Method for evaluating the burning appearance of cigarettes

Abstract

The present invention discloses a method for evaluating the combustion appearance of a cigarette. It includes step S1 of collecting data on the external attributes of the cigarette, step S2 of standardizing the data of the cigarette, step S3 of constructing a fuzzy comprehensive evaluation matrix for the external attributes of the cigarette, and step S4 of determining the weight vector of each attribute by factor analysis to perform the calculation of the fuzzy comprehensive evaluation model and determining the final evaluation level based on the calculated fuzzy vector B. For the problem of analyzing and evaluating the combustion appearance of a cigarette, the present invention can determine the factor weight vector of data from different attributes and different samples by establishing a fuzzy comprehensive model of the combustion appearance of the cigarette, and obtain the level evaluation of the combustion appearance of the cigarette in combination with the subjective evaluation of each appearance attribute of the cigarette sample.

Description

Technical Field

[0001] The present invention relates to the technical field of the combustion appearance of cigarettes, and specifically to a method for evaluating the combustion appearance of cigarettes.

Background Art

[0002] With the rapid development of the market economy, the cigarette industry has shown a good development trend, and gradually realized the computer-aided production of tobacco leaves, environmentally friendly prevention and control technologies, and the systematization of quality traceability. As people's consumption level gradually improves, the manufacturing quality standards of cigarettes are becoming increasingly high. In the production of cigarettes, technology updates and iterations continue, and a manufacturing process that reduces the harm of cigarettes and improves product quality is being explored.

[0003] The research and development of cigarettes must pass review procedures such as clinical trial observation, pharmacological and toxicological testing, and component analysis by the hygiene department. The manufacture of cigarettes must undergo quality assessment in accordance with national standards GB5606~5610-85 and pass quality evaluation. Quality inspection mainly focuses on evaluating the internal quality, external quality, and tar content of cigarettes. The burning appearance of cigarettes is an important visual form that smokers intuitively perceive during the burning process of cigarette products, and its quality directly influences smokers' judgment of the quality of cigarette products. Furthermore, during the smoking process of cigarettes, the dropping and scattering of cigarette ash frequently occurs, and in serious cases, the burning cone may detach. This not only pollutes the environment but is also one of the factors that causes discomfort and dissatisfaction to smokers and passive smokers regarding the environment they are in, and can also affect smokers' loyalty to cigarette brands and even pose safety risks. Meanwhile, with the development of cigarette technology and the continuous improvement of consumption levels and quality, smokers' demands for cigarette quality have also increased, and attention and selection of cigarette products are gradually expanding from taste, price, and packaging to more factors. The appearance of a cigarette after burning is attracting increasing attention from smokers due to its intuitive nature. From a smoker's perspective, the burning time, the length of time the ash remains on the cigarette, and the experience provided by burning and smoking a cigarette are the most important evaluation criteria.

[0004] In recent years, the tobacco industry has conducted research on the combustion appearance of cigarettes, mainly focusing on two areas: ash astringency and ash dispersion, and detection methods. However, there is no perfect method for evaluating the combustion appearance of cigarettes during the smoking process. Objective evaluation of the combustion appearance of cigarettes is crucial for objective analysis of cigarette appearance and for comparing the combustion appearance of different cigarettes.

[0005] To solve the above problems, we propose the present invention. [Overview of the project]

[0006] The present invention aims to comprehensively evaluate the combustion appearance during the smoking process of a cigarette, to form a perfect analytical evaluation method, to provide effective quality information for cigarettes to the cigarette manufacturing industry, and to provide effective feedback for improving cigarette manufacturing methods.

[0007] To fill in the gaps in the above evaluation items, the present invention provides a method for analyzing and evaluating the combustion appearance of a cigarette, which establishes a fuzzy overall evaluation model of combustion appearance applicable to all types of cigarettes, and ultimately obtains an overall score that reflects the combustion appearance of the cigarette under test by recording each external attribute of the cigarette using a panoramic camera and inputting it into the model.

[0008] A first aspect of the present invention provides a method for evaluating the burning appearance of a cigarette, comprising the following steps.

[0009] S1, a step of collecting data on the external attributes of a cigarette, wherein the external attributes of the cigarette include a total of 11 attributes in real time while the cigarette is being smoked, including crack rate, ash column displacement, ash column grayness, ash column length, ash column area, ash shrinkage rate, ash scattering value, ash retention rate, carbonization line width, carbonization line uniformity, and combustion rate.

[0010] While collecting external attributes of a cigarette can be done using devices and methods commonly used in the prior art, it is preferable to collect external attributes of a cigarette by capturing the real-time smoking state of the cigarette with a camera system using a manipulator that simulates the entire process of a human smoking a cigarette. Here, the manipulator that simulates the process and environment of a human smoking a cigarette can be referred to in the patent application filed by the present applicant on April 20, 2020, application number 202010329623.9, with the title of the invention "Manipulator for simulating the entire process of a human smoking a cigarette and a simulation method thereof." Furthermore, a method for collecting external attributes of a cigarette using this manipulator can be referenced from a patent application filed by the present applicant on April 26, 2020, application number 202010340747.7, with the title of the invention "Method for measuring omnidirectional ash astringency during cigarette combustion by simulating the entire process of a human smoking a cigarette using a manipulator." In addition, the technical means of the manipulator and the method for measuring the ash astringency of a cigarette are incorporated into this application.

[0011] The meanings of the above external attributes of cigarettes are as follows: The cracking rate is the ratio of the area of ​​fallen ash or cracks on the ash column to the total surface area of ​​the ash column. The ash column displacement is the maximum displacement angle between the burnt ash column of the cigarette and the axis of the cigarette's filter rod. The grayness of the ash column reflects the relative color difference in the same reference system of the remaining part of the cigarette after it has burned, excluding the cracks; a higher value indicates a whiter ash column. The length of the ash column is the length of the ash column formed after the cigarette has burned. The area of ​​the ash column is the width of the ash column formed after the cigarette has burned. The ash shrinkage rate is the ratio of the area of ​​the ash column after the cigarette has burned to the area before unburning. The ash scattering value is the area of ​​ash that fell to locations other than the designated ash-beating position. The ash retention rate is the ratio of the maximum length of the ash column that can be held during the burning process of a cigarette to the length of the cigarette that has burned down to the butt. The width of the carbonization line is the width of the carbonization line at the bottom of the tobacco burning cone. The uniformity of the carbonization line is the maximum deviation angle between the carbonization line at the bottom of the tobacco burning cone and the cross-section of the tobacco. The burning rate is the ratio of the burning time to the burning length.

[0012] S2 is a step to standardize the data for cigarettes, and the data is standardized using the following formula: Steps to obtain TIFF0007850820000001.tif5170. TIFF0007850820000002.tif31170

[0013] S3 is a step in constructing a fuzzy overall evaluation matrix of the external attributes of cigarettes, Attribute set Each element in TIFF0007850820000003.tif5170 is defined as representing a different external attribute of a cigarette, evaluation set TIFF0007850820000004.tif5170 represents a different evaluation result) TIFF0007850820000005.tif18170 The fuzzy set of single-factor evaluations for all attributes is: Steps to create TIFF0007850820000006.tif5170.

[0014] S4, the weight vectors for each attribute were determined by factor analysis, and the fuzzy composite evaluation model was calculated. A step in which the final evaluation level is determined based on TIFF0007850820000007.tif5170, If we establish a factor analysis model of the following form for each attribute, The above factor analysis model, TIFF0007850820000008.tif31170, is represented in the following matrix format: TIFF0007850820000009.tif11170 Each common factor is an independent normally distributed random variable with a mean of 0 and a variance of 1, and its covariance matrix is Assuming that TIFF0007850820000010.tif5170 is true, each special factor and each special factor and each common factor are independent of each other, i.e. Assuming it is TIFF0007850820000011.tif24170, If the result is close to TIFF0007850820000012.tif58170, i.e., if the residual is less than 5%, the model fit is judged to be good and the process continues. By rotating the factors counterclockwise, the interpretability and uniqueness of the transformed factors become clear, and when an orthogonal rotation of the plane is performed on two factors, TIFF0007850820000013.tif5170 has the following format: Taking the orthogonal matrix TIFF0007850820000014.tif5170, TIFF0007850820000015.tif13170TIFF0007850820000016.tif5170 are rotation factor loading matrices, and the model takes the following form: TIFF0007850820000017.tif6170 In this case, the common factor is The result is TIFF0007850820000018.tif6170, and the current purpose of rotating the two factors is to split the variable into two parts that are explained by different factors, We need to maximize the variance with TIFF0007850820000019.tif6170, and the relative variance between the two is, It can be represented as TIFF0007850820000020.tif15170, The influence of TIFF0007850820000021.tif5170 and the influence of different dependencies of each attribute variable on common factors have been eliminated. next, Satisfying TIFF0007850820000022.tif24170, For TIFF0007850820000023.tif50170, rotate these factors in pairs, and in one complete rotation process, Select two elements at a time from TIFF0007850820000024.tif5170 and rotate them counterclockwise, and in total... It has TIFF0007850820000025.tif5170, and once one rotation is complete it proceeds to the next cycle, and in this iteration TIFF0007850820000026.tif5170 The rotation gradually increases until the difference between the total variance of one cycle and the total variance of the previous cycle is 5% or less, at this point By multiplying TIFF0007850820000027.tif5170 with the standard orthogonalized eigenvectors TIFF0007850820000028.tif12170 Then, through fuzzy transformation, TIFF0007850820000029.tif111701 corresponds to a 1:1 ratio, and the evaluation corresponding to the largest element is the step, which is the final evaluation level of the fuzzy overall evaluation of cigarettes.

[0015] Preferably, in step S1, a manipulator that simulates the entire process of human smoking of a cigarette is used to capture the real-time smoking state of the cigarette by a camera system, so as to collect the external attributes of the cigarette.

[0016] Preferably, in step S3, TIFF0007850820000030.tif5170 in descending order of level, TIFF0007850820000031.tif5170.

[0017] Preferably, in step S3, TIFF0007850820000032.tif5170 is obtained by the scoring of experts.

[0018] Compared with the prior art, the present invention has the following beneficial effects. 1. For the problem of analyzing and evaluating the combustion appearance of cigarettes, the present invention establishes a fuzzy comprehensive model of the combustion appearance of cigarettes, determines the factor weight vector of data of different attributes from different samples, and can obtain the level evaluation of the combustion appearance of cigarettes. 2. The evaluation method of the present invention is consistent with the actual appearance performance results during the combustion of cigarettes, and the evaluation results can objectively reflect the appearance performance during the combustion of cigarettes, which is very important for the objective analysis of the ash convergence and ash scattering properties of cigarettes and the comparison of the external attributes of different cigarettes.

Brief Description of the Drawings

[0019] [Figure 1] It is an overall flowchart of the method in the present invention.

Embodiments for Carrying Out the Invention

[0020] The present invention will be described in detail below with reference to specific examples, but the embodiments of the present invention are not limited to these. Experimental methods in the examples where specific conditions are not specified generally follow conventional conditions and conditions described in manuals, or conditions suggested by the manufacturer. The general equipment, materials, reagents, etc. used are all commercially available unless otherwise specified. The raw materials required in the following examples and comparative examples are all commercially available.

[0021] Example 1 The present invention provides a method for analyzing and evaluating the combustion appearance of a cigarette, comprising the following steps.

[0022] S1 collects data on the external attributes of the cigarette. Under simulated smoking conditions and environmental simulations performed by a robot, the real-time state of the cigarette, including crack rate, ash column displacement, ash column grayness, ash column length, ash column area, ash shrinkage rate, ash scattering value, ash retention rate, carbonization line width, carbonization line uniformity, and combustion rate, is captured from various angles using multiple cameras.

[0023] Two sets of industrial multi-axis manipulators are used to clamp a cigarette and simulate the arm and wrist movements of a smoker while smoking, with the manipulator's movement speed and angle adjustable within a predetermined range. The manipulator's fingers simulate the action of ash beating, with the strength, interval, and number of repetitions adjustable within a predetermined range. A suction device is used to simulate the smoking process, with the suction mode, suction time, suction volume, and suction interval adjustable within a predetermined range. Data acquisition is performed using a high-speed, high-precision smart camera set with a resolution of 1600*1200 and a frame rate of 50fps or higher.

[0024] The external attributes of collected cigarettes include a total of 11 attributes: cracking rate, ash column displacement, ash column grayness, ash column length, ash column area, ash shrinkage rate, ash scattering value, ash retention rate, carbonization line width, carbonization line uniformity, and burning rate. The meaning of these external attributes is as follows: The cracking rate is the ratio of the area of ​​ash fallout or cracks on the ash column to the total surface area of ​​the ash column. The ash column displacement is the maximum displacement angle between the burnt ash column of the cigarette and the axis of the cigarette's filter rod. The ash column grayness reflects the relative color difference in the same reference frame of the remaining part of the ash column after the cigarette has burned, excluding the cracks; a higher value indicates a whiter ash column. The ash column length is the length of the ash column formed after the cigarette has burned. The ash column area is the width of the ash column formed after the cigarette has burned. The ash shrinkage rate is the ratio of the area of ​​the ash column after the cigarette has burned to the area before unburning. The ash scattering value is the area of ​​ash that falls outside the designated ash-beating location. The ash retention rate is the ratio of the maximum ash column that can be held during the burning process of a cigarette to the length of the cigarette that burns down to the butt. The carbonization line width is the width of the carbonization line at the bottom of the cigarette's burning cone. The uniformity of the carbonization line is the maximum deviation angle between the carbonization line at the bottom of the cigarette's burning cone and the cross-section of the cigarette. The burning rate is the ratio of the burning time to the burning length.

[0025] In this example, 390 samples of standard-thickness cigarettes were collected, and the 11 appearance indicators described above were collected for each sample. The collected values ​​for each indicator for each sample are shown in Table 1. Here, This represents TIFF0007850820000033.tif18170.

[0026] TIFF0007850820000034.tif249170TIFF0007850820000035.tif252170TIFF0007850820000036.t if252170TIFF0007850820000037.tif252170TIFF0007850820000038.tif252170TIFF00078508200 00039.tif252170TIFF0007850820000040.tif252170TIFF0007850820000041.tif252170TIFF000 7850820000042.tif252170TIFF0007850820000043.tif252170TIFF0007850820000044.tif155170

[0027] S2, standardize the data for cigarettes. TIFF0007850820000045.tif5170 defines the data of cigarettes collected in step S1, TIFF0007850820000046.tif5170 represents the measured values ​​of 11 external attributes of tobacco. The data is standardized using the following formula so that the mean and unit variance are zero. Obtain TIFF0007850820000047.tif5170. TIFF0007850820000048.tif24170

[0028] Table 2 shows the standardized data for each indicator, using four cigarette samples (1, 100, 167, and 300) from the examples.

[0029] TIFF0007850820000049.tif40170

[0030] S3. Construct a fuzzy overall evaluation matrix of the external attributes of cigarettes. TIFF0007850820000050.tif31170

[0031] Obtained through expert scoring TIFF0007850820000051.tif18170 The fuzzy set of single-factor evaluations for all attributes is: TIFF0007850820000052.tif5170

[0032] Using four cigarette samples (1, 100, 167, and 300) from the examples, the fuzzy overall evaluation matrices for each sample are shown in Tables 3-6.

[0033] TIFF0007850820000053.tif87170

[0034] TIFF0007850820000054.tif87170

[0035] TIFF0007850820000055.tif87170

[0036] TIFF0007850820000056.tif87170

[0037] S4. Factor analysis is used to determine the weight vectors of each attribute and calculate the fuzzy composite evaluation model. A factor analysis model of the following form is established for each attribute, TIFF0007850820000057.tif31170 In this case, the factor analysis model can be represented in the following matrix form: TIFF0007850820000058.tif11170 Each common factor is an independent normally distributed random variable with a mean of 0 and a variance of 1. TIFF0007850820000059.tif5170 Each special factor and each special factor and the common factor are independent of each other, that is Assuming it is TIFF0007850820000060.tif24170, It is TIFF0007850820000061.tif12170, It is broken down into something like TIFF0007850820000062.tif12170, The filename becomes TIFF0007850820000063.tif22170.

[0038] Up to this point, regarding the examples, the correlation coefficient matrix calculated using 390 samples is shown in Table 7, and the principal component factor loading matrix is ​​shown in Table 8.

[0039] TIFF0007850820000064.tif87170

[0040] TIFF0007850820000065.tif87170

[0041] The variance of special factors is, It is estimated using the diagonal elements of TIFF0007850820000066.tif5170, i.e. The filename is TIFF0007850820000067.tif9170, The residual matrix is It can be represented as TIFF0007850820000068.tif6170.

[0042] Since the explanations of the principal components are not strictly unique, rotating the factors counterclockwise clarifies the interpretability and uniqueness of the transformed factors. When a plane orthogonal rotation is performed on two factors, TIFF0007850820000069.tif5170 has the following format: Taking the orthogonal matrix TIFF0007850820000070.tif5170, TIFF0007850820000071.tif13170TIFF0007850820000072.tif5170 are rotation factor loading matrices, and the model takes the following form: TIFF0007850820000073.tif6170 In this case, the common factor is As of TIFF0007850820000074.tif6170, We need to maximize the variance with TIFF0007850820000075.tif6170. The relative variance between the two is: It can be represented as TIFF0007850820000076.tif15170. This format allows, This eliminates the influence of TIFF0007850820000077.tif5170 and the influence of different dependencies on common factors for each attribute variable. Next, The condition TIFF0007850820000078.tif69170 should be met. TIFF0007850820000079.tif5170 These factors are rotated in pairs. In one complete rotation process, TIFF0007850820000080.tif5170 Select two factors and rotate them counterclockwise, TIFF0007850820000081.tif5170 Once one rotation is complete, proceed to the next cycle, and in this repetition TIFF0007850820000082.tif5170 The rotation stops until the difference between the total variance of a given cycle and the total variance of the previous cycle is 5% or less. TIFF0007850820000083.tif5170 Factor weight vectors for fuzzy overall evaluation obtained by multiplying with standard orthogonalized eigenvectors: Obtain TIFF0007850820000084.tif6170.

[0043] Regarding the examples, the rotation factor loading matrices calculated using 390 samples are shown in Table 9.

[0044] TIFF0007850820000085.tif87170

[0045] Correspondingly, the factor weight vector for the fuzzy overall evaluation is as follows: TIFF0007850820000086.tif18170, in other words The filename is TIFF0007850820000087.tif5170. TIFF0007850820000088.tif51701 corresponds to a 1:1 ratio, and the evaluation corresponding to the largest element is the final evaluation level of the fuzzy overall evaluation of the cigarette.

[0046] Using four cigarette samples 1, 100, 167, and 300 from the example, the fuzzy vector for sample 1 is The data is TIFF0007850820000089.tif5170, and the burning appearance of the cigarette is rated as "fairly good," and the fuzzy vector for sample 100 is The filename is TIFF0007850820000090.tif5170, and the burning appearance of the cigarette is rated as "good," and the fuzzy vector for sample 167 is The data is TIFF0007850820000091.tif5170, and the burning appearance of the cigarette is rated as "normal", and the fuzzy vector for sample 300 is The product code is TIFF0007850820000092.tif5170, and the burning appearance of this cigarette is rated as "very good".

Claims

1. A method for evaluating the combustion appearance of a cigarette, performed by a computer, S1, a step of collecting data on the external attributes of a cigarette, wherein the external attributes of the cigarette include a total of 11 attributes in real time during the smoking state of the cigarette, including crack rate, ash column displacement, ash column grayness, ash column length, ash column area, ash shrinkage rate, ash scattering value, ash retention rate, carbonization line width, carbonization line uniformity, and combustion rate. S2, a step to standardize the data for cigarettes, wherein the data is standardized using the following formula: Steps to obtain, (In the formula, This is the data on cigarettes collected in step S1. This shows the measured values ​​of 11 external attributes of tobacco. (That is the case.) S3 is a step in constructing a fuzzy overall evaluation matrix of the external attributes of cigarettes, Attribute set It is stipulated that, evaluation set The fuzzy set of single-factor evaluations for all attributes is: The steps that make up the structure, S4. Factor analysis was used to determine the weight vectors of each attribute, and the fuzzy composite evaluation model was calculated. A step in which the final evaluation level is determined based on the following: If we establish a factor analysis model of the following form for each attribute, (Here, This represents the number of common factors (i.e., the number of variables), These are common factors that each variable possesses. It is a special factor that is related only to, It is a factor loading, (It constitutes...) The above factor analysis model is represented in the following matrix format: Each common factor is an independent normally distributed random variable with a mean of 0 and a variance of 1, and its covariance matrix is Assuming that each special factor is independent of the others and between special factors and common factors, that is (Here, (This represents the variance of special factors.) Assuming that is the case, And, It is broken down as follows: When defined as the corresponding standard orthogonalized eigenvector, And so, The variance of special factors is, It is estimated using the diagonal elements, i.e. And, The residual matrix is It can be expressed as, By rotating the factors counterclockwise, the interpretability and uniqueness of the transformed factors become clear, and when an orthogonal rotation of the plane is performed on two factors, It has the following form: Taking the following orthogonal matrix, This is the rotation factor loading matrix, and the model takes the following form: At this time, the common factor is Therefore, the current purpose of rotating the two factors is to divide the variable into two parts that are explained by different factors. It is necessary to maximize the variance between the two, and the relative variance between the two is, It can be expressed as, The influence of the various factors and the different degrees of dependence of each attribute variable on the common factor are eliminated. next, Satisfying the conditions, By rotating these factors in pairs, in one complete rotation process, Select two at a time and rotate them counterclockwise, and in total It has a rotation, and once one rotation is completed it moves on to the next cycle, and in this repetition It gradually increases, and when the difference between the total variance of one cycle and the total variance of the previous cycle becomes 5% or less, the rotation stops, and at this point the final By obtaining this and multiplying it by the standard orthogonalized eigenvectors, we obtain a fuzzy overall evaluation. Subsequently, through fuzzy transformation, The elements correspond one-to-one, and the evaluation corresponding to the most significant element is the final evaluation level of the Fuzzy Overall Evaluation of cigarettes, which is the step and A method for evaluating the burning appearance of a cigarette, characterized by including [a certain element].

2. The method for evaluating the combustion appearance of a cigarette according to claim 1, characterized in that in step S1, an external attribute of the cigarette is collected by capturing the real-time smoking state of the cigarette with a camera system using a manipulator that simulates the entire process of a person smoking a cigarette.

3. In step S3, In order of highest level: A method for evaluating the burning appearance of a cigarette according to claim 1, characterized in that...

4. In step S3, A method for evaluating the burning appearance of a cigarette according to claim 1, characterized in that the result is obtained by scoring by an expert.

Images