Tobacco shred thermophysical property online detection method based on infrared spectrum and application

The online detection method using infrared spectroscopy and thermal property prediction models addresses the challenge of rapid, accurate thermal property measurement in tobacco processing, enhancing quality control and thermal processing efficiency.

CN119064410APending Publication Date: 2024-12-03HUBEI CHINA TOBACCO INDUSTRY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411352341.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-26
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Existing methods for measuring the thermal properties of tobacco are limited to laboratory settings and cannot achieve rapid, real-time detection, failing to accurately reflect the processing characteristics of tobacco materials due to their varying thermophysical properties.

Method used

An online detection method using infrared spectroscopy that couples infrared thermometers and infrared moisture meters with thermal conductivity prediction models to predict thermal properties of cut tobacco, incorporating a thermal property prediction model that integrates leaf group formula type, wet basis moisture content, and temperature as inputs, and specific heat capacity and thermal conductivity as outputs, enabling real-time detection.

Benefits of technology

Enables fast and accurate online detection of thermal properties of cut tobacco, improving quality control and thermal processing technology by reducing detection time and enhancing detection accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BDA0005062684450000101
    Figure BDA0005062684450000101
  • Figure BDA0005062684450000111
    Figure BDA0005062684450000111
  • Figure HDA0005062684460000011
    Figure HDA0005062684460000011
Patent Text Reader

Abstract

The invention provides a tobacco shred thermophysical property detection method and application thereof. According to the tobacco shred thermophysical property detection method based on the infrared spectrum technology provided by the invention, equipment such as an infrared thermometer and an infrared moisture meter and the thermophysical property prediction model are coupled, so that rapid and accurate online real-time detection of the tobacco shred thermophysical property can be realized, and a tobacco product quality regulation and control method is greatly improved; and a new path is provided for regulation and control of a hot working process in a cut tobacco making process. According to the method, real-time and efficient detection of the thermophysical property is realized by relying on a tobacco shred thermophysical property database established in a laboratory. According to the method, tobacco shred sampling and thermophysical property measurement test are carried out in the database establishment stage, then online detection of the thermophysical property of the tobacco shred can be realized by using online temperature, moisture content and other detection equipment, the detection time is short, and the detection means fully meet the timeliness required by production.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of online detection of thermal properties of tobacco, and relates to a detection method and application of thermal properties of tobacco, and in particular to an online detection method and application of thermal properties of tobacco based on infrared spectroscopy. Background Art

[0002] Before and after the drying process, the temperature and moisture content of tobacco will fluctuate. The amplitude of the fluctuation reflects the intensity of the processing in the process, that is, the calorific value. However, due to different types of tobacco materials have different thermophysical properties, different leaf group formulas will have different energy changes during the processing, even if the temperature and moisture content change by the same value. Therefore, in order to obtain the energy changes of tobacco during the drying process, it is necessary to test the thermophysical parameters before and after processing.

[0003] During the processing, the internal components of tobacco shreds change, and accordingly, their thermal properties such as specific heat capacity and thermal conductivity also change. The thermal properties of tobacco shreds are strongly correlated with parameters such as temperature and moisture content. By constructing a functional relationship between temperature, moisture content and thermal properties, a thermal property prediction model is obtained, and the thermal properties of tobacco shreds can be predicted by detecting temperature and moisture content. The thermal processing steps in the tobacco making process are particularly critical to the sensory quality of cigarette products. However, the existing online detection devices for tobacco materials mainly detect the temperature or moisture content of the material in a single dimension, lack the detection of the thermal properties of the material before and after the processing, and cannot well reflect the processing characteristics of tobacco materials.

[0004] However, the existing method for measuring the thermal properties of tobacco needs to be carried out in the laboratory using instruments such as thermal constant analyzers. It cannot be directly applied on the production line and cannot achieve rapid and real-time detection of thermal properties.

[0005] Therefore, how to find a suitable way to improve the quality control method of tobacco products and solve the above-mentioned problems in the existing measurement of thermal properties of tobacco has become one of the urgent issues to be solved by many front-line researchers in the industry. Summary of the invention

[0006] In view of this, the technical problem to be solved by the present invention is to provide a detection method and application of the thermal properties of cut tobacco, in particular, an online detection method of the thermal properties of cut tobacco based on infrared spectroscopy. The detection method provided by the present invention can realize fast and accurate online thermal property detection of cut tobacco by coupling equipment such as infrared thermometers and infrared moisture meters with thermal property prediction models, providing a new path for the thermal processing technology control of the cut tobacco process.

[0007] The present invention provides a method for detecting thermal properties of cut tobacco, comprising the following steps:

[0008] 1) Measure the thermal properties of tobacco samples with various leaf group formulas and establish a tobacco thermal physical database;

[0009] The thermal property measurement includes the specific heat capacity, thermal conductivity, leaf group formula type, wet basis moisture content and temperature of the cut tobacco sample;

[0010] 2) Based on the tobacco cut thermophysical database constructed in the above steps, a thermophysical property prediction model of tobacco cut is constructed using a data-driven method;

[0011] The thermal property prediction model includes a prediction model that takes leaf group formula type, wet basis moisture content and temperature as inputs and specific heat capacity and thermal conductivity as outputs;

[0012] 3) coupling the detection device and the prediction model of the thermal properties of the tobacco obtained in the above steps to detect the thermal properties of the tobacco;

[0013] The detection equipment includes an infrared moisture meter and / or an infrared temperature meter.

[0014] Preferably, the shredded tobacco sample is a shredded tobacco sample after being cut;

[0015] The width of the tobacco leaves in the cut tobacco sample is 0.9 to 1.1 mm;

[0016] The step 1) specifically comprises the following steps:

[0017] (1) Balancing the tobacco sample to a uniform moisture content, and then measuring the moisture content of the tobacco sample;

[0018] (2) setting multiple temperatures and multiple wet-based moisture contents to conduct full factorial experimental design, using a constant temperature and humidity chamber to subject the tobacco samples to a constant temperature and humidity process, and obtaining tobacco samples with the above-set temperature and wet-based moisture content;

[0019] (3) Using a TPS thermal constant analyzer, the thermal physical properties of the tobacco samples with different combinations of wet-basis moisture content and temperature parameters obtained in the above steps were measured to construct a tobacco thermal physical database covering specific heat capacity, thermal conductivity, leaf group formula category, wet-basis moisture content and temperature.

[0020] Preferably, the temperature range of the multiple temperatures includes 25 to 75°C;

[0021] The multiple wet-basis moisture contents range from 12% to 19%.

[0022] Preferably, the thermophysical property prediction model is a thermophysical property prediction model that uses the tobacco cut thermophysical database as the data set required for model construction, uses leaf group formula category, wet basis moisture content, and temperature as model inputs, and uses specific heat capacity and thermal conductivity as model outputs.

[0023] Preferably, the data set is divided into a training set and a test set for machine learning;

[0024] In the data set, the proportion of the test set is not less than 20%.

[0025] Preferably, the data drive further comprises the following steps:

[0026] The data set is imported into the computing software, and different machine learning methods are compared to evaluate the training effect and / or accuracy, and the optimal model is obtained as the thermophysical property prediction model for subsequent use.

[0027] Preferably, the evaluation index includes one or more of root mean square error, mean absolute error and coefficient of determination;

[0028] The computing software includes Matlab.

[0029] Preferably, the data drive further comprises the following steps:

[0030] When multiple machine learning methods cannot achieve the desired detection results, the sample size of tobacco samples is increased, the tobacco thermophysical database is expanded, and the thermophysical property prediction model is rebuilt;

[0031] The detection method is specifically a real-time online detection method.

[0032] Preferably, the detection device is an online detection device;

[0033] The detection method also includes verification and correction steps;

[0034] The verification and correction step specifically includes taking online samples of the tobacco shreds, comparing the predicted thermophysical property parameters given by the thermophysical property prediction model of the tobacco shreds with the actual measured values ​​of the thermophysical properties of the samples, and verifying and correcting the model.

[0035] The present invention provides application of the detection method described in any one of the above technical solutions in the field of online real-time detection of thermal properties of tobacco.

[0036] The present invention provides a method for detecting the thermal properties of cut tobacco, comprising the following steps: firstly, measuring the thermal properties of cut tobacco samples of various leaf group formulas to establish a cut tobacco thermal physical database; the thermal property measurement includes the specific heat capacity, thermal conductivity, leaf group formula category, wet basis moisture content and temperature of the cut tobacco samples; then, based on the cut tobacco thermal physical database constructed in the above steps, a cut tobacco thermal property prediction model is constructed using a data-driven method; the thermal property prediction model includes a prediction model with leaf group formula category, wet basis moisture content and temperature as inputs and specific heat capacity and thermal conductivity as outputs; finally, coupling a detection device and the cut tobacco thermal property prediction model obtained in the above steps to realize the detection of the cut tobacco thermal properties; the detection device includes an infrared moisture meter and / or infrared thermometer. Compared with the prior art, the present invention provides a cut tobacco thermal property detection method based on infrared spectroscopy technology, which can realize fast and accurate online cut tobacco thermal property real-time detection by coupling infrared thermometers, infrared moisture meters and other equipment and thermal property prediction models, greatly improving the tobacco product quality control method and providing a new path for the thermal processing technology control of the cut tobacco process.

[0037] The present invention relies on the tobacco thermal property database established in the laboratory to realize real-time and efficient detection of thermal properties. It is only necessary to perform tobacco sampling and thermal property measurement tests in the database establishment stage, and then only by using online temperature, moisture content and other detection equipment, the online detection of tobacco thermal properties can be realized, the detection time is short, and the detection means fully meet the timeliness required for production. Compared with the existing tobacco thermal property detection technology mainly through non-steady-state method, such as transient plane heat source (TPS) method, which is the most widely used method at present, the TPS method is to supply instantaneous heat to the sample to be tested through a disc-shaped heat source, and at the same time, measure the response of the sample to thermal interference to obtain the sample temperature rise data. Then, the material thermal property parameters are determined based on the non-steady-state heat conduction differential equation. However, the disadvantages of the TPS method for measuring the thermal properties of tobacco are: sampling is required in advance, and the temperature of the sample to be tested is maintained constant; the detection takes a long time, which takes about tens of seconds. The present invention does not need to prepare samples or contact samples, and the detection time is short, which can realize efficient real-time online detection of tobacco thermal properties. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The present invention provides a simplified schematic diagram of the process of detecting the thermal properties of cut tobacco. DETAILED DESCRIPTION

[0039] In order to further understand the present invention, preferred embodiments of the present invention are described below in conjunction with examples, but it should be understood that these descriptions are only for further illustrating the features and advantages of the present invention rather than limiting the patent claims of the present invention.

[0040] All raw materials of the present invention have no particular limitation on their sources, and can be purchased from the market or prepared according to conventional methods known to those skilled in the art.

[0041] There is no particular limitation on the purity of all raw materials in the present invention. The present invention preferably uses analytically pure materials or materials with purity requirements conventional in the field of thermal property testing of cut tobacco.

[0042] All noun expressions and abbreviations in the present invention are conventional noun expressions and abbreviations in the field. Each noun expression and abbreviation is clear and unambiguous in its relevant application field. Those skilled in the art can clearly, accurately and uniquely understand them based on the noun expressions and abbreviations.

[0043] The present invention provides a method for detecting thermal properties of cut tobacco, comprising the following steps:

[0044] 1) Measure the thermal properties of tobacco samples with various leaf group formulas and establish a tobacco thermal physical database;

[0045] The thermal property measurement includes the specific heat capacity, thermal conductivity, leaf group formula type, wet basis moisture content and temperature of the cut tobacco sample;

[0046] 2) Based on the tobacco cut thermophysical database constructed in the above steps, a thermophysical property prediction model of tobacco cut is constructed using a data-driven method;

[0047] The thermal property prediction model includes a prediction model that takes leaf group formula type, wet basis moisture content and temperature as inputs and specific heat capacity and thermal conductivity as outputs;

[0048] 3) coupling the detection device and the prediction model of the thermal properties of the tobacco obtained in the above steps to detect the thermal properties of the tobacco;

[0049] The detection equipment includes an infrared moisture meter and / or an infrared temperature meter.

[0050] The present invention firstly measures the thermal properties of tobacco samples of various leaf group formulas to establish a tobacco thermal physical database;

[0051] The thermal property measurements include the specific heat capacity, thermal conductivity, leaf group formulation type, wet basis moisture content and temperature of the cut tobacco sample.

[0052] In the present invention, the shredded tobacco sample is preferably a shredded tobacco sample that has been cut.

[0053] In the present invention, the width of the tobacco shreds in the cut tobacco sample is preferably about 1 mm, can be 0.9 to 1.1 mm, or can be 0.95 to 1.05 mm.

[0054] In the present invention, the step 1) preferably comprises the following steps:

[0055] (1) Balancing the tobacco sample to a uniform moisture content, and then measuring the moisture content of the tobacco sample;

[0056] (2) setting multiple temperatures and multiple wet-based moisture contents to conduct full factorial experimental design, using a constant temperature and humidity chamber to subject the tobacco samples to a constant temperature and humidity process, and obtaining tobacco samples with the above-set temperature and wet-based moisture content;

[0057] (3) Using a TPS thermal constant analyzer, the thermal physical properties of the tobacco samples with different combinations of wet-basis moisture content and temperature parameters obtained in the above steps were measured to construct a tobacco thermal physical database covering specific heat capacity, thermal conductivity, leaf group formula category, wet-basis moisture content and temperature.

[0058] The present invention firstly balances the tobacco cut sample to a uniform moisture content, then measures the moisture content of the tobacco cut sample, and then sets multiple temperatures and multiple wet-based moisture contents to perform a full factorial experimental design. A constant temperature and humidity chamber is used to subject the tobacco cut sample to a constant temperature and humidity process to obtain tobacco cut samples with the above-set temperature and wet-based moisture content.

[0059] In the present invention, this step is based on the moisture content of the tobacco cut sample. According to actual needs, different wet-base moisture contents and temperatures are set, and then the tobacco cut sample is kept at a constant temperature and humidity to the specified wet-base moisture content and temperature.

[0060] In the present invention, the temperature range of the plurality of temperatures preferably includes 25 to 75°C, more preferably 35 to 65°C, and more preferably 45 to 55°C.

[0061] In the present invention, the ranges of the plurality of wet-basis moisture contents preferably include 12% to 19%, more preferably 13% to 18%, more preferably 14% to 17%, and more preferably 15% to 16%.

[0062] Finally, the present invention uses a TPS thermal constant analyzer to measure the thermal properties of tobacco samples with different wet basis moisture content and temperature parameter combinations obtained in the above steps, and constructs a tobacco thermal physical database covering specific heat capacity, thermal conductivity, leaf group formula category, wet basis moisture content and temperature.

[0063] In the present invention, a specific experimental scheme is designed based on the moisture content of the tobacco sample before step (1), and a uniform sampling experiment is performed for the temperature and moisture range described in (3). For example, the temperature between 25 and 75°C is 5°C as a gradient, and 11 levels of 25, 30, 35...75 are obtained, and the moisture content between 12% and 19% is 1% as a gradient, and 8 levels of 12, 13...19% are obtained. Then, a total of 11*8 samplings are required to make the measured sample tobacco temperature and moisture cover 25-75°C and 12%-19%. Therefore, the role of the constant temperature and humidity chamber is to obtain samples with specified moisture and temperature.

[0064] Therefore, in the next step, the thermal properties data of the sample at the specified temperature and moisture content are measured.

[0065] In the present invention, the thermophysical property prediction model is preferably a thermophysical property prediction model that uses the tobacco thermal physics database as the data set required for model construction, uses leaf group formula category, wet basis moisture content, and temperature as model inputs, and uses specific heat capacity and thermal conductivity as model outputs.

[0066] In the present invention, the data set is preferably divided into a training set and a test set for machine learning.

[0067] In the present invention, the proportion of the test set in the data set is preferably not less than 20%, more preferably not less than 19%, and more preferably not less than 18%. Specifically, the ratio of the training set and the test set for machine learning is preferably (7-8): (3-2), more preferably (7.2-7.8): (2.8-2.2), and more preferably (7.4-7.6): (2.6-2.4).

[0068] The present invention then constructs a prediction model for the thermal properties of tobacco cuts based on the tobacco cuts thermal physical database constructed in the above steps and using a data-driven method;

[0069] The thermophysical property prediction model includes a prediction model that takes leaf group formula type, wet basis moisture content and temperature as inputs and specific heat capacity and thermal conductivity as outputs.

[0070] In the present invention, the data driving preferably further comprises the following steps:

[0071] The data set is imported into the computing software, and different machine learning methods are compared to evaluate the training effect and / or accuracy, and the optimal model is obtained as the thermophysical property prediction model for subsequent use.

[0072] In the present invention, the evaluation index preferably includes one or more of root mean square error, mean absolute error and determination coefficient, and more preferably includes root mean square error, mean absolute error or determination coefficient.

[0073] In the present invention, the computing software preferably includes Matlab.

[0074] In the present invention, the data driving preferably further comprises the following steps:

[0075] When multiple machine learning methods cannot achieve the desired detection results, the sample size of tobacco samples is increased, the tobacco thermal physics database is expanded, and the thermal property prediction model is rebuilt.

[0076] Finally, the present invention couples the detection device with the thermal property prediction model of the tobacco obtained in the above steps to realize the detection of the thermal property of the tobacco;

[0077] The detection equipment includes an infrared moisture meter and / or an infrared temperature meter.

[0078] In the present invention, the detection method is preferably a real-time online detection method.

[0079] In the present invention, the detection device is preferably an online detection device.

[0080] In the present invention, the detection method preferably further comprises verification and correction steps.

[0081] In the present invention, the verification and correction step is preferably performed by sampling the tobacco shreds online, comparing the predicted thermophysical property parameters given by the thermophysical property prediction model of the tobacco shreds with the actual measured values ​​of the thermophysical properties of the samples, and verifying and correcting the model.

[0082] See also Figure 1 , Figure 1 The present invention provides a simplified schematic diagram of the process of detecting the thermal properties of cut tobacco.

[0083] The present invention is to complete and refine the overall technical solution, and better improve the stability, repeatability and accuracy of the online real-time detection method of the thermal properties of tobacco. The above-mentioned online detection method of the thermal properties of tobacco based on infrared spectroscopy can specifically include the following contents:

[0084] Description of the method for measuring the specific heat capacity of cut tobacco based on infrared spectroscopy:

[0085] (1) The thermal properties of cut tobacco with different leaf composition were measured in the laboratory.

[0086] The first step is to take the cut tobacco and equilibrate it in a sealed bag to a uniform moisture content, and then measure the moisture content of the sample;

[0087] The second step is to conduct a full factorial experimental design. According to actual needs, the moisture content of the tobacco wet basis is set to 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19% and other levels, and the temperature is set to 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75℃ and other levels;

[0088] The third step is to set the constant temperature and humidity chamber parameters according to different wet-base moisture content and temperature, and place the sample sealed bag in the constant temperature and humidity chamber to balance to the specified wet-base moisture content and temperature and keep it for use;

[0089] The fourth step is to use the TPS thermal constant analyzer to measure the thermal properties of tobacco under different combinations of wet-basis moisture content and temperature parameters, and to build a tobacco thermal physics database covering specific heat capacity, thermal conductivity, leaf group formula category, wet-basis moisture content, and temperature.

[0090] (2) Establishing a prediction model for the thermal properties of cut tobacco based on a data-driven approach

[0091] Based on the above-mentioned tobacco thermophysical database, a prediction model for the specific heat capacity and thermal conductivity of tobacco is constructed using a data-driven approach. The model takes leaf group formula category, wet basis moisture content, and temperature as inputs, and specific heat capacity and thermal conductivity as outputs. The specific implementation steps are as follows:

[0092] The first step is to use the tobacco thermophysical database as the data set required for model construction, with leaf group formula category, wet basis moisture content, and temperature as model inputs, and specific heat capacity and thermal conductivity as model outputs. The data set is divided into a training set and a test set for machine learning;

[0093] In the second step, the data set is imported into Matlab2023a software, and different machine learning methods are compared to select the model with the best training effect and the highest test set accuracy (evaluation indicators: root mean square error, mean absolute error, and determination coefficient) as the thermophysical property prediction model to be used in the next step. If each machine learning method fails to achieve the expected effect, the sample size will be increased to enrich the tobacco thermophysical database and re-construct the thermophysical property prediction model.

[0094] (3) Online detection and correction of thermal property prediction model

[0095] The detection equipment on the production line (infrared moisture meter, infrared temperature meter) and the prediction model of thermal properties of tobacco are coupled to realize online detection of thermal properties of tobacco. Online sampling is carried out to compare the predicted thermal property parameters with the actual measured values ​​of the thermal properties of the samples, and the model is verified and corrected.

[0096] The present invention provides application of the detection method described in any one of the above technical solutions in the field of online real-time detection of thermal properties of tobacco.

[0097] The above content of the present invention provides an online detection method and application of thermal properties of cut tobacco based on infrared spectroscopy. The thermal properties detection method of cut tobacco based on infrared spectroscopy technology provided by the present invention can realize fast and accurate online real-time detection of thermal properties of cut tobacco by coupling infrared thermometers, infrared moisture meters and other equipment and thermal properties prediction models, greatly improving the quality control method of tobacco products and providing a new path for the thermal processing technology control of the cut tobacco process.

[0098] The present invention relies on the tobacco thermal property database established in the laboratory to realize real-time and efficient detection of thermal properties. It is only necessary to perform tobacco sampling and thermal property measurement tests in the database establishment stage, and then only by using online temperature, moisture content and other detection equipment, the online detection of tobacco thermal properties can be realized, the detection time is short, and the detection means fully meet the timeliness required for production. Compared with the existing tobacco thermal property detection technology mainly through non-steady-state method, such as transient plane heat source (TPS) method, which is the most widely used method at present, the TPS method is to supply instantaneous heat to the sample to be tested through a disc-shaped heat source, and at the same time, measure the response of the sample to thermal interference to obtain the sample temperature rise data. Then, the material thermal property parameters are determined based on the non-steady-state heat conduction differential equation. However, the disadvantages of the TPS method for measuring the thermal properties of tobacco are: sampling is required in advance, and the temperature of the sample to be tested is maintained constant; the detection takes a long time, which takes about tens of seconds. The present invention does not need to prepare samples or contact samples, and the detection time is short, which can realize efficient real-time online detection of tobacco thermal properties.

[0099] In order to further illustrate the present invention, a method for detecting thermal properties of tobacco provided by the present invention and its application are described in detail below in combination with the embodiments. However, it should be understood that these embodiments are implemented on the premise of the technical scheme of the present invention, and detailed implementation methods and specific operating procedures are given only to further illustrate the features and advantages of the present invention, rather than to limit the claims of the present invention. The protection scope of the present invention is not limited to the embodiments described below.

[0100] Example

[0101] Taking the Huanghelou (hard treasure) product as an example, after sampling the dried tobacco online, the sample was balanced in a constant temperature and humidity chamber for 12 hours to the specified temperature and moisture content on a wet basis according to the set conditions. Then, the thermal conductivity and specific heat capacity of the tobacco at the specified temperature and moisture content on a wet basis were measured using a TPS thermal constant analyzer. The measured thermal properties of the tobacco are shown in Table 1.

[0102] Table 1 Test data of thermal conductivity and specific heat capacity of tobacco (partial)

[0103]

[0104] Based on the thermophysical property data of tobacco, a thermophysical property prediction model for the leaf group formula of Huanghelou (Hard Treasure) product was established. The above sample thermophysical property data were imported into MATLAB, and the Gaussian process regression model was used to train the sample data. The kernel function of the Gaussian process regression model was Matern5 / 2.

[0105] y1(λ)=f(x1,x2) (1)

[0106] y2(c)=f(x1,x2) (2)

[0107] Formula (1) and Formula (2) are the prediction formulas of thermal conductivity (λ) and specific heat capacity (c) obtained after training the Gaussian process regression model, respectively, where the moisture content of tobacco (x1) and tobacco temperature (x2) are input variables, and y1(λ) and y2(c) are the predicted values ​​of thermal conductivity and specific heat capacity.

[0108] Through real-time monitoring of production process data by infrared moisture meter and infrared thermometer, the real-time data of moisture content and temperature of tobacco after drying are obtained. According to formula (1) and formula (2), the thermal physical property data of tobacco on the current production line are calculated in real time. The comparison between the thermal physical property results of tobacco tested by transient plane heat source method (TPS) is shown in Table 2.

[0109] Table 2 Comparison of thermal properties of tobacco cut by TPS method and examples

[0110]

[0111] It can be seen that the thermal property data of tobacco cut measured by the embodiment and the TPS method have a small error, showing the high detection accuracy of the method. At the same time, the method has the advantage of real-time calculation of data, realizes on-site real-time detection, and improves efficiency.

[0112] The above is a detailed introduction to an online detection method and application of thermal properties of cut tobacco based on infrared spectroscopy provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method and its core ideas of the present invention, including the best mode, and also enables any technician in the field to practice the present invention, including the manufacture and use of any device or system, and the implementation of any combined method. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention. The scope of patent protection of the present invention is defined by the claims and may include other embodiments that can be thought of by those skilled in the art. If these other embodiments have structural elements that are not different from the text of the claims, or if they include equivalent structural elements that are not substantially different from the text of the claims, then these other embodiments should also be included in the scope of the claims.

Claims

1. A method for detecting thermal properties of cut tobacco, characterized in that: The following steps are involved: 1) Measure the thermal properties of tobacco samples with various leaf group formulas and establish a tobacco thermal physical database; The thermal property measurement includes the specific heat capacity, thermal conductivity, leaf group formula type, wet basis moisture content and temperature of the cut tobacco sample; 2) Based on the tobacco cut thermophysical database constructed in the above steps, a thermophysical property prediction model of tobacco cut is constructed using a data-driven method; The thermal property prediction model includes a prediction model that takes leaf group formula type, wet basis moisture content and temperature as inputs and specific heat capacity and thermal conductivity as outputs; 3) coupling the detection device and the prediction model of the thermal properties of the tobacco obtained in the above steps to detect the thermal properties of the tobacco; The detection equipment includes an infrared moisture meter and / or an infrared temperature meter.

2. The detection method according to claim 1, characterized in that The shredded tobacco sample is specifically a shredded tobacco sample after being cut; The width of the tobacco leaves in the cut tobacco sample is 0.9 to 1.1 mm; The step 1) specifically comprises the following steps: (1) Balancing the tobacco sample to a uniform moisture content, and then measuring the moisture content of the tobacco sample; (2) setting multiple temperatures and multiple wet-based moisture contents to conduct full factorial experimental design, using a constant temperature and humidity chamber to subject the tobacco samples to a constant temperature and humidity process, and obtaining tobacco samples with the above-set temperature and wet-based moisture content; (3) Using a TPS thermal constant analyzer, the thermal physical properties of the tobacco samples with different combinations of wet-basis moisture content and temperature parameters obtained in the above steps were measured to construct a tobacco thermal physical database covering specific heat capacity, thermal conductivity, leaf group formula category, wet-basis moisture content and temperature.

3. The detection method according to claim 2, characterized in that: The temperature range of the multiple temperatures includes 25 to 75°C; The multiple wet-basis moisture contents range from 12% to 19%.

4. The detection method according to claim 1, characterized in that: The thermophysical property prediction model specifically uses the cut tobacco thermophysical database as the data set required for model construction, uses leaf group formula category, wet basis moisture content, and temperature as model inputs, and uses specific heat capacity and thermal conductivity as model outputs.

5. The detection method according to claim 4, characterized in that: The data set is divided into a training set and a test set for machine learning; In the data set, the proportion of the test set is not less than 20%.

6. The detection method according to claim 4, characterized in that: The data drive also includes the following steps: The data set is imported into the computing software, and different machine learning methods are compared to evaluate the training effect and / or accuracy, and the optimal model is obtained as the thermophysical property prediction model for subsequent use.

7. The detection method according to claim 6, characterized in that: The evaluation index includes one or more of root mean square error, mean absolute error and determination coefficient; The computing software includes Matlab.

8. The detection method according to claim 6, characterized in that: The data drive also includes the following steps: When multiple machine learning methods cannot achieve the desired detection results, the sample size of tobacco samples is increased, the tobacco thermophysical database is expanded, and the thermophysical property prediction model is rebuilt; The detection method is specifically a real-time online detection method.

9. The detection method according to claim 1, characterized in that: The detection device is an online detection device; The detection method also includes verification and correction steps; The verification and correction step specifically includes taking online samples of the tobacco shreds, comparing the predicted thermophysical property parameters given by the thermophysical property prediction model of the tobacco shreds with the actual measured values ​​of the thermophysical properties of the samples, and verifying and correcting the model.

10. Application of the detection method according to any one of claims 1 to 9 in the field of online real-time detection of thermal properties of cut tobacco.