Electronic atomized liquid viscosity prediction method, model and model construction method

Abstract

The invention provides an electronic atomized liquid viscosity prediction method, a model and a model construction method, and the prediction method comprises the steps: providing a pre-constructed electronic atomized liquid viscosity prediction model, the electronic atomized liquid viscosity prediction model being configured with a function relationship between the viscosity of the electronic atomized liquid and the components and temperature of the electronic atomized liquid; acquiring viscosity detection data, component data and target temperature of the target sample; adjusting parameters of the electronic atomized liquid viscosity prediction model according to the viscosity detection data to obtain a target viscosity prediction model; and determining the predicted viscosity of the target sample according to the target viscosity prediction model, the component data and the target temperature. By establishing the function relationship between the viscosity of the electronic atomized liquid and the data of each component of the electronic atomized liquid and the temperature data, the viscosity change rule of the target sample can be determined from numerical values, reference is provided for subsequent adjustment of a new formula of the electronic atomized liquid, the experiment frequency and cost of viscosity detection are reduced, and the detection accuracy is improved. And a theoretical basis is provided for setting storage and transportation conditions.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of electronic atomization liquids, and in particular, to a method for predicting the viscosity of an electronic atomization liquid, a model, and a model construction method. Background Art

[0002] Electronic cigarettes are the mainstream products of new tobacco products. Because they are similar to traditional cigarettes in terms of physiological feelings and smoking methods, they are favored by the majority of consumers. The viscosity of the electronic atomization liquid is an important physical property that affects its fluidity and atomization effect. The atomization liquid with a lower viscosity has good fluidity, but it may cause oil leakage and popping sounds, affecting the user experience; while the atomization liquid with a higher viscosity can produce more smoke, but it may cause poor oil guiding in the atomizer.

[0003] During the production process of the electronic atomization liquid, the change in the content of each component will cause the change in the viscosity of the product. In addition, temperature has a great influence on viscosity. At different storage temperatures, the viscosity of the same electronic atomization liquid varies greatly; the higher the temperature, the lower the viscosity. This viscosity property of the electronic atomization liquid may cause large fluctuations in viscosity due to temperature changes during transportation or storage, resulting in problems such as dripping of the electronic atomization liquid.

[0004] However, the viscosity of the electronic atomization liquid is affected by various factors such as the components, content, and temperature of the electronic atomization liquid, which is relatively complex. It is difficult to directly calculate the value through the existing viscosity formula, making it lack a basis for regulating the viscosity of the electronic atomization liquid when the manufacturer transports or stores the electronic atomization liquid. Summary of the Invention

[0005] The present application aims to provide a method for predicting the viscosity of an electronic atomization liquid, a model, and a model construction method to solve the technical problem of lacking a basis for regulating the viscosity of the electronic atomization liquid.

[0006] In a first aspect, the embodiments of the present application provide a method for predicting the viscosity of an electronic atomization liquid, including:

[0007] Providing a pre-constructed electronic atomization liquid viscosity prediction model, where the electronic atomization liquid viscosity prediction model is configured with a functional relationship between the viscosity of the electronic atomization liquid, the components of the electronic atomization liquid, and the temperature;

[0008] Obtaining the viscosity detection data, component data, and target temperature of the target sample;

[0009] Adjusting the parameters of the electronic atomization liquid viscosity prediction model according to the viscosity detection data to obtain a target viscosity prediction model;

[0010] Determining the predicted viscosity of the target sample according to the target viscosity prediction model, the component data, and the target temperature.

[0011] Optionally, the electronic atomization liquid viscosity prediction model includes a first correction coefficient and a second correction coefficient. Adjusting the parameters of the electronic atomization liquid viscosity prediction model according to the viscosity detection data to obtain the target viscosity prediction model includes:

[0012] Determine one component data or the target temperature of the target sample as the target parameter;

[0013] Provide viscosity detection data of the target parameter under at least two different conditions;

[0014] Calculate the first correction coefficient and the second correction coefficient according to the viscosity detection data under the at least two different conditions.

[0015] Optionally, the electronic atomization liquid viscosity prediction model is configured with at least one of a temperature viscosity coefficient, a glycerol / propylene glycol viscosity coefficient, a moisture viscosity coefficient, and an ethanol viscosity coefficient. The component data includes at least one of glycerol component data, propylene glycol component data, moisture component data, and ethanol component data; the predicted viscosity includes a first predicted viscosity; determining the predicted viscosity of the target sample according to the target viscosity prediction model, the component data, and the target temperature includes:

[0016] Calculate at least one of the temperature viscosity coefficient, the glycerol / propylene glycol viscosity coefficient, the moisture viscosity coefficient, and the ethanol viscosity coefficient according to at least one of the glycerol component data, the propylene glycol component data, the moisture component data, the ethanol component data, and the electronic atomization liquid viscosity prediction model;

[0017] Calculate the first predicted viscosity according to at least one of the temperature viscosity coefficient, the glycerol / propylene glycol viscosity coefficient, the moisture viscosity coefficient, the ethanol viscosity coefficient, and the electronic atomization liquid viscosity prediction model.

[0018] Optionally, the electronic atomization liquid viscosity prediction model is further configured with at least one of a moisture correction value and an ethanol correction value. The predicted viscosity further includes a second predicted viscosity. After calculating the first predicted viscosity, it further includes:

[0019] Calculate the moisture correction value according to the moisture component data, the glycerol / propylene glycol viscosity coefficient, and the electronic atomization liquid viscosity prediction model;

[0020] Calculate the ethanol correction value according to the ethanol component data, the glycerol / propylene glycol viscosity coefficient, and the electronic atomization liquid viscosity prediction model;

[0021] Calculate the second predicted viscosity according to the first viscosity coefficient, the moisture correction value, and the ethanol correction value.

[0022] Optionally, the predicted viscosity further includes a third predicted viscosity. After calculating the second predicted viscosity, it further includes:

[0023] Calculating the third predicted viscosity according to the first correction coefficient, the second correction coefficient, and the second predicted viscosity.

[0024] In a second aspect, an embodiment of the present application further provides a method for constructing an electronic atomization liquid viscosity prediction model according to the first aspect, including:

[0025] Determining a plurality of relevant parameters related to the viscosity of the target sample electronic atomization liquid;

[0026] Obtaining the viscosity detection data of each of the relevant parameters under different conditions;

[0027] Performing data fitting according to the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity;

[0028] Constructing the electronic atomization liquid viscosity prediction model according to the plurality of relevant parameters and the influence coefficient of the viscosity.

[0029] Optionally, the relevant parameters include temperature. The performing data fitting according to the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity includes:

[0030] Constructing a temperature factor according to the temperature;

[0031] Performing data fitting according to the temperature factor and the viscosity detection data to obtain a temperature viscosity coefficient calculation formula.

[0032] Optionally, the relevant parameters include glycerol component data and propylene glycol component data. The performing data fitting according to the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity includes:

[0033] Determining the proportion of the glycerol component data in the sum of the glycerol component data and the propylene glycol component data;

[0034] Constructing a glycerol / propylene glycol factor according to the proportion;

[0035] Performing data fitting according to different glycerol / propylene glycol factors and the viscosity detection data to obtain a glycerol / propylene glycol viscosity coefficient calculation formula.

[0036] Optionally, the relevant parameters include moisture component data. The performing data fitting according to the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity includes:

[0037] Construct a moisture factor based on the moisture component data;

[0038] Perform data fitting based on the moisture factor and the viscosity detection data to obtain a calculation formula for the moisture viscosity coefficient.

[0039] Optionally, the relevant parameters include ethanol component data. Performing data fitting based on the viscosity detection data and the relevant parameters under different conditions, the influence coefficients of the relevant parameters on viscosity include:

[0040] Construct an ethanol factor based on the ethanol component data;

[0041] Perform data fitting based on the ethanol factor and the viscosity detection data to obtain a calculation formula for the ethanol viscosity coefficient.

[0042] Optionally, the relevant parameters include moisture component data, propylene glycol component data, and glycerol component data. Performing data fitting based on the viscosity detection data and the relevant parameters under different conditions, the influence coefficients of the relevant parameters on viscosity include:

[0043] Perform data fitting based on the glycerol / propylene glycol viscosity coefficient, the moisture component data, and the viscosity detection data to obtain a calculation formula for the moisture correction value.

[0044] Optionally, the relevant parameters include ethanol component data, propylene glycol component data, and glycerol component data. Performing data fitting based on the viscosity detection data and the relevant parameters under different conditions, the influence coefficients of the relevant parameters on viscosity include:

[0045] Perform data fitting based on the glycerol / propylene glycol viscosity coefficient, the ethanol component data, and the viscosity detection data to obtain a calculation formula for the ethanol correction value.

[0046] In a third aspect, an embodiment of the present application further provides an electronic atomization liquid viscosity prediction model obtained according to the construction method described in the second aspect.

[0047] By establishing the functional relationship between the viscosity of the electronic atomization liquid and the component data and temperature data of the electronic atomization liquid, an embodiment of the present application can first numerically determine the viscosity change law of the target sample. Secondly, after combining the component data, target temperature, and viscosity detection data of the target sample to be measured, the model can be fine-tuned according to the target sample to improve the prediction accuracy of the model and provide a reference for the subsequent adjustment of the new formula of the electronic atomization liquid. Finally, predicting the viscosity of the electronic atomization liquid through the model can reduce the experimental cost of viscosity detection and provide a theoretical basis for setting storage and transportation conditions. Description of the Drawings

[0048] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments of the present application. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0049] Figure 1 Schematic diagram of the construction method flow of the electronic atomization liquid viscosity prediction model provided by an embodiment of the present application;

[0050] Figure 2 Schematic diagram of the method flow for calculating the influence coefficient of temperature on viscosity provided by an embodiment of the present application;

[0051] Figure 3 Schematic diagram of the data fitting result of the temperature factor and viscosity provided by an embodiment of the present application;

[0052] Figure 4 Schematic diagram of the method flow for calculating the influence coefficients of glycerol and propylene glycol on viscosity provided by an embodiment of the present application;

[0053] Figure 5 Schematic diagram of the data fitting result of the VG ratio factor and viscosity provided by an embodiment of the present application;

[0054] Figure 6 Schematic diagram of the method flow for calculating the influence coefficient of moisture on viscosity provided by an embodiment of the present application;

[0055] Figure 7 Schematic diagram of the data fitting result of the moisture factor and viscosity provided by an embodiment of the present application;

[0056] Figure 8 Schematic diagram of the method flow for calculating the influence coefficient of ethanol on viscosity provided by an embodiment of the present application;

[0057] Figure 9 Schematic diagram of the data fitting result of the ethanol factor and viscosity provided by an embodiment of the present application;

[0058] Figure 10 Schematic diagram of the method flow for calculating the moisture correction value provided by an embodiment of the present application;

[0059] Figure 11 Schematic diagram of the data fitting result of the temperature factor, VG ratio factor, moisture factor, ethanol factor and ideal viscosity provided by an embodiment of the present application;

[0060] Figure 12 Schematic diagram of introducing the calculation deviation of moisture and viscosity provided by an embodiment of the present application;

[0061] Figure 13 Schematic diagram of the influence of moisture on calculation deviation in an aqueous system provided by an embodiment of the present application;

[0062] Figure 14 Schematic diagram of the influence of PG and VG on calculation deviation in an aqueous system provided by an embodiment of the present application;

[0063] Figure 15 Schematic flow chart of the method for calculating the ethanol correction value provided by an embodiment of the present application;

[0064] Figure 16 Schematic diagram of introducing the calculation deviation of ethanol and viscosity provided by an embodiment of the present application;

[0065] Figure 17 Schematic diagram of the influence of ethanol on calculation deviation in an ethanol-containing system provided by an embodiment of the present application;

[0066] Figure 18 Schematic diagram of the influence of PG and VG on calculation deviation in an ethanol-containing system provided by an embodiment of the present application;

[0067] Figure 19 Schematic flow chart of the method for calculating the correction coefficient of other components provided by an embodiment of the present application;

[0068] Figure 20 Schematic flow chart of the prediction method for the viscosity of e-liquid provided by an embodiment of the present application. Detailed implementation manners

[0069] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts fall within the protection scope of the present application.

[0070] It should be noted that if there is no conflict, the various features in the embodiments of the present application can be combined with each other, and all are within the protection scope of the present application. In addition, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flow chart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the flow chart. Furthermore, the terms "first", "second", "third", etc. used in the present application do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0071] First, the technical background of the viscosity in the e-liquid provided by the embodiments of the present application will be introduced.

[0072] Viscosity is an important physical property that affects the fluidity and atomization effect of e-liquid. E-liquid with low viscosity has good fluidity, but it may cause leakage problems due to this; e-liquid with high viscosity has good atomization effect, but it is prone to cause problems of poor oil conduction, affecting the operation of the electronic atomizer.

[0073] The main components of e-liquid usually include propylene glycol, glycerol, nicotine, and other flavor substances. Among them, propylene glycol and glycerol, as solvents for e-liquid, have a relatively large impact on viscosity. Other components such as water and ethanol also have a certain impact on viscosity. In addition to the components of e-liquid, temperature also has a great impact on viscosity.

[0074] According to the literature "Research on the Relationship between Viscosity and Temperature of Several Organic Liquids", the authors Shi Baoping et al. used an NDJ-97 type rotational viscometer to measure the viscosities of quinoline, castor oil, liquid paraffin, decahydronaphthalene and their mixtures at different temperatures respectively. The relationship between the viscosities and temperatures of these organic substances and their mixtures was discussed in detail, and the data points were fitted and regression analyzed using excel to obtain a good power function curve. At the same time, a regression significance test was carried out on the correlation coefficient r of the power function equation, and the results showed that its regression effect was highly significant, providing a reference for establishing the relationship between the viscosity and temperature of e-liquid in this model.

[0075] In the Chinese patent application CN111737874A "A Method, Device, Equipment and Readable Storage Medium for Predicting the Viscosity of Facial Mask Liquid", a method for predicting the viscosity of facial mask liquid is disclosed. This method uses computational fluid dynamics to simulate the flow field distribution inside the emulsifying pot during the production process to obtain a distribution model of the shear rate related to the homogenization speed; for the target type of thickener, through data fitting, the relationships between the thickener concentration, homogenization time, shear rate and the viscosity of the facial mask liquid are determined respectively, and then a viscosity prediction model for the target type of thickener is constructed; finally, the prediction of the viscosity of the facial mask liquid is realized jointly using the shear rate distribution model and the viscosity prediction model.

[0076] In the Chinese patent application CN118298965A, "A Method and System for Detecting the Viscosity of the Latex Matrix of Semi-finished Emulsion Explosives", a method and system for detecting the viscosity of the latex matrix of semi-finished emulsion explosives are disclosed. The method includes obtaining the viscosity data of the latex matrix of semi-finished emulsion explosives; preprocessing the viscosity data of the latex matrix using the Cook distance method to obtain a training data set; constructing a viscosity prediction model using a multi-layer perceptron neural network; training the viscosity prediction model based on the training data set, and obtaining a trained viscosity prediction model through the firefly algorithm; and analyzing the latex matrix of semi-finished emulsion explosives using the trained viscosity prediction model to obtain the viscosity data of the latex matrix. This solves the problem that it is difficult to accurately predict the viscosity of the latex matrix of semi-finished emulsion explosives.

[0077] For ease of understanding, the following introduces the construction method of the electronic atomization liquid viscosity prediction model provided by the embodiments of the present application. Please refer to Figure 1 , Figure 1 which is a schematic flow chart of the construction method of the electronic atomization liquid viscosity prediction model provided by an embodiment of the present application, and specifically includes:

[0078] S11. Determine multiple relevant parameters related to the viscosity of the target sample electronic atomization liquid.

[0079] In this step, the target sample is the electronic atomization liquid to be measured, and the viscosity of the target sample electronic atomization liquid is the viscosity value of the target sample. Specifically, the number of target samples can be one or multiple, which is specifically determined according to the detection purpose. For example, when the detection purpose is to determine the viscosity of a certain electronic atomization liquid within a specific temperature range, the target sample can be one; while when the detection purpose is to regulate the components of the electronic atomization liquid and determine their respective viscosities, the target samples can be multiple.

[0080] In this step, the relevant parameters are other physical parameters associated with the viscosity of the target sample, such as the component data of the target sample and the temperature, etc. Specifically in this embodiment, the component data mainly includes the component data of water, the component data of propylene glycol and glycerol, the component data of ethanol, etc., which have relatively large effects on the viscosity of the electronic atomization liquid. In addition to components such as water, propylene glycol, glycerol, and ethanol that have relatively large effects on viscosity, it can also include other components that have relatively small effects on viscosity, such as nicotine components, sweetener substance components, etc.

[0081] It should be noted that since other substance components account for a relatively small proportion in the e-liquid, their influence on viscosity is relatively small. Therefore, in this embodiment, the influence of other substance components on the viscosity of the e-liquid is summarized by a correction coefficient. In some other embodiments, those skilled in the art can determine the influence of other substance components on viscosity according to the treatment methods of water, propylene glycol, glycerol, and ethanol components in the embodiments of this application, so as to further improve the accuracy of the e-liquid viscosity prediction model. It can be understood that the more component data considered and statistically analyzed by the e-liquid viscosity prediction model, the more accurate the prediction effect of the model, the more complex the model, and the more difficult it is to construct. Those skilled in the art can select and configure an appropriate number of component data for the prediction model according to the actual situation, and no limitation is made thereto.

[0082] S12. Obtain the viscosity detection data of each relevant parameter under different conditions.

[0083] In this step, the viscosity detection data is the viscosity value of the target sample obtained by a viscosity detection instrument. Specifically, in this embodiment, the viscosity detection instrument is an Anton Paar ViscoQC100-L viscometer. After the instrument is calibrated and stable, select a suitable rotor, wait for the temperature to stabilize, then put the rotor into a container filled with the target sample e-liquid for automatic detection, and record the viscosity detection data of the target sample at the same time.

[0084] In this step, the relevant parameters under different conditions represent different detection experimental conditions. For example, when the relevant parameter is temperature, it represents the viscosity detection data of the target sample at different temperatures; when the relevant parameter is the water component data, it represents the viscosity detection data of the target sample at different water contents, and so on for different conditions of other relevant parameters.

[0085] S13. Perform data fitting according to the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameter on viscosity.

[0086] In this step, the influence coefficient is the influence relationship of the relevant parameter on the viscosity of the target sample. For example, when the temperature changes to {x1, x2... xn}, the viscosity of the target sample changes to {y1, y2... yn} accordingly. By performing data fitting, the functional relationship y = f(x) can be obtained, and f(x) is the influence coefficient of temperature on viscosity. Similarly, in this embodiment, in addition to the influence coefficient f(x) of temperature on viscosity, it also includes the influence coefficient f(a) of propylene glycol and glycerol on viscosity, the influence coefficient f(b) of water on viscosity, and the influence coefficient f(c) of ethanol on viscosity.

[0087] It should be noted that, in addition to the influence coefficients of each correlation coefficient directly affecting viscosity, there are also influence coefficients that indirectly affect viscosity through the direct interaction of different correlation coefficients. For example, the interactions between water and propylene glycol, glycerol, and between ethanol and propylene glycol, glycerol, will all affect the viscosity of the target sample. Thus, in this embodiment, the influence coefficients also include a moisture correction value f(d) that takes into account the interaction between water and propylene glycol, glycerol and affects viscosity, and an ethanol correction value f(e) that takes into account the interaction between ethanol and propylene glycol, glycerol and affects viscosity.

[0088] It should also be noted that, in addition to the influence coefficients of the above components on viscosity, the influence of other substance components in the e-liquid on viscosity is represented by a number of correction coefficients. The calculation of the specific correction coefficients needs to be determined according to the viscosity detection data of the target sample.

[0089] S14. Construct an e-liquid viscosity prediction model based on the influence coefficients of multiple correlation parameters and viscosity.

[0090] In this step, the e-liquid viscosity prediction model is a functional relationship for calculating the viscosity of the target sample based on the various correlation parameters of the target sample. Specifically, the e-liquid viscosity model is configured with a temperature-viscosity coefficient, a glycerol / propylene glycol viscosity coefficient, a moisture viscosity coefficient, an ethanol viscosity coefficient, a moisture correction value, and an ethanol correction value. In some embodiments, it is also configured with a first correction coefficient and a second correction coefficient determined after correction according to the viscosity detection data of the target sample, thereby obtaining a target viscosity prediction model (i.e., the corrected e-liquid viscosity model). Specifically in the embodiment, the functional relationship provided by the e-liquid viscosity prediction model is a multivariate function including parameters such as the temperature-viscosity coefficient, the glycerol / propylene glycol viscosity coefficient, the moisture viscosity coefficient, the ethanol viscosity coefficient, the moisture correction value, and the ethanol correction value. The specific functional relationship will be described in detail later and will not be elaborated here.

[0091] The calculation of the viscosity coefficients of each relevant parameter will be described separately below.

[0092] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of a method for calculating the influence coefficient of temperature on viscosity provided by an embodiment of the present application, specifically including:

[0093] S21. Construct a temperature factor based on temperature.

[0094] S22. Perform data fitting based on the temperature factor and the viscosity detection data to obtain a temperature-viscosity coefficient calculation formula.

[0095] In step S21, during data fitting, in order to simplify the calculated formula obtained by fitting into a linear function of one variable, a temperature factor x is introduced to replace the temperature, so that the calculated formula of the temperature viscosity coefficient is simplified into the form of f(x) = kx + m, where the temperature factor x = 10 58 / (t + 273.15) 23 , both k and m are constants.

[0096] In step S22, first, viscosity detection data of the target sample at different temperatures needs to be obtained. Specifically in this embodiment, the target sample is an e-liquid whose solvent only includes propylene glycol and glycerol, and the ratio of propylene glycol to glycerol is 1:1. Specifically in this embodiment, the solvent of the target sample only includes propylene glycol and glycerol, and the ratio of the two is 1:1. The data fitting result shows that k is 16.052, m is 9.6759, and the regression coefficient R 2 is 0.9999. The calculated formula of the temperature viscosity coefficient obtained by data fitting is: f(x) = 16.052x + 9.6759. It should be noted that under different target sample systems, the calculated k values and m values are different, that is, the calculated formula of f(x) is also different. Specifically, those skilled in the art can determine the corresponding k values and m values according to the actual target sample system.

[0097] Furthermore, the result of data fitting is verified by comparing it with the actual viscosity detection data. The specific result of data fitting is as shown in Table 1 below and Figure 3 shown.

[0098] Table 1 Data fitting results of temperature viscosity coefficient

[0099]

[0100] According to the above table, it can be determined that under the target sample system, when predicting the viscosity of the target sample at temperatures from 15 °C to 70 °C through the model, the calculated results obtained by prediction are 94.66% to 101.75% of the actual detection results, with good accuracy. Therefore, the prediction model provided in the embodiment of this application can be well used to predict the viscosity of the target sample at different temperatures.

[0101] Please refer to Figure 4 , Figure 4 which is a schematic flowchart of the method for calculating the influence coefficient of glycerol and propylene glycol on viscosity provided by an embodiment of this application, specifically including:

[0102] S41. Determine the proportion of the glycerol component data in the sum of the glycerol component data and the propylene glycol component data.

[0103] S42. Construct a glycerol / propylene glycol factor according to the proportion.

[0104] S43. Perform data fitting based on the glycerol / propylene glycol factor and viscosity detection data to obtain the calculation formula for the glycerol / propylene glycol viscosity coefficient.

[0105] In step S41, the solvent of the e-liquid usually only includes glycerol (i.e., vegetable glycerin, abbreviated as VG) and propylene glycol (abbreviated as PG). According to different product requirements, the ratio of glycerol to propylene glycol in the solvent may be adjusted accordingly. Therefore, the influence of glycerol and propylene glycol on the viscosity of the target sample mainly depends on the ratio between the two. The ratio of the glycerol component to the solvent (i.e., the sum of the glycerol and propylene glycol components) can be denoted as VG ratio = VG / (VG + PG).

[0106] In step S42, during data fitting, in order to further simplify the obtained calculation formula, the VG ratio factor a is introduced to replace the VG ratio, where the VG ratio factor a = VG / [10*(VG + PG)].

[0107] In step S43, after data fitting, the polynomial fitting function for the glycerol / propylene glycol viscosity coefficient is f(a) = 1.1345a 3 - 4.2982a 2 + 23.519a + 43.55, where f(a) is the glycerol / propylene glycol viscosity coefficient, a is the VG ratio factor, and the regression coefficient R 2 is 0.9982, and the fitting effect is good.

[0108] Further, the result of data fitting is verified by comparing it with the actual viscosity detection data. The specific result of data fitting is shown in Table 2 below and Figure 5 as follows.

[0109] Table 2 Data fitting results of glycerol / propylene glycol viscosity coefficient

[0110]

[0111] According to the above table, it can be determined that in the system of the target sample, when predicting the viscosity of the target sample with a VG ratio ranging from 0% to 100% through the model, the calculated results obtained are 95.77% to 109.73% of the actual detection results, with good accuracy. Therefore, the prediction model provided in the embodiments of the present application can be well used to predict the viscosity of the target sample at different VG ratios.

[0112] Please refer to Figure 6 , Figure 6 which is a schematic flowchart of the method for calculating the influence coefficient of moisture on viscosity provided by an embodiment of the present application, specifically including:

[0113] S61. Construct a moisture factor based on the moisture component data.

[0114] S62. Perform data fitting based on the moisture factor and the viscosity detection data to obtain a calculation formula for the moisture viscosity coefficient.

[0115] In step S61, during data fitting, in order to further simplify the calculation formula obtained by fitting, a moisture factor b is introduced 1.7 to replace the moisture component data b, that is, the moisture factor is b 1.7 .

[0116] In step S62, the e-liquid contains relatively little moisture, but during the storage of the e-liquid, water vapor in the air may enter the e-liquid, resulting in an increase in the moisture content of the e-liquid, and moisture has a great influence on the viscosity of the e-liquid. Therefore, it is still necessary to detect the viscosity detection data corresponding to different moisture components for data fitting to determine the moisture viscosity coefficient. After data fitting, the moisture influence coefficient is f(b) = -35.945b 1.7 +202.7, where f(b) is the moisture viscosity coefficient, b is the moisture component data, and b 1.7 is the moisture factor, and the regression coefficient R 2 is 0.9968, and the fitting effect is good.

[0117] Furthermore, the result of data fitting is verified by comparing it with the actual viscosity detection data. The specific result of data fitting is as shown in Table 3 below and Figure 7 .

[0118] Table 3 Data fitting results of moisture viscosity coefficient

[0119]

[0120] It can be determined from the above table that under the system of the target sample, the viscosity of the target sample with a moisture content ranging from 0% to 10% is predicted by the model, and the calculated result obtained by the prediction is 96.00% to 102.59% of the actual detection result, with good accuracy. Therefore, the prediction model provided by the embodiments of the present application can be well used to predict the viscosity of the target sample at different moisture contents.

[0121] Please refer to Figure 8 , Figure 8 , which is a schematic flowchart of a method for calculating the influence coefficient of ethanol on viscosity provided by an embodiment of the present application, specifically including:

[0122] S81. Construct an ethanol factor based on the ethanol component data.

[0123] S82. Perform data fitting based on the ethanol component data and viscosity detection data to obtain the calculation formula for the ethanol viscosity coefficient.

[0124] In step S81, during data fitting, in order to further simplify the calculation formula obtained by fitting, an ethanol factor c is introduced 2 / 3 to replace the ethanol component data c, that is, the ethanol factor is c 2 / 3 .

[0125] In step S82, according to the formula of some e-liquids, a certain amount of ethanol may be added to the e-liquid to improve the taste of the e-liquid, and the ethanol content also has a great influence on the concentration of the e-liquid. Therefore, it is necessary to detect the viscosity detection data corresponding to different ethanol components for data fitting to determine the ethanol viscosity coefficient. After data fitting, the water influence coefficient is f(c)= -26.236c 2 / 3 +202.7, where f(c) is the ethanol viscosity coefficient, c 2 / 3 is the ethanol factor, c is the ethanol component data, and the regression coefficient R 2 is 0.9963, and the fitting effect is good.

[0126] Furthermore, the result of data fitting is verified by comparing it with the actual viscosity detection data. The specific result of data fitting is as shown in Table 4 below and Figure 9 shown.

[0127] Table 4 Data fitting results of ethanol viscosity coefficient

[0128]

[0129] It can be determined from the above table that under the system of the target sample, when predicting the viscosity of the target sample with an ethanol proportion ranging from 0% to 10% through the model, the calculated results obtained by prediction are 96.63% to 102.34% of the actual detection results, with good accuracy. Therefore, the prediction model provided in the embodiment of the present application can be well used to predict the viscosity of the target sample at different ethanol proportions.

[0130] Please refer to Figure 10 , Figure 10 , which is a schematic flowchart of the method for calculating the moisture correction value provided by an embodiment of the present application, specifically including:

[0131] S101. Perform data fitting based on the glycerol / propylene glycol viscosity coefficient, moisture component data, and viscosity detection data to obtain the calculation formula for the moisture correction value.

[0132] In step S101, when there is moisture in the e-liquid, there is an interaction between the moisture in the solvent and glycerol and propylene glycol, which will affect the viscosity value of the target sample. For example, when not considering the viscosity change due to the interaction between moisture and glycerol and propylene glycol, the ideal viscosity is:

[0133] η 理想 = 1.1672 * f(x) * f(a)f(b) * f(c) / 10 7 + 0.5613, R 2 = 0.9926, and its fitting effect is good. For the specific data fitting effect, please refer to Figure 11 . However, after introducing moisture, there is a certain deviation between the ideal viscosity and the actually detected viscosity. For example, in the PG / VG system containing 1% moisture, the viscosity coefficient of glycerol / propylene glycol is 106.05. At this time, the ideal viscosity η 理想 is 86.86, but the actually detected viscosity data is 99.55, with a deviation of 12.69, and the error has a great impact. For the specific deviation, please refer to Figure 12 . Therefore, in addition to calculating the individual influence of each component on the viscosity, it is also necessary to introduce a moisture correction value to calculate the viscosity changed due to the interaction between moisture and glycerol and propylene glycol. Similarly, ethanol also has an interaction with glycerol and propylene glycol, so an ethanol correction value needs to be introduced accordingly. Finally, the viscosity after data fitting and correction should be η1 = η 理想 + f(d) + f(e), where η1 is the viscosity after data fitting and correction, f(d) is the moisture correction value, and f(e) is the ethanol correction value.

[0134] Specifically, first, the calculation formula for the moisture correction value is explained. Through data fitting of different moisture component data with the glycerol / propylene glycol viscosity coefficient and the viscosity detection data, the regression coefficient is R 2 = 0.9397. For the specific data fitting effect, please refer to Figure 12 . The calculation formula for the moisture correction value is obtained as: f(d) = 0.9178 * (f2(b) + f1(f(a))) - 7.9747, where f(d) is the moisture correction value, f2(b) is the influence of moisture on the calculation deviation in the system containing moisture, PG, and VG, and f1(f(a)) is the influence of PG and VG on the calculation deviation in the system containing moisture, PG, and VG. The influence of moisture on the calculation deviation f2(b) = -0.0806 * b 2 - 0.9585 * b + 13.808. For the specific fitting effect, please refer to Figure 13 . The influence of PG and VG on the calculation deviation is: f1(f(a)) = -3.7329 * (f(a)) 2 + 0.7489 * f(a) + 14.108. For the specific fitting effect, please refer toFigure 14 。

[0135] Furthermore, the result of data fitting is verified by comparing it with the actual viscosity detection data. The specific result of data fitting is shown in Table 5 below.

[0136] Table 5 Data fitting result of moisture correction value

[0137]

[0138]

[0139] It can be determined from the above table that under the system of the target sample, when the moisture content of the target sample ranges from 1% to 10%, the viscosities of different glycerol / propylene glycol viscosity coefficients are predicted through the model, and the calculated results obtained by the prediction are 95.12% to 104.28% of the actual detection results, with good accuracy. Therefore, the prediction model provided in the embodiment of the present application can be preferably used to predict the viscosity of the target sample in the system where moisture, glycerol, and propylene glycol coexist.

[0140] Please refer to Figure 15 , Figure 15 which is a schematic flow chart of the method for calculating the ethanol correction value provided by an embodiment of the present application, specifically including:

[0141] S151. Perform data fitting on the glycerol / propylene glycol viscosity coefficient, ethanol component data, and viscosity detection data to obtain a calculation formula for the moisture correction value.

[0142] In step S151, when ethanol exists in the e-liquid, there is an interaction between ethanol and glycerol and propylene glycol in the solvent, which will affect the viscosity value of the target sample. Therefore, in addition to calculating the individual influence of each component on the viscosity, an ethanol correction value needs to be introduced to calculate the viscosity changed due to the interaction between ethanol and glycerol and propylene glycol. Specifically, data fitting is performed on different ethanol component data, glycerol / propylene glycol viscosity coefficient, and viscosity detection data, and the regression coefficient is R 2 = 0.9659. For the specific data fitting effect, please refer to Figure 16 . The calculation formula for the ethanol correction value is: f(e)= -0.0223*(f2(c)+f2(f(a))) 2 -0.1787*(f2(c)+f2(f(a)))+6.8507, where f(e) is the ethanol correction value, f2(c) is the influence of ethanol on the calculation deviation in the system containing ethanol, PG, and VG, and f2(f(a)) is the influence of PG and VG on the calculation deviation in the system containing ethanol, PG, and VG. The influence of ethanol on the calculation deviation f2(c)= 0.2761*c 2-3.7694*c - 6.7367, for the specific data fitting effect, please refer to Figure 17 。The influence of PG and VG on the calculation deviation is: f2(f(a)) = -3.2061*f(a) 2 + 6.7167*f(a) + 2.0752, for the specific data fitting effect, please refer to Figure 18 。

[0143] Furthermore, the results of data fitting are verified by comparing with the actual viscosity detection data. The specific results of data fitting are shown in Table 5 below.

[0144] Table 5 Data fitting results of moisture correction value

[0145]

[0146] According to the above table, it can be determined that under the system of the target sample, when the ethanol content in the target sample ranges from 1% to 10%, the viscosities of different viscosity coefficients of glycerol / propylene glycol are predicted through the model, and the calculated results obtained are 96.17% to 102.22% of the actual detection results, with good accuracy. Therefore, the prediction model provided by the embodiments of the present application can be well used to predict the viscosity of the target sample in the system where ethanol, glycerol, and propylene glycol coexist.

[0147] In some embodiments, in addition to the above factors such as temperature, glycerol, propylene glycol, moisture, and ethanol affecting the viscosity of the target sample, some other components will also affect the viscosity of the target sample, such as flavors, sweeteners, nicotine, etc. Specifically, please refer to Figure 19 , Figure 19 which is a schematic flowchart of the method for calculating the correction coefficient of other components provided by an embodiment of the present application, specifically including:

[0148] S191. Calculate the first correction coefficient and the second correction coefficient according to the viscosity detection data and the corrected viscosity data of different target samples.

[0149] In step S191, for a given plurality of target samples, by obtaining the temperature, propylene glycol component data, glycerol component data, moisture component data, and ethanol component data of the target samples, first calculate the predicted viscosities of at least two target samples through η1 = η 理想 + f(d) + f(e), and then perform data fitting on the predicted viscosities and the viscosity detection data of at least two target samples, so that the viscosity calculation formula after fitting is η 实际= p * η1 + q, where p is the first correction coefficient and q is the second correction coefficient. Thus, the first correction coefficient p and the second correction coefficient q can be determined by data fitting. The first correction coefficient p and the second correction coefficient q have no actual physical meaning and are used to relatively vaguely express the comprehensive influence of other substance components in the target sample on the viscosity. Thus, the viscosity calculation relationship can be obtained as follows:

[0150] η 实际 = p * (1.1672 * f(x) * f(a)f(b) * f(c) / 10 7 + 0.5613 + f(d) + f(e)) + q

[0151] where η 实际 is the predicted viscosity calculated according to the viscosity prediction model, p is the first correction coefficient, q is the second correction coefficient, f(x) is the temperature-viscosity coefficient, f(a) is the glycerol / propylene glycol viscosity coefficient, f(b) is the water viscosity coefficient, f(c) is the ethanol viscosity coefficient, f(d) is the water correction value, and f(e) is the ethanol correction value.

[0152] An embodiment of the present application further provides an e-liquid viscosity prediction model constructed based on the model construction method in the above embodiments. The e-liquid viscosity prediction model can be stored in an electronic device in the form of a computer-readable storage medium. The electronic device includes a personal computer, a smart phone, and other memories, such as random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); the memory can also include a combination of the above types of memories.

[0153] Based on the above e-liquid viscosity prediction model, the e-liquid viscosity prediction method provided by the embodiment of the present application is introduced below.

[0154] Please refer to Figure 20 , Figure 20 which is a schematic flowchart of the e-liquid viscosity prediction method provided by an embodiment of the present application, specifically including:

[0155] S201. Provide a pre-constructed e-liquid viscosity prediction model, which is configured with a functional relationship between the viscosity of the e-liquid, the components of the e-liquid, and the temperature.

[0156] S202. Obtain the viscosity detection data, component data, and target temperature of the target sample.

[0157] S203. Adjust the parameters of the electronic atomization liquid viscosity prediction model according to the viscosity detection data to obtain the target viscosity prediction model.

[0158] S204. Determine the predicted viscosity of the target sample according to the target viscosity prediction model, the component data, and the target temperature.

[0159] In step S201, the electronic atomization liquid viscosity prediction model is pre-constructed and stored in the electronic device, which can be directly called by the user.

[0160] In step S92, the viscosity detection data, the component data, and the target temperature of the target sample are pre-detected, and they can be input into the electronic atomization liquid viscosity prediction model.

[0161] In step S203, the target viscosity prediction model is a pre-stored model, which does not consider the influence of other substance components in the target sample on the viscosity except for temperature, glycerol, propylene glycol, water, and ethanol. Therefore, it is necessary to correct it with the actual viscosity detection data to obtain the target viscosity prediction model (i.e., the corrected electronic atomization liquid viscosity prediction model). Specifically, it includes:

[0162] S2031. Determine one component data or the target temperature of the target sample as the target parameter.

[0163] S2032. Provide the viscosity detection data of the target parameter under at least two different conditions.

[0164] S2033. Calculate the first correction coefficient and the second correction coefficient according to the viscosity detection data under at least two different conditions.

[0165] In step S2031, first, it is necessary to determine one component data or the target temperature of the target sample as the target parameter, such as the water component data or the target temperature. When the water component data is the target parameter, the predicted viscosity of the target sample at different water components can be calculated subsequently; when the temperature is the target parameter, the predicted viscosity of the target sample at different temperatures can be calculated subsequently.

[0166] In step S2032, obtain the viscosity detection data of the corresponding target parameter under at least two different conditions. For example, when the target parameter is the water component data, the viscosity detection data when the water component data is 1.6787% and the water component data is 8.6439% can be provided, and the viscosity detection data are 195.29 and 92.47 respectively.

[0167] In step S2033, input the temperature, glycerol component data, propylene glycol component data, moisture component data, and ethanol component data into the e-liquid viscosity prediction model, and the temperature viscosity coefficient f(x), glycerol / propylene glycol viscosity coefficient f(a), moisture viscosity coefficient f(b), and ethanol viscosity coefficient f(c) can be calculated respectively. Then, according to the formula η 理想 = 1.1672 * f(x) * f(a) * f(b) * f(c) / 107 + 0.5613, calculate the first predicted viscosity, where η 理想 is the first predicted viscosity.

[0168] After calculating the first predicted viscosity, it is also necessary to calculate the moisture correction value f(d) and the ethanol correction value f(e) respectively according to the moisture component data, ethanol component data, and glycerol / propylene glycol viscosity coefficient f(a). Then, according to the formula η1 = η 理想 + f(d) + f(e), calculate the second predicted viscosity, where η1 is the second predicted viscosity.

[0169] After calculating the second predicted viscosity, according to the target parameters, corresponding second predicted viscosity, and viscosity detection data under different selected conditions, calculate the first correction coefficient p and the second correction coefficient q according to the formula η 实际 = p * η1 + q, thus completing the target viscosity detection model, that is, η 实际 = p * (1.1672 * f(x) * f(a) * f(b) * f(c) / 10 7 + 0.5613 + f(d) + f(e)) + q, where η 实际 is the third predicted viscosity.

[0170] In step S204, according to the third predicted viscosity formula, substitute the temperature viscosity coefficient f(x), glycerol / propylene glycol viscosity coefficient f(a), moisture viscosity coefficient f(b), ethanol viscosity coefficient f(c), moisture correction value f(d), ethanol correction value f(e), first correction coefficient p, and second correction coefficient q, and the predicted viscosity of the target sample can be calculated.

[0171] The application of the e-liquid viscosity prediction method provided by the embodiments of the present application will be described below with reference to Examples 1 to 4.

[0172] Example 1

[0173] The target sample used in Example 1 is an e-liquid with a cola flavor, and the selected target parameter is the moisture component data, which is used to calculate the viscosity of the cola-flavored e-liquid when it contains different moisture contents. Specifically, the temperature of the target sample is t = 25 °C, the propylene glycol (PG) content is 21.34%, the glycerol (VG) content is 38.80%, the ethanol content is 15.14%, and the moisture content increases with the storage time, and the viscosity also changes. The first correction coefficient p and the second correction coefficient q are calculated with the moisture component data being 5.7624% and 11.1686% respectively. The first correction coefficient p is 0.9289, and the second correction coefficient q is 26.695. For the specific predicted viscosity results, please refer to Table 6 below.

[0174] Table 6 Predicted Viscosities of Cola-Flavored E-Liquids with Different Moisture Contents

[0175]

[0176]

[0177] As can be seen from the above table, for cola-flavored e-liquids with different moisture contents, the viscosity is predicted by the prediction method provided in the embodiments of the present application, and the calculated results obtained by the prediction are 96.22% to 101.35% of the actual detection results, showing good accuracy.

[0178] Example 2

[0179] The target sample used in Example 2 is an e-liquid with a blue raspberry flavor, and the selected target parameter is the moisture component data, which is used to calculate the viscosity of the blue raspberry-flavored e-liquid when it contains different moisture contents. For the specific predicted viscosity results, please refer to Table 7 below.

[0180] Table 7 Predicted Viscosities of Blue Raspberry-Flavored E-Liquids with Different Moisture Contents

[0181]

[0182] As can be seen from the above table, for blue raspberry-flavored e-liquids with different moisture contents, the viscosity is predicted by the prediction method provided in the embodiments of the present application, and the calculated results obtained by the prediction are 96.61% to 102.96% of the actual detection results, showing good accuracy.

[0183] Example 3

[0184] The target sample used in Example 3 is an e-liquid with a watermelon flavor, and the selected target parameter is the moisture component data, which is used to calculate the viscosity of the watermelon-flavored e-liquid when it contains different moisture contents. For the specific predicted viscosity results, please refer to Table 8 below.

[0185] Table 8 Predicted Viscosity of Watermelon Flavored E-Liquid at Different Moisture Contents

[0186]

[0187] As can be seen from the above table, for watermelon flavored e-liquids with different moisture contents, the viscosity is predicted by the prediction method provided in the embodiments of the present application, and the calculated results obtained by the prediction are 97.82% to 108.50% of the actual detection results, showing good accuracy.

[0188] Example 4

[0189] The target sample used in Example 4 is H'BUBBA flavored e-liquid, and the selected target parameter is temperature, which is used to calculate the viscosity of H'BUBBA flavored e-liquid at different temperatures. According to η 实际 = p*(1.1672*f(x)*f(a)*f(b)*f(c) / 10 7 + 0.5613 + f(d) + f(e)) + q, after transformation, we can get

[0190] η 实际 = (p*1.1672*f(a)*f(b)*f(c) / 10 7 )*f(x) + p*(0.5613 + f(d) + f(e)) + q. For the same

[0191] kind of e-liquid, its components remain unchanged, that is, the glycerol / propylene glycol viscosity coefficient f(a), moisture viscosity coefficient f(b), ethanol viscosity coefficient f(c), moisture correction value f(d), and ethanol correction value f(e) calculated according to the components are all fixed values. Therefore, p*1.1672*f(a)f(b)*f(c) / 10 7 and p*(0.5613 + f(d) + f(e)) + q are both fixed constants, that is, the viscosity of the e-liquid has a linear relationship with f(x). At this time, η 实际 has a linear relationship with the temperature factor x = 10 58 / (t + 273.15) 23 , that is, η 实际 = kx + m. The specific predicted viscosity results are shown in Table 9 below.

[0192] Table 9 Predicted Viscosity of H'BUBBA Flavored E-Liquid at Different Temperatures

[0193]

[0194] As can be seen from the above table, calculated based on the data at 25°C and 50°C, according to the formula η 实际= kx + m, we can calculate that k = 6.5783 and m = 12.548. That is, the formula for the change of viscosity with the temperature factor is η 实际

[0195] = 6.5783x + 12.548. For the H'BUBBA flavored e-liquid at different temperatures, the viscosity is predicted by the prediction method provided in the embodiments of the present application, and the calculated results obtained by the prediction are 96.02% to 104.99% of the actual detection results, showing good accuracy.

[0196] Example 5

[0197] The target sample used in Example 5 is the CHERRY BERRY flavored e-liquid, and the selected target parameter is temperature, which is used to calculate the viscosity of the CHERRY BERRY flavored e-liquid at different temperatures. For the specific predicted viscosity results, please refer to Table 10 below.

[0198] Table 10 Predicted Viscosities of CHERRY BERRY Flavored E-Liquid at Different Temperatures

[0199]

[0200]

[0201] According to the above table, calculating based on the data at 25°C and 50°C, according to the formula η 实际 = kx + m, we can calculate that k = 8.9306 and m = 12.205. That is, the formula for the change of viscosity with the temperature factor is η 实际

[0202] = 8.9306x + 12.205. For the CHERRY BERRY flavored e-liquid at different temperatures, the viscosity is predicted by the prediction method provided in the embodiments of the present application, and the calculated results obtained by the prediction are 97.91% to 103.79% of the actual detection results, showing good accuracy.

[0203] Example 6

[0204] The target sample used in Example 6 is the BLUE RAZZ CHERRY flavored e-liquid, and the selected target parameter is temperature, which is used to calculate the viscosity of the BLUE RAZZ CHERRY flavored e-liquid at different temperatures. For the specific predicted viscosity results, please refer to Table 11 below.

[0205] Table 11 Predicted Viscosities of BLUE RAZZ CHERRY Flavored E-Liquid at Different Temperatures

[0206]

[0207] According to the above table, based on the data at temperatures of 25°C and 50°C, according to the formula η 实际 = kx + m, it can be calculated that k = 8.1791 and m = 13.068. That is, the variation formula of viscosity with the temperature factor is η 实际

[0208] = 8.1791x + 13.068. For the BLUE RAZZ CHERRY flavored e-liquid at different temperatures, the viscosity is predicted by the prediction method provided in the embodiments of the present application. The calculated results obtained by the prediction are 96.84% to 104.49% of the actual detection results, showing good accuracy.

[0209] Example 7

[0210] The target sample used in Example 7 is the BERRY MELON SPLASH flavored e-liquid, and the selected target parameter is temperature, which is used to calculate the viscosity of the BERRY MELON SPLASH flavored e-liquid at different temperatures. For the specific predicted viscosity results, please refer to Table 11 below.

[0211] Table 12 Predicted viscosities of the BERRY MELON SPLASH flavored e-liquid at different temperatures

[0212]

[0213] According to the above table, based on the data at temperatures of 25°C and 50°C, according to the formula η 实际 = kx + m, it can be calculated that k = 8.8969 and m = 13.177. That is, the variation formula of viscosity with the temperature factor is η 实际

[0214] = 8.8969x + 13.177. For the BERRY MELON SPLASH flavored e-liquid at different temperatures, the viscosity is predicted by the prediction method provided in the embodiments of the present application. The calculated results obtained by the prediction are 97.21% to 102.08% of the actual detection results, showing good accuracy.

[0215] In summary, the electronic atomization liquid viscosity prediction method provided by the embodiments of the present application comprehensively considers the effects of temperature, propylene glycol, glycerol, moisture, ethanol, and other components on the viscosity of the electronic atomization liquid, and establishes a corresponding functional relationship based on data fitting. Compared with other functional relationships of exponential type, power function type, double logarithmic type, or polynomial type, by introducing a temperature factor, a glycerol / propylene glycol factor, a moisture factor, and an ethanol factor, the functional relationship of each viscosity-related parameter in the present application is a linear or quadratic function of one variable, the calculation process is simpler, and the predicted results output during the model extrapolation prediction process are relatively stable with small deviations. Generally speaking, when using the electronic atomization liquid viscosity prediction method provided by the embodiments of the present application, the predicted viscosity is 95% to 105% of the actually detected viscosity, with a small deviation from the actual value and good prediction effect.

[0216] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them; under the idea of the present disclosure, the technical features in the above embodiments or different embodiments can also be combined, and the steps can be implemented in any order, and there are many other variations in different aspects of the present disclosure as described above. For the sake of brevity, they are not provided in detail; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A method for predicting the viscosity of an electronic atomization liquid, characterized in that: include: Providing a pre-built electronic atomization liquid viscosity prediction model, wherein the electronic atomization liquid viscosity prediction model is configured with a functional relationship between the viscosity of the electronic atomization liquid and the electronic atomization liquid components and temperature; Obtaining viscosity test data, component data, and target temperature of a target sample; Adjusting the parameters of the electronic atomized liquid viscosity prediction model according to the viscosity detection data to obtain a target viscosity prediction model; The predicted viscosity of the target sample is determined according to the target viscosity prediction model, the component data, and the target temperature.

2. The prediction method according to claim 1, characterized in that: The electronic atomized liquid viscosity prediction model includes a first correction coefficient and a second correction coefficient. The parameters of the electronic atomized liquid viscosity prediction model are adjusted according to the viscosity detection data to obtain a target viscosity prediction model including: Determine a component data or a target temperature of the target sample as a target parameter; Providing viscosity detection data of the target parameter under at least two sets of different conditions; The first correction coefficient and the second correction coefficient are calculated according to the at least two groups of viscosity detection data under different conditions.

3. The prediction method according to claim 2, characterized in that: The electronic atomization liquid viscosity prediction model is configured with at least one of a temperature viscosity coefficient, a glycerol / propylene glycol viscosity coefficient, a water viscosity coefficient, and an ethanol viscosity coefficient; the component data includes at least one of glycerol component data, propylene glycol component data, water component data, and ethanol component data; the predicted viscosity includes a first predicted viscosity, and the predicted viscosity of the target sample is determined according to the target viscosity prediction model, the component data, and the target temperature, including: Calculate at least one of the temperature viscosity coefficient, the glycerol / propylene glycol viscosity coefficient, the water viscosity coefficient and the ethanol viscosity coefficient according to at least one of the glycerol component data, the propylene glycol component data, the water component data and the ethanol component data and the electronic atomization liquid viscosity prediction model; The first predicted viscosity is calculated according to at least one of the temperature viscosity coefficient, the glycerol / propylene glycol viscosity coefficient, the water viscosity coefficient, the ethanol viscosity coefficient, and the electronic atomization liquid viscosity prediction model.

4. The prediction method according to claim 3, characterized in that: The electronic atomized liquid viscosity prediction model is further configured with at least one of a water correction value and an ethanol correction value, the predicted viscosity further includes a second predicted viscosity, and after calculating the first predicted viscosity, further includes: Calculate the moisture correction value according to the moisture component data, the glycerol / propylene glycol viscosity coefficient and the electronic atomization liquid viscosity prediction model; Calculating the ethanol correction value according to the ethanol component data, the glycerol / propylene glycol viscosity coefficient and the electronic atomization liquid viscosity prediction model; A second predicted viscosity is calculated based on the first viscosity coefficient, the water correction value, and the ethanol correction value.

5. The prediction method according to claim 4, characterized in that: The predicted viscosity further includes a third predicted viscosity, and after calculating the second predicted viscosity, further includes: The third predicted viscosity is calculated according to the first correction coefficient, the second correction coefficient, and the second predicted viscosity.

6. A method for constructing a viscosity prediction model for electronic atomized liquid according to any one of claims 1 to 5, characterized in that: include: Determining a plurality of parameters related to the viscosity of the target sample electronic atomized liquid; Obtaining the viscosity detection data of each of the relevant parameters under different conditions; Performing data fitting based on the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity; The electronic atomization liquid viscosity prediction model is constructed according to the multiple related parameters and the viscosity influence coefficient.

7. The construction method according to claim 6, characterized in that: The relevant parameters include temperature, and the data fitting is performed based on the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on viscosity, which includes: Construct temperature factor based on temperature; Data fitting is performed based on the temperature factor and the viscosity detection data to obtain a temperature viscosity coefficient calculation formula.

8. The construction method according to claim 6, characterized in that: The relevant parameters include propylene glycol component data and propylene glycol component data. The data fitting is performed based on the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity, which includes: Determining the ratio of the glycerol component data to the sum of the glycerol component data and the propylene glycol component data; constructing a glycerol / propylene glycol factor according to the ratio; Data fitting is performed based on the propylene glycol / glycerol factor and the viscosity detection data to obtain a calculation formula for the propylene glycol / glycerol viscosity coefficient.

9. The construction method according to claim 6, characterized in that: The relevant parameters include water component data, and the data fitting is performed based on the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on viscosity, which includes: constructing a moisture factor according to the moisture component data; Data fitting is performed based on the moisture factor and the viscosity detection data to obtain a calculation formula for the moisture viscosity coefficient.

10. The construction method according to claim 6, characterized in that: The relevant parameters include ethanol component data, and the data fitting is performed based on the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity, which includes: constructing an ethanol factor according to the ethanol component data; Data fitting is performed based on the ethanol factor and the viscosity detection data to obtain a calculation formula for the ethanol viscosity coefficient.

11. The construction method according to claim 8, characterized in that: The relevant parameters include water component data, propylene glycol component data and propylene glycol component data. The data fitting is performed based on the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity, which includes: Data fitting is performed based on the propylene glycol / glycerol viscosity coefficient, the water component data and the viscosity detection data to obtain a water correction value calculation formula.

12. The construction method according to claim 8, characterized in that: The relevant parameters include ethanol component data, propylene glycol component data and propylene glycol component data. The data fitting is performed based on the viscosity detection data and the relevant parameters under different conditions to obtain the influence coefficient of the relevant parameters on the viscosity, which includes: Data fitting is performed based on the propylene glycol / glycerol viscosity coefficient, the ethanol component data and the viscosity detection data to obtain an ethanol correction value calculation formula.

13. A prediction model for the viscosity of electronic atomized liquid obtained according to the construction method according to any one of claims 6 to 12.

Images