CN113869641A - Tobacco shred quality comprehensive evaluation method based on principal component analysis method - Google Patents
Tobacco shred quality comprehensive evaluation method based on principal component analysis method Download PDFInfo
- Publication number
- CN113869641A CN113869641A CN202110989940.8A CN202110989940A CN113869641A CN 113869641 A CN113869641 A CN 113869641A CN 202110989940 A CN202110989940 A CN 202110989940A CN 113869641 A CN113869641 A CN 113869641A
- Authority
- CN
- China
- Prior art keywords
- tobacco
- quality
- tobacco shred
- shreds
- sampling
- 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.)
- Withdrawn
Links
- 235000002637 Nicotiana tabacum Nutrition 0.000 title claims abstract description 156
- 238000011156 evaluation Methods 0.000 title claims abstract description 25
- 238000012847 principal component analysis method Methods 0.000 title claims abstract description 14
- 244000061176 Nicotiana tabacum Species 0.000 title 1
- 241000208125 Nicotiana Species 0.000 claims abstract description 155
- 238000000034 method Methods 0.000 claims abstract description 42
- 238000013210 evaluation model Methods 0.000 claims abstract description 21
- 238000013441 quality evaluation Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000005070 sampling Methods 0.000 claims description 46
- 235000019504 cigarettes Nutrition 0.000 claims description 44
- 230000008569 process Effects 0.000 claims description 23
- 238000001514 detection method Methods 0.000 claims description 18
- 238000004458 analytical method Methods 0.000 claims description 11
- 238000002156 mixing Methods 0.000 claims description 11
- 238000000513 principal component analysis Methods 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 10
- 239000000428 dust Substances 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000007789 sealing Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 abstract description 28
- 238000005096 rolling process Methods 0.000 abstract description 13
- 238000004519 manufacturing process Methods 0.000 abstract description 8
- 238000012795 verification Methods 0.000 abstract description 3
- 238000002474 experimental method Methods 0.000 abstract description 2
- 230000000007 visual effect Effects 0.000 abstract 1
- 230000001953 sensory effect Effects 0.000 description 6
- 238000010220 Pearson correlation analysis Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000001035 drying Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000003137 locomotive effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 240000009235 Rubia cordifolia Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000010025 steaming Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- 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
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] 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
- G06Q10/06395—Quality analysis or management
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Manufacturing Of Cigar And Cigarette Tobacco (AREA)
- Manufacture Of Tobacco Products (AREA)
Abstract
The invention discloses a tobacco shred quality comprehensive evaluation method based on a principal component analysis method. The results show that: the tobacco shred quality comprehensive evaluation model and the quality verification experiment result based on the invention are as follows: test III > test IV > test I > test II > test V, the rolling quality results are as follows: the deviation of the test III and the test IV is low and is consistent with the model evaluation result. Therefore, the comprehensive tobacco shred quality evaluation model and the evaluation method constructed by the method are simple and visual, are suitable for the current production situation of the tobacco shred processing at the current stage, and can provide a new idea for tobacco shred quality evaluation in the tobacco shred production process.
Description
Technical Field
The invention relates to the field of tobacco shred manufacturing, in particular to a comprehensive tobacco shred quality evaluation method based on a principal component analysis method.
Background
The core of cigarette enterprise production is that the re-baked tobacco leaves are processed by a series of procedures of steaming, adding, cutting, baking, mixing and the like by adopting advanced automatic equipment to produce the tobacco shreds meeting the requirements of the shred making process, and the tobacco shreds are rolled into qualified cigarettes in a rolling workshop. The tobacco mixing and flavoring process is an important process and a last process in the tobacco shred production process, and the stability and uniformity of the quality of the tobacco shreds directly influence the internal quality of subsequent cigarettes and the rolling quality of the tobacco shreds.
The evaluation of the quality of the cut tobacco at the present stage is based on the evaluation and optimization of the cut tobacco making process based on the national standard cigarette process specification, and comprises the following aspects: firstly, the stability of the quality of the cut tobacco is improved, in the research of the aspect, the method is characterized in that a similarity matching identification model of the cut tobacco and a cut tobacco mixing uniformity and quality stability evaluation model are established by applying a Fourier transform near infrared (FT-NIR) spectral analysis technology to the eagle and the Zhang Jiayun, so that the accurate and rapid evaluation of the cut tobacco quality stability is realized; the method for evaluating the quality stability of the silk production process is established by utilizing a statistical method and giving different weights to different processes and different variables in the processes; on the one hand, the Duyunpeng and Schroe training respectively establishes a tobacco moisture prediction model of the whole process flow before tobacco drying and a tobacco moisture prediction model after perfuming by adopting a normal distribution statistical method and an LSTM machine learning method, so that the batch homogeneity of the tobacco moisture is improved; the Fangyi grasps the change rule of the moisture content of the tobacco shreds in each process by researching the change trend of the moisture content of the tobacco shreds in the tobacco shred conveying process, and ensures the stability of the quality of the cigarettes; on the one hand, the Rubia cordifolia is based on a time sequence database InfluxDB, and a cubic exponential smoothing algorithm, a long and short term memory network (LSTM) and sequence-to-sequence learning (Seq2Seq) are used for storing, inquiring, counting, analyzing, early warning and modeling prediction of production data in real time, so that the stability of moisture at an outlet of a cut tobacco drying process is improved; according to the method, index weights are calculated on different sections, processes and process parameters in a multi-level mode by adopting a hierarchical analysis method, an index characterization function is established through a fuzzy algorithm, a batch comprehensive score Sbtatch model is used for process quality evaluation, and the process control level of the whole process of a batch is accurately evaluated. The research on the three aspects can effectively improve the stability and the quality of the cut tobacco, but the influence of the cut tobacco quality on the rolling quality is not considered, so that certain limitation exists.
The principal component analysis is an analysis method for concentrating data by adopting a dimension reduction technology. The principle is to try to recombine the original variables into a group of new several comprehensive variables which are not related to each other, and to extract several smaller sum variables from the group according to the actual needs, wherein the sum variables reflect the information of the original variables as much as possible. The principal component analysis has the advantages of convenient application, suitability for various occasions and the like.
The method adopts a principal component analysis method, establishes a new tobacco shred quality evaluation model in the mixed tobacco shred flavoring process by analyzing indexes related to the tobacco shred quality, and verifies the model through rolling quality standard deviation analysis and sensory evaluation. The method aims to realize the comprehensive evaluation of the quality of the cut tobacco in the mixed tobacco and flavoring process through a few irrelevant variables and improve the applicability requirement of the cut tobacco on the rolling quality.
Disclosure of Invention
In order to overcome the defects, the application provides a comprehensive tobacco shred quality evaluation method based on a principal component analysis method.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a method for comprehensively evaluating the quality of tobacco shreds based on principal component analysis method includes
S1, sampling
S1.1, tobacco shred sampling
Randomly selecting M batches of cut tobacco, randomly sampling for N times in each batch, and taking the cut tobacco from a discharge port of a cut tobacco mixing and flavoring machine by using a sampling container, wherein the sampling quality of a cut tobacco sample to be tested is (1000.0 +/-100.0) g/time, and the time interval of single sampling is 15 +/-5 min;
s1.2, sampling of cigarette finished products
Sampling finished cigarettes at the outlets of cigarette making machines corresponding to 5 batches of tobacco shred sampling by adopting a fixed sampling machine table and fixed sampling personnel, randomly taking 150 samples of each group of samples each time, continuously sampling for N times under the condition of time interval of 30 +/-5 min, placing the samples in a constant temperature and humidity box for balancing for 24h, sealing and labeling, and reserving samples for later use;
s2, data detection
Detecting the tobacco shred structure and the physical indexes of the finished cigarette product;
the tobacco shred structure comprises 4 indexes of long shreds, medium shreds, short shreds and broken shreds which are related to the quality of the tobacco shreds;
the physical indexes of the finished cigarette product comprise 7 indexes of single cigarette weight, dust content, suction resistance, hardness, length, circumference and end cut;
wherein:
in the tobacco shred structure detection, the tobacco shred length definition standard is as follows: shredding: the length of the tobacco shreds is less than 1.0 mm; short silk: the length of the tobacco shreds is more than 1.0mm and less than 2.5 mm; medium silk: the length of the tobacco shreds is more than 2.5mm and less than 3.5 mm; filament: the length of the tobacco shreds is more than 3.5 mm;
in the detection of physical indexes of finished cigarette products, 150 samples extracted at a single time are as follows: 100 samples are used for measuring the end part filament falling and circumference, 20 samples are used for measuring the dust content, and the other 30 samples are used for measuring the cigarette weight, length, hardness and suction resistance, and the average values are respectively taken for recording;
s3, principal component analysis
S3.1, data normalization
The data is standardized, and the data standardization formula is as follows:
wherein x isnpIs the original data ( n 1, 2.., 11; p 1,2, …,6),is the average of the n-th index, SnIs data standard deviation, x'npThe data is normalized;
s3.2, extracting main components
Calculating a characteristic value and a characteristic vector:
the normalized data is represented as an n x p matrix with its corresponding eigenvalues λiOrthogonalized unit feature vector Z for component variance contributioniFor variance contribution rate, the calculation formula is:
Zithe size of (2) represents the capability of the component to react with information, and the principal component is extracted according to the numerical value of the component;
determining the principle of the number m of the main components: the characteristic value is greater than 1; the number m of the main components is less than the number of the detection indexes n;
s3.3, establishing an evaluation model
Creating a tobacco shred quality comprehensive evaluation model F, wherein the tobacco shred quality comprehensive evaluation model F comprises the following steps:
F=(a11+a12+a13+......a1m)×X1×Z1+(a21+a22+a23+......a2m)×X2×Z2+(a31+
a32+a33+......a3m)×X3×Z3+......+(an1+an2+an3+......anm)×Xn×Zm
wherein, a11、a12、a13....anmScoring coefficients for the principal components, i.e., feature vectors of the principal components;
Z1、Z2、Z3...Zmvariance contribution rate for extracting principal component;
X=(X1,X2,...,Xn) Is an n-dimensional random variable, X1,X2,X3...X11Sequentially representing filament, medium filament, short filament, broken filament, length, single weight, circumference, suction resistance, hardness, end part filament drop and end content;
m is the number of main components, and m is less than n;
s4, results and analysis
And calculating the comprehensive score of the main components of the quality of the cut tobacco in the 5 batches of cut tobacco mixing and flavoring processes through the model F, and then performing comprehensive evaluation on the quality of the cut tobacco according to the score.
The invention has the following beneficial effects:
the quality of the cut tobacco in the mixed tobacco perfuming process is comprehensively evaluated by adopting principal component analysis, the rolling physical index and the cigarette sensory quality of the cut tobacco are evaluated, the cut tobacco quality and the rolling physical index are both in a proper range, and a comprehensive evaluation model established based on the principal component analysis is consistent with the analysis result and the sensory evaluation result of the physical index of the cigarette (wherein, the score of a test III and a test IV is the highest, which indicates that the quality of the cut tobacco in the batch is the best, the corresponding physical index result is obviously superior to that of other batches, which indicates that the suitability of the cut tobacco in the batch on the computer is relatively good).
The comprehensive tobacco shred quality model established by the application can be used for comprehensively evaluating the quality of tobacco shreds produced by shred manufacturing on one hand; on the other hand, the method can also be used for predicting the computer suitability of rolling the cut tobacco. The model is simple and practical to operate, blindness in evaluation can be effectively avoided, and an idea is provided for tobacco shred quality evaluation.
The method can also be popularized to the tobacco shred quality evaluation of other brands and procedures, and has good popularization value and applicability.
Drawings
The invention will be further described with reference to the accompanying drawings and specific embodiments,
fig. 1 is a lithotripsy graph showing the relationship between the eigenvalues and the number of principal components in example 2.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
A method for comprehensively evaluating the quality of tobacco shreds based on principal component analysis method includes
S1, sampling
S1.1, tobacco shred sampling
Randomly selecting 5 batches of cut tobacco, randomly sampling 6 times in each batch, and taking the cut tobacco from a discharge port of a cut tobacco mixing and flavoring machine by using a sampling container, wherein the sampling quality of a cut tobacco sample to be tested is (1000.0 +/-100.0) g/time, and the time interval of single sampling is 15 +/-5 min;
s1.2, sampling of cigarette finished products
Sampling finished cigarettes at the outlets of cigarette making machines corresponding to 5 batches of tobacco shred sampling by adopting a fixed sampling machine and fixed sampling personnel, randomly taking 150 samples of each group of samples each time, continuously sampling for 6 times under the condition of time interval of 30 +/-5 min, placing the samples in a constant temperature and humidity box (20 +/-1) DEG C and RH (60 +/-2)%) for balancing for 24 hours, sealing and labeling the samples, and reserving the samples for later use;
s2, data detection
Detecting the tobacco shred structure and the physical indexes of the finished cigarette product;
the tobacco shred structure comprises 4 indexes of long shreds, medium shreds, short shreds and broken shreds which are related to the quality of the tobacco shreds;
the physical indexes of the finished cigarette product comprise 7 indexes of single cigarette weight, dust content, suction resistance, hardness, length, circumference and end cut;
wherein:
in the tobacco shred structure detection, the tobacco shred length definition standard is as follows: shredding: the length of the tobacco shreds is less than 1.0 mm; short silk: the length of the tobacco shreds is more than 1.0mm and less than 2.5 mm; medium silk: the length of the tobacco shreds is more than 2.5mm and less than 3.5 mm; filament: the length of the tobacco shreds is more than 3.5 mm;
in the detection of physical indexes of finished cigarette products, 150 samples extracted at a single time are as follows: 100 samples are used for measuring the end part filament falling and circumference, 20 samples are used for measuring the dust content, and the other 30 samples are used for measuring the cigarette weight, length, hardness and suction resistance, and the average values are respectively taken for recording;
s3, principal component analysis
S3.1, data normalization
The data is standardized, and the data standardization formula is as follows:
wherein x isnpIs the original data ( n 1, 2.., 11; p 1,2, …,6),is the average of the n-th index, SnIs data standard deviation, x'npThe data is normalized;
s3.2, extracting main components
Calculating a characteristic value and a characteristic vector:
the normalized data is represented as an n x p matrix with its corresponding eigenvalues λiOrthogonalized unit feature vector Z for component variance contributioniFor variance contribution rate, the calculation formula is:
Zithe size of (2) represents the capability of the component to react with information, and the principal component is extracted according to the numerical value of the component;
determining the principle of the number m of the main components:
the characteristic value is greater than 1; the number m of the main components is less than the number of the detection indexes n;
s3.3, establishing an evaluation model
Creating a tobacco shred quality comprehensive evaluation model F, wherein the tobacco shred quality comprehensive evaluation model F comprises the following steps:
F=(a11+a12+a13+......a1m)×X1×Z1+(a21+a22+a23+......a2m)×X2×Z2+(a31+a32+a33+......a3m)×X3×Z3+......+(an1+an2+an3+......anm)×Xn×Zm
wherein, a11、a12、a13....anmScoring coefficients for the principal components, i.e., feature vectors of the principal components;
Z1、Z2、Z3...Zmvariance contribution rate for extracting principal component;
X=(X1,X2,...,Xn) Is an n-dimensional random variable, X1,X2,X3...X11Sequentially representing filament, medium filament, short filament, broken filament, length, single weight, circumference, suction resistance, hardness, end part filament drop and end content;
m is the number of main components, and m is less than n;
s4, results and analysis
And calculating the comprehensive score of the main components of the quality of the cut tobacco in the 5 batches of cut tobacco mixing and flavoring processes through the model F, and then performing comprehensive evaluation on the quality of the cut tobacco according to the score.
Example 2
Sampling, data detection and data standardization processing are carried out based on the embodiment 1;
performing Pearson correlation analysis by using SPSS software;
11 indexes of the tobacco shred quality of each test batch are respectively recorded as: filament X1Middle filament X2Short filament X3Shredded X4Length X5Mass X6Circumference X7Resistance X8Hardness X9End portion filament falling X10The content of the residue X11。
Pearson correlation analysis results:
the long yarn is obviously related to the medium yarn and the short yarn on the level of 0.01 (P <0.01), and is obviously related to the broken yarn and the hardness on the level of 0.05 (P < 0.05); the length and the hardness of the cut tobacco and the cigarette are obviously and negatively related (P <0.05), and the hardness is obviously related on the 0.01 level (P < 0.01); short filaments were significantly associated with end-cut rates (P < 0.05); the mass is obviously related to the circumference and the suction resistance; the significant correlation between the tobacco shred structure and the physical indexes of the cigarette is shown, and the indexes are more.
Performing principal component analysis by using SPSS software;
the principle of determining the number m of the principal components is that the characteristic value is larger than 1, and the number m of the principal components is smaller than the number of the detection indexes n.
Main component extraction was performed based on example 1;
the calculation results are shown in table 1, the variance contribution rates of the first 4 eigenvectors are respectively 24.73%, 20.42%, 15.58% and 10.15%, and the variance cumulative contribution rate is 70.87%, and meanwhile, the eigenvalues of the first 4 principal components are all larger than 1 with reference to fig. 1 (a lithograph showing the relationship between the eigenvalue and the number of principal components). Since the first 4 main components contain most of the information of 11 indexes of the tobacco shred quality, the comprehensive evaluation of the tobacco shred quality of different batches by using the first 4 main components is feasible.
The number m of the finally determined principal components is 4.
TABLE 1 correlation matrix eigenvalues and cumulative contribution ratios of principal components
The score coefficient of each principal component, i.e., the feature vector of the principal component, was calculated based on example 1, and the results are shown in table 2.
TABLE 2 tobacco shred quality principal component eigenvectors
Substituting the score coefficients of the principal components into the comprehensive evaluation model F, so that the evaluation model equations of the principal components are respectively as follows:
F1=0.422X1-0.492X2-0.292X3+0.046X4+0.335X5+0.213X6+0.378X7+0.144X8-0.227X9-0.312X10-0.146X11
F2=-0.348X1+0.033X2+0.208X3+0.381X4+0.119X5+0.508X6+0.238X7+0.469X8+0.342X9+0.117X10-0.098X11
F3=-0.088X1-0.356X2+0.382X3+0.448X4+0.029X5+0.044X6-0.041X7-0.228X8-0.405X9+0.100X10+0.537X11
F4=-1.981X1-0.183X2+1.018X3-0.678X4+23.390X5-0.655X6+3.588X7+0.189X8-1.040X9-2.358X10+0.646X11
establishing a new tobacco shred quality comprehensive evaluation model based on 4 main component evaluation models:
F=24.727F1+20.417F2+15.579F3+10.15F4
obtaining a tobacco shred quality comprehensive evaluation model:
F=-18.145X1-18.908X2+13.309X3+9.022X4+248.574X5+9.676X6+49.991X7+11.500X8-15.496X9-27.716X10+9.309X11
the model is utilized to calculate the comprehensive score of the tobacco shred quality in 5 batches of tobacco shred mixing and flavoring processes, and then the tobacco shred quality is evaluated according to the score, and the result is shown in table 3. The result of the comprehensive evaluation of the tobacco shred quality by using the model is as follows: test III > test IV > test I > test II > test V.
Table 35 batches of tobacco shred quality comprehensive score and ranking
Example 3
Verification experiment
The quality of each batch of tobacco shreds is subjected to data statistics, standard deviations of the physical indexes of the cigarette, such as length, quality, circumference, suction resistance and hardness, are calculated, single-factor variance analysis is carried out through SPSS software, and the result is shown in Table 4:
TABLE 45 analysis results of physical index software for batch tobacco shred rolling
As can be seen from the above table, there is no significant difference between the batches of length standard deviation, mass standard deviation and suction resistance standard deviation, the circumference standard deviation and hardness standard deviation are the lowest batch in test IV, and the highest batch in test III.
In order to more accurately evaluate the applicability of each batch of cut tobacco rolling and a locomotive and sort and score the length, the quality, the circumference, the suction resistance and the hardness standard deviation average value, the scoring condition is shown in a table 5:
TABLE 5 cigarette physical index grading table
As can be seen from the above table, test III and test IV have higher scores and lower deviation of length, quality, circumference, suction resistance and hardness, which indicates that the two batches of cut tobacco rolling suitability are better, the deviation values of test I, test V and test II are relatively larger, and the applicability of the cut tobacco to a locomotive in the rolling process is poorer.
The verification result is consistent with the result of the comprehensive evaluation model established by the principal component analysis of example 2.
In order to further test the evaluation effect of the tobacco shred quality main component analysis method, sensory evaluation is carried out on 5 batches of cigarettes in a dark evaluation mode, and the evaluation result is shown in table 6.
Sensory evaluation result of 65 batches of cigarettes in table
As can be seen from the above table, the scores of the sensory evaluation of 5 batches of cigarettes are ranked as follows:
test III > test IV > test I > test II > test V, the ranking being in accordance with the results of example 2.
Example 4
A tobacco shred quality comprehensive evaluation system based on a principal component analysis method.
The system is used for implementing the comprehensive tobacco shred quality evaluation method based on the principal component analysis method, and comprises the following steps:
firstly, a sampling unit:
1, tobacco shred sampling: randomly selecting 5 batches of cut tobacco, randomly sampling 6 times in each batch, and taking the cut tobacco from a discharge port of a cut tobacco mixing and flavoring machine by using a sampling container, wherein the sampling quality of a cut tobacco sample to be tested is (1000.0 +/-100.0) g/time, and the time interval of single sampling is 15 +/-5 min; if the quality of the tobacco shred sample exceeds the standard range, discarding the sample and resampling;
2, sampling of cigarette finished products:
sampling finished cigarettes at the outlets of cigarette making machines corresponding to 5 batches of tobacco shred sampling by adopting a fixed sampling machine and fixed sampling personnel, randomly taking 150 samples of each group of samples each time, continuously sampling for 6 times under the condition of time interval of 30 +/-5 min, placing the samples in a constant temperature and humidity box for balancing for 24h, sealing and labeling, and reserving samples for later use;
II, a data detection unit:
detecting the tobacco shred structure and the physical indexes of the finished cigarette product;
the tobacco shred structure comprises 4 indexes of long shreds, medium shreds, short shreds and broken shreds which are related to the quality of the tobacco shreds;
the physical indexes of the finished cigarette product comprise 7 indexes of single cigarette weight, dust content, suction resistance, hardness, length, circumference and end cut;
wherein:
in the tobacco shred structure detection, the tobacco shred length definition standard is as follows: shredding: the length of the tobacco shreds is less than 1.0 mm; short silk: the length of the tobacco shreds is more than 1.0mm and less than 2.5 mm; medium silk: the length of the tobacco shreds is more than 2.5mm and less than 3.5 mm; filament: the length of the tobacco shreds is more than 3.5 mm;
in the detection of physical indexes of finished cigarette products, 150 samples extracted at a single time are as follows: 100 samples are used for measuring the end part filament falling and circumference, 20 samples are used for measuring the dust content, and the other 30 samples are used for measuring the cigarette weight, length, hardness and suction resistance, and the average values are respectively taken for recording;
thirdly, a principal component analysis unit:
1, a data standardization module:
the data processing device is used for carrying out standardization processing on the data;
2 principal component extraction module
For extracting principal components and calculating eigenvalues and eigenvectors:
determining the principle of the number m of the main components: the characteristic value is greater than 1; the number m of the main components is less than the number of the detection indexes n;
3, an evaluation model construction module:
creating a tobacco shred quality comprehensive evaluation model F, wherein the tobacco shred quality comprehensive evaluation model F comprises the following steps:
F=(a11+a12+a13+......a1m)×X1×Z1+(a21+a22+a23+......a2m)×X2×Z2+(a31+a32+a33+......a3m)×X3×Z3+......+(an1+an2+an3+......anm)×Xn×Zm
wherein, a11、a12、a13....anmScoring coefficients for the principal components, i.e., feature vectors of the principal components;
Z1、Z2、Z3...Zmvariance contribution rate for extracting principal component;
X=(X1,X2,...,Xn) Is an n-dimensional random variable, X1,X2,X3...X11Sequentially representing filament, medium filament, short filament, broken filament, length, single weight, circumference, suction resistance, hardness, end part filament drop and end content;
m is the number of main components, and m is less than n;
fourthly, a result and analysis unit:
and calculating the comprehensive score of the main components of the quality of the cut tobacco in the 5 batches of cut tobacco mixing and flavoring processes through the model F, and then performing comprehensive evaluation on the quality of the cut tobacco according to the score.
It should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments or portions thereof without departing from the spirit and scope of the invention.
Claims (4)
1. The comprehensive tobacco shred quality evaluation method based on the principal component analysis method is characterized by comprising the following steps of: comprises that
S1, sampling
S1.1, tobacco shred sampling
Randomly selecting M batches of cut tobacco, randomly sampling each batch for N times, wherein the sampling quality of a cut tobacco sample to be detected is (1000.0 +/-100.0) g/time, and the single sampling time interval is 15 +/-5 min;
s1.2, sampling of cigarette finished products
Adopting a fixed sampling machine table and fixed sampling personnel form, randomly taking 150 samples in each group, continuously sampling for N times under the condition of time interval of 30 +/-5 min, placing the samples in a constant temperature and humidity chamber for balancing for 24h, sealing and labeling, and reserving the samples for later use;
s2, data detection
Detecting the tobacco shred structure and the physical indexes of the finished cigarette product;
the tobacco shred structure comprises 4 indexes of long shreds, medium shreds, short shreds and broken shreds which are related to the quality of the tobacco shreds;
the physical indexes of the finished cigarette product comprise 7 indexes of single cigarette weight, dust content, suction resistance, hardness, length, circumference and end cut;
s3, principal component analysis
S3.1, data normalization
Carrying out standardization processing on the data;
s3.2, extracting main components
Calculating a characteristic value and a characteristic vector:
the normalized data are represented as an N × p matrix (N ═ 1, 2.., 11; p ═ 1,2, …, N), the corresponding eigenvalues λiOrthogonalized unit feature vector Z for component variance contributioniFor variance contribution rate, the calculation formula is:
Zithe size of (2) represents the capability of the component to react with information, and the principal component is extracted according to the numerical value of the component;
s3.3, establishing an evaluation model
Creating a tobacco shred quality comprehensive evaluation model F, wherein the tobacco shred quality comprehensive evaluation model F comprises the following steps:
F=(a11+a12+a13+......a1m)×X1×Z1+(a21+a22+a23+......a2m)×X2×Z2+(a31+a32+a33+......a3m)×X3×Z3+......+(an1+an2+an3+......anm)×Xn×Zm
wherein, a11、a12、a13....anmScoring the principal component by a factor;
Z1、Z2、Z3...Zmvariance contribution rate for extracting principal component;
X=(X1,X2,...,Xn) Is an n-dimensional random variable, X1,X2,X3...X11Sequentially representing filament, medium filament, short filament, broken filament, length, single weight, circumference, suction resistance, hardness, end part filament drop and end content;
m is the number of main components;
s4, results and analysis
And calculating the comprehensive score of the main components of the quality of the cut tobacco in the 5 batches of cut tobacco mixing and flavoring processes through the model F, and then performing comprehensive evaluation on the quality of the cut tobacco according to the score.
2. The comprehensive tobacco shred quality evaluation method based on the principal component analysis method is characterized by comprising the following steps of:
step S2:
in the tobacco shred structure detection, the tobacco shred length definition standard is as follows:
shredding: the length of the tobacco shreds is less than 1.0 mm; short silk: the length of the tobacco shreds is more than 1.0mm and less than 2.5 mm; medium silk: the length of the tobacco shreds is more than 2.5mm and less than 3.5 mm; filament: the length of the tobacco shreds is more than 3.5 mm;
in the detection of physical indexes of finished cigarette products, 150 samples extracted at a single time are as follows:
100 samples are used for measuring the end part filament falling and circumference, 20 samples are used for measuring the dust content, and the other 30 samples are used for measuring the cigarette weight, the length, the hardness and the suction resistance, and the average values are respectively taken for recording.
3. The comprehensive tobacco shred quality evaluation method based on the principal component analysis method is characterized by comprising the following steps of:
in step S3.1, the data normalization formula is:
4. The comprehensive tobacco shred quality evaluation method based on the principal component analysis method is characterized by comprising the following steps of:
step S3.2, determining the number m of the main components:
the characteristic value is greater than 1; the number m of the principal components is smaller than the number of the detection indexes n.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110989940.8A CN113869641A (en) | 2021-08-26 | 2021-08-26 | Tobacco shred quality comprehensive evaluation method based on principal component analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110989940.8A CN113869641A (en) | 2021-08-26 | 2021-08-26 | Tobacco shred quality comprehensive evaluation method based on principal component analysis method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113869641A true CN113869641A (en) | 2021-12-31 |
Family
ID=78988363
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110989940.8A Withdrawn CN113869641A (en) | 2021-08-26 | 2021-08-26 | Tobacco shred quality comprehensive evaluation method based on principal component analysis method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113869641A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116183834A (en) * | 2023-03-06 | 2023-05-30 | 江苏中烟工业有限责任公司 | Method for evaluating applicability of tobacco leaf raw materials to cigarettes of different circumferences |
-
2021
- 2021-08-26 CN CN202110989940.8A patent/CN113869641A/en not_active Withdrawn
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116183834A (en) * | 2023-03-06 | 2023-05-30 | 江苏中烟工业有限责任公司 | Method for evaluating applicability of tobacco leaf raw materials to cigarettes of different circumferences |
CN116183834B (en) * | 2023-03-06 | 2023-09-19 | 江苏中烟工业有限责任公司 | Method for evaluating applicability of tobacco leaf raw materials to cigarettes of different circumferences |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107796782B (en) | Redrying quality stability evaluation method based on tobacco leaf characteristic spectrum consistency measurement | |
CN107330555A (en) | Silk making process parameter weighting method based on random forest regression | |
CN102509243A (en) | Method and system for evaluating quality in process of manufacturing cigarette | |
CN105699437B (en) | A kind of quick nondestructive differentiating method of the different freshness hickory nuts based on electronic nose | |
CN110196265B (en) | Method for evaluating distribution stability of cut tobacco in cigarette branches in batch based on distribution probability | |
CN105717066B (en) | A kind of near infrared spectrum identification model based on weighted correlation coefficient | |
CN111950795B (en) | Random forest-based prediction method for loosening and conditioning water adding proportion | |
CN113869641A (en) | Tobacco shred quality comprehensive evaluation method based on principal component analysis method | |
CN105334738B (en) | A kind of method of evaluating performance suitable for tobacco processing course pid control circuit | |
CN110163479A (en) | Method for establishing main cause model of quality abnormity problem of tobacco rolling workshop | |
CN115099457A (en) | On-line predicting and analyzing system for tobacco shred structure | |
CN111642782A (en) | Tobacco leaf raw material efficacy positioning method based on cigarette formula requirements | |
CN115078380A (en) | Method for evaluating cigarette combustion appearance | |
CN113449792A (en) | Method for nondestructive rapid detection of food quality | |
CN110887809B (en) | Method for measuring stem content in tobacco shreds based on near infrared spectrum technology | |
CN113030007B (en) | Method for rapidly testing quality stability of tobacco essence based on similarity learning algorithm | |
CN118114149A (en) | Detection method and system for liquid crystal display panel | |
CN118190859A (en) | Rapid identification method for wool and cashmere fibers | |
CN116383623A (en) | Method for selecting rejection rate characteristics of cigarette machine based on machine learning | |
CN113310937A (en) | Method for rapidly identifying high-temperature sterilized milk, pasteurized fresh milk of dairy cow and reconstituted milk of milk powder | |
CN113310929A (en) | Soybean powder doped in high-temperature sterilized milk and spectral identification method of doping proportion thereof | |
CN115859784A (en) | Method for establishing production process parameter and cigarette sensory quality characteristic correlation model | |
CN113030008B (en) | Near-infrared online quality detection method for cattail pollen charcoal processed product | |
CN112101817B (en) | Evaluation method for homogenization degree of finished cured tobacco products processed in batches | |
CN113310933A (en) | Spectrum identification method for number of days for storing raw buffalo milk |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211231 |
|
WW01 | Invention patent application withdrawn after publication |