.92-0.275*(X 3 -5.62)/2.67+1.727*(X 4 -0.025)/0.012；
the score calculation model of the third target principal component is:
Y 3 ＝-0.978*(X 1 -77.57)/6.37+2.695*(X 2 -16.78)/3.92-1.630*(X 3 -5.62)/2.67-1.887*(X 4 -0.025)/0.012。
(2) Regression model
A first layer: h 1 ＝-8.941*Y 1 +18.860；
A second layer: the method is free;
third layer: the method is free;
fourth layer: h 4 ＝0.480*Y 1 +8.063；
Fifth layer: h 5 ＝1.501*Y 1 +14.323；
Sixth layer: h 6 ＝5.244*Y 1 +1.602*Y 2 +23.646；
Seventh layer: h 7 ＝1.656*Y 1 +4.884；
Eighth layer: h 8 ＝0.045*Y 1 +0.151。
And combining the score calculation model and the regression model to obtain the tobacco shred structure prediction model of the leaf structure.
10. Model verification
Two groups of leaves of Sichuan and Hunan are selected respectively, the leaf structures obtained by detection are shown in a table 55, after the leaf structures are cut according to the cutting width of 1.0mm, the two groups of tobacco structures are detected by using a vibrating screen RetschAS400 (Germany), the results are shown in a table 56, and the absolute deviation range of a predicted value and a true value is 0.02% -6.87% as can be seen from the verification result of a tobacco structure prediction model in the table 56. Therefore, the tobacco shred structure prediction model in the embodiment of the application has higher accuracy.
Table 55 blade structure
Table 56 tobacco shred structure prediction model verification result based on leaf structure
The above is a specific application example of the method for constructing a tobacco shred structure prediction model provided by the application, and the following is an embodiment of the device for constructing a tobacco shred structure prediction model provided by the application.
Referring to fig. 2, an embodiment of a device for constructing a tobacco shred structure prediction model provided by the present application includes:
the acquisition unit 301 is configured to acquire a plurality of tobacco flake samples, and detect a leaf structure of each tobacco flake sample and a shredded tobacco shred structure.
The first analysis unit 302 is configured to perform principal component analysis on the leaf structure of each smoke sheet sample, select a preset number of principal components with a cumulative variance contribution rate greater than or equal to 95% as target principal components, and output a score coefficient matrix of the target principal components.
And a construction unit 303, configured to construct a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculate a score of each target principal component based on the score calculation model.
And the second analysis unit 304 is configured to perform stepwise multiple regression analysis on the analysis of each target principal component and the tobacco shred structure, and output a regression model of the analysis of the target principal component and the tobacco shred structure.
And the combining unit 305 is configured to combine the score calculation model and the regression model to obtain a tobacco shred structure prediction model of the leaf structure.
The above is an embodiment of a device for constructing a tobacco shred structure prediction model provided by the present application, and the following is another embodiment of a device for constructing a tobacco shred structure prediction model provided by the present application, including:
The acquisition unit 401 is configured to acquire a plurality of tobacco flake samples, and detect a leaf structure of each tobacco flake sample and a shredded tobacco shred structure.
The first analysis unit 402 is configured to perform principal component analysis on a leaf structure of each smoke sheet sample, select a preset number of principal components with a cumulative variance contribution rate greater than or equal to 95% as target principal components, and output a score coefficient matrix of the target principal components.
A construction unit 403, configured to construct a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculate a score of each target principal component based on the score calculation model.
And the second analysis unit 404 is configured to perform stepwise multiple regression analysis on the analysis of each target principal component and the tobacco shred structure, and output a regression model of the analysis of the target principal component and the tobacco shred structure.
And the combining unit 405 is configured to combine the score calculation model and the regression model to obtain a tobacco shred structure prediction model of the leaf structure.
As a further improvement, the acquisition unit 401 specifically includes:
an acquisition subunit 4011, configured to acquire a plurality of tobacco flake samples;
the first detection subunit 4012 is configured to detect a value of each vane structural index in each tobacco flake sample to obtain a vane structure of each tobacco flake sample, where the vane structural indexes include a large-medium flake rate, a large flake rate, a medium flake rate, a small flake rate, a fragment rate and a small fragment rate;
The second detecting subunit 4013 is configured to detect a proportion of each tobacco shred structural index in a tobacco shred sample obtained based on shredding the tobacco shred sample, so as to obtain a tobacco shred structure of each tobacco shred sample.
As a further improvement, further comprising:
the third analysis unit 406 is configured to perform correlation analysis on the tobacco shred structural index and the leaf structural index in each tobacco flake sample based on a correlation analysis method of binary distance variables, so as to obtain correlation coefficients of each tobacco shred structural index and each leaf structural index;
a first retaining unit 407, configured to retain the tobacco shred structural index and the leaf structural index in the tobacco flake sample when the correlation coefficient between the leaf structural index and any tobacco shred structural index is within a first preset range;
the first removing unit 408 is configured to remove the leaf structural index when the correlation coefficients of the leaf structural index and all tobacco shred structural indexes are not within the first preset range.
As a further improvement, further comprising:
a second reserving unit 409 for reserving the detection index preferentially when detecting that the detection index and the feature index coexist in the reserved blade structure index;
the detection indexes comprise a large piece rate, a medium piece rate, a small piece rate, a piece rate and a piece breaking rate, and the characteristic indexes comprise a large piece rate, a medium piece rate and a small piece breaking rate.
As a further improvement, further comprising:
a fourth analysis unit 410, configured to perform correlation analysis on the retained blade structural indexes based on a correlation analysis method of binary distance variables, so as to obtain correlation coefficients between every two blade structural indexes;
the second removing unit 411 is configured to remove, when the correlation coefficients between every two blade structural indexes are not within the second preset range, the tobacco flake samples corresponding to the correlation coefficients not within the second preset range.
As a further improvement, further comprising:
the prediction unit 412 is configured to perform leaf structure detection on the obtained tobacco leaf to be detected, input the leaf structure of the tobacco leaf to be detected to the tobacco shred structure prediction model to perform tobacco shred structure prediction, and output the tobacco shred structure of the tobacco leaf to be detected.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. The construction method of the tobacco shred structure prediction model is characterized by comprising the following steps of:
obtaining a plurality of tobacco flake samples, and detecting the leaf structure of each tobacco flake sample and the cut tobacco structure after shredding;
carrying out principal component analysis on the leaf structures of the tobacco flake samples, selecting a preset number of principal components with accumulated variance contribution rate greater than or equal to 95% as target principal components, and outputting a scoring coefficient matrix of the target principal components;
constructing a score calculation model of each target principal component and the blade structure based on the score coefficient matrix, and calculating the score of each target principal component based on the score calculation model;
step-by-step multiple regression analysis is carried out on the score of each target principal component and the tobacco shred structure, and a regression model of the score of each target principal component and the tobacco shred structure is output;
and combining the score calculation model and the regression model to obtain the tobacco shred structure prediction model of the leaf structure.
2. The method of constructing a tobacco shred structure prediction model according to claim 1, wherein detecting a leaf structure of each of the tobacco shred samples comprises:
and detecting the numerical value of each leaf structural index in each tobacco flake sample to obtain the leaf structure of each tobacco flake sample, wherein the leaf structural indexes comprise a large leaf rate, a medium leaf rate, a small leaf rate, a fragment rate and a small fragment rate.
3. The method for constructing a tobacco shred structure prediction model according to claim 2, wherein detecting the shredded tobacco shred structure of each of the tobacco shred samples comprises:
detecting the proportion of each tobacco shred structural index in the tobacco shred samples obtained based on the tobacco shred sample shredding, and obtaining the tobacco shred structure of each tobacco shred sample.
4. The method for constructing a tobacco shred structure prediction model according to claim 3, wherein the steps of obtaining a plurality of tobacco shred samples, detecting leaf structures of the tobacco shred samples and shredded tobacco shred structures, and then further comprise:
performing correlation analysis on the tobacco shred structural indexes and the leaf structural indexes in each tobacco flake sample based on a correlation analysis method of binary distance variables to obtain correlation coefficients of the tobacco shred structural indexes and the leaf structural indexes;
when the correlation coefficient of the leaf structural index and any tobacco shred structural index is in a first preset range, reserving the tobacco shred structural index and the leaf structural index in the tobacco flake sample;
and when the correlation coefficients of the leaf structural indexes and all the tobacco shred structural indexes are not in the first preset range, removing the leaf structural indexes.
5. The method for constructing a tobacco shred structure prediction model according to claim 4, wherein when the correlation coefficients of the leaf structure index and all the tobacco shred structure indexes are not within the first preset range, removing the leaf structure index, and further comprising:
when the detection index and the characteristic index exist in the reserved blade structure index at the same time, preferentially reserving the detection index;
the detection indexes comprise a large piece rate, a medium piece rate, a small piece rate, a piece rate and a piece breaking rate, and the characteristic indexes comprise a large piece rate, a medium piece rate and a small piece breaking rate.
6. The method for constructing a tobacco shred structure prediction model according to claim 5, wherein when it is detected that there are both a detection index and a feature index in the retained leaf structure index, the detection index is preferentially retained, and further comprising:
carrying out correlation analysis on the reserved blade structural indexes based on a correlation analysis method of binary distance variables to obtain correlation coefficients between every two blade structural indexes;
and when the correlation coefficient between every two blade structural indexes is not in a second preset range, removing the tobacco flake sample corresponding to the correlation coefficient which is not in the second preset range.
7. The method for constructing a tobacco shred structure prediction model according to claim 1, wherein the combining the score calculation model and the regression model to obtain the tobacco shred structure prediction model of the leaf structure further comprises:
and detecting leaf structures of the obtained tobacco leaves to be detected, inputting the leaf structures of the tobacco leaves to be detected into the tobacco shred structure prediction model to predict the tobacco shred structures, and outputting the tobacco shred structures of the tobacco leaves to be detected.
8. The construction device of the tobacco shred structure prediction model is characterized by comprising the following components:
the tobacco shred cutting device comprises an acquisition unit, a cutting unit and a cutting unit, wherein the acquisition unit is used for acquiring a plurality of tobacco flake samples and detecting the leaf structure of each tobacco flake sample and the cut tobacco structure after cutting;
the first analysis unit is used for carrying out principal component analysis on the leaf structures of the tobacco flake samples, selecting a preset number of principal components with the accumulated variance contribution rate being more than or equal to 95% as target principal components, and outputting a score coefficient matrix of the target principal components;
a construction unit, configured to construct a score calculation model of each of the target principal components and the blade structure based on the score coefficient matrix, and calculate a score of each of the target principal components based on the score calculation model;
The second analysis unit is used for carrying out stepwise multiple regression analysis on the scores of the target principal components and the tobacco shred structures and outputting regression models of the scores of the target principal components and the tobacco shred structures;
and the combining unit is used for combining the score calculation model and the regression model to obtain the tobacco shred structure prediction model of the leaf structure.
9. The device for constructing a tobacco shred structure prediction model according to claim 8, wherein the obtaining unit specifically comprises:
the acquisition subunit is used for acquiring a plurality of tobacco flake samples;
the first detection subunit is used for detecting the numerical value of each leaf structural index in each tobacco flake sample to obtain the leaf structure of each tobacco flake sample, wherein the leaf structural indexes comprise a large leaf rate, a medium leaf rate, a small leaf rate, a fragment rate and a small fragment rate;
and the second detection subunit is used for detecting the proportion of the tobacco shred structural indexes in the tobacco shred samples obtained based on the shredding of the tobacco flake samples to obtain the tobacco shred structures of the tobacco flake samples.
10. The apparatus for constructing a tobacco shred structure prediction model according to claim 8, further comprising:
The prediction unit is used for detecting leaf structures of the obtained tobacco leaves to be detected, inputting the leaf structures of the tobacco leaves to be detected into the tobacco shred structure prediction model for tobacco shred structure prediction, and outputting the tobacco shred structures of the tobacco leaves to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010710238.9A CN112016046B (en) | 2020-07-22 | 2020-07-22 | Construction method and device of tobacco shred structure prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010710238.9A CN112016046B (en) | 2020-07-22 | 2020-07-22 | Construction method and device of tobacco shred structure prediction model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112016046A CN112016046A (en) | 2020-12-01 |
| CN112016046B true CN112016046B (en) | 2023-11-28 |
Family
ID=73498792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010710238.9A Active CN112016046B (en) | 2020-07-22 | 2020-07-22 | Construction method and device of tobacco shred structure prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112016046B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113576016A (en) * | 2021-07-15 | 2021-11-02 | 河南中烟工业有限责任公司 | Shredding adjusting method based on tobacco shred structure |
| CN114098136A (en) * | 2021-12-20 | 2022-03-01 | 河南中烟工业有限责任公司 | Method and processing device for improving tobacco shred structure on line |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330467A (en) * | 2017-07-03 | 2017-11-07 | 云南昆船设计研究院 | A kind of Forecasting Methodology influenceed based on piece cigarette morphological feature on tobacco structure |
| WO2018040302A1 (en) * | 2016-08-31 | 2018-03-08 | 上海创和亿电子科技发展有限公司 | Method and system for measuring width of cut tobacco piece or cut stem, and server having said system |
-
2020
- 2020-07-22 CN CN202010710238.9A patent/CN112016046B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018040302A1 (en) * | 2016-08-31 | 2018-03-08 | 上海创和亿电子科技发展有限公司 | Method and system for measuring width of cut tobacco piece or cut stem, and server having said system |
| CN107330467A (en) * | 2017-07-03 | 2017-11-07 | 云南昆船设计研究院 | A kind of Forecasting Methodology influenceed based on piece cigarette morphological feature on tobacco structure |
Non-Patent Citations (1)
| Title |
|---|
| 主成分回归分析法在永州烟叶评吸质量与外观质量的关系模型构建中的应用;张延军;李旭华;林锐锋;文俊;叶为民;;中国农学通报(18);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112016046A (en) | 2020-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112016046B (en) | Construction method and device of tobacco shred structure prediction model | |
| CN106897570B (en) | Chronic obstructive pulmonary disease testing system based on machine learning | |
| WO2019153595A1 (en) | Method for predicting risk of chronic obstructive pulmonary disease, server, and computer readable storage medium | |
| Ferrão et al. | Accurate genomic prediction of Coffea canephora in multiple environments using whole-genome statistical models | |
| CN104346406B (en) | Training corpus expanding device and training corpus extending method | |
| CN108181263A (en) | The extraction of tobacco leaf genius loci and method of discrimination based near infrared spectrum | |
| CN102650630A (en) | Method for building distinguishing model for evaluating odor type of flue-cured tobacco raw material on the basis of note components | |
| CN110013048B (en) | Method and device for determining tobacco shred structure suitable for fine cigarette | |
| CN102866127A (en) | Method for assisting cigarette formula by adopting SIMCA (Soft Independent Modeling of Class Analogy) based on Near-infrared spectral information | |
| CN107767079A (en) | A kind of objective integrated evaluating method of tobacco style feature | |
| CN106770303A (en) | Cigarette shreds structure characterization methods based on graphical analysis | |
| CN105092789A (en) | Analysis method for tobacco leaf style characteristic variation trend in lamina aging process | |
| CN103892435A (en) | Method for preparing tobacco powder raw materials and stabilizing quality of reconstituted tobacco | |
| CN109324016A (en) | A kind of determination method of redried odor type style | |
| CN113010739B (en) | Video tag auditing method and device and electronic equipment | |
| CN109856139B (en) | Tree leaf quantity estimation method and system | |
| CN115931738A (en) | Method and system for evaluating quality stability of finished tobacco flakes | |
| CN114647582A (en) | Automatic driving scene generation method | |
| CN114998732A (en) | Plant disease severity assessment method and device | |
| CN110866221B (en) | Detection method for threshing and redrying data accuracy | |
| CN114627963A (en) | Protein data filling method, system, computer device and readable storage medium | |
| CN107934712A (en) | Elevator health degree detection method and device based on statistical models | |
| CN110082251B (en) | Method for defining and detecting tobacco stem labels | |
| CN110432539A (en) | A kind of raw material of cigarette surplus and deficiency alternative | |
| CN111553613B (en) | Intelligent grading evaluation method and system for quality of seismic acquisition data |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |