Method, device and computer readable medium for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod

Abstract

The application provides a method, device and medium for establishing a prediction model of a cigarette filter rod for aerosol particle interception rate, which comprises the following steps: firstly, determining a reference interception coefficient k based on the detection value of a standard filter rod, and obtaining a relationship formula between the interception coefficient k of the standard filter rod and a flow rate v; then, obtaining the relationship between the unknown filter rod and the standard filter rod in terms of the interception coefficient according to the relationship between the filter rod pressure drop and the length and the feature that the interception rate of the unit length pressure drop is the same under the same flow rate; and further obtaining the prediction model of the interception coefficient of the unknown filter rod. The scheme has the advantage that the prediction of the aerosol particle interception rate of the filter rod can be relatively accurately realized.

Description

Technical Field

[0001] This invention relates to the field of cigarette filter performance analysis technology, specifically to a method, apparatus, and medium for establishing a predictive model of the retention rate of aerosol particles by a cigarette filter rod. Background Technology

[0002] Cigarette smoke is an aerosol. Based on its transport properties, aerosol components can be divided into permanent gases, vapors of volatile and semi-volatile components, and aerosol particles.

[0003] Of these, permanent gases are almost not retained by the filter; aerosol particles are retained on the filter during transport through mechanical filtration, while the vapors of volatile and semi-volatile components are retained by the filter through condensation and adsorption mechanisms. Since tar and nicotine mainly exist in particulate form, the retention of particulate phases in flue gas aerosols has been extensively studied.

[0004] The main mechanisms by which filter tips retain aerosol particles are diffusion deposition, direct interception, and inertial impaction. Diffusion deposition occurs when small particles move irregularly and are adsorbed and retained by the filter material. Direct interception is the most common filter mechanism, occurring when the flue gas flow is perpendicular to the cylindrical cross-section of the filter material. During the flow, particles collide with the flue gas, condense, and are then adsorbed and retained by the filter. Inertial impaction occurs when larger particles move due to their inertia and eventually come to rest due to viscosity, thus being filtered out by the filter material. Retention efficiency depends not only on the properties of the filter tip but also on the size of the aerosol particles and the aerosol transport rate.

[0005] Although there has been much research on the filtration of aerosol particles based on fluid theory, there are still many difficulties in accurately predicting filtration efficiency, including: the interaction between adjacent fibers, the difference between the shape and orientation of fibers and classical models, and the polydispersity of flue gas aerosol particles.

[0006] Therefore, although empirical or semi-empirical retention models have always been of practical value in flue gas aerosol filtration research, since flue gas aerosol filtration involves both particulate filtration and vapor compound retention, empirical or semi-empirical retention models can only be used for overall aerosol retention research and cannot distinguish the differences in retention between particulate phase and vapor compounds.

[0007] To make the above distinction, predicting and analyzing the aerosol particle retention rate is a crucial breakthrough. Currently, there are methods for predicting and analyzing the filter rod length and aerosol retention rate, such as the scheme disclosed in the patent application number CN202211090250.X, entitled "A Method for Predicting the Retention Efficiency of Cigarette Smoke by a Filter." However, the parameters of the cigarette filter rod also include pressure drop, circumference, flow velocity, and other parameters. It is necessary to study the influence of these parameters on the aerosol particle retention rate and establish relevant prediction models in order to conduct a more in-depth study on the relationship between the filter rod and the aerosol retention rate.

[0008] In order to solve the above problems, people have been seeking an ideal technological solution. Summary of the Invention

[0009] The purpose of this invention is to address the shortcomings of existing technologies by providing a method, system, device, and medium for establishing a prediction model of the aerosol particle retention rate of cigarette filter rods, which can relatively accurately predict the retention rate of aerosol particles by filter rods and form a prediction model.

[0010] To achieve the above objectives, the technical solution adopted by this invention is: a method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod, implemented through the following steps:

[0011] Step 1) Set a standard filter rod, which has a certain rejection coefficient k for a preset aerosol particles under a set suction capacity;

[0012] Step 2) Based on the dataset showing the effect of the standard filter rod's suction capacity on the aerosol particle retention rate, a polynomial fitting method is used to obtain the formula relating the retention coefficient k to the flow velocity v:

[0013] k=av 3 +bv 2 -cv+d (1)

[0014] Where a, b, c, and d are the coefficients of the corresponding relational formulas; specifically, the standard filter rod has a fiber bundle specification of 3.0Y / 35000, a pressure drop of 2903Pa, a length of 100mm, a circumference of 24.2mm, and a retention coefficient of k at a suction capacity of 10-70ml. The formula (1) is expressed as:

[0015] k = -1.36 * 10 -6 v 3 +2.54*10 -4 v 2 -1.60*10 -2 v+0.582

[0016] Step 3) Since the pressure drop of the filter rod is directly proportional to its length, let kp The natural logarithmic change of aerosol penetration per unit length pressure drop is given by:

[0017] Ln(1-E)=k p △P (2)

[0018] Where E represents the filter rod's rejection rate; 1-E represents the aerosol's penetration rate; and ΔP represents the pressure drop per unit length.

[0019] Step 4) At the same flow rate, the retention rate per unit length of pressure drop is the same, therefore:

[0020] k x / k =△P x / △P0 (3)

[0021] Where, k x ΔP represents the rejection coefficient of the unknown filter rod; k represents the rejection coefficient of the standard filter rod; ΔP x △P0 represents the suction resistance of the unknown filter rod; △P0 represents the suction resistance of the standard filter rod.

[0022] Step 5) Substitute formula (3) into formula (1) to obtain the prediction model formula for the unknown filter rod:

[0023] k x = (av) 3 +bv 2 -cv+d)*( △P x / △P0 ) (4)

[0024] Specifically: k x = (-1.36*10) -6 v 3 +2.54*10 -4 v 2 -1.60*10 -2 v+0.582)*( △P x / △P0 );

[0025] The flow rate v and suction resistance ΔP of the unknown filter rod were obtained respectively. x After determining the suction resistance ΔP0 of the standard filter rod, the required retention coefficient of the unknown filter rod can be obtained according to formula (4).

[0026] Based on the above, for unknown filter rods of different diameters, under the same suction capacity, the linear velocity is inconsistent with that of the standard filter rod. This can be corrected using the following method:

[0027] Since pressure drop is proportional to flow rate, we have the following formula:

[0028] △P = (Q / Q0)△P0(5)

[0029] For the same circular filter rod, the formula for converting to linear velocity becomes:

[0030] △P = (v / v0)△P0 (6)

[0031] ΔP0 is calculated using formula (6) for standard filter rods and unknown filter rods of different diameters at the same flow rate. Then, the retention coefficient k of the unknown medium or fine-branched filter rods is calculated using formula (4). x .

[0032] Based on the above, the rejection rate of the unknown filter rod is calculated using the following logarithmic formula:

[0033] = L (7)

[0034] Where L is the length of the filter rod; k x A is the rejection coefficient; A0 is the amount of particles entering the filter rod; A p The amount of particles that penetrate the L-length filter rod; A p / A0 indicates the filter rod penetration rate;

[0035] E=1-A p / A0(8)

[0036] Where E represents the retention rate.

[0037] Based on the above, the aerosol particles are dioctyl phthalate.

[0038] Based on the above, the ratio of the amount of dioctyl phthalate retained by the filter rod to the total amount of dioctyl phthalate in the aerosol is the retention rate of the filter rod.

[0039] A device for predicting the retention rate of aerosol particles by a cigarette filter rod, comprising:

[0040] Memory, used to store computer programs;

[0041] The processor, when executing the computer program, implements the steps of the method for establishing a prediction model for the retention rate of aerosol particles by the cigarette filter rod.

[0042] A computer-readable medium storing a computer program, which, when executed by a processor, implements the steps of the method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod.

[0043] This invention has outstanding substantive features and significant progress compared with the prior art. Specifically, this invention first sets a standard filter rod as a reference to obtain its true retention coefficient under set conditions. At the same time, due to the direct correspondence between suction capacity and flow rate, the relationship between retention coefficient and flow rate can be obtained from the data on the influence of suction capacity on retention system efficiency through polynomial fitting.

[0044] Then, by utilizing the direct proportional relationship between filter rod pressure drop and length, and the characteristic that the retention efficiency per unit pressure drop is the same at the same flow rate, the relationship between the unknown filter rod and the standard filter rod with respect to the retention coefficient and pressure drop value is found.

[0045] Finally, by substituting and transforming the formula, a predictive model based on flow rate and pressure drop as input factors is obtained to predict the retention coefficient of an unknown filter rod. In actual use, the retention coefficient of the filter rod under test can be obtained based on the flow rate, suction resistance and suction resistance value of the standard filter rod, and the retention efficiency can be easily calculated.

[0046] This method thoroughly analyzes the relationship between the retention efficiency of the filter rod and its length, pressure drop, circumference, and flow rate, and establishes a more comprehensive prediction model for input factors, providing a foundation for the study of cigarette filter rod retention efficiency. Attached Figure Description

[0047] Figure 1 This relates the aerosol rejection rate to the filter rod length in this invention.

[0048] Figure 2 This relates to the relationship between aerosol retention and flow rate in this invention.

[0049] Figure 3 This relates to the relationship between aerosol retention and pressure drop in this invention.

[0050] Figure 4 This means that the model predictions in this invention are consistent with the measured values. Detailed Implementation

[0051] The technical solution of the present invention will be further described in detail below through specific embodiments.

[0052] Example 1

[0053] A method for establishing a predictive model for the retention rate of aerosol particles by cigarette filter rods is implemented through the following steps:

[0054] Step 1) Set up a standard filter rod. The standard filter rod has a fiber bundle specification of 3.0Y / 35000, a pressure drop of 2903Pa, a length of 100mm, a circumference of 24.2mm, and a definite rejection coefficient k for a preset aerosol particles at a suction capacity of 10-70ml. In this embodiment, the aerosol particles are dioctyl phthalate.

[0055] Step 2) Based on the dataset showing the effect of the standard filter rod's suction capacity on the aerosol particle retention rate, such as... Figure 2 As shown, using polynomial fitting, the relationship between the interception coefficient k and the flow velocity v is obtained as follows:

[0056] k = -1.36 * 10 -6 v 3 +2.54*10 -4 v 2 -1.60*10 -2 v+0.582 (1)

[0057] Step 3) Since the pressure drop of the filter rod is directly proportional to its length, let k p The natural logarithmic change of aerosol penetration per unit length pressure drop is given by:

[0058] Ln(1-E)=k p △P (2)

[0059] Where E represents the filter rod's rejection rate, which is the ratio of the amount of dioctyl phthalate rejected by the filter rod to the total amount of dioctyl phthalate in the aerosol; 1-E represents the aerosol's penetration rate; ΔP represents the pressure drop per unit length; E=1-A p / A0.

[0060] Step 4) At the same flow rate, the retention rate per unit length of pressure drop is the same, therefore:

[0061] k x / k =△P x / △P0 (3)

[0062] Where, k x ΔP represents the rejection coefficient of the unknown filter rod; k represents the rejection coefficient of the standard filter rod; ΔP x △P0 represents the suction resistance of the unknown filter rod; △P0 represents the suction resistance of the standard filter rod.

[0063] Step 5) Substitute formula (3) into formula (1) to obtain the prediction model formula for the unknown filter rod:

[0064] k x = (-1.36*10) -6 v 3 +2.54*10 -4 v2 -1.60*10 -2 v+0.582)*( △P x / △P0 )(4);

[0065] The flow rate v and suction resistance ΔP of the unknown filter rod were obtained respectively. x After determining the suction resistance ΔP0 of the standard filter rod, the required retention coefficient of the unknown filter rod can be obtained according to formula (4).

[0066] Furthermore, assuming that the amount of aerosol particles retained by the filter rod is proportional to the initial amount of particles entering the filter nozzle, the process can be expressed by the following formula.

[0067] ,

[0068] = L, (7)

[0069] In the formula, L is the length of the filter rod, dL is the infinitesimal element of the filter rod; A is the amount of particles entering the infinitesimal element of the filter rod; and k is the retention coefficient.

[0070] Integrating both sides of the first formula yields equation (7).

[0071] Where A0 is the amount of particles entering the filter rod; A p Let A be the number of particles that penetrate the L-length filter rod. Then, from equation (3), we can know that the filter rod penetration rate A... p The natural logarithm of A0 should be negatively proportional to the length of the filter rod.

[0072] Finally, the retention rate is calculated according to formula (8).

[0073] E=1-A p / A0(8)

[0074] Where E represents the retention rate.

[0075] For unknown filter rods of different diameters, such as medium and slim cigarettes, the linear velocity is inconsistent with that of a standard filter rod under the same draw volume. This can be corrected using the following methods:

[0076] Since pressure drop is proportional to flow rate, we have the following formula:

[0077] △P = (Q / Q0)△P0(5)

[0078] For the same circular filter rod, the formula for converting to linear velocity becomes:

[0079] △P = (v / v0)△P0 (6)

[0080] ΔP0 is calculated using formula (6) for standard filter rods and unknown filter rods with different diameters at the same flow rate. Then, the retention coefficient k of the unknown medium or fine-branched filter rods is calculated using formula (4). x .

[0081] Tables 1 and 2 show the comparison between the measured values ​​of cigarette filter rods with different parameters obtained by traditional methods and the predicted values ​​obtained by this method.

[0082] Figure 4 This demonstrates the consistency between the model's predicted values ​​and the actual test values.

[0083] Figure 1 and Figure 3 This demonstrates the corresponding relationship between aerosol rejection rate and filter rod length and pressure drop, which serves as the fundamental basis for this method.

[0084] In this embodiment, taking a conventional filter rod A016 as an example, its filament specifications are 3.0Y / 34000, length 100mm, circumference 24.1mm, and its pressure drop measured on the comprehensive testing platform is 2700pa, with a pressure drop per unit length ΔP. x Given a pressure of 27 Pa, the calculated linear velocity v at a suction volume of 35 ml is 37.8 m / s. The pressure drop per unit length ΔP0 of the modeled standard filter rod is 29.03. (The remaining text appears to be incomplete and requires further context.) x Substituting △P0 into Formula 4, we can find its retention coefficient k. x The value of k is 0.249. Through the experimental method for calculating k in this invention, the actual measured value of k for A016 is 0.240, which shows good consistency.

[0085] Example 2

[0086] Taking the B001 medium-sized filter rod as an example, its filament specifications are 5.8Y / 22000, length 120mm, circumference 19.8mm. Its pressure drop, tested on the comprehensive testing platform, is 2900pa, with a pressure drop per unit length ΔP. x The linear velocity v at a suction volume of 35 ml is calculated to be 56.1 cm / s, which is 29 Pa. Using formula (9), the pressure drop per unit length ΔP0 of the modeling standard filter rod at a velocity of 56.1 m / s is calculated to be 56.1 m / s. The values ​​of v and ΔP are then used to calculate the pressure drop per unit length ΔP0. x Substitute △P0 into Formula 7 to find its retention coefficient k. x The value of k is 0.138. Using the experimental method for calculating k in this invention, the actual measured value of k for B001 is 0.133, which shows good consistency.

[0087] Example 3

[0088] Taking the F012 fine filter rod as an example, its filament specifications are 6.0Y / 17000, length 120mm, circumference 17mm. On the comprehensive testing platform, its pressure drop is 3800pa, and the pressure drop per unit length ΔP is... x Given a pressure of 38 Pa, the linear velocity v at a suction volume of 35 ml is calculated to be 76.1 cm / s. Using formula (9), the pressure drop per unit length ΔP0 of the modeling standard filter rod at a velocity of 76.1 m / s is calculated to be 58.35 m / s. Substituting v and ΔP... x Substitute △P0 into Formula 7 to find its retention coefficient k. x The value of k is 0.129. Using the experimental method for calculating k in this invention, the actual measured value of k for F012 is 0.121, which shows good consistency.

[0089] serial number filament specifications Pressure drop (Pa) Length (mm) Circumference (mm) A008 3.0Y / 34000 2800 100 24.1 A016 3.0Y / 34000 2700 100 24.1 A021 3.0Y / 35000 2900 100 24.1 A023 3.0Y / 34000 2800 100 24.1 A028 3.0Y / 35000 2900 100 24.1 B001 5.8Y / 22000 2900 120 19.8 C001 8.0Y / 15000 3200 120 16.9 F-001 4.8Y / 22000 3300 120 19.8 F-002 5.1Y / 22000 3300 120 19.8 F-003 5.5Y / 22000 3300 120 19.8 F-004 5.8Y / 22000 3300 120 19.8 F-005 5.8Y / 26000 3300 120 19.8 F-006 5.8Y / 26000 3400 120 19.8 F-007 5.8Y / 26000 3600 120 19.8 F-008 5.8Y / 26000 3800 120 19.8 F-009 5.8Y / 26000 4200 120 19.8 F-010 3.0Y / 32000 3500 108 24.1 F-011 3.0Y / 32000 2800 108 24.1 F-012 6.0Y / 17000 3800 120 17 F-013 7.5Y / 16000 3800 120 17 F-014 8.0Y / 15000 3800 120 17 F-015 8.0Y / 15000 3200 120 17 GS-01 3.0Y / 35000 3180 120 24.2 GS-02 3.0Y / 35000 2620 100 24.2 GS-03 8.0Y / 15000 3140 120 17 HN-1 3.0Y / 35000 2903 100 24.2

[0090] Table 1. Cigarette Filter Rod Sample Information

[0091] Sample number Pressure drop per unit length ΔP (Pa) Linear velocity (cm / s) <![CDATA[Pressure drop ΔP0 (Pa) of constant flow rate standard rod]]> <![CDATA[Pressure drop ratio (ΔP / ΔP0)]]> <![CDATA[K x Predicted value (cm) -1 )]]> <![CDATA[K x Measured value (cm) -1 )]]> relative deviation A008 28.0 37.8 29.03 0.96440 0.258 0.253 1.98% A016 27.0 37.8 29.03 0.92996 0.249 0.240 3.75% A021 29.0 37.8 29.03 0.99885 0.268 0.259 3.47% A023 28.0 37.8 29.03 0.96440 0.258 0.243 6.17% A028 29.0 37.8 29.03 0.99885 0.268 0.265 1.13% B001 24.2 56.1 43.01 0.56184 0.138 0.133 3.76% C001 26.7 77.0 59.04 0.45166 0.101 0.098 5.94% F-001 27.5 56.1 43.01 0.63934 0.157 0.149 5.37% F-002 27.5 56.1 43.01 0.63934 0.157 0.163 -3.68% F-003 27.5 56.1 43.01 0.63934 0.157 0.153 2.61% F-004 27.5 56.1 43.01 0.63934 0.157 0.151 3.97% F-005 27.5 56.1 43.01 0.63934 0.157 0.149 5.37% F-006 28.3 56.1 43.01 0.65871 0.152 0.161 -5.59% F-007 30.0 56.1 43.01 0.69746 0.171 0.165 3.64% F-008 31.7 56.1 43.01 0.73621 0.180 0.174 3.45% F-009 35.0 56.1 43.01 0.81370 0.199 0.189 5.29% F-010 32.4 37.8 29.03 1.11621 0.299 0.288 3.82% F-011 25.9 37.8 29.03 0.89297 0.239 0.237 0.84% F-012 31.7 76.1 58.35 0.54271 0.129 0.121 6.61% F-013 31.7 76.1 58.35 0.54271 0.129 0.123 4.88% F-014 31.7 76.1 58.35 0.54271 0.129 0.121 6.61% F-015 26.7 76.1 58.35 0.45702 0.108 0.114 -5.26% GS-01 26.5 37.5 28.79 0.92033 0.247 0.253 -2.37% GS-02 26.2 37.5 28.79 0.90991 0.245 0.246 -0.41% GS-03 26.2 76.1 58.35 0.44845 0.106 0.099 7.07%

[0092] Table 2. Predicted results of the filter rod's aerosol rejection coefficient (suction volume 35ml)

[0093] Example 4

[0094] A device for predicting the retention rate of aerosol particles by a cigarette filter rod, comprising:

[0095] Memory, used to store computer programs;

[0096] The processor, when executing the computer program, implements the steps of the method for establishing a prediction model for the retention rate of aerosol particles by the cigarette filter rod.

[0097] The device can perform its functions on any analytical device with a processor and memory.

[0098] Example 5

[0099] A computer-readable medium storing a computer program, which, when executed by a processor, implements the steps of the method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod.

[0100] This computer-readable medium can run on any device capable of executing the corresponding computer program.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in the present invention.

Claims

1. A method for establishing a predictive model for the retention rate of aerosol particles by a cigarette filter rod, characterized in that: This will be implemented through the following steps: Step 1) Set up a standard filter rod. The standard filter rod has a fiber bundle specification of 3.0Y / 35000, a pressure drop of 2903Pa, a length of 100mm, a circumference of 24.2mm, and a definite retention coefficient k at a suction capacity of 10-70ml. Step 2) Based on the dataset showing the effect of the standard filter rod's suction capacity on the aerosol particle retention rate, a polynomial fitting method is used to obtain the formula relating the retention coefficient k to the flow velocity v: k= -1.36*10 -6 v 3 +2.54*10 -4 v 2 -1.60*10 -2 v+0.582 (1) Step 3) Since the pressure drop of the filter rod is directly proportional to its length, let k p The natural logarithmic change of aerosol penetration per unit length pressure drop is given by: Ln(1-E)=k p △P (2) Where E represents the filter rod's rejection rate; 1-E represents the aerosol's penetration rate; and ΔP represents the pressure drop per unit length. Step 4) At the same flow rate, the rejection rate per unit length of pressure drop is the same, therefore: k x / k =△P x / △P0 (3) Where, k x ΔP represents the rejection coefficient of the unknown filter rod; k represents the rejection coefficient of the standard filter rod; ΔP x △P0 represents the suction resistance of the unknown filter rod; △P0 represents the suction resistance of the standard filter rod. Step 5) Substitute formula (3) into formula (1) to obtain the prediction model formula for the unknown filter rod: k x = (-1.36*10 -6 v 3 +2.54*10 -4 v 2 -1.60*10 -2 v+0.582)*( △P x / △P0 )(4); The flow rate v and suction resistance ΔP of the unknown filter rod were obtained respectively. x After determining the suction resistance ΔP0 of the standard filter rod, the required retention coefficient of the unknown filter rod can be obtained according to formula (4).

2. The method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod according to claim 1, characterized in that: For unknown filter rods of different diameters, under the same suction capacity, the linear velocity is inconsistent with that of a standard filter rod. This can be corrected using the following methods: Since pressure drop is proportional to flow rate, we have the following formula: △P = (Q / Q0)△P0(5) For the same circular filter rod, the formula for converting to linear velocity becomes: △P = (v / v0)△P0 (6) ΔP0 is calculated using formula (6) for standard filter rods and unknown filter rods of different diameters at the same flow rate. Then, the retention coefficient k of the unknown medium or fine-branched filter rods is calculated using formula (4). x .

3. The method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod according to claim 1 or 2, characterized in that: The rejection rate of the unknown filter rod is calculated using the following logarithmic formula: = L ， （7） Where L is the length of the filter rod; k x A is the rejection coefficient; A0 is the amount of particles entering the filter rod; A p The amount of particles that penetrate the L-length filter rod; A p / A0 indicates the filter rod penetration rate; E=1-A p / A0(8) Where E represents the retention rate.

4. The method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod according to claim 1, 2, or 3, characterized in that: The aerosol particles are dioctyl phthalate.

5. The method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod according to claim 1, 2, or 3, characterized in that: The retention rate of the filter rod is the ratio of the amount of dioctyl phthalate retained by the filter rod to the total amount of dioctyl phthalate in the aerosol.

6. A device for predicting the retention rate of aerosol particles by a cigarette filter rod, characterized in that: include: Memory, used to store computer programs; The processor, when executing the computer program, implements the steps of the method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod as described in any one of claims 1-5.

7. A computer-readable medium, characterized in that: The computer-readable medium stores a computer program that, when executed by a processor, implements the steps of the method for establishing a prediction model for the retention rate of aerosol particles by a cigarette filter rod as described in any one of claims 1-5.

Images