CN107589228A - The method that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index - Google Patents
The method that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index Download PDFInfo
- Publication number
- CN107589228A CN107589228A CN201711091967.5A CN201711091967A CN107589228A CN 107589228 A CN107589228 A CN 107589228A CN 201711091967 A CN201711091967 A CN 201711091967A CN 107589228 A CN107589228 A CN 107589228A
- Authority
- CN
- China
- Prior art keywords
- index
- tobacco leaf
- sample
- industrial
- industrial usability
- 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.)
- Granted
Links
Landscapes
- Manufacture Of Tobacco Products (AREA)
Abstract
The invention discloses a kind of method that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, comprise the step of, sample;Prepared by industrial sorted sample, the measuring and calculating of graded samples ratio and quality index appearance quantify;Industrial sorted sample sensory evaluating smoking, quantification of targets and industrial usability judge;The tobacco leaf industrial usability classification forecast model based on appearance index is established, determines characteristic look index and desired value;Establish its corresponding part by weight matched curve of external appearance characteristic index;According to external appearance characteristic index and matched curve, batch primary industry applicability quantitative prediction value is calculated.The present invention sorts to the industry for instructing raw tobacco material to quantify, and it is significant to improve raw tobacco material industrial usability.
Description
Technical field
The present invention relates to a kind of batch based on industrial requirement to sort tobacco leaf external appearance characteristic Index, particularly relates to
And a kind of method that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, belong to quality of tobacco assessment technique neck
Domain.
Background technology
Tobacco leaf industry sorting purpose is quality of tobacco characteristic is more refined, and quality boundary is apparent, and qualitative character becomes apparent from.
Accelerate with the paces in tobacco industrial enterprise famous brand, big market, using industrial enterprise's raw materials requirement to be oriented to, carry out tobacco leaf industry
Sorting is to lift the inevitable choice of composition of raw materials applicability and stability.Tobacco leaf presentation quality is the external reflection of inherent quality,
Most of industrial enterprise's sorting presentation qualities perform according to GB 2635 at present, including color, maturity, blade construction, body
Part, oil, colourity, it is residual wound etc. factors, assorting room need to pay close attention to compared with multi objective, to operating personnel's hierarchical level requirement compared with
It is high;At present also without method more quick, that industrial usability is accurately predicted according to batch tobacco leaf presentation quality.It is thus determined that work
Industry separation characteristic appearance index, a kind of science, easily batch industry sorting tobacco leaf industrial usability Forecasting Methodology are established, for
Strengthen the specific aim of industrial assorting room, improve the industrial usability for selecting tobacco leaf, enriching quality of tobacco appraisement system has weight
The meaning wanted.
The content of the invention
It is a primary object of the present invention to, there is provided a kind of industry sorting tobacco leaf characteristic look determined according to formulary requirements refers to
Mark method, in assorting room, specific aim strengthens grasp of the sorting operating personnel to characteristic look index, is lifted and is applicable with industry
Property for the separating effect that is oriented to, further improve tobacco leaf industry sorting quality and efficiency.Simultaneously, there is provided one kind passes through tobacco leaf feature
The method of appearance index prediction sorting tobacco leaf industrial usability, is easy to that raw tobacco material industrial usability by the gross is predicted and measured
Change.
In order to solve the above technical problems, the present invention, which provides a kind of predicted by tobacco leaf characteristic look index, sorts tobacco leaf industry
The method of applicability, it is characterised in that specifically comprise the following steps:
Step SS1:Sampling;
Step SS2:Prepared by industrial sorted sample, the measuring and calculating of graded samples ratio and quality index appearance quantify;
Step SS3:Industrial sorted sample sensory evaluating smoking, quantification of targets and industrial usability judge;
Step SS4:The tobacco leaf industrial usability classification forecast model based on appearance index is established, determines characteristic look index
And desired value;
Step SS5:Establish its corresponding part by weight matched curve of external appearance characteristic index;
Step SS6:According to external appearance characteristic index and matched curve, batch primary industry applicability quantitative prediction value is calculated.
As a kind of preferred embodiment, the step SS1 is specifically included:The first cured tobacco leaf sample of collection, samples bag number
Determined according within the 0.5% of arrival lophophore number;If batch sample extracts 3 bags, often wrapped from center to surrounding less than 1000 loads
Some places of sample are extracted, if often place takes dry plate tobacco leaf.
As a kind of preferred embodiment, the step SS2 is specifically included:According to sample appearance quality, then it is grouped,
And each group sample appearance index is quantified, the appearance index includes:Color, maturity, blade construction, identity, oil,
Colourity.
As a kind of preferred embodiment, the step SS3 is specifically included:Sense organ industry is carried out to each graded samples to comment
Valency, Appreciation gist YC/T 530-2015《Cured tobacco leaf liquor style characteristic sensory evaluation method》Related request performs, and to every
Class sample industrial usability carries out qualitative evaluation, and A represents that industrial usability is strong, and it is stronger that B represents industrial usability.
As a kind of preferred embodiment, the step SS4 is specifically included:Build all kinds of sample industrial usability classification moulds
Type, give information gain-ratio and determine characteristic look index, the disaggregated model output result include the appearance index filtered out and
Desired value.
As a kind of preferred embodiment, the step SS4 also includes:According to sense organ industrial evaluation result of determination, with reference to
Industry sorting classification, establishes the industrial sorted sample categorised decision tree based on commercial formulations use demand, each appearance classification sample
Product have the judgement if appropriate for formulary requirements after corresponding appearance index score value, and both combine, and establish decision tree, sieve
Select the representative appearance index for meeting formulary requirements and corresponding index score value.
As a kind of preferred embodiment, the process of establishing of the decision tree includes:
Index of the comentropy as measurement sample set purity is introduced, the value of comentropy is smaller, then the purity of sample set is got over
Height, its definition are:
Wherein D is current sample set, pkFor the ratio shared by kth class sample in current sample setDvIt is a for all values on attribute a in DvSample, D is calculated according to formula 1vComentropy, then
The sample number included according to different branch nodes is different, and weight is assigned to branch node | Dv|/| D |, more points of sample
Zhi Jiedian influence is bigger, and the formula for calculating " information gain " (the information gain) of a attributes is:
Category division is carried out based on information gain-ratio, easily the attribute more to value number has preferred, inclined to reduce
The adverse effect that may be brought well, " ratio of profit increase " is introduced to select optimal dividing attribute, ratio of profit increase definition is:
Wherein IV (a) is referred to as attribute a " eigenvalue ", and attribute a possibility value number is more, then eigenvalue is bigger, its
Calculation formula is:
To prevent that it is preferred that information gain-ratio from having to the possible small numbers of attribute of value, when dividing attribute, inspiration is used
Formula, the attribute that information gain is found out in attribute and is higher than average level is first divided from candidate, then therefrom select ratio of profit increase highest.
As a kind of preferred embodiment, the step SS5 is specifically included:According to presentation quality quantify require, with it is N number of by
Gradually incremental external appearance characteristic desired value is the upper limit, and calculating is cumulative weight ratio less than the tobacco leaf weight ratio of the upper limit, its
In, N is positive integer.
As a kind of preferred embodiment, the step SS5 also includes:By to outward appearance index feature value and accumulative ratio
Scatter diagram, it is known that both are substantially in " S " type curved line relation, that is, the General Expression form of curvilinear equation to be fitted for G (x)=
B0+b1*EXP (b2+b3/x), x are that corresponding external appearance characteristic index adds up higher limit, and G (x) is its corresponding cumulative weight ratio
Example;Wherein b0, b1, b2, b3 are equation parameter, and with reference to the cumulative weight ratio, G (x), which is carried out curve fitting, to be obtained
B0, b1, b2, b3, so that it is determined that fit curve equation.
As a kind of preferred embodiment, the step SS6 is specifically included:According to the appearance index characteristic value and plan of determination
The curvilinear equation of conjunction, substitute into Formulas I AE=(1-G (x+0.5))/100, calculate industrial applicable characteristic value, the value can be considered special
Fixed regional, specific grade raw material meets the predicted estimate ratio of suitable particular industry formulary requirements;IAE values are bigger, show this batch
The tobacco leaf ratio of secondary samples met industrial requirement is bigger, and industrial usability is better.
The beneficial effect that the present invention is reached:The present invention proposes that a kind of industry sorting tobacco leaf determined according to formulary requirements is special
Levy appearance index method, in assorting room, specific aim strengthens grasp of the sorting operating personnel to characteristic look index, lifted with
Industrial usability is the separating effect being oriented to, and further improves tobacco leaf industry sorting quality and efficiency.Simultaneously, there is provided a kind of basis
The method that characteristic look index predicts batch tobacco leaf industrial usability, is easy to be predicted raw tobacco material industrial usability by the gross
And quantify.
Brief description of the drawings
Fig. 1 is the scatter chart in the typical appearance characteristic index oil index difference numerical intervals of the present invention.
Embodiment
The invention will be further described below in conjunction with the accompanying drawings.Following examples are only used for clearly illustrating the present invention
Technical scheme, and can not be limited the scope of the invention with this.
1) sample
It is test specimen to select certain producing region C2F grades, and sampling bag number determines (such as bulk production according to the 0.1% of arrival lophophore number
Sample is 20000 loads, then samples bag number=20000*0.1%, that is, chooses 20 bag samplings);Often wrap the extraction sample from center to surrounding
At 5-7, often place takes 10~15 tobacco leaves.Sample 14kg is extracted in this experiment altogether.
2) sample appearance industry sorting
Industrial sorting is carried out to experiment C2F rank tests sample, according to《Flue-cured tobacco GB2635》Corresponding grade factor
Regulation, classifies, and weigh to test specimen presentation quality;Presentation quality scoring is carried out according to ocular estimate index, according to
Ocular estimate index and standard shown in table 1 carry out score value quantization, wherein, appearance index includes color, maturity, identity, oil
Point, blade construction and colourity.The classification of final test specimen and corresponding proportion are shown in Table 2.
Table 1
Table 2
3) sample sense organ industrial evaluation
According to graded samples determined by outward appearance industry sorting, according to YC/T 530-2015《Cured tobacco leaf liquor style is special
Color sensation official's evaluation method》Sensory quality assessment is carried out respectively, and according to formulary requirements, is obtained corresponding industrial usability and characterized.Portion
Sample Analyses Methods for Sensory Evaluation Results is divided to be shown in Table 3.
The part test sample Analyses Methods for Sensory Evaluation Results of table 3
4) disaggregated model is built
According to sense organ industrial evaluation result of determination, classification is sorted with reference to industry, is established based on commercial formulations use demand
Industrial sorted sample categorised decision tree.Each appearance classification sample after corresponding appearance index score value, have if appropriate for
The judgement of formulary requirements, both combine, establish decision tree, filter out the representative appearance index for meeting formulary requirements and correspondingly refer to
Mark score value.
The foundation of decision tree is based on comentropy." comentropy " (information entropy) is that measurement sample set is pure
A kind of the most frequently used index is spent, the value of comentropy is smaller, then the purity of sample set is higher, and its definition is:
Wherein D is current sample set, pkFor the ratio shared by kth class sample in current sample set
DvIt is a for all values on attribute a in DvSample, D is calculated according to formula 1vComentropy, further according to difference
The sample number that is included of branch node it is different, give branch node to assign weight | Dv|/| D |, the more branch node of sample
Influence is bigger, and the formula for calculating " information gain " (the information gain) of a attributes is:
Category division is carried out based on information gain-ratio, easily the attribute more to value number has preferred, inclined to reduce
The adverse effect that may be brought well, " ratio of profit increase " (gain ratio) is introduced to select optimal dividing attribute, ratio of profit increase definition
For:
Wherein IV (a) is referred to as attribute a " eigenvalue " (intrinsic value), and attribute a possibility value number is got over
More, then eigenvalue is bigger, and its calculation formula is:
To prevent that it is preferred that information gain-ratio from having to the possible small numbers of attribute of value, when dividing attribute, inspiration is used
Formula, the attribute that information gain is found out in attribute and is higher than average level is first divided from candidate, then therefrom select ratio of profit increase highest.
Decision tree finally establishes result;
Final screening index is oil, and it can be seen from decision tree result, oil index can be used as differentiation tobacco leaf to meet formula
Whether the outward appearance industry separation characteristic index that desirability judges, as a result shows, according to oil index to meeting formulary requirements and entering
Row judges that classification accuracy is up to 83.7%, in ideal level.
5) it is fitted distribution curve
Determine the distribution curve in typical appearance characteristic index oil index difference numerical intervals.Curve is carried out first to estimate
Meter, as a result as shown in Figure 1.According to the variation tendency of figure, and according to models fitting result, it is known that serpentine curve can preferably be intended
The accumulative ratio of oil difference score value is closed, as shown in table 4.
Table 4
The ANOVA of table 5
Independent variable is x.
As shown in Table 5, it is homogeneous equal to possess variance.Directly operated using " non-linear " model, as a result such as table 6 below institute
Show:
The iteration history table of table 6
Derivative is by numerical calculation;Wherein, a:Main number of iterations shows that secondary number of iterations is on the right side of decimal on the left of decimal
Display;b:It is out of service after 20 iteration, find optimal solution.
The estimates of parameters of table 7
The correlation of the estimates of parameters of table 8
The analysis of results table of table 9
ANOVA′
From above-mentioned analysis result (i.e. table 7, table 8, table 9):So that the minimum principle of residual error, after iteration 20 times, equation
Parameter has searched out optimal solution, and this model can explain 96.7% variation, and fitting degree is preferable.
Fit curve equation is:G (X)=- 30.143+52.168*exp (2.297-12.488/X).
6) industrial applicable characteristic value is calculated
Industrial applicable characteristic value (Industrial Applicable Eigenvalues), it can be considered given area, specific
Grade raw material meets the predicted estimate ratio of suitable particular industry formulary requirements.IAE values are bigger, show that the batch sample meets work
Industry demand A tobacco leaf ratio is bigger, meets that industrial requirement A applicability is better.
IAE=(100-G (x+0.5))/100
Wherein x is the external appearance characteristic index boundary value filtered out according to decision tree, in this example x=6.5, and G (x) is difference
Add up the curvilinear equation of ratio fitting corresponding to oil score value, in this example, the lower ratio of prediction choosing is G (x+0.5)=57, IAE=
0.43, that is, predict that the batch raw material meets that industrial usability A ratio is about 43%, with final congruent level industry separation results
45.44% difference is little, can be as the reference index for judging the sorting Leakage in Value evaluation of certain batch primary industry.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, some improvement and deformation can also be made, these are improved and deformation
Also it should be regarded as protection scope of the present invention.
Claims (10)
1. pass through tobacco leaf characteristic look index prediction sorting tobacco leaf industrial usability method, it is characterised in that specifically include as
Lower step:
Step SS1:Sampling;
Step SS2:Prepared by industrial sorted sample, the measuring and calculating of graded samples ratio and quality index appearance quantify;
Step SS3:Industrial sorted sample sensory evaluating smoking, quantification of targets and industrial usability judge;
Step SS4:The tobacco leaf industrial usability classification forecast model based on appearance index is established, characteristic look index is determined and refers to
Scale value;
Step SS5:Establish its corresponding part by weight matched curve of external appearance characteristic index;
Step SS6:According to external appearance characteristic index and matched curve, batch primary industry applicability quantitative prediction value is calculated.
2. the method according to claim 1 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS1 is specifically included:The first cured tobacco leaf sample of collection, sampling bag number is according to the 0.5% of arrival lophophore number
Within determine;If batch sample extracts 3 bags, often wraps and some places of sample are extracted from center to surrounding, often place takes less than 1000 loads
If dry plate tobacco leaf.
3. the method according to claim 1 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS2 is specifically included:According to sample appearance quality, then it is grouped, and to each group sample appearance index
Quantified, the appearance index includes:Color, maturity, blade construction, identity, oil, colourity.
4. the method according to claim 1 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS3 is specifically included:Sense organ industrial evaluation, Appreciation gist YC/T 530- are carried out to each graded samples
2015《Cured tobacco leaf liquor style characteristic sensory evaluation method》Related request is performed, and every class sample industrial usability is carried out
Qualitative evaluation, A represent that industrial usability is strong, and it is stronger that B represents industrial usability.
5. the method according to claim 1 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS4 is specifically included:All kinds of sample industrial usability disaggregated models are built, it is true to give information gain-ratio
Determine characteristic look index, the disaggregated model output result includes appearance index and the desired value filtered out.
6. the method according to claim 5 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS4 also includes:According to sense organ industrial evaluation result of determination, classification is sorted with reference to industry, foundation is based on
The industrial sorted sample categorised decision tree of commercial formulations use demand, each appearance classification sample is in corresponding appearance index score value
Afterwards, there is the judgement if appropriate for formulary requirements, both combine, and establish decision tree, filter out the representative for meeting formulary requirements
Property appearance index and corresponding index score value.
7. the method according to claim 6 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the process of establishing of the decision tree includes:
Index of the comentropy as measurement sample set purity is introduced, the value of comentropy is smaller, then the purity of sample set is higher, its
Definition is:
Wherein D is current sample set, pkFor the ratio shared by kth class sample in current sample set
DvIt is a for all values on attribute a in DvSample, D is calculated according to formula 1vComentropy, saved further according to different branch
The included sample number of point is different, and weight is assigned to branch node | Dv|/| D |, the influence of the more branch node of sample is bigger,
The formula of " information gain " (information gain) for calculating a attributes is:
Category division is carried out based on information gain-ratio, easily the attribute more to value number has preferred, can to reduce preference
The adverse effect that can be brought, " ratio of profit increase " is introduced to select optimal dividing attribute, ratio of profit increase definition is:
Wherein IV (a) is referred to as attribute a " eigenvalue ", and attribute a possibility value number is more, then eigenvalue is bigger, and it is calculated
Formula is:
To prevent that it is preferred that information gain-ratio from having to the possible small numbers of attribute of value, when dividing attribute, use is heuristic, first
The attribute that information gain is found out in attribute and is higher than average level is divided from candidate, then therefrom selects ratio of profit increase highest.
8. the method according to claim 1 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS5 is specifically included:Quantify to require according to presentation quality, with N number of gradually incremental external appearance characteristic index
It is cumulative weight ratio less than the tobacco leaf weight ratio of the upper limit to be worth for the upper limit, calculating, wherein, N is positive integer.
9. the method according to claim 8 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS5 also includes:Pass through the scatter diagram to outward appearance index feature value and accumulative ratio, it is known that Liang Zhe great
Cause is in " S " type curved line relation, that is, the General Expression form of curvilinear equation to be fitted is G (x)=b0+b1*EXP (b2+b3/x),
X is that corresponding external appearance characteristic index adds up higher limit, and G (x) is its corresponding cumulative weight ratio;Wherein b0, b1, b2, b3 are
Equation parameter, with reference to the cumulative weight ratio, G (x), which is carried out curve fitting, can obtain b0, b1, b2, b3, so that it is determined that
Fit curve equation.
10. the method according to claim 1 that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index, its
It is characterised by, the step SS6 is specifically included:According to the appearance index characteristic value of determination and the curvilinear equation of fitting, formula is substituted into
In IAE=(1-G (x+0.5))/100, industrial applicable characteristic value is calculated, the value can be considered given area, specific grade raw material
Meet the predicted estimate ratio of suitable particular industry formulary requirements;IAE values are bigger, show that the batch sample meets industrial requirement
Tobacco leaf ratio is bigger, and industrial usability is better.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711091967.5A CN107589228B (en) | 2017-11-08 | 2017-11-08 | Method for predicting industrial applicability of sorted tobacco leaves through tobacco leaf characteristic appearance indexes |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711091967.5A CN107589228B (en) | 2017-11-08 | 2017-11-08 | Method for predicting industrial applicability of sorted tobacco leaves through tobacco leaf characteristic appearance indexes |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107589228A true CN107589228A (en) | 2018-01-16 |
CN107589228B CN107589228B (en) | 2019-12-13 |
Family
ID=61046785
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711091967.5A Active CN107589228B (en) | 2017-11-08 | 2017-11-08 | Method for predicting industrial applicability of sorted tobacco leaves through tobacco leaf characteristic appearance indexes |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107589228B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108354219A (en) * | 2018-04-24 | 2018-08-03 | 湖北中烟工业有限责任公司 | A method of tobacco mellowing degree is judged by tobacco leaf presentation quality |
CN111402963A (en) * | 2020-03-30 | 2020-07-10 | 上海烟草集团有限责任公司 | Tobacco formula generation method, system, medium and device |
CN111667176A (en) * | 2020-06-05 | 2020-09-15 | 江苏中烟工业有限责任公司 | Tobacco leaf style characteristic highlighting degree evaluation method based on cigarette brand requirements |
CN111680910A (en) * | 2020-06-05 | 2020-09-18 | 江苏中烟工业有限责任公司 | Tobacco application range positioning method based on cigarette formula efficacy |
CN113408945A (en) * | 2021-07-15 | 2021-09-17 | 广西中烟工业有限责任公司 | Method and device for detecting purity of flue-cured tobacco, electronic equipment and storage medium |
CN114376259A (en) * | 2022-02-14 | 2022-04-22 | 河南中烟工业有限责任公司 | Automatic sorting method for raw material tobacco leaves |
CN114766703A (en) * | 2022-04-07 | 2022-07-22 | 河南中烟工业有限责任公司 | Upper tobacco leaf sorting method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105445421A (en) * | 2015-11-11 | 2016-03-30 | 江苏中烟工业有限责任公司 | Method for predicting sensory quality in lamina alcoholization process via appearance indexes |
CN105929115A (en) * | 2016-06-24 | 2016-09-07 | 浙江中烟工业有限责任公司 | Evaluation method for ecological adaptability of tobacco leaves |
-
2017
- 2017-11-08 CN CN201711091967.5A patent/CN107589228B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105445421A (en) * | 2015-11-11 | 2016-03-30 | 江苏中烟工业有限责任公司 | Method for predicting sensory quality in lamina alcoholization process via appearance indexes |
CN105929115A (en) * | 2016-06-24 | 2016-09-07 | 浙江中烟工业有限责任公司 | Evaluation method for ecological adaptability of tobacco leaves |
Non-Patent Citations (3)
Title |
---|
汤洋: "上部烟叶一次性采烤技术对烟叶可用性的影响", 《湖南农业科学》 * |
罗思维等: "配方模块打叶技术对提高吉林烟叶工业适用性的作用", 《广东化工》 * |
过伟民等: "烤烟中部烟叶外观区域特征分布及其与外观品质和物理特性的关系", 《烟草科技》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108354219A (en) * | 2018-04-24 | 2018-08-03 | 湖北中烟工业有限责任公司 | A method of tobacco mellowing degree is judged by tobacco leaf presentation quality |
CN111402963A (en) * | 2020-03-30 | 2020-07-10 | 上海烟草集团有限责任公司 | Tobacco formula generation method, system, medium and device |
CN111667176A (en) * | 2020-06-05 | 2020-09-15 | 江苏中烟工业有限责任公司 | Tobacco leaf style characteristic highlighting degree evaluation method based on cigarette brand requirements |
CN111680910A (en) * | 2020-06-05 | 2020-09-18 | 江苏中烟工业有限责任公司 | Tobacco application range positioning method based on cigarette formula efficacy |
CN113408945A (en) * | 2021-07-15 | 2021-09-17 | 广西中烟工业有限责任公司 | Method and device for detecting purity of flue-cured tobacco, electronic equipment and storage medium |
CN114376259A (en) * | 2022-02-14 | 2022-04-22 | 河南中烟工业有限责任公司 | Automatic sorting method for raw material tobacco leaves |
CN114766703A (en) * | 2022-04-07 | 2022-07-22 | 河南中烟工业有限责任公司 | Upper tobacco leaf sorting method |
Also Published As
Publication number | Publication date |
---|---|
CN107589228B (en) | 2019-12-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107589228A (en) | The method that sorting tobacco leaf industrial usability is predicted by tobacco leaf characteristic look index | |
CN104063599B (en) | Index screening and processing method for evaluating quality of tobacco leaves | |
Lach | Patents and productivity growth at the industry level: A first look | |
CN105445421B (en) | A kind of method of aesthetic quality during predictive slice nicotinic alcohol by appearance index | |
CN106600455A (en) | Electric charge sensitivity assessment method based on logistic regression | |
CN110288484B (en) | Insurance classification user recommendation method and system based on big data platform | |
CN108898479A (en) | The construction method and device of Credit Evaluation Model | |
CN103424542A (en) | Tobacco leaf quality evaluation method based on sensory evaluation | |
CN106202267A (en) | The Citrus chachiensis Hort. integrated quality evaluation methodology of a kind of yellow fruit | |
CN107767079A (en) | A kind of objective integrated evaluating method of tobacco style feature | |
CN108399255A (en) | A kind of input data processing method and device of Classification Data Mining model | |
CN111596010A (en) | Quantitative evaluation and classification method for tobacco industry availability | |
CN107784452A (en) | A kind of objective integrated evaluating method of tobacco style characteristic similarity | |
CN109409579A (en) | The method of BP neural network prediction Raw material processing suitability | |
CN107341613A (en) | A kind of method for aiding in tobacco leaf formulation balance to replace | |
CN102488309B (en) | Intelligent tobacco formulation method | |
CN110852600A (en) | Method for evaluating dynamic risk of market subject | |
CN104323412A (en) | Building and application method for functional tobacco module | |
CN110132879A (en) | A kind of flue-cured tobacco note discrimination method based near infrared spectrum | |
CN101539517A (en) | Method for evaluating high-gloss coating surfaces by fuzzy logic model | |
CN111680910A (en) | Tobacco application range positioning method based on cigarette formula efficacy | |
CN118485465A (en) | Agricultural batch market product sales prediction analysis system and method based on artificial intelligence | |
CN107897995A (en) | A kind of segmentation split based on former cigarette formula module is homogenized regulation and control method | |
CN106651167A (en) | Biological information engineer skill rating system | |
CN110046934A (en) | A kind of idle clothing valuation system and method based on data mining |
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 |