A method and system for mixed control of surfactants

Abstract

The application discloses a kind of mixing control method and system of surfactant, it is related to active agent mixing control technical field.The method comprises the following steps: initialization is carried out to the kind data set and performance data set of surfactant, obtains surfactant solution space;Multiple kind data are obtained by traversing surfactant solution space;A roulette model is constructed, multiple kind data are input into the roulette model, obtain first mixing mode;Mixing is carried out based on first mixing mode, generate first mixed surfactant, carry out mixing evaluation, generate first mixing score;First mixing mode is optimized, obtain second mixing mode, carry out the mixing control of surfactant.The technical problem that the single mixing mode in the prior art is difficult to adapt to different application scenarios, resulting in unsatisfactory mixing effect, achieves the technical effect that mixing mode is intelligently adjusted according to target scene demand, improves the mixing performance of surfactant.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of active agent mixing control, and in particular to a surfactant mixing control method and system. BACKGROUND

[0002] In modern chemical industry and daily life, surfactants play a crucial role in detergent, cosmetics, medicine, paint, oil field chemistry and other fields by reducing liquid surface tension, changing interface properties and other ways. However, with the progress of science and technology and the increasing demand of consumers for product performance, a single surfactant often cannot meet the complex and changing performance requirements. In order to meet the performance requirements of surfactants in different application scenarios, different types of surfactants are mixed to achieve performance optimization. Mixing surfactants not only combines the advantages of different types of surfactants, but also produces synergistic effects through intermolecular interactions, thereby improving overall performance. In the mixing process of surfactants, different types and performance of surfactants will interact when mixed, affecting the performance of the final mixed product. Traditional surfactant mixing control methods are often based on fixed formulations and mixing ratios, which are difficult to adapt to the diversified needs of different application scenarios. SUMMARY

[0003] The surfactant mixing control method and system provided by the embodiments of the present application solve the technical problems that the mixing mode is single in the prior art, it is difficult to adapt to the needs of different application scenarios, and the mixing effect is not ideal.

[0004] In view of the above problems, the embodiments of the present application provide a surfactant mixing control method and system.

[0005] The first aspect of the embodiments of the present application provides a surfactant mixing control method, which comprises:

[0006] The category data set and the performance data set of the surfactant are initialized to obtain a surfactant solution space; the target scene demand is obtained based on the application scene, the plurality of category data of the surfactant is obtained by traversing the surfactant solution space; a roulette model is constructed, the plurality of category data is input into the roulette model, and a first mixing mode is obtained; the first mixed surfactant is generated based on the first mixing mode; the first mixed surfactant is mixed and evaluated to generate a first mixing score; the first mixing mode is optimized based on the first mixing score to obtain a second mixing mode, and the surfactant is mixed and controlled according to the second mixing mode.

[0007] The second aspect of the embodiments of the present application provides a surfactant mixing control system, which comprises:

[0008] The solution space acquisition module is configured to extract a kind data set of surfactants and a performance data set for initialization, and obtain a surfactant solution space; the data acquisition module is configured to obtain target scene requirements based on an application scene, and traverse the surfactant solution space to obtain multiple kind data of surfactants; the model construction module is configured to construct a roulette model, input the multiple kind data into the roulette model, and obtain a first mixing mode; the mixed evaluation module is configured to mix based on the first mixing mode, generate a first mixed surfactant, perform mixed evaluation on the first mixed surfactant, and generate a first mixed score; and the optimization module is configured to optimize the first mixing mode based on the first mixed score, obtain a second mixing mode, and perform mixing control of surfactants according to the second mixing mode.

[0009] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0010] First, a kind data set of surfactants and a performance data set are extracted for initialization to obtain a surfactant solution space. Then, target scene requirements are obtained based on an application scene, and multiple kind data of surfactants are obtained by traversing the surfactant solution space. A roulette model is constructed, the multiple kind data are input into the roulette model, and a first mixing mode is obtained. Next, mixing is performed based on the first mixing mode, a first mixed surfactant is generated, mixed evaluation is performed on the first mixed surfactant, and a first mixed score is generated. Finally, the first mixing mode is optimized based on the first mixed score, a second mixing mode is obtained, and mixing control of surfactants is performed according to the second mixing mode. The technical problem of single mixing mode in the prior art, which is difficult to adapt to different application scene requirements, resulting in unsatisfactory mixing effect, is solved, and the technical effect of intelligently adjusting the mixing mode according to target scene requirements and improving the mixing performance of surfactants is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0012] Figure 1 A flowchart of a surfactant mixing control method provided by the embodiments of the present application is shown in the figure.

[0013] Figure 2A surface active agent mixing control system structure schematic diagram provided by an embodiment of the application.

[0014] Label explanation: solution space acquisition module 11, data acquisition module 12, model construction module 13, mixing evaluation module 14, optimization module 15. DETAILED DESCRIPTION

[0015] The embodiment of the application provides a surface active agent mixing control method and system, which solves the technical problems of single mixing mode, difficulty in adapting to different application scene requirements, and resulting in unsatisfactory mixing effect in the prior art.

[0016] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0017] It should be noted that the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units does not have to be limited to the clearly listed steps or units, but can include other steps or modules that are not clearly listed or inherent to the process, method, product or device.

[0018] Embodiment one

[0019] As shown in Figure 1 The embodiment of the application provides a surface active agent mixing control method, and the method comprises the following steps:

[0020] The kind data set and the performance data set of the surface active agent are extracted and initialized to obtain a surface active agent solution space.

[0021] In the research and application process of the surface active agent, in order to optimize the performance of the surface active agent and meet the requirements of a specific application scene, detailed analysis of the kind and performance of the surface active agent is required. Specifically, by extracting the kind data set and the performance data set of the surface active agent and performing initialization processing, a surface active agent solution space is constructed. The kind data set contains different chemical kind surface active agents; the performance data set contains the performance of different kind surface active agents, such as cleaning power, foam stability, emulsifying ability, biodegradability and the like; and the surface active agent solution space contains all surface active agent kinds and corresponding performance data.

[0022] Based on the application scenario, the target scene requirement is obtained, and the surfactant solution space is traversed to obtain multiple kinds of surfactant data.

[0023] Different application scenarios correspond to different requirements. For example, in detergent applications, high cleaning power, low foam, and environmental biodegradability may be required; while in cosmetic applications, more attention may be paid to mildness, stability, and compatibility with the skin. The target scene requirement usually requires one or more surfactants to meet the requirements. According to the target scene requirement, multiple kinds of surfactant data are obtained by traversing the surfactant solution space. In the process of traversing the solution space, surfactant kind data that meets the requirements is screened out according to the target scene requirement. The surfactant kind data includes one or more surfactants, each of which has its unique performance characteristics and advantages.

[0024] Further, based on the application scenario, the target scene requirement is obtained, and multiple kinds of surfactant data are obtained by traversing the surfactant solution space. The method comprises:

[0025] Based on the target scene requirement, the performance data interval is set; the surfactant solution space is traversed to perform random iteration extraction, a data list is constructed based on the extracted data, the data list contains a first kind of surfactant, a second kind of surfactant,..., and an n-th kind of surfactant; it is judged whether the performance data of the first kind of surfactant meets the performance data interval; if the performance data of the first kind of surfactant meets the performance data interval, the first kind of surfactant is identified and added to the candidate data group, and iteration is performed to obtain the candidate data group, the candidate data group is integrated to obtain the multiple kinds of surfactant data.

[0026] To obtain the combination of surfactants according to the requirements of the target scene, the performance data interval needs to be set first, and random iterative extraction is performed in the surfactant solution space to construct a list of surfactant species data that meet the performance requirements. Specifically, according to the requirements of the target scene, the performance data interval of each key performance indicator (such as cleaning power, foam stability, biodegradability, etc.) is set, and the performance data interval reflects the specific requirements of the target scene for surfactant performance; random iterative extraction is performed in the surfactant solution space until all solutions in the solution space are extracted; for each iteration of the extracted data, a data list is constructed, which contains the first kind of surfactant, the second kind of surfactant,..., and the nth kind of surfactant, wherein each kind of surfactant corresponds to at least one performance data; for each kind of surfactant in the data list (e.g., the first kind of surfactant), check whether its performance data meets the previously set performance data interval; if the performance data of a certain kind of surfactant meets the performance data interval, mark it and add it to a candidate data group, which will contain all surfactant species that meet the performance requirements; repeat the above process from the first kind of surfactant to the nth kind of surfactant; integrate the data in the candidate data group to obtain a list containing multiple surfactant species data that meet the requirements of the target scene.

[0027] A roulette model is constructed, and the multiple species data are input into the roulette model to obtain a first mixed mode.

[0028] In the mixing control of surfactants, the roulette model is a commonly used selection mechanism for randomly but non-uniformly selecting the most likely combination from a group of candidate solutions (multiple kinds of surfactants). Through the roulette model, the most likely surfactant combination that meets the requirements of the target scene can be selected from the multiple species data in a random but non-uniform manner, thereby constructing the first mixed mode. The first mixed mode is also the mixing scheme, including the initial mixing ratio and mixing method.

[0029] Further, the method of constructing the roulette model includes:

[0030] The history mixing information base is called, and the plurality of surfactant category data is traversed to match, obtain a mixing ratio data set and a mixing concentration data set; the mixing ratio data set and the mixing concentration data set are used to define a mixing ratio optimization variable and a mixing concentration optimization variable; weight training is performed according to the plurality of category data and the mixing ratio optimization variable to obtain a weight coefficient of the mixing ratio; the weight coefficient of the mixing ratio, the mixing ratio optimization variable and the mixing concentration optimization variable are combined with the plurality of category data to construct an optimization objective function; the optimization objective function is used to perform mixing performance calculation to obtain a target performance index; the mixing ratio optimization variable, the mixing concentration optimization variable and the target performance index are combined to construct a fitness function; a plurality of area sizes on a roulette wheel are calculated according to the target performance index, and the plurality of area sizes have a corresponding relationship with the plurality of category data; the roulette wheel gambling model is constructed according to the plurality of area sizes and the fitness function.

[0031] In surfactant mixing control, to obtain optimized mixing formulations, a historical mixing information database can be utilized, combined with matching and training of data from multiple categories, to construct an optimization objective function and a roulette wheel model. Specifically, surfactant mixing records similar to the current application scenario are extracted from the historical mixing information database, including mixing ratio datasets and mixing concentration datasets. Data from multiple surfactant categories is traversed and matched with data in the historical mixing information database. Through matching, historical mixing ratio and mixing concentration data related to the current category data can be obtained. Based on the mixing ratio and mixing concentration datasets, mixing ratio optimization variables and mixing concentration optimization variables are defined. These variables will be used in subsequent optimization processes to adjust the mixing ratio and concentration of surfactants. A machine learning algorithm is used to train the model with weights according to multiple categories of data and the mixing ratio optimization variables. Through training, weight coefficients for the mixing ratio can be obtained, reflecting the importance of different types of surfactants in the mixing. Based on the weight coefficients of the mixing ratio, the mixing ratio optimization variables, and the mixing concentration optimization variables, combined with data from multiple categories, an optimization objective function is constructed. Based on the optimization objective function, mixing performance is calculated to find the optimal mixing ratio and concentration, i.e., the target performance index. Based on the optimization variables of mixing ratio and mixing concentration, and the target performance index, a fitness function is constructed to evaluate the merits of different mixing formulations. According to the target performance index, the areas of multiple regions on the roulette wheel are calculated; that is, based on the mixing ratio of each surfactant, the area occupied by each surfactant on the roulette wheel is calculated. These multiple region areas correspond to multiple data types, and the size of the area reflects the importance of the corresponding data type in the mixture. Combining the multiple region areas and the fitness function, a roulette wheel betting model is constructed. In the roulette wheel betting model, each region represents a data type, and the area of ​​the region is proportional to its importance in the mixture. By simulating the roulette wheel rotation process, a mixing formulation can be randomly but non-uniformly selected from multiple data types.

[0032] Furthermore, the optimization objective function is:

[0033] ;in, For target performance indicators, Data for multiple types of surfactants, For the first The weighting coefficients for the mixing ratio of the surfactants. For the first The mixing ratio of the surfactants, For the first The mixed concentration of the surfactants.

[0034] A set of optimal mixing ratios and concentrations can be found to maximize the target performance indicator using an optimization objective function. In the optimization objective function, is a target performance indicator, is a plurality of species data of surfactants, is a weight coefficient of a mixing ratio of a first surfactant, is a mixing ratio of a first surfactant, is a mixing concentration of a first surfactant.

[0035] Further, the plurality of species data is input into the roulette model to obtain a first mixing mode, and the method comprises:

[0036] The fitness of the plurality of species data is calculated by the fitness function to generate a plurality of selected probabilities, and the plurality of selected probabilities and the plurality of species data have a one-to-one correspondence relationship; the plurality of selected probabilities are accumulated to obtain a cumulative probability.

[0037] The roulette selection step based on the cumulative probability is:

[0038] A1: generating a pseudo-random number randomly based on a preset interval; A2: comparing the pseudo-random number with the cumulative probability, selecting the species data containing the pseudo-random number in the cumulative probability, repeating A1 and A2 until the number of species data meets a preset threshold, obtaining a species data set; based on the mixing ratio optimization variable, the mixing concentration optimization variable, and the species data set, the first mixing mode is obtained.

[0039] In the surfactant mixing control, a first mixing mode is obtained using a roulette model. Specifically, the fitness of each category data is calculated by a fitness function, and the higher the fitness value, the higher the probability of being selected. The fitness values of all category data are normalized to generate a plurality of selected probabilities, each of which has a one-to-one correspondence with a category data. Starting from the first category data, its selected probability is added to the selected probabilities of all previous category data to obtain the cumulative probability of the category data. The cumulative probability of the first category data is its selection probability itself, and the cumulative probability of the subsequent category data is the sum of its selection probability and the cumulative probabilities of all previous category data. A random number generation interval (such as between 0 and 1) is preset, and then a pseudo-random number is randomly generated. The generated pseudo-random number is compared with the cumulative probability to find the category data whose cumulative probability value is just greater than or equal to the pseudo-random number, and the category data is selected. Repeat steps A1 and A2 until the number of selected category data meets the preset threshold, and the selected category data is combined to form a category data set. According to the mixing ratio optimization variable and the mixing concentration optimization variable, the proportion and concentration of each surfactant in the mixture are determined based on the category data set, and a first mixing mode is constructed.

[0040] Based on the first mixing mode, a first mixed surfactant is generated, and a first mixing score is generated by evaluating the first mixed surfactant.

[0041] According to the proportion and concentration of each surfactant determined in the first mixing mode, a first mixed surfactant is generated. The first mixed surfactant is evaluated for the effect of mixing and the stability of the mixed surfactant. According to the evaluation results, a first mixing score is generated, which will be used as a basis for evaluating the performance of the first mixing mode.

[0042] Further, based on the first mixing mode, a first mixed surfactant is generated, and a first mixing score is generated by evaluating the first mixed surfactant. The method comprises:

[0043] The first mixed surfactant is generated by executing the first mixing mode through a mixing device; an evaluation test scheme is designed, and the performance of the first mixed surfactant is tested according to the evaluation test scheme to obtain first evaluation test results; the first evaluation test results are analyzed for score conversion to generate a plurality of score values, and the first mixing score is calculated based on the plurality of score values.

[0044] The first mixed mode input is mixed by the mixing device according to the first mixed mode to generate a first mixed surfactant. The performance of the first mixed surfactant is tested by designing an evaluation test scheme, and the first evaluation test result is obtained. Specifically, the test target is determined, such as evaluating the cleaning effect, foam stability, biodegradability, etc. of the first mixed surfactant; according to the test target, appropriate performance indicators are selected, such as detergency, foam generation and durability, stability, safety, etc.; test conditions close to actual application scenarios are designed, such as water temperature, water quality, washing time, and washing species; experimental operations are performed according to the test process, and the data of each performance indicator is recorded to obtain the first evaluation test result. According to the importance and weight of each performance indicator, a score weight is assigned to each performance indicator, the experimental result of each performance indicator is converted into a corresponding score value, and finally the score value is multiplied by the score weight. All multiplication results are added as the total score of the first mixed surfactant, i.e. the first mixed score.

[0045] Based on the first mixed score, the first mixed mode is optimized to obtain a second mixed mode, and the mixing control of the surfactant is performed according to the second mixed mode.

[0046] According to the first mixed score, the first mixed mode is optimized to obtain a more optimal second mixed mode, and the mixing control of the surfactant is performed according to the second mixed mode. Specifically, according to the analysis result of the first mixed score, the parameters in the first mixed mode are adjusted, such as component allocation ratio, mixing time, mixing temperature, etc. The performance of each second mixed mode candidate is tested according to the previously designed evaluation test scheme, and the data is recorded. According to the test data, the score of each second mixed mode candidate is calculated to obtain the corresponding second mixed score, the scores of each second mixed mode are compared, and the mode with the highest score is selected as the optimal second mixed mode. The mixing device will mix the surfactant according to the second mixed mode.

[0047] Further, based on the first mixed score, the first mixed mode is optimized to obtain a second mixed mode, and the mixing control of the surfactant is performed according to the second mixed mode. The method comprises:

[0048] The first mixed score is evaluated, and the optimization target data and the constraint condition are defined according to the evaluation result;

[0049] Based on the constraint condition, the optimization target data is taken as the optimization target, and the first mixed mode is iteratively optimized and searched through the optimization search channel to determine a plurality of mixed modes;

[0050] The plurality of mixed modes are cross-validated, the mixed modes that fail the validation are identified as variation data and added to the optimization search channel for correction;

[0051] The mixed mode that passes the verification is mixedly evaluated to generate a mixed score set;

[0052] The mixed score set is arranged in descending order, and the mixed score of the first order is compared with the first mixed score. When the mixed score of the first order is greater than the first mixed mode, it is considered that the stability of the mixed mode corresponding to the mixed score of the first order is higher than the stability of the first mixed mode, and the mixed mode corresponding to the mixed score of the first order is output as the second mixed mode.

[0053] By evaluating the first mixed score, the performance indicators that should be improved, i.e., the optimization target data, can be determined, and the constraint conditions can be determined according to professional knowledge and expert experience. A suitable optimization algorithm, such as a genetic algorithm, a particle swarm optimization algorithm, a simulated annealing algorithm, etc., is selected as an optimization search channel, and the search range and parameters of the algorithm are set according to the constraint conditions. The optimization target data is used as an optimization target, and the first mixed mode is iteratively optimized and searched through the optimization search channel. In each iteration, a plurality of new mixed mode candidate schemes are generated. The plurality of generated mixed mode candidate schemes are cross-validated to ensure that they can all perform well under different conditions. The mixed mode that fails the verification is identified as variation data and is added to the optimization search channel so as to correct these variations in subsequent iterations. The mixed mode that passes the cross-validation is mixedly evaluated, and performance tests are performed according to the evaluation test scheme designed previously. According to the test results, the mixed score of each mixed mode is generated, and a mixed score set is formed. The mixed score set is arranged in descending order, and the mixed score of the first order is compared with the first mixed score. If the mixed score of the first order is greater than the first mixed score, it is considered that the stability of the mixed mode is higher than that of the first mixed mode, and the mixed mode corresponding to the mixed score of the first order is output as the second mixed mode. In other words, when the mixed score of the first order is less than or equal to the first mixed mode, it is possible that the first mixed mode found for the first time is already the global optimum, or it is possible that the global optimum is not traversed because of other abnormalities, so secondary calculation is needed. If the score value is still higher than the first mixed mode, the first mixed mode is used as the optimum for mixed control.

[0054] In summary, the embodiments of the present application have at least the following technical effects:

[0055] Firstly, a surfactant category dataset and a performance dataset are extracted for initialization to obtain a surfactant solution space. Then, a target scene requirement is obtained based on an application scenario, and multiple category data of surfactants are obtained by traversing the surfactant solution space. A roulette model is constructed, the multiple category data are input into the roulette model, and a first mixing mode is obtained. Then, the first mixing mode is mixed to generate a first mixed surfactant, the first mixed surfactant is mixed and evaluated to generate a first mixed score. Finally, the first mixed mode is optimized based on the first mixed score to obtain a second mixed mode, and the mixing control of the surfactant is performed according to the second mixed mode. The technical problems that the mixing mode is single and difficult to adapt to different application scene requirements in the prior art, resulting in unsatisfactory mixing effect are solved, and the technical effects of intelligently adjusting the mixing mode according to the target scene requirement and improving the mixing performance of the surfactant are achieved.

[0056] Embodiment two

[0057] Based on the same inventive concept as the mixing control method of one surfactant in the foregoing embodiments, as shown in Figure 2 The application provides a surfactant mixing control system, and the system and method embodiments in the application are based on the same inventive concept. The system comprises:

[0058] A solution space acquisition module 11 is configured to extract a surfactant category dataset and a performance dataset for initialization to obtain a surfactant solution space. A data acquisition module 12 is configured to obtain a target scene requirement based on an application scenario, and obtain multiple category data of surfactants by traversing the surfactant solution space. A model construction module 13 is configured to construct a roulette model, input the multiple category data into the roulette model, and obtain a first mixing mode. A mixing evaluation module 14 is configured to mix based on the first mixing mode to generate a first mixed surfactant, and perform mixing evaluation on the first mixed surfactant to generate a first mixed score. An optimization module 15 is configured to optimize the first mixed mode based on the first mixed score to obtain a second mixed mode, and perform mixing control of the surfactant according to the second mixed mode.

[0059] Further, the data acquisition module 12 is configured to perform the following method:

[0060] Setting performance data interval based on the target scene requirement; traversing the surfactant solution space for random iteration extraction, constructing a data list based on the extracted data, the data list containing the first kind of surfactant, the second kind of surfactant... the nth kind of surfactant; judging whether the performance data of the first kind of surfactant meets the performance data interval; if the performance data of the first kind of surfactant meets the performance data interval, identifying the first kind of surfactant and adding it to the candidate data group, thereby iterating to obtain the candidate data group, and integrating the data of the candidate data group to obtain the multiple kinds of data of the surfactant.

[0061] Further, the model construction module 13 is configured to perform the following method:

[0062] Accessing the historical mixing information library, traversing the multiple kinds of data of the surfactant for matching to obtain a mixing ratio data set and a mixing concentration data set; defining a mixing ratio optimization variable and a mixing concentration optimization variable based on the mixing ratio data set and the mixing concentration data set; performing weight training according to the multiple kinds of data and the mixing ratio optimization variable to obtain a weight coefficient of the mixing ratio; constructing an optimization objective function according to the weight coefficient of the mixing ratio, the mixing ratio optimization variable, the mixing concentration optimization variable, and the multiple kinds of data; performing mixing performance calculation based on the optimization objective function to obtain a target performance index; constructing a fitness function based on the mixing ratio optimization variable, the mixing concentration optimization variable, and the target performance index; calculating multiple area sizes on a roulette according to the target performance index, the multiple area sizes having a corresponding relationship with the multiple kinds of data; and constructing the roulette betting model according to the multiple area sizes and the fitness function.

[0063] Further, the model construction module 13 is configured to perform the following method:

[0064] ; wherein, is a target performance index, is multiple kinds of data of a surfactant, is a weight coefficient of a mixing ratio of an nth kind of surfactant, is a mixing ratio of an nth kind of surfactant, is a mixing concentration of an nth kind of surfactant. Further, the model construction module 13 is configured to perform the following method:

[0065] Further, the model construction module 13 is configured to perform the following method:

[0066] ​​The fitness function is used to perform fitness calculation on the plurality of category data to generate a plurality of selected probabilities, which have a one-to-one correspondence with the plurality of category data; and the plurality of selected probabilities are accumulated to obtain a cumulative probability;

[0067] The roulette selection based on the cumulative probability comprises:

[0068] A1: generating a pseudo-random number based on a preset interval; A2: comparing the pseudo-random number with the cumulative probability, and selecting category data containing the pseudo-random number in the cumulative probability; repeating A1 and A2 until the number of category data meets a preset threshold to obtain a category data set; and obtaining the first mixing mode based on the mixing ratio optimization variable, the mixing concentration optimization variable, and the category data set.

[0069] Further, the mixing evaluation module 14 is configured to perform the following method:

[0070] The first mixed surfactant is generated by executing the first mixing mode by a mixing device; an evaluation test scheme is designed, and the performance of the first mixed surfactant is tested according to the evaluation test scheme to obtain a first evaluation test result; the first evaluation test result is analyzed for score conversion to generate a plurality of score values, and the first mixed score is calculated based on the plurality of score values.

[0071] Further, the optimization module 15 is configured to perform the following method:

[0072] The first mixed score is evaluated, and an optimization target data and a constraint condition are defined according to the evaluation result; the optimization target data is taken as an optimization target based on the constraint condition, the first mixing mode is iteratively optimized and searched by an optimization search channel to determine a plurality of mixing modes; the plurality of mixing modes are cross-validated, the mixing modes that fail the validation are identified as mutation data and added to the optimization search channel for correction; the mixing modes that pass the validation are mixed and evaluated to generate a mixed score set; the mixed score set is arranged in descending order, and the first mixed score is compared with a first-order mixed score; when the first-order mixed score is greater than the first mixed mode, it is considered that the stability of the first-order mixed score corresponding to the mixing mode is higher than that of the first mixed mode, and the first-order mixed score corresponding to the mixing mode is output as the second mixing mode.

[0073] It should be noted that the above-mentioned embodiment sequences of the present application are merely for description only, but not for representing the advantages and disadvantages of the embodiments. And the above-mentioned embodiments of the present specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0074] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0075] The specification and drawings are merely exemplary of the present application, and any and all modifications, variations, combinations or equivalents that are within the scope of the present application should be included. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method of mixed control of surfactants, characterized by, The method comprises: initializing the surfactant category data set and the performance data set to obtain a surfactant solution space; obtaining target scene requirements based on an application scenario, and traversing the surfactant solution space to obtain a plurality of surfactant category data; constructing a roulette model, inputting the plurality of category data into the roulette model, and obtaining a first mixing mode; mixing based on the first mixing mode to generate a first mixed surfactant, performing mixing evaluation on the first mixed surfactant, and generating a first mixing score; optimizing the first mixing mode based on the first mixing score to obtain a second mixing mode, and performing surfactant mixing control according to the second mixing mode; calling a historical mixing information library, matching the plurality of surfactant category data, and obtaining a mixing ratio data set and a mixing concentration data set; defining a mixing ratio optimization variable and a mixing concentration optimization variable based on the mixing ratio data set and the mixing concentration data set; weight training according to the plurality of category data and the mixing ratio optimization variable to obtain a weight coefficient of the mixing ratio; constructing an optimization objective function according to the weight coefficient of the mixing ratio, the mixing ratio optimization variable, and the mixing concentration optimization variable, in combination with the plurality of category data; performing mixing performance calculation based on the optimization objective function to obtain a target performance indicator; constructing a fitness function based on the mixing ratio optimization variable, the mixing concentration optimization variable, and in combination with the target performance indicator; calculating a plurality of area sizes on a roulette according to the target performance indicator, wherein the plurality of area sizes have a corresponding relationship with the plurality of category data; constructing the roulette model according to the plurality of area sizes in combination with the fitness function.

2. The method of claim 1, wherein, Based on the target scene requirements, the plurality of surfactant category data in the surfactant solution space is obtained, and the method comprises: setting a performance data interval based on the target scene requirements; traversing the surfactant solution space to perform random iteration extraction, constructing a data list based on the extracted data, and the data list contains a first category surfactant to an nth category surfactant; determining whether the performance data of the first category surfactant meets the performance data interval; if the performance data of the first category surfactant meets the performance data interval, identifying the first category surfactant and adding it to a candidate data group, thereby iterating to obtain the candidate data group, integrating the data of the candidate data group, and obtaining the plurality of surfactant category data.

3. The method of claim 1, wherein, The optimization objective function is: ； wherein, is a target performance index, is a plurality of kinds data of surfactants, is a weight coefficient of a mixing ratio of the th surfactant, is a mixing ratio of the th surfactant, is a mixing concentration of the th surfactant.

4. The method of claim 1, wherein, inputting the plurality of category data into the roulette model to obtain a first mixing mode, the method comprising: performing fitness calculation on the plurality of category data through the fitness function to generate a plurality of selected probabilities, wherein the plurality of selected probabilities have a one-to-one corresponding relationship with the plurality of category data; accumulating the plurality of selected probabilities to obtain a cumulative probability; the steps of roulette selection based on the cumulative probability are: A1: generating a pseudo-random number based on a preset interval; A2: comparing the pseudo-random number with the cumulative probability, selecting category data containing the pseudo-random number in the cumulative probability, repeating A1 and A2 until the number of category data meets a preset threshold, and obtaining a category data set; obtaining the first mixing mode based on the mixing ratio optimization variable, the mixing concentration optimization variable, and the category data set.

5. The method of claim 1, wherein, Based on the first mixing mode, a first mixed surfactant is generated, and a first mixing score is generated by mixing evaluation of the first mixed surfactant. The method comprises: generating the first mixed surfactant by executing the first mixing mode through a mixing device; designing an evaluation test scheme, testing the performance of the first mixed surfactant according to the evaluation test scheme, and obtaining first evaluation test results; analyzing the first evaluation test results to generate a plurality of score values, and calculating the first mixing score according to the plurality of score values.

6. The method of claim 1, wherein, Based on the first mixing score, a second mixing mode is obtained by optimizing the first mixing mode. The method comprises: evaluating the first mixing score, and defining optimization target data and constraint conditions according to the evaluation results; based on the constraint conditions, taking the optimization target data as an optimization target, and iteratively optimizing and searching the first mixing mode through an optimization search channel to determine a plurality of mixing modes; cross-validation of the plurality of mixing modes, identification of the mixing mode that fails to pass the verification as variation data, and addition to the optimization search channel for correction; mixing evaluation of the mixing mode that passes the verification to generate a mixing score set; arranging the mixing score set in descending order, comparing the first mixing score with the first mixing score corresponding to the first sequence, and when the first mixing score is greater than the first mixing score, regarding the stability of the mixing mode corresponding to the first mixing score as higher than the stability of the first mixing mode, and outputting the mixing mode corresponding to the first mixing score as the second mixing mode.

7. A mixed control system of surfactants, characterized by, A method for implementing the mixing control method of any one of claims 1-6, the system comprising: a solution space acquisition module, configured to extract a category data set and a performance data set of surfactants for initialization, and obtain a surfactant solution space; a data acquisition module, configured to obtain target scene requirements based on an application scene, and traverse the surfactant solution space to obtain a plurality of category data of surfactants; a model construction module, configured to construct a roulette model, input the plurality of category data into the roulette model, and obtain a first mixing mode; a mixing evaluation module, configured to generate a first mixed surfactant based on the first mixing mode, and generate a first mixing score by mixing evaluation of the first mixed surfactant. An optimization module is configured to optimize the first mixing mode based on the first mixing score, and obtain a second mixing mode, and the surfactant mixing control is performed according to the second mixing mode.

Images