Body lotion preparation method based on intelligent optimization and product
Through intelligent optimization methods, based on the multivariate linear regression model and similarity calculation, the homogeneous parameters of the emollient are dynamically adjusted, which solves the product instability caused by empirical settings in traditional preparation, and achieves more efficient production and quality control.
Patent Information
- Application Number
- CN202510462886.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-14
- Publication Date
- 2025-07-08
AI Technical Summary
In the preparation of traditional emollicular milk, homogeneity parameter setting depends on personal experience, resulting in unstable product quality caused by differences in raw material particle size, lack of scientific data analysis and records, which affects product texture and stability.
The intelligent optimization method is adopted to establish the mapping relationship between particle size information and homogeneous parameters through a multivariate linear regression model, and use similarity calculation and threshold judgment to dynamically adjust the homogeneous parameters, combining data cleaning and historical production data optimization and preparation process.
The scientific and precise setting of homogeneous parameters is achieved, the texture uniformity and stability of the moisturizing lotion is improved, the product quality fluctuations caused by changes in raw material particle size are reduced, and the production efficiency and product consistency are improved.
Smart Images
Figure CN120267571A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of cosmetic production, and more particularly, to a method and product for preparing a moisturizing lotion based on intelligent optimization. Background Art
[0002] In the field of traditional moisturizing lotion preparation, the setting of homogenization parameters (such as rotation speed, homogenization time, homogenization pressure, and homogenization temperature) mainly depends on the personal experience of operators or fixed parameters. Different operators may set parameters based on personal habits or experience, resulting in significant differences in homogenization parameters during the production of the same formula in different batches. In actual production, due to different sources and batches of raw materials, there may be significant variations in their particle sizes. If these particle size differences are not considered and fixed homogenization parameters are still used for production, it may lead to insufficient homogenization of raw materials in some batches.
[0003] The particle size and distribution of raw material particles have a crucial impact on the quality of the moisturizing lotion. However, traditional preparation processes often overlook this key factor and do not incorporate the particle size information of raw material particles into the setting of homogenization parameters.
[0004] When the raw material particles are relatively large, homogenization according to conventional parameters may not be able to refine the particles to an ideal degree, resulting in a rough texture of the moisturizing lotion and affecting the usability and stability of the product. Conversely, when the raw material particles are relatively small, excessive homogenization may damage the structure of the raw materials and reduce the efficacy of the product.
[0005] Although enterprises have accumulated a large amount of historical production data during long-term production processes, these data are often not effectively sorted and analyzed. On the one hand, data records are not standardized, lacking a unified standard and format, resulting in poor readability and usability of the data. On the other hand, the lack of scientific data cleaning and analysis methods causes a large amount of valuable information to be buried in the vast amount of data.
[0006] Therefore, it is particularly important to design a method for preparing a moisturizing lotion that can intelligently optimize the homogenization process. Summary of the Invention
[0007] In view of this, the present invention proposes a method and product for preparing a moisturizing lotion based on intelligent optimization, aiming to solve the problem of unreasonable setting of homogenization parameters in the current technology.
[0008] On the one hand, the present invention proposes a method and product for preparing a moisturizing lotion based on intelligent optimization, including:
[0009] Adding the first raw material group, PEG-100 stearate, glyceryl stearate, and propylparaben to an oil-phase pot and stirring and heating to 80°C;
[0010] Add water to the emulsifying pot, then add the second raw material group to the emulsifying pot, and finally add methylparaben and disodium EDTA to the emulsifying pot, stir and heat to 85°C; add the materials in the oil phase pot to the emulsifying pot, collect the current particle size information of the particles in the first raw material group and the second raw material group, and determine the initial homogenization parameters according to the current particle size information; obtain the historical particle size information of the particles in the first raw material group and the second raw material group in the historical batches, calculate the similarity between the current particle size information and each of the historical particle size information, obtain the maximum value of the similarity and use it as the maximum similarity; determine whether to adjust the initial homogenization parameters according to the maximum similarity; when it is determined that adjustment is required, use the adjusted homogenization parameters as the final homogenization parameters; when it is determined that no adjustment is required, use the initial homogenization parameters as the final homogenization parameters;
[0011] After homogenization, keep warm and stir; then add dipotassium glycyrrhizinate into the emulsifying pot, start the vacuum pump to evacuate, and keep warm at 80°C for 20 minutes;
[0012] The temperature was lowered and stirred in the emulsifying pot. After the temperature dropped to 45°C, phenoxyethanol was added into the emulsifying pot, and then stirred for 10 minutes. After the temperature dropped below 35°C, the solid phase was filtered to obtain a moisturizing lotion.
[0013] Furthermore, the first raw material group includes: cetearyl alcohol, caprylic / capric triglyceride, cyclopentasiloxane, liquid paraffin, ethylhexyl palmitate, dimethicone, evening primrose oil, cocoyl glucoside and tocopherol;
[0014] The second raw material group includes: xanthan gum, glycerin, propylene glycol and allantoin.
[0015] Furthermore, the historical data is stored by the following method:
[0016] The particle size information of each batch is recorded as historical particle size information, and the initial homogenization parameters and final homogenization parameters of each batch are recorded based on the time series;
[0017] Clean the recorded historical particle size information, initial homogenization parameters and final homogenization parameters and store them as historical data;
[0018] The historical particle size information and the current particle size information both include: D10 value, D50 value and D90 value, and the initial homogenization parameters and the final homogenization parameters both include: rotation speed parameter, homogenization time, homogenization pressure and homogenization temperature.
[0019] Furthermore, the data cleaning includes:
[0020] Judge whether there is any missing in the historical particle size information, initial homogenization parameters, and final homogenization parameters in each piece of the historical data. If it is judged that there is a missing situation in the historical data, then delete the current historical data.
[0021] Further, when determining the initial homogenization parameters according to the particle size information of the current time, it includes:
[0022] Based on multiple linear regression, establish the mapping relationship between the current particle size parameters and the initial homogenization parameters. The specific relationship is as follows:
[0023] P j =α j ·D 50 +β j ·(D 90 -D 10 )+γ j ;
[0024] Among them, α j , β j and γ j are all model coefficients; P j is the jth homogenization parameter, j = 1, 2, 3, 4; P1 is the rotational speed parameter, P2 is the homogenization time, P3 is the homogenization pressure, and P4 is the homogenization temperature.
[0025] Further, the α j , β j and γ j are obtained through the following relationship:
[0026] Convert the historical particle size information of each batch into matrix form, with each row corresponding to the historical particle size information of one batch, and the third column being all 1s, to obtain matrix X; construct the initial homogenization parameters of each batch into an observation matrix Y; construct α j , β j and γ j into matrix θ; calculate and solve matrix θ according to the following formula:
[0027]
[0028] Among them, n is the total number of batches, is the D50 value in the nth batch, is the jth homogenization parameter in the nth batch.
[0029] Further, the similarity is calculated and obtained through the following relationship:
[0030]
[0031] Among them, Sim(X, Y) is the similarity between the current particle size information and any historical particle size information, where X represents the current particle size information and Y represents the historical particle size information; x i is the i-th current particle size information, and y i is the i-th historical particle size information; i = 1, 2, 3; when i = 1, it represents the D10 value, when i = 2, it represents the D50 value, and when i = 3, it represents the D90 value.
[0032] Further, when determining whether to adjust the initial homogenization parameters according to the maximum similarity, it includes:
[0033] Preset a similarity threshold. When the maximum similarity is greater than or equal to the similarity threshold, it is determined that no adjustment is required; when the maximum similarity is less than the similarity threshold, it is determined that adjustment is required.
[0034] Further, when it is determined that adjustment is required, the adjustment is performed by the following method:
[0035] Calculate the difference between the current particle size information and the D90 value in the historical particle size information. When the difference is negative, adjust it through the following formula to obtain the final homogenization parameter:
[0036] P' j = P j *(1 - L);
[0037] When the difference is positive, adjust it through the following formula to obtain the final homogenization parameter:
[0038] P' j = P j *(1 + L);
[0039] Among them, P' j is the final homogenization parameter of the j-th homogenization parameter, and L is the difference between the current particle size information and the D90 value in the historical particle size information.
[0040] On the other hand, the present invention also protects a skin lotion obtained by the above skin lotion preparation method based on intelligent optimization.
[0041] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0042] The present invention establishes an accurate mapping relationship between the raw material particle size information and the initial homogenization parameters through a multiple linear regression model. This method breaks away from the bondage of traditional experience and makes the determination of the initial homogenization parameters more scientific and accurate. In practical applications, by collecting a large amount of historical production data, including the particle size information of the raw materials and the corresponding homogenization parameters, the model coefficients are solved using the least squares method. These coefficients reflect the specific influence degree of the particle size parameters on the homogenization parameters, enabling the accurate calculation of the initial homogenization parameters based on the real-time collected raw material particle size information in a new production batch.
[0043] Moreover, the present invention establishes a dynamic optimization system based on similarity calculation and threshold judgment. By calculating the similarity between the current particle size information and the historical particle size information and comparing it with a pre-set threshold, it can be intelligently judged whether the initial homogenization parameters need to be adjusted. When the similarity is greater than or equal to the threshold, it indicates that the particle size characteristics of the current raw material are similar to those of a certain batch in the historical data, and at this time, the initial homogenization parameters can be directly used for production, avoiding unnecessary adjustments. When the similarity is less than the threshold, it indicates that the particle size characteristics of the current raw material are significantly different from the historical data, and parameter adjustment is required according to the particle size difference.
[0044] In terms of data recording, the recording standards and formats for historical particle size information, initial homogenization parameters, and final homogenization parameters are specified to ensure the integrity and accuracy of the data. In the data cleaning process, incomplete data is deleted by judging whether there are missing values, improving the data quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:
[0046] Figure 1 It is a flowchart of the method for preparing a skin lotion based on intelligent optimization provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0047] Exemplary embodiments disclosed in the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments.
[0048] During the preparation of the emollient lotion, the length of the homogenization time has an important impact on the quality, stability, and usage effect of the product. Excessive homogenization time may bring the following adverse effects: The oil droplet size is too small, affecting the skin feel. Prolonged homogenization will continuously break up the oil droplets, making the particle size too small (less than 100 nm);
[0049] Although smaller particle sizes contribute to the transparency of the emulsion, overly small particle sizes may affect the skin feel; the skin feel of the emulsion becomes thinner, lacking a smooth and thick feeling, and is prone to a false smooth but not moisturizing feeling, making it difficult to form a protective film after use;
[0050] Prolonged homogenization will cause the emulsifier encapsulation ability in the emulsion system to weaken; the emulsifier may denature or fail under prolonged high shear, resulting in emulsion instability, prone to layering or oil-water separation, and oil separation will also occur during long-term storage, reducing the storage stability of the product and shortening the shelf life.
[0051] Refer to Figure 1 As shown, an emollient lotion preparation method based on intelligent optimization provided by an embodiment of the present invention includes:
[0052] S1: Add the first raw material group, PEG-100 stearate, glyceryl stearate, and propylparaben to the oil phase pot and stir and heat to 80 °C;
[0053] S2: Add water to the emulsification pot, then add the second raw material group to the emulsification pot, and finally add methylparaben and disodium EDTA to the emulsification pot and stir and heat to 85 °C;
[0054] S3: Add the materials in the oil phase pot to the emulsification pot, collect the current particle size information of the particles in the first raw material group and the second raw material group, and determine the initial homogenization parameters according to the current particle size information; obtain the historical particle size information of the particles in the first raw material group and the second raw material group in the historical batches, calculate the similarity between the current particle size information and each historical particle size information, obtain the maximum value of the similarity as the maximum similarity; judge whether to adjust the initial homogenization parameters according to the maximum similarity; when it is judged that adjustment is needed, use the adjusted homogenization parameters as the final homogenization parameters; when it is judged that no adjustment is needed, use the initial homogenization parameters as the final homogenization parameters;
[0055] S4: After homogenization, heat preservation and stirring are performed; then dipotassium glycyrrhizinate is added to the emulsification pot, and a vacuum pump is turned on to evacuate the mixture, and the mixture is kept at 80°C for 20 minutes;
[0056] S5: Cooling and stirring in the emulsifying pot, after the temperature drops to 45°C, adding phenoxyethanol into the emulsifying pot, and then stirring for 10 minutes; after the temperature drops below 35°C, filtering and taking the solid phase to obtain a moisturizing lotion.
[0057] In each production batch, the added amounts of the first raw material and the second raw material are fixed.
[0058] It should be noted that for the determination and adjustment of homogenization parameters: first, collect the particle size information, add the materials in the oil phase pot to the emulsification pot, and collect the particle size information of the particles in the first raw material group and the second raw material group. The particle size and distribution of the raw material particles will affect the texture, stability and feel of the lotion.
[0059] Determine initial parameters: Determine initial homogenization parameters based on the collected particle size information. Homogenization parameters may include homogenization speed, time, pressure, etc. Appropriate homogenization parameters can fully mix the raw materials to achieve ideal texture and stability.
[0060] Comparison of similarity: Obtain the historical particle size information of raw material particles in historical batches, calculate the similarity between the current particle size information and each historical particle size information, and select the maximum similarity as the maximum similarity.
[0061] Adjust parameters: Determine whether to adjust the initial homogenization parameters based on the maximum similarity. If the maximum similarity is high, it means that the particle size of the current raw material is similar to that of a historical batch, and there may be no need to adjust the initial parameters; if the similarity is low, the parameters need to be adjusted to ensure stable product quality. The parameters finally determined are the final homogenization parameters.
[0062] In some embodiments of the present application, the first raw material group includes: cetearyl alcohol, caprylic / capric triglyceride, cyclopentasiloxane, liquid paraffin, ethylhexyl palmitate, dimethicone, evening primrose oil, coconut glucoside and tocopherol;
[0063] The second raw material group includes: xanthan gum, glycerin, propylene glycol and allantoin.
[0064] In some embodiments of the present application, historical data is stored by the following method:
[0065] The particle size information of each batch is recorded as historical particle size information, and the initial homogenization parameters and final homogenization parameters of each batch are recorded based on the time series;
[0066] Clean the recorded historical particle size information, initial homogenization parameters, and final homogenization parameters, and store them as historical data.
[0067] Both the historical particle size information and the current particle size information include the D10 value, D50 value, and D90 value. Both the initial homogenization parameters and the final homogenization parameters include the rotational speed parameter, homogenization time, homogenization pressure, and homogenization temperature.
[0068] Among them, the D10 value is the particle size corresponding to when the cumulative particle size distribution reaches 10%, the D50 value is the particle size corresponding to when the cumulative particle size distribution reaches 50%, and the D90 value is the particle size corresponding to when the cumulative particle size distribution reaches 90%.
[0069] It should be noted that the current particle size information collected during each batch of production is retained and saved as historical particle size information. These particle size information cover the D10 value, D50 value, and D90 value. The D10 value means the particle size corresponding to when the cumulative particle size distribution reaches 10% among all particles; the D50 value represents the particle size corresponding to when the cumulative particle size distribution reaches 50%, which is the so-called median particle size; the D90 value is the particle size corresponding to when the cumulative particle size distribution reaches 90%. Through these three characteristic values, the particle size distribution of the raw material particles can be described more comprehensively.
[0070] Homogenization parameter recording: Record the initial homogenization parameters (including rotational speed parameter, homogenization time, homogenization pressure, and homogenization temperature) and the adjusted final homogenization parameters during the production of each batch in chronological order. This recording method based on time series can reflect the change of parameters over time during the production process.
[0071] After the recording is completed, data cleaning should be performed on the historical particle size information, initial homogenization parameters, and final homogenization parameters. Data cleaning is to ensure the accuracy and integrity of the data, and remove possible incorrect data, missing values, etc. The data after cleaning will be stored to form historical data for subsequent analysis and utilization.
[0072] The particle size information and homogenization parameter records in the historical data provide a strong basis for the intelligent adjustment of homogenization parameters in subsequent production. When a new batch of production is carried out, by comparing the current particle size information with the historical particle size information and calculating their similarity, the homogenization parameters adopted in the historical similar particle size cases can be referred to, so as to more scientifically determine the initial homogenization parameters and judge whether parameter adjustment is needed according to the similarity to achieve better homogenization effect.
[0073] A large amount of historical data helps to establish the mapping relationship between particle size information and homogenization parameters. For example, machine learning algorithms can be used to analyze the historical data to find the most suitable combination of homogenization parameters under different particle size distributions, and then achieve the accurate prediction and optimization of homogenization parameters.
[0074] Through long-term accumulation and analysis of historical data, the product quality performance of raw materials with different particle sizes under specific homogenization parameters can be summarized. Production personnel can, based on these empirical rules, adjust the homogenization parameters in advance in new production batches to avoid product quality fluctuations caused by changes in the particle size of raw materials, thereby improving the stability and consistency of the quality of the skin lotion products.
[0075] As the historical data continues to be enriched, the production process can be continuously optimized and improved. For example, if it is found that specific homogenization parameters can significantly improve the stability or usability of the product within a certain particle size range, these optimized parameters can be applied to subsequent production to gradually enhance the overall quality of the product.
[0076] Historical data can serve as an important reference for production process monitoring. By comparing the particle size information and homogenization parameters of different batches, abnormal situations in the production process can be detected in a timely manner, such as a sudden change in the particle size of a certain batch or the deviation of the homogenization parameters from the normal range, so as to take measures for adjustment in a timely manner to ensure the smooth progress of the production process.
[0077] In some embodiments of the present application, data cleaning includes:
[0078] Judge whether there is any missing historical particle size information, initial homogenization parameters, and final homogenization parameters in each historical data. If it is judged that there is a missing situation in the historical data, the current historical data will be deleted.
[0079] It should be noted that when establishing the mapping relationship between particle size information and homogenization parameters, calculating similarity, and making parameter adjustment judgments using historical data, complete and accurate data is the basis for obtaining reliable conclusions. If data with missing values is used for analysis, it may introduce biases, resulting in incorrect parameter adjustment suggestions or inaccurate quality predictions. For example, if the homogenization time data of a certain batch is missing, incorrect correlation results may be obtained when analyzing the relationship between particle size and homogenization time, which will in turn affect the setting of the homogenization time parameter in subsequent production. If data with missing values is used for training machine learning models or statistical analysis, the reliability and prediction accuracy of the models will be reduced. Deleting missing data can ensure that the data used for analysis and modeling has a high quality, so that the established model can more accurately reflect the true relationship between particle size information and homogenization parameters, providing more reliable support for intelligent optimization.
[0080] In some embodiments of the present application, when determining the initial homogenization parameters according to the current particle size information, it includes:
[0081] Based on multiple linear regression, establish the mapping relationship between the current particle size parameters and the initial homogenization parameters. The specific relationship is as follows:
[0082] Pj = α j ·D 50 + β j ·(D 90 - D 10 ) + γ j ;
[0083] Wherein, α j , β j and γ j are all model coefficients; P j is the j-th homogenization parameter, j = 1, 2, 3, 4; P1 is the rotational speed parameter, P2 is the homogenization time, P3 is the homogenization pressure, and P4 is the homogenization temperature.
[0084] It should be noted that the determination of each homogenization parameter (including rotational speed, homogenization time, homogenization pressure, and homogenization temperature) is related to two particle size characteristics of the raw material particles and a basic value. Specifically, based on the particle size at which the cumulative particle size distribution reaches 50%, combined with the difference in particle sizes corresponding to when the cumulative particle size distribution reaches 90% and 10%, and then adding a fixed basic value, through a specific calculation logic, the corresponding homogenization parameter is finally obtained.
[0085] The above content changes the traditional extensive mode of setting homogenization parameters by experience, replaces subjective judgment with a scientific calculation model, and makes the determination of parameters such as rotational speed, time, pressure, and temperature more accurate. Calculating parameters based on the true particle size characteristics of the raw material particles makes the production process more adaptable to the raw material characteristics, reduces product quality fluctuations caused by unreasonable parameters, and ensures key quality indicators such as the texture and stability of the emollient lotion. Training the model coefficients with historical data can deeply explore the internal relationship between particle size and homogenization parameters, provide a scientific basis for optimizing the production process, and help improve production efficiency and product standardization level.
[0086] In some embodiments of the present application, α j , β j and γ j are obtained through the following relationship:
[0087] Convert the historical particle size information of each batch into matrix form, with each row corresponding to the historical particle size information of one batch, and the third column being 1 for all, to obtain matrix X; construct the initial homogenization parameters of each batch into an observation matrix Y; construct α j , β j and γ j into matrix θ; calculate and solve matrix θ according to the following formula:
[0088]
[0089] Wherein, n is the total number of batches, is the D50 value in the n-th batch, is the j-th homogenization parameter in the n-th batch.
[0090] It should be noted that, first, the historical production data is sorted: the historical particle diameter information of each batch (including the particle diameter characteristic data of different batches) is formed into a specific matrix according to the rule that each row corresponds to one batch, and all elements in the third column of the matrix are filled with 1. At the same time, the initial homogenization parameters of each batch (such as rotation speed, time, pressure, temperature, etc.) are sorted into another matrix. Finally, through mathematical operation steps such as matrix transpose, matrix multiplication, and inverse matrix calculation, a matrix containing three coefficients is calculated, and these three coefficients are used to accurately establish the quantitative relationship between the particle diameter information and the homogenization parameters.
[0091] The above content calculates the coefficients reflecting the internal relationship between the particle diameter and the homogenization parameters by using rigorous matrix operations on historical production data, making the determination of the initial homogenization parameters more scientific and accurate. For example, it can accurately match the rotation speed, time and other parameters corresponding to different particle diameters.
[0092] Setting the homogenization parameters based on the accurate coefficient relationship makes the production process more adaptable to the characteristics of raw material particles, reduces the product quality fluctuations caused by unreasonable parameters, and ensures that the texture of the lotion is uniform and has strong stability.
[0093] Mining the value of historical data, optimizing the production process in a data-driven manner, providing a scientific reference for parameter setting in subsequent production, promoting the upgrading of the lotion preparation process towards intelligence and precision, and improving the production efficiency and standardization level.
[0094] In some embodiments of the present application, the similarity is calculated and obtained through the following relationship:
[0095]
[0096] where Sim(X,Y) is the similarity between the current particle size information and any historical particle size information, X represents the current particle size information, and Y represents the historical particle size information; x i is the i-th current particle size information, and y i is the i-th historical particle size information; i = 1, 2, 3; when i = 1, it represents the D10 value, when i = 2, it represents the D50 value, and when i = 3, it represents the D90 value.
[0097] In the preparation of lotion, by quantifying the similarity between the current and historical particle size information, a basis is provided for judging whether to adjust the homogenization parameters. If the similarity is high, it indicates that the current raw material particle size characteristics are similar to those of a certain historical batch, and the historical parameters can be referred to; if the similarity is low, the parameters need to be further adjusted to ensure that the production adapts to the raw material characteristics and improves the product quality stability.
[0098] In some embodiments of the present application, when determining whether to adjust the initial homogenization parameters based on the maximum similarity, it includes:
[0099] A similarity threshold is preset in advance. When the maximum similarity is greater than or equal to the similarity threshold, it is determined that no adjustment is required; when the maximum similarity is less than the similarity threshold, it is determined that adjustment is required.
[0100] In some embodiments of the present application, when it is determined that adjustment is required, the adjustment is performed by the following method:
[0101] Calculate the difference between the D90 values of the current particle size information and the historical particle size information. When the difference is negative, the following formula is used for adjustment to obtain the final homogenization parameter:
[0102] P' j = P j *(1 - L);
[0103] When the difference is positive, the following formula is used for adjustment to obtain the final homogenization parameter:
[0104] P' j = P j *(1 + L);
[0105] Where P' j is the final homogenization parameter of the j-th homogenization parameter, and L is the difference between the D90 values of the current particle size information and the historical particle size information.
[0106] It should be noted that the rule for determining whether to adjust the initial homogenization parameter is as follows.
[0107] Preset standard: Set a similarity threshold in advance as the judgment benchmark.
[0108] Comparison and judgment: Compare the maximum similarity calculated from the current particle size information and the historical particle size information with the preset threshold.
[0109] If the maximum similarity ≥ threshold, it indicates that the current raw material particle size characteristics are highly similar to those of a certain historical batch, and there is no need to adjust the initial homogenization parameters.
[0110] If the maximum similarity < threshold, it indicates that the current raw material particle size characteristics are quite different, and the initial homogenization parameters need to be adjusted.
[0111] By presetting the similarity threshold and comparing the maximum similarity, it can accurately determine whether to adjust the initial homogenization parameters. When the maximum similarity is higher than the threshold, it indicates that the current raw material particle size characteristics match well with the historical data. No adjustment can avoid ineffective operations and save production resources and time; while when the maximum similarity is lower than the threshold, the adjustment mechanism is triggered in a timely manner to ensure that the homogenization parameters can adapt to the actual differences in the raw material particle size, and avoid affecting the product quality due to fixed parameters.
[0112] Taking the D90 value difference as the adjustment basis, different formulas are used to calculate the final homogenization parameters in different cases (positive or negative difference). If the difference is negative (the current D90 value is less than the historical value), the parameter influence is reduced through the corresponding formula to adapt to the change of smaller particle size; if the difference is positive (the current D90 value is greater than the historical value), the parameter strength is enhanced through another formula. This refined adjustment enables the homogenization parameters (such as rotation speed, time, etc.) to scientifically respond to the change of raw material particle size, ensuring the full homogenization of raw materials, improving the texture uniformity and stability of the emollient lotion, reducing the quality fluctuation caused by particle size difference, and guaranteeing the product quality.
[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the specific implementation manners of the present invention can still be modified or equivalently replaced, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.
Claims
1. A method for preparing a skin lotion based on intelligent optimization, characterized in that, include: Add the first raw material group, PEG-100 stearate, glyceryl stearate and propylparaben to the oil phase pot, stir and heat to 80°C; Add water to the emulsifying pot, then add the second raw material group to the emulsifying pot, and finally add methylparaben and disodium EDTA to the emulsifying pot, stir and heat to 85°C; add the materials in the oil phase pot to the emulsifying pot, collect the current particle size information of the particles in the first raw material group and the second raw material group, and determine the initial homogenization parameters according to the current particle size information; obtain the historical particle size information of the particles in the first raw material group and the second raw material group in the historical batches, calculate the similarity between the current particle size information and each of the historical particle size information, obtain the maximum value of the similarity and use it as the maximum similarity; determine whether to adjust the initial homogenization parameters according to the maximum similarity; When it is determined that adjustment is required, the adjusted homogenization parameters are used as the final homogenization parameters; when it is determined that adjustment is not required, the initial homogenization parameters are used as the final homogenization parameters; After homogenization, keep warm and stir; then add dipotassium glycyrrhizinate into the emulsifying pot, start the vacuum pump to evacuate, and keep warm at 80°C for 20 minutes; The temperature was lowered and stirred in the emulsifying pot. After the temperature dropped to 45°C, phenoxyethanol was added into the emulsifying pot, and then stirred for 10 minutes. After the temperature dropped below 35°C, the solid phase was filtered to obtain a moisturizing lotion.
2. The method for preparing a skin lotion based on intelligent optimization according to claim 1, wherein The first raw material group includes: cetearyl alcohol, caprylic / capric triglyceride, cyclopentasiloxane, liquid paraffin, ethylhexyl palmitate, polydimethylsiloxane, evening primrose oil, cocoyl glucoside and tocopherol; The second raw material group includes: xanthan gum, glycerin, propylene glycol and allantoin.
3. The method for preparing a skin lotion based on intelligent optimization according to claim 1, characterized in that, The historical data is stored in the following way: The particle size information of each batch is recorded as historical particle size information, and the initial homogenization parameters and final homogenization parameters of each batch are recorded based on the time series; Clean the recorded historical particle size information, initial homogenization parameters and final homogenization parameters and store them as historical data; The historical particle size information and the current particle size information both include: D10 value, D50 value and D90 value, and the initial homogenization parameters and the final homogenization parameters both include: rotation speed parameter, homogenization time, homogenization pressure and homogenization temperature.
4. The method for preparing the skin lotion based on intelligent optimization according to claim 3, characterized in that, The data cleaning includes: It is determined whether the historical particle size information, initial homogenization parameters and final homogenization parameters in each of the historical data are missing. If it is determined that there are missing information in the historical data, the current historical data is deleted.
5. The method for preparing a skin lotion based on intelligent optimization according to claim 4, wherein When the initial homogenization parameters are determined according to the secondary particle size information, it includes: The mapping relationship between the current particle size parameters and the initial homogeneity parameters is established based on multivariate linear regression. The specific relationship is as follows: P j = α j · D 50 + β j · (D 90 - D 10 ) + γ j ; Among them, α j , β j and γ j are all model coefficients; P j is the j-th homogenization parameter, j = 1, 2, 3, 4; P1 is the rotational speed parameter, P2 is the homogenization time, P3 is the homogenization pressure, and P4 is the homogenization temperature.
6. The method for preparing a skin lotion based on intelligent optimization according to claim 5, wherein Said α j , β j and γ j are obtained through the following relationship: Convert the historical particle size information of each batch into a matrix form, where each row corresponds to the historical particle size information of a batch, and the third column is all 1, to obtain matrix X; construct the initial homogenization parameters of each batch into an observation matrix Y; for α j , β j and γ j construct them into matrix θ; calculate and solve matrix θ according to the following formula: where n is the total number of batches, is the D50 value in the n-th batch, is the j-th homogenization parameter in the n-th batch.
7. The method for preparing a skin lotion based on intelligent optimization according to claim 6, wherein The similarity is calculated by the following relationship: Among them, Sim(X, Y) is the similarity between the current particle size information and any historical particle size information, X represents the current particle size information, and Y represents the historical particle size information; x i is the i-th current particle size information, and y i is the i-th historical particle size information; i = 1, 2, 3; when i = 1, it represents the D10 value, when i = 2, it represents the D50 value, and when i = 3, it represents the D90 value.
8. The method for preparing a skin lotion based on intelligent optimization according to claim 7, wherein When judging whether to adjust the initial homogenization parameters according to the maximum similarity, it includes: A similarity threshold is preset, and when the maximum similarity is greater than or equal to the similarity threshold, it is determined that no adjustment is required; when the maximum similarity is less than the similarity threshold, it is determined that adjustment is required.
9. The method for preparing a skin lotion based on intelligent optimization according to claim 8, wherein When it is determined that adjustments are necessary, make them in the following ways: Calculate the difference between the D90 values in the current particle size information and the historical particle size information. When the difference is negative, adjust it through the following formula to obtain the final homogenization parameter: P' j = P j *(1 - L); When the difference is positive, adjust it through the following formula to obtain the final homogenization parameter: P' j = P j *(1 + L); where P' j is the final homogenization parameter of the j-th homogenization parameter, and L is the difference between the D90 values in the current particle size information and the historical particle size information.
10. A skin lotion obtained by the method for preparing a skin lotion based on intelligent optimization according to any one of claims 1-9.