Batching optimization method and device for antibacterial material production

Abstract

The invention discloses a batching optimization method and device for antibacterial material production, and relates to the technical field of antibacterial materials.The method comprises the steps that the powder particle size space of sepiolite powder for antibacterial material production and the solution concentration space of an antibacterial metal ion solution are obtained; randomly generating a first powder particle size and a first solution concentration, and carrying out specific surface area analysis and agglomeration probability analysis to obtain a first specific surface area and a first agglomeration probability; performing toxicity probability analysis and metal particle concentration analysis according to the first solution concentration to obtain a first toxicity probability and a first metal particle concentration; performing antibacterial performance analysis to obtain first antibacterial performance information, calculating to obtain first antibacterial fitness of the first ingredient parameters, and continuing optimization to obtain an antibacterial material ingredient optimization result. The technical problems that in the prior art, ingredients of the antibacterial material are generally set through production experience, ingredient adjustment and optimization are not accurate, and the performance of the antibacterial material is affected are solved.

Description

Technical Field

[0001] The invention relates to the technical field of antibacterial materials, and in particular to a method and a device for optimizing ingredients for producing antibacterial materials. Background Art

[0002] Existing antibacterial materials are widely used in the fields of medical treatment, food packaging, etc., and the antibacterial effect is achieved by loading metal nanoparticles (such as silver and copper) on porous carrier powders. However, in the process of optimizing the ingredients, there are some key issues that are not fully considered, resulting in a poor balance between antibacterial performance and biosafety. First, the particle size of the carrier powder directly affects the specific surface area. Too large a particle size reduces the metal loading and affects the antibacterial effect; while too small a particle size may cause particle agglomeration, resulting in uneven metal distribution and inhibiting the effective performance of the antibacterial performance. Secondly, if the concentration of the metal ion solution is too low, the effective antibacterial concentration cannot be reached. If it is too high, it may cause biological toxicity, especially in applications that are in direct contact with the human body, the toxicity risk cannot be ignored. The ingredients of antibacterial materials in the prior art are generally set based on production experience, and there are technical problems such as inaccurate ingredient adjustment and optimization and affecting the performance of antibacterial materials. Summary of the invention

[0003] The present application provides a method and device for optimizing ingredients for antibacterial material production, which is used to solve the technical problem that the ingredients of antibacterial materials in the prior art are generally set based on production experience, resulting in inaccurate ingredient adjustment and optimization, which affects the performance of the antibacterial materials.

[0004] In a first aspect of the present application, a method for optimizing ingredients for producing an antibacterial material is provided, the method comprising: obtaining a powder particle size space of sepiolite powder for producing the antibacterial material, and a solution concentration space of an antibacterial metal ion solution, wherein the antibacterial material is prepared and produced by using sepiolite powder and antibacterial metal ion solution as ingredients; Randomly generating a first powder particle size and a first solution concentration in the powder particle size space and the solution concentration space as a first batching parameter, performing a specific surface area analysis according to the first powder particle size to obtain a first specific surface area, and performing an agglomeration probability analysis to obtain a first agglomeration probability, wherein the agglomeration probability analysis result is corrected through an agglomeration probability stability analysis; According to the first solution concentration, a toxicity probability analysis is performed to obtain a first toxicity probability, and a metal particle concentration analysis is performed to obtain a first metal particle concentration; The antibacterial performance is analyzed according to the first specific surface area and the first metal particle concentration to obtain the first antibacterial performance information, and the first antibacterial fitness of the first batching parameters is calculated in combination with the first agglomeration probability and the first toxicity probability, and the batching parameters are further optimized to obtain the optimal batching parameters as the antibacterial material batching optimization result.

[0005] The second aspect of the present application provides a batching optimization device for antibacterial material production, the device comprising: a batching space acquisition module, used to acquire the powder particle size space of sepiolite powder for producing the antibacterial material, and the solution concentration space of the antibacterial metal ion solution, wherein the antibacterial material is prepared and produced by using sepiolite powder and the antibacterial metal ion solution as batching ingredients; a powder particle size analysis module, for randomly generating a first powder particle size and a first solution concentration in the powder particle size space and the solution concentration space as a first batching parameter, performing a specific surface area analysis according to the first powder particle size to obtain a first specific surface area, and performing an agglomeration probability analysis to obtain a first agglomeration probability, wherein the agglomeration probability analysis result is corrected by an agglomeration probability stability analysis; A solution concentration analysis module, configured to perform a toxicity probability analysis based on the first solution concentration to obtain a first toxicity probability, and to perform a metal particle concentration analysis to obtain a first metal particle concentration; The ingredient optimization module is used to analyze the antibacterial performance according to the first specific surface area and the first metal particle concentration to obtain the first antibacterial performance information, calculate the first antibacterial fitness of the first ingredient parameter in combination with the first agglomeration probability and the first toxicity probability, and continue to optimize the ingredient parameters to obtain the optimal ingredient parameters as the antibacterial material ingredient optimization result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application provides a method and device for optimizing ingredients for the production of antibacterial materials, which realizes the optimization and regulation of antibacterial properties by comprehensively considering the particle size of sepiolite powder and the concentration of antibacterial metal ion solution. By obtaining the randomly generated parameters in the powder particle size and solution concentration space, and combining the specific surface area analysis, agglomeration probability analysis, toxicity probability analysis and metal particle concentration analysis, the key performance of the antibacterial material is accurately predicted. Specifically, the antibacterial performance is analyzed in multiple dimensions, and the agglomeration phenomenon and toxicity risk are comprehensively considered to ensure that the material reduces biological toxicity while improving performance. In addition, agglomeration stability analysis is performed, and the agglomeration probability analysis is corrected to improve the rationality and accuracy of agglomeration risk analysis. Through this ingredient optimization method, the balance between powder particle size, metal concentration and antibacterial effect can be achieved, which effectively solves the problem of poor ingredient optimization in the prior art. Finally, the method can output the optimal ingredient parameters, improve the production efficiency and antibacterial effect of the antibacterial material, and ensure its safety and stability in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic flow chart of a method for optimizing ingredients for producing antibacterial materials provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a batching optimization device for antibacterial material production provided in an embodiment of the present application.

[0009] Explanation of the reference numerals: ingredient space acquisition module 11 , powder particle size analysis module 12 , solution concentration analysis module 13 , ingredient optimization module 14 . DETAILED DESCRIPTION

[0010] The present application provides a method and device for optimizing ingredients for antibacterial material production, which is used to solve the technical problem that the ingredients of antibacterial materials in the prior art are generally set based on production experience, resulting in inaccurate ingredient adjustment and optimization, which affects the performance of the antibacterial materials.

[0011] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0013] Embodiment 1, as Figure 1 As shown, the present application provides a method for optimizing ingredients for the production of antibacterial materials, comprising: S10: obtaining a powder particle size space of sepiolite powder used to produce an antibacterial material and a solution concentration space of an antibacterial metal ion solution, wherein the antibacterial material is prepared and produced by using sepiolite powder and antibacterial metal ion solution as ingredients; In the embodiment of the present application, the antibacterial material is an antibacterial material produced by using sepiolite powder as a carrier powder and an antibacterial metal ion solution as a source of antibacterial metal particles, and using the above sepiolite powder and antibacterial metal ion solution as ingredients. Exemplarily, the antibacterial metal ion solution can be silver nitrate (AgNO3) or copper sulfate (CuSO4), which respectively provide silver ions or copper ions for the preparation of silver-based antibacterial materials or copper-based antibacterial materials.

[0014] Obtaining the powder particle size space of sepiolite powder for the production of antibacterial materials, as well as the solution concentration space of antibacterial metal ion solution, is the basic step in the entire ingredient optimization process.

[0015] The powder particle size space refers to the different particle size ranges of sepiolite powder. As a natural porous silicate material, sepiolite has good adsorption properties, and its particle size directly affects its specific surface area and metal ion loading capacity. The solution concentration space refers to the concentration range of the antibacterial metal ion solution. The concentration of antibacterial metal ions (such as silver ions, copper ions, etc.) is a core factor affecting the performance of antibacterial materials. Exemplarily, the powder particle size space is 0.1μm-100μm, and the solution concentration space is 0.001mol / L-1mol / L. The powder particle size space can be constructed based on the minimum particle size and maximum particle size of sepiolite powder in the production process of antibacterial materials as the endpoint values ​​of the powder particle size space, and the solution concentration space can be constructed using the minimum concentration and maximum concentration of the antibacterial metal ion solution as the endpoint values ​​of the solution concentration space.

[0016] By obtaining the above powder particle size space and solution concentration space, a variety of parameter combinations can be provided for subsequent optimization to find the best balance between antibacterial performance, material stability and biosafety. This process lays a solid technical foundation for ingredient optimization and ensures that the selected particle size and concentration range can effectively cover all possibilities in actual production.

[0017] S20: randomly generating a first powder particle size and a first solution concentration in the powder particle size space and the solution concentration space as a first batching parameter, performing a specific surface area analysis according to the first powder particle size to obtain a first specific surface area, and performing an agglomeration probability analysis to obtain a first agglomeration probability, wherein the agglomeration probability analysis result is corrected through an agglomeration probability stability analysis; In the embodiment of the present application, in the powder particle size space and the solution concentration space, the first powder particle size and the first solution concentration are first randomly generated, and these two parameters are used as the initial batching optimization input, i.e., the first batching parameters. The random generation can be produced by an existing random algorithm to obtain a random first powder particle size and a first solution concentration.

[0018] Furthermore, the specific surface area of ​​the sepiolite is analyzed according to the generated first powder particle size to obtain the first specific surface area. For example, the first specific surface area can be obtained by classifying the specific surface area according to the first powder particle size.

[0019] In addition, an agglomeration probability analysis is performed according to the first powder particle size. Agglomeration refers to the powder particles agglomerating together due to interaction forces to form larger particle clusters, which in turn affects the uniformity of the material and the release effect of metal ions. For example, the probability of agglomeration under different powder particle sizes can be analyzed through historical sample data. The occurrence of agglomeration is accidental, so an agglomeration probability stability analysis is also performed to correct the agglomeration probability analysis results to improve the accuracy of the agglomeration probability analysis.

[0020] Through the above steps, the specific surface area and agglomeration probability at the first powder particle size can be analyzed and obtained as basic data for subsequent analysis of the antibacterial material performance. Through agglomeration probability analysis and stability analysis, the accuracy of agglomeration probability analysis can be improved to avoid the impact of accidental agglomeration on analysis accuracy.

[0021] Step S20 in the method provided in the embodiment of the present application includes: According to the production record data of antibacterial materials, a set of sample powder particle sizes of sepiolite powder is collected, and a set of sample specific surface areas under different sample powder particle sizes is obtained; Constructing a mapping relationship between the sample powder particle size set and the sample specific surface area set to obtain a specific surface area classifier; The first powder particle size is input into the specific surface area classifier, and the first specific surface area is obtained by classification.

[0022] In the embodiment of the present application, a sample powder particle size set of sepiolite powder is collected based on the antibacterial material production record data. The sample powder particle size set includes different particle size values ​​of sepiolite powder used in historical production, generally covering the micron and nanometer ranges, such as a variety of particle sizes from 0.1 μm to 100 μm.

[0023] Furthermore, the specific surface area set of samples corresponding to these sample powder particle sizes is obtained. Specific surface area refers to the surface area of ​​a material per unit mass, usually expressed in m² / g. Different sample powder particle sizes correspond to different specific surface areas, and the specific surface area of ​​sepiolite powder with different powder particle sizes can be obtained by testing using methods such as the BET method. Alternatively, the specific surface area can be calculated, for example, specific surface area = 6 / density*particle size, for example, the specific surface area corresponding to a sample powder particle size of 0.1 μm is approximately 10 m² / g.

[0024] In this way, a sample powder particle size set and a sample specific surface area set are obtained. A mapping relationship between the sample powder particle size set and the sample specific surface area set is constructed, such as a mapping table, in which each sample powder particle size in the mapping table is mapped to a corresponding sample specific surface area, so as to obtain a specific surface area classifier.

[0025] Furthermore, the current first powder particle size is input into the specific surface area classifier, and the first specific surface area corresponding to the first powder particle size is classified and output based on the previously constructed mapping relationship. Exemplarily, if the first powder particle size exists in the specific surface area classifier, the sample specific surface area corresponding to the first powder particle size is directly output as the first specific surface area. If it does not exist, the sample powder particle size with the smallest difference from the first powder particle size is calculated and the sample specific surface area corresponding to the sample powder particle size with the smallest difference is output as the first specific surface area.

[0026] By using the historical data of sepiolite powder to construct a specific surface area classifier, the specific surface area corresponding to the powder particle size can be output quickly and accurately, providing basic data support for subsequent antibacterial performance evaluation.

[0027] Step S20 in the method provided in the embodiment of the present application further includes: Collect a set of sample powder particle sizes, and obtain the probability of agglomeration under different sample powder particle sizes to obtain a set of sample agglomeration probabilities; Using the sample powder particle size set and the sample agglomeration probability set, training an agglomeration probability analyzer including a plurality of agglomeration probability analysis branches, wherein the supervised training data of each agglomeration probability analysis branch is different; Input the first powder particle size into the agglomeration probability analyzer, output a plurality of first analysis agglomeration probabilities, and perform fusion processing to obtain a first basic agglomeration probability; The first clustering stability information is calculated according to the multiple first analysis clustering probabilities, and the first basic clustering probability is corrected and calculated according to the ratio of the first clustering stability information to the average clustering stability information to obtain the first clustering probability.

[0028] In the embodiment of the present application, based on the aforementioned steps, the same sample powder particle size set can be collected and obtained. Furthermore, based on the historical production data of the antibacterial material, the probability of agglomeration in the production of antibacterial materials under different sample powder particle sizes is collected. For example, if 10 productions of antibacterial materials are recorded under a sample powder particle size, and agglomeration occurs 4 times, then the sample agglomeration probability of the sample powder particle size is 0.4. In this way, a sample agglomeration probability set is collected and obtained.

[0029] On this basis, the sample powder particle size set and the sample agglomeration probability set are used to train an agglomeration probability analyzer containing multiple agglomeration probability analysis branches. Each agglomeration probability analysis branch of the agglomeration probability analyzer uses different supervised training data, which are derived from different production batches, experimental conditions or material types. Through this diverse data set, multiple branches are trained, and the accuracy of the agglomeration probability analysis is improved by integrating multiple branches. Therefore, the performance of each agglomeration probability analysis branch is different, and different outputs may be generated for the same input. Fusion of the outputs of multiple agglomeration probability analysis branches can improve the accuracy of the overall analysis of the agglomeration probability.

[0030] In the embodiment of the present application, the sample powder particle size set and the sample agglomeration probability set are used to train an agglomeration probability analyzer including multiple agglomeration probability analysis branches, including: Dividing the sample powder particle size set and the sample agglomeration probability set to obtain multiple groups of agglomeration supervision training data; The multiple groups of clustering supervision training data are respectively used to train multiple clustering probability analysis branches, and the clustering probability analyzer is obtained by combining them.

[0031] In an embodiment of the present application, the sample powder particle size set is used as the input data of the agglomeration probability analyzer, and the sample agglomeration probability set is used as the output data of the agglomeration probability analyzer. As the whole supervised training data, they are divided to obtain multiple groups of agglomeration supervised training data.

[0032] Among them, the number of supervised training data in the divided multiple groups of reunion supervised training data is the same, and the data in each group of reunion supervised training data is different. Multiple groups of reunion supervised training data are used to train multiple reunion probability analysis branches. The supervised training data of different reunion probability analysis branches are different, and the trained reunion probability analysis branches are also different. When performing reunion probability analysis, the outputs of multiple reunion probability analysis branches are combined to obtain the final output, which can improve the accuracy of the analysis. In addition, when multiple groups of reunion supervised training data are divided for training, the amount of data in each group of reunion supervised training data is small, and the training convergence speed is fast.

[0033] Exemplarily, based on a feedforward neural network, each agglomeration probability analysis branch is trained, and each agglomeration probability analysis branch includes an input layer, a hidden layer, and an output layer. The input layer inputs powder particle size data, which is a single node, and multiple hidden layers are set. The node parameters are set during training, and the activation function is set to Sigmoid. The output layer is a single node, which represents the predicted agglomeration probability value. The Sigmoid activation function is used to limit the output to between 0 and 1, which meets the definition of the probability value. During the training process, each sample powder particle size data is input into the network, and the predicted agglomeration probability is generated through the input layer, the hidden layer, and the output layer in turn. The predicted agglomeration probability of the output is compared with the actual sample agglomeration probability, and the loss value is calculated. Then, through the back propagation algorithm, the partial derivative of the loss function to each layer weight is calculated, and the network weight is updated. Back propagation starts from the output layer, and error information is transmitted backward layer by layer, and the weight is adjusted to minimize the loss function. Repeat the above forward propagation and back propagation process, and perform multiple iterative training of multiple agglomeration supervision training data until the loss is less than the requirement, for example, less than 5%, and the training is completed.

[0034] Until multiple groups of clustering supervised training data are used, multiple clustering probability analysis branches are obtained through training, and the multiple clustering probability analysis branches are combined to obtain a clustering probability analyzer.

[0035] By training multiple agglomeration probability analysis branches, the agglomeration probability analyzer can effectively analyze the agglomeration probability under powder particle size, and the analysis results of multiple branches can analyze the stability of the agglomeration probability, provide accurate agglomeration probability prediction, and help optimize the selection of powder particle size in the production of antibacterial materials, reducing the negative impact of agglomeration on material properties.

[0036] Based on the trained agglomeration probability analyzer, the current first powder particle size is input into multiple agglomeration probability analysis branches in the agglomeration probability analyzer, and the analysis results of multiple agglomeration probability analysis branches are output, that is, multiple first analysis agglomeration probabilities. The multiple first analysis agglomeration probabilities may be the same or different.

[0037] Furthermore, a plurality of first analysis clustering probabilities are fused, for example, a mean is calculated, to obtain a first basic clustering probability.

[0038] Furthermore, the occurrence of clustering is accidental. For example, clustering may occur due to the influence of ambient temperature and humidity. Therefore, the stability of multiple first analysis clustering probabilities is further analyzed to correct the first basic clustering probability to avoid abnormal data in the sample clustering probability set due to the accidental clustering, which affects the accuracy of the clustering probability analysis.

[0039] Exemplarily, first agglomeration stability information is calculated based on multiple first analysis agglomeration probabilities, for example, the variance of multiple first analysis agglomeration probabilities is calculated as the first agglomeration stability information. The variance can reflect the stability of the agglomeration probability. If the probability of agglomeration occurring under a sample powder particle size is relatively stable, then multiple first analysis agglomeration probabilities are similar, and the variance is small. Otherwise, the variance is large.

[0040] Further, according to the ratio of the first agglomeration stability information and the average agglomeration stability information, the first basic agglomeration probability is corrected and calculated to obtain the final first agglomeration probability, for example, the ratio is multiplied by the first basic agglomeration probability to obtain the final first agglomeration probability. Exemplarily, the average agglomeration stability information can be used as the average agglomeration stability information by calculating the mean of the variance of the agglomeration probability of all sample powder particle sizes produced under different conditions, reflecting the stability of the overall agglomeration probability of different powder particle sizes. The ratio is calculated by comparing the first agglomeration stability information with the average agglomeration stability information of all sample powder particle sizes. If the ratio is high, it means that the agglomeration phenomenon of the first powder particle size shows instability under different environments, then the agglomeration probability that may be generated by the first powder particle size may be greater, and the performance in terms of agglomeration probability is poor, and the first basic agglomeration probability needs to be corrected and amplified. On the contrary, the probability of agglomeration of the first powder particle size is relatively stable, and the performance in terms of agglomeration probability is relatively stable. The first basic agglomeration probability is corrected and reduced to reflect the agglomeration situation of the first powder particle size. Through this correction process, a more reliable first agglomeration probability is obtained.

[0041] Through the above steps, the probability of agglomeration at a specific particle size can be accurately predicted as a performance parameter reflecting the agglomeration of powder particles, providing a reliable basis for the subsequent performance optimization of antibacterial materials.

[0042] S30: performing a toxicity probability analysis according to the first solution concentration to obtain a first toxicity probability, and performing a metal particle concentration analysis to obtain a first metal particle concentration; In an embodiment of the present application, the toxicity of the antibacterial material is evaluated based on the concentration of the first solution generated. This step requires a toxicity probability analysis, that is, predicting the potential toxicity probability of the antibacterial material to an organism or the environment at the first solution concentration. Toxicity usually comes from excessive release of metal ions. For example, silver ions or copper ions may have adverse effects on human cells or the ecological environment at high concentrations. The first toxicity probability can be obtained by experimental testing or analysis.

[0043] And, the metal particle concentration is analyzed according to the concentration of the first solution. The metal particle concentration is an important indicator of the performance of the antibacterial material and directly affects the antibacterial effect. According to the historical production data of the antibacterial material, the metal particle concentration under different solution concentrations under the same weight ratio of sepiolite powder and antibacterial metal ion solution can be obtained by classification.

[0044] A higher metal particle concentration means that the material has stronger antimicrobial ability because it can release more metal ions to inhibit or kill microorganisms. However, too high a metal particle concentration may also increase the toxicity risk of the material. A comprehensive analysis combining the toxicity probability and metal particle concentration is performed to achieve a balance between performance and safety.

[0045] Through toxicity probability analysis and metal particle concentration analysis, the toxicity risk and metal loading capacity of antibacterial materials at different solution concentrations can be evaluated, providing key data support for subsequent material performance optimization and ingredient selection.

[0046] Step S30 in the method provided in the embodiment of the present application includes: Collecting a set of sample solution concentrations, and collecting the probability of toxicity of antibacterial materials at different solution concentrations, to obtain a set of sample toxicity probabilities; Using the sample solution concentration set and the sample toxicity probability set, training a toxicity probability analyzer including a plurality of toxicity probability analysis branches, wherein the supervised training data of each toxicity probability analysis branch is different; Inputting the first solution concentration into the toxicity probability analyzer, outputting a plurality of first analysis toxicity probabilities, and fusing them to obtain a first basic toxicity probability; The first toxicity stability information is calculated according to the multiple first analysis toxicity probabilities, and the first basic toxicity probability is corrected and calculated according to the ratio of the first toxicity stability information to the average toxicity stability information to obtain the first toxicity probability.

[0047] In the embodiment of the present application, according to the different concentrations of the metal ion solution used in the production process of the antibacterial material, a set of sample solution concentrations is collected. The set of sample solution concentrations covers multiple values ​​from low concentration to high concentration. For example, the concentration of the silver ion solution may range from 0.001M to 1M, and the copper ion solution may also have a similar concentration range.

[0048] Furthermore, the probability of toxicity of antibacterial materials at different solution concentrations is collected to form a set of sample toxicity probabilities. For example, the toxicity probability is based on experimentally measured data, reflecting the possibility that the antibacterial material will produce a toxic reaction to an organism or the environment at a certain solution concentration. For example, at lower silver ion concentrations (such as 0.01 M), the toxicity may be low, with a probability of around 0.1; while at high concentrations (such as above 0.5 M), the toxicity probability may increase significantly to 0.8 or even higher. In this way, the test obtains a set of sample toxicity probabilities at different sample solution concentrations.

[0049] Among them, during the production process of antibacterial materials, MTT / CCK-8 cell viability test can be used for sampling experimental testing, and multiple antibacterial materials produced at different solution concentrations can be tested to evaluate cell survival by measuring the metabolic activity of cells. When the cell survival rate decreases, toxicity is considered to exist, and the toxicity probability is calculated based on the proportion of toxic antibacterial materials, for example, 0.1. In this way, the test obtains a set of sample toxicity probabilities.

[0050] Furthermore, a sample solution concentration set and a sample toxicity probability set are used as supervised training data to train a toxicity probability analyzer including multiple toxicity probability analysis branches, wherein the supervised training data for each toxicity probability analysis branch is different. Optionally, the training process is the same as that of the agglomeration probability analyzer in the aforementioned content, but the training data is different. The sample solution concentration set and the sample toxicity probability set are divided to obtain multiple groups of toxicity supervised training data, and each group of toxicity supervised training data is used to train a toxicity probability analysis branch. The training is based on a feedforward neural network, and the training process is the same as the agglomeration probability analysis branch, but the specific training data is different and will not be repeated here. After training to obtain multiple toxicity probability analysis branches, a toxicity probability analyzer is obtained by combining them.

[0051] The concentration of the first solution is input into multiple toxicity probability analysis branches in the toxicity probability analyzer to perform toxicity probability prediction, and multiple first analysis toxicity probabilities are output. The training data of each toxicity probability analysis branch is different, and the output first analysis toxicity probabilities may also be different. Integrating multiple first analysis toxicity probabilities can improve the accuracy of toxicity probability prediction.

[0052] Exemplarily, a plurality of first analysis toxicity probabilities are fused to obtain a first basic toxicity probability, for example, an average of the plurality of first analysis toxicity probabilities is calculated as the first basic toxicity probability.

[0053] Optionally, based on multiple first analysis toxicity probabilities, the variance of multiple first analysis toxicity probabilities is calculated as the first toxicity stability information, and then the mean of the variance of multiple toxicity probabilities for producing antibacterial materials under different environments with all sample solution concentrations is calculated as the average toxicity stability information.

[0054] According to the ratio of the first toxicity stability information and the average toxicity stability information, the first basic toxicity probability is corrected and calculated, for example, multiplied by the first basic toxicity probability to obtain the first toxicity probability. If the ratio is close to 1, it means that the toxicity probability prediction of this concentration is consistent with the overall stability level, and the prediction result is relatively reliable; if the ratio is significantly higher than 1, it means that the prediction at this concentration is not stable enough, the fluctuation is large, and a greater toxicity probability may be generated. The prediction result may need to be corrected to improve the reliability of the final prediction. By correcting the first basic toxicity probability to obtain the first toxicity probability, the performance of the antibacterial material in terms of biological toxicity at the first solution concentration can be reflected. The smaller the first toxicity probability, the better the performance of the antibacterial material at the first solution concentration.

[0055] The probability of antimicrobial materials producing biological toxicity is accidental and may change due to the influence of conditions such as volatilization. Therefore, by analyzing the stability of the toxicity probability and correcting the toxicity probability, the accuracy and reliability of the toxicity probability analysis can be improved, providing an important reference for the subsequent performance evaluation and ingredient optimization of antimicrobial materials.

[0056] Step S30 in the method provided in the embodiment of the present application further includes: Collecting a set of sample solution concentrations of the antibacterial metal ion solution, and collecting a set of sample metal particle concentrations of the antibacterial material at different sample solution concentrations and preset ratios, wherein the preset ratio includes a ratio of sepiolite powder and the antibacterial metal ion solution; Constructing a mapping relationship between the sample solution concentration set and the sample metal particle concentration set to obtain a metal particle concentration classifier; The metal particle concentration classifier is used to classify the first solution concentration to obtain a first metal particle concentration.

[0057] In the embodiment of the present application, a sample solution concentration set of an antibacterial metal ion solution is collected, and the set includes data of antibacterial metal ion solutions (such as silver ion or copper ion solutions) at different concentrations, covering a variety of values ​​from low concentration to high concentration. For example, the solution concentration range may vary from 0.001 M to 1 M.

[0058] Furthermore, according to the preset ratio of sepiolite powder and antibacterial metal ion solution in the production of antibacterial materials, the concentration of metal particles in the antibacterial materials produced by different sample solution concentrations is collected. The preset ratio includes the ratio of sepiolite powder and antibacterial metal ion solution. Exemplarily, the preset ratio is a mass ratio of sepiolite powder to antibacterial metal ion solution of 1:10. In this way, the concentration of metal particles in the antibacterial materials produced by different solution concentrations under the preset ratio is obtained, and a set of sample solution concentrations is obtained. For example, under the preset ratio, when the sample solution concentration is 0.1M, the corresponding sample metal particle concentration is 0.5%.

[0059] Furthermore, a mapping relationship between the sample solution concentration set and the sample metal particle concentration set is constructed, and each sample solution concentration corresponds to a sample metal particle concentration, forming a mapping table of the sample solution concentration set and the sample metal particle concentration set as a metal particle concentration classifier.

[0060] The current first solution concentration is input into the metal particle concentration classifier, and the corresponding sample metal particle concentration is obtained to obtain the first metal particle concentration. If there is no first solution concentration in the metal particle concentration classifier, the metal particle concentration corresponding to the sample solution concentration with the smallest difference from the first solution concentration is obtained as the first metal particle concentration.

[0061] By constructing a metal particle concentration classifier, the metal particle concentration in the antibacterial material under different solution concentrations can be accurately classified. This is crucial for the antibacterial performance of the antibacterial material, thereby improving the performance of the antibacterial material in the optimization of antibacterial material formulation.

[0062] S40: Analyze the antibacterial performance according to the first specific surface area and the first metal particle concentration to obtain first antibacterial performance information, calculate the first antibacterial fitness of the first batching parameters in combination with the first agglomeration probability and the first toxicity probability, and continue to optimize the batching parameters to obtain optimal batching parameters as the antibacterial material batching optimization result.

[0063] In the embodiment of the present application, the antibacterial performance analysis is performed based on the first specific surface area and the first metal particle concentration to obtain the first antibacterial performance information. The specific surface area refers to the total surface area of ​​the sepiolite powder per unit mass. Generally, a larger specific surface area can adsorb more metal particles and improve the metal loading capacity of the antibacterial material. At the same time, the concentration of metal particles determines the amount of antibacterial metal ions (such as silver ions, copper ions, etc.) loaded in the material, which directly affects the antibacterial ability of the material. Exemplarily, the antibacterial performance analysis can be performed based on the first specific surface area and the first metal particle concentration based on machine learning means.

[0064] Furthermore, the first antibacterial performance information obtained by the analysis is combined with the first agglomeration probability and the first toxicity probability to calculate the first antibacterial fitness of the first batching parameter to reflect the quality of the antibacterial material prepared by the first batching parameter. The greater the first antibacterial performance information, the greater the first antibacterial fitness, and the smaller the first agglomeration probability and the first toxicity probability, the first antibacterial fitness is negatively correlated.

[0065] After calculating and obtaining the first antibacterial fitness, continue to optimize the batching parameters in the powder particle size space and the solution concentration space, for example, continue to randomly generate powder particle size and solution concentration, analyze and calculate through the above steps, obtain the antibacterial fitness, until the optimization converges, output the batching parameters with the largest antibacterial fitness, and obtain the optimal batching parameters as the antibacterial material batching optimization result.

[0066] The optimization of the antibacterial material formulation parameters through the above steps ensures that while ensuring the antibacterial effect, the agglomeration and toxicity of the antibacterial material are minimized, the comprehensive performance of the material is improved, and the accuracy and reliability of the antibacterial material formulation optimization are improved.

[0067] Step S40 in the method provided in the embodiment of the present application includes: Collect a set of sample specific surface areas and a set of sample metal particle concentrations, and test to obtain the minimum inhibitory concentration of the antibacterial material under different sample specific surface areas and sample metal particle concentrations, and mark them as a sample antibacterial performance information set; Using the sample specific surface area set, the sample metal particle concentration set and the sample antibacterial performance information set, based on a decision tree, an antibacterial performance classifier is constructed; Using the antibacterial performance classifier, the first specific surface area and the first metal particle concentration are input into the classification to obtain first antibacterial performance information; Calculating and obtaining a first antibacterial fitness according to the first antibacterial performance information, the first agglomeration probability and the first toxicity probability; In the powder particle size space and solution concentration space, the optimization of the batching parameters is continued until convergence, and the optimal batching parameters with the maximum antibacterial fitness are output.

[0068] In the embodiment of the present application, based on the record data of the production of antibacterial materials, the specific surface area of ​​the powder particles and the metal particle concentration obtained in each test of the production of antibacterial materials in the past are collected to obtain a sample specific surface area set and a sample metal particle concentration set. The combination of the specific surface area and the metal particle concentration determines the antibacterial performance of the antibacterial material.

[0069] Furthermore, the minimum inhibitory concentration (MIC) of the antibacterial materials with different sample specific surface areas and sample metal particle concentrations was tested to obtain the minimum inhibitory concentration (MIC) of the antibacterial materials with different sample specific surface areas and sample metal particle concentrations, and the MIC was marked as a sample antibacterial performance information set. The larger the minimum inhibitory concentration, the greater the sample antibacterial performance information.

[0070] Furthermore, the sample specific surface area set and the sample metal particle concentration set are used as decision input data, the sample antibacterial performance information set is used as decision output data, and the decision tree algorithm is used to construct an antibacterial performance classifier. The constructed antibacterial performance classifier can obtain the antibacterial performance information of the antibacterial material according to the input specific surface area and metal particle concentration classification.

[0071] Optionally, the antibacterial property classifier can also be trained based on a feedforward neural network.

[0072] Exemplarily, in the decision tree model, specific surface area and metal particle concentration are used as input features (or variables) to predict the antibacterial performance (output) of the material. Specific surface area is usually expressed as m² / g, while metal particle concentration is usually expressed as a percentage. During the construction process, a feature (such as specific surface area) is selected as the root node, and then another feature (such as metal particle concentration) is selected to construct subnodes, and multi-layer subnodes are constructed. Multi-layer nodes can classify the input specific surface area and metal particle concentration, and finally obtain intervals of different specific surface areas and metal particle concentrations. The specific surface area and metal particle concentration in each interval correspond to a sample of antibacterial performance information. The division process of the decision tree will terminate when certain conditions are met, such as the data cannot be further divided (for example, the minimum division interval of the metal particle concentration reaches 0.1%) or the preset maximum depth has been reached (for example, 50 layers of nodes are constructed, and the last layer is a leaf node), and the construction is completed.

[0073] The constructed antibacterial performance classifier is used to input the current first specific surface area and the first metal particle concentration to perform decision classification of multi-layer nodes, and finally the first antibacterial performance information corresponding to the specific surface area and the metal particle concentration in the last layer of nodes is divided.

[0074] Furthermore, the first antibacterial performance information, the first agglomeration probability and the first toxicity probability reflect the quality of the antibacterial material produced by the first specific surface area and the first metal particle concentration. Combined with the first antibacterial performance information, the first agglomeration probability and the first toxicity probability, the first antibacterial fitness is calculated to reflect the quality of the first ingredient parameters.

[0075] Exemplarily, according to the antimicrobial fitness = weight * antimicrobial performance information + weight * (1 / agglomeration probability) + weight * (1 / toxicity probability), the first antimicrobial performance information, the first agglomeration probability and the first toxicity probability are weighted and calculated, and the inverse of the agglomeration probability and the toxicity probability are weighted and calculated, and the size of the agglomeration probability and the toxicity probability is negatively correlated with the size of the antimicrobial fitness. Exemplarily, the weights of the antimicrobial performance information, the agglomeration probability and the toxicity probability are set according to the importance of the three to the antimicrobial material. For example, if the antimicrobial material needs to be used in a product that contacts the human body, the weight of the toxicity probability is larger, otherwise it is smaller. Exemplarily, the weights of the antimicrobial performance information, the agglomeration probability and the toxicity probability are 0.6, 0.1 and 0.3.

[0076] Furthermore, the optimization of the batching parameters is continued in the powder particle size space and the solution concentration space. For example, an iterative optimization algorithm (such as genetic algorithm, particle swarm optimization, etc.) is used to continuously generate new combinations of powder particle size and solution concentration, and the above steps are used for analysis and calculation to obtain new antibacterial fitness until the optimization converges. For example, when the generated powder particle size and solution concentration combinations reach 100 or 500, the optimization is completed, and the optimal batching parameters with the maximum antibacterial fitness are output as the optimal batching parameters and the optimization results of the antibacterial material batching.

[0077] In summary, the embodiments of the present application have at least the following technical effects: The embodiment of the present application provides a method and device for optimizing ingredients for the production of antibacterial materials, which realizes the optimization and regulation of antibacterial performance by comprehensively considering the particle size of sepiolite powder and the concentration of antibacterial metal ion solution. By obtaining the randomly generated parameters in the powder particle size and solution concentration space, and combining the specific surface area analysis, agglomeration probability analysis, toxicity probability analysis and metal particle concentration analysis, the key performance of the antibacterial material is accurately predicted. Specifically, the antibacterial performance is analyzed in multiple dimensions, and the agglomeration phenomenon and toxicity risk are comprehensively considered to ensure that the material reduces biological toxicity while improving performance. In addition, an agglomeration stability analysis is performed, and the agglomeration probability analysis is corrected to improve the rationality and accuracy of the agglomeration risk analysis. Through this ingredient optimization method, a balance between powder particle size, metal concentration and antibacterial effect can be achieved, which effectively solves the problem of poor ingredient optimization in the prior art. Finally, the method can output the optimal ingredient parameters, improve the production efficiency and antibacterial effect of the antibacterial material, and ensure its safety and stability in practical applications.

[0078] Embodiment 2, based on the same inventive concept as the method for optimizing ingredients for producing antibacterial materials in the above embodiment, Figure 2 As shown, the present application provides a batching optimization device for antibacterial material production, and the device and method embodiments in the present application embodiments are based on the same inventive concept. The device includes: The ingredient space acquisition module 11 is used to acquire the powder particle size space of the sepiolite powder used to produce the antibacterial material, and the solution concentration space of the antibacterial metal ion solution, wherein the antibacterial material is prepared and produced by using the sepiolite powder and the antibacterial metal ion solution as ingredients; A powder particle size analysis module 12 is used to randomly generate a first powder particle size and a first solution concentration in the powder particle size space and the solution concentration space as a first batching parameter, perform a specific surface area analysis according to the first powder particle size to obtain a first specific surface area, and perform an agglomeration probability analysis to obtain a first agglomeration probability, wherein the agglomeration probability analysis result is corrected by an agglomeration probability stability analysis; A solution concentration analysis module 13, configured to perform a toxicity probability analysis according to the first solution concentration to obtain a first toxicity probability, and to perform a metal particle concentration analysis to obtain a first metal particle concentration; The ingredient optimization module 14 is used to perform antibacterial performance analysis based on the first specific surface area and the first metal particle concentration to obtain first antibacterial performance information, calculate the first antibacterial fitness of the first ingredient parameter in combination with the first agglomeration probability and the first toxicity probability, and continue to optimize the ingredient parameters to obtain optimal ingredient parameters as the antibacterial material ingredient optimization result.

[0079] Furthermore, the powder particle size analysis module 12 is also used to perform the following steps: According to the production record data of antibacterial materials, a set of sample powder particle sizes of sepiolite powder is collected, and a set of sample specific surface areas under different sample powder particle sizes is obtained; Constructing a mapping relationship between the sample powder particle size set and the sample specific surface area set to obtain a specific surface area classifier; The first powder particle size is input into the specific surface area classifier, and the first specific surface area is obtained by classification.

[0080] Furthermore, the powder particle size analysis module 12 is also used to perform the following steps: Collect a set of sample powder particle sizes, and obtain the probability of agglomeration under different sample powder particle sizes to obtain a set of sample agglomeration probabilities; Using the sample powder particle size set and the sample agglomeration probability set, training an agglomeration probability analyzer including a plurality of agglomeration probability analysis branches, wherein the supervised training data of each agglomeration probability analysis branch is different; Input the first powder particle size into the agglomeration probability analyzer, output a plurality of first analysis agglomeration probabilities, and perform fusion processing to obtain a first basic agglomeration probability; The first clustering stability information is calculated according to the multiple first analysis clustering probabilities, and the first basic clustering probability is corrected and calculated according to the ratio of the first clustering stability information to the average clustering stability information to obtain the first clustering probability.

[0081] The sample powder particle size set and the sample agglomeration probability set are used to train an agglomeration probability analyzer including multiple agglomeration probability analysis branches, including: Dividing the sample powder particle size set and the sample agglomeration probability set to obtain multiple groups of agglomeration supervision training data; The multiple groups of clustering supervision training data are respectively used to train multiple clustering probability analysis branches, and the clustering probability analyzer is obtained by combining them.

[0082] Furthermore, the solution concentration analysis module 13 is also used to perform the following steps: Collecting a set of sample solution concentrations, and collecting the probability of toxicity of antibacterial materials at different solution concentrations, to obtain a set of sample toxicity probabilities; Using the sample solution concentration set and the sample toxicity probability set, training a toxicity probability analyzer including a plurality of toxicity probability analysis branches, wherein the supervised training data of each toxicity probability analysis branch is different; Inputting the first solution concentration into the toxicity probability analyzer to obtain a plurality of first analysis toxicity probabilities, and fusing the obtained first basic toxicity probability; The first toxicity stability information is calculated according to the multiple first analysis toxicity probabilities, and the first basic toxicity probability is corrected and calculated according to the ratio of the first toxicity stability information to the average toxicity stability information to obtain the first toxicity probability.

[0083] Furthermore, the solution concentration analysis module 13 is also used to perform the following steps: Collecting a set of sample solution concentrations of the antibacterial metal ion solution, and collecting a set of sample metal particle concentrations of the antibacterial material at different sample solution concentrations and preset ratios, wherein the preset ratio includes a ratio of sepiolite powder and the antibacterial metal ion solution; Constructing a mapping relationship between the sample solution concentration set and the sample metal particle concentration set to obtain a metal particle concentration classifier; The metal particle concentration classifier is used to classify the first solution concentration to obtain a first metal particle concentration.

[0084] Furthermore, the ingredient optimization module 14 is also used to perform the following steps: Collect a set of sample specific surface areas and a set of sample metal particle concentrations, and test to obtain the minimum inhibitory concentration of the antibacterial material under different sample specific surface areas and sample metal particle concentrations, and mark them as a sample antibacterial performance information set; Using the sample specific surface area set, the sample metal particle concentration set and the sample antibacterial performance information set, based on a decision tree, an antibacterial performance classifier is constructed; Using the antibacterial performance classifier, the first specific surface area and the first metal particle concentration are input into the classification to obtain first antibacterial performance information; Calculating and obtaining a first antibacterial fitness according to the first antibacterial performance information, the first agglomeration probability and the first toxicity probability; In the powder particle size space and solution concentration space, the optimization of the batching parameters is continued until convergence, and the optimal batching parameters with the maximum antibacterial fitness are output.

[0085] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, 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, multitasking and parallel processing are also possible or may be advantageous.

[0086] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0087] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may 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 fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for optimizing ingredients for the production of antibacterial materials, characterized in that: The method comprises: Obtaining a powder particle size space of sepiolite powder for producing an antibacterial material and a solution concentration space of an antibacterial metal ion solution, wherein the antibacterial material is produced by using sepiolite powder and antibacterial metal ion solution as ingredients; Randomly generating a first powder particle size and a first solution concentration in the powder particle size space and the solution concentration space as a first batching parameter, performing a specific surface area analysis according to the first powder particle size to obtain a first specific surface area, and performing an agglomeration probability analysis to obtain a first agglomeration probability, wherein the agglomeration probability analysis result is corrected through an agglomeration probability stability analysis; According to the first solution concentration, a toxicity probability analysis is performed to obtain a first toxicity probability, and a metal particle concentration analysis is performed to obtain a first metal particle concentration; The antibacterial performance is analyzed according to the first specific surface area and the first metal particle concentration to obtain the first antibacterial performance information, and the first antibacterial fitness of the first batching parameters is calculated in combination with the first agglomeration probability and the first toxicity probability, and the batching parameters are further optimized to obtain the optimal batching parameters as the antibacterial material batching optimization result.

2. The ingredient optimization method for antibacterial material production according to claim 1, characterized in that: Performing a specific surface area analysis according to the first powder particle size to obtain a first specific surface area includes: According to the production record data of antibacterial materials, a set of sample powder particle sizes of sepiolite powder is collected, and a set of sample specific surface areas under different sample powder particle sizes is obtained; Constructing a mapping relationship between the sample powder particle size set and the sample specific surface area set to obtain a specific surface area classifier; The first powder particle size is input into the specific surface area classifier, and the first specific surface area is obtained by classification.

3. The ingredient optimization method for antibacterial material production according to claim 1, characterized in that: Perform a reunion probability analysis to obtain the first reunion probability, including: Collect a set of sample powder particle sizes, and obtain the probability of agglomeration under different sample powder particle sizes to obtain a set of sample agglomeration probabilities; Using the sample powder particle size set and the sample agglomeration probability set, training an agglomeration probability analyzer including a plurality of agglomeration probability analysis branches, wherein the supervised training data of each agglomeration probability analysis branch is different; Input the first powder particle size into the agglomeration probability analyzer, output a plurality of first analysis agglomeration probabilities, and perform fusion processing to obtain a first basic agglomeration probability; The first clustering stability information is calculated according to the multiple first analysis clustering probabilities, and the first basic clustering probability is corrected and calculated according to the ratio of the first clustering stability information to the average clustering stability information to obtain the first clustering probability.

4. The method for optimizing ingredients for producing antibacterial materials according to claim 3, characterized in that: The sample powder particle size set and the sample agglomeration probability set are used to train an agglomeration probability analyzer including multiple agglomeration probability analysis branches, including: Dividing the sample powder particle size set and the sample agglomeration probability set to obtain multiple groups of agglomeration supervision training data; The multiple groups of clustering supervision training data are respectively used to train multiple clustering probability analysis branches, and then a clustering probability analyzer is obtained by combining them.

5. The ingredient optimization method for antibacterial material production according to claim 1, characterized in that: Performing a toxicity probability analysis according to the concentration of the first solution to obtain a first toxicity probability includes: Collecting a set of sample solution concentrations, and collecting the probability of toxicity of antibacterial materials at different solution concentrations, to obtain a set of sample toxicity probabilities; Using the sample solution concentration set and the sample toxicity probability set, training a toxicity probability analyzer including a plurality of toxicity probability analysis branches, wherein the supervised training data of each toxicity probability analysis branch is different; Inputting the first solution concentration into the toxicity probability analyzer to obtain a plurality of first analysis toxicity probabilities, and fusing the obtained first basic toxicity probability; The first toxicity stability information is calculated according to the multiple first analysis toxicity probabilities, and the first basic toxicity probability is corrected and calculated according to the ratio of the first toxicity stability information to the average toxicity stability information to obtain the first toxicity probability.

6. The method for optimizing ingredients for producing antibacterial materials according to claim 1, characterized in that: Performing a metal particle concentration analysis to obtain a first metal particle concentration includes: Collecting a set of sample solution concentrations of the antibacterial metal ion solution, and collecting a set of sample metal particle concentrations of the antibacterial material at different sample solution concentrations and preset ratios, wherein the preset ratio includes a ratio of sepiolite powder and the antibacterial metal ion solution; Constructing a mapping relationship between the sample solution concentration set and the sample metal particle concentration set to obtain a metal particle concentration classifier; The metal particle concentration classifier is used to classify the first solution concentration to obtain a first metal particle concentration.

7. The method for optimizing ingredients for producing antibacterial materials according to claim 1, characterized in that: The first specific surface area and the first metal particle concentration are used to analyze the antibacterial performance to obtain first antibacterial performance information, and the first antibacterial fitness of the first batching parameter is calculated by combining the first agglomeration probability and the first toxicity probability, including: Collect a set of sample specific surface areas and a set of sample metal particle concentrations, and test to obtain the minimum inhibitory concentration of the antibacterial material under different sample specific surface areas and sample metal particle concentrations, and mark them as a sample antibacterial performance information set; Using the sample specific surface area set, the sample metal particle concentration set and the sample antibacterial performance information set, based on a decision tree, an antibacterial performance classifier is constructed; Using the antibacterial performance classifier, the first specific surface area and the first metal particle concentration are input into the classification to obtain first antibacterial performance information; Calculating and obtaining a first antibacterial fitness according to the first antibacterial performance information, the first agglomeration probability and the first toxicity probability; In the powder particle size space and solution concentration space, the optimization of the batching parameters is continued until convergence, and the optimal batching parameters with the maximum antibacterial fitness are output.

8. A batching optimization device for antibacterial material production, characterized in that: The device is used to implement the ingredient optimization method for antibacterial material production according to any one of claims 1 to 7, and the device comprises: An ingredient space acquisition module, used to acquire the powder particle size space of sepiolite powder and the solution concentration space of antibacterial metal ion solution for producing antibacterial material, wherein the antibacterial material is prepared and produced by using sepiolite powder and antibacterial metal ion solution as ingredients; a powder particle size analysis module, for randomly generating a first powder particle size and a first solution concentration in the powder particle size space and the solution concentration space as a first batching parameter, performing a specific surface area analysis according to the first powder particle size to obtain a first specific surface area, and performing an agglomeration probability analysis to obtain a first agglomeration probability, wherein the agglomeration probability analysis result is corrected by an agglomeration probability stability analysis; A solution concentration analysis module, configured to perform a toxicity probability analysis based on the first solution concentration to obtain a first toxicity probability, and to perform a metal particle concentration analysis to obtain a first metal particle concentration; The ingredient optimization module is used to analyze the antibacterial performance according to the first specific surface area and the first metal particle concentration to obtain the first antibacterial performance information, calculate the first antibacterial fitness of the first ingredient parameter in combination with the first agglomeration probability and the first toxicity probability, and continue to optimize the ingredient parameters to obtain the optimal ingredient parameters as the antibacterial material ingredient optimization result.

Images