Preparation method of novel silane coupling agent
By obtaining and optimizing the preparation requirements data of silane coupling agents and accurately controlling the raw material formula and process parameters, the problem of inaccurate raw material selection and proportion in the preparation of silane coupling agents is solved, the purity and stability of the product are improved, and the Preparation of high-quality silane coupling agents.
Patent Information
- Application Number
- CN202411072599.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-22
AI Technical Summary
In the prior art, the raw material selection and inaccurate ratio are inaccurate during the preparation process of silane coupling agent, resulting in low purity of silane coupling agent and poor use stability and unstable product quality.
By obtaining the preparation requirements data of the silane coupling agent to be prepared, performing data analysis and processing, optimizing raw material formula and preparation process parameters, and controlling the preparation device using a computer program to improve the reaction rate and purity.
It improves the purity and stability of silane coupling agent, ensures the stability of product quality, and meets the diversified market needs.
Smart Images

Figure CN119007847A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of chemical manufacturing technology, and in particular relates to a method for preparing a novel silane coupling agent. Background Art
[0002] Silane coupling agents are a class of organic silicon compounds with special structures and properties; coupling agents usually have two different functional groups: one functional group can react with inorganic materials to form chemical bonds; the other functional group can react with organic materials.
[0003] In the related art, when preparing silane coupling agents, the selection of raw materials is usually not accurate and the ratio is not precise enough, so that the reaction yield between the raw materials during preparation is low, the purity of the prepared silane coupling agent is low, and the stability in use is poor, resulting in unstable quality of the prepared silane coupling agent product. Summary of the invention
[0004] The embodiment of the present application provides a novel method for preparing a silane coupling agent, which can solve the problems of low purity and poor stability of the prepared silane coupling agent due to inaccurate raw material selection and inaccurate ratio.
[0005] In a first aspect, the present invention provides a novel method for preparing a silane coupling agent, comprising:
[0006] Acquiring preparation requirement data of a new type of silane coupling agent to be prepared; wherein the preparation requirement data is used to reflect the target performance that the new type of silane coupling agent to be prepared needs to meet;
[0007] Performing data analysis based on the preparation demand data to obtain first preparation data; wherein the first preparation data is used to reflect the raw material formula required for preparing the new silane coupling agent;
[0008] Performing analysis and processing based on the first preparation data to obtain second preparation data; wherein the second preparation data is used to reflect the preparation process required for preparing the novel silane coupling agent;
[0009] Optimizing the first preparation data and the second preparation data to obtain total preparation data; wherein the total preparation data is used to reflect the raw material formula parameters and preparation process parameters that can meet the target performance of the new silane coupling agent;
[0010] According to the total preparation data, the silane coupling agent preparation device is controlled to prepare the novel silane coupling agent.
[0011] The novel silane coupling agent preparation method provided in the present application, firstly, by obtaining the preparation requirement data of the novel silane coupling agent to be prepared, the required performance of the novel silane coupling agent can be understood, and data analysis is performed based on the preparation requirement data to obtain first preparation data, so as to obtain preparation raw materials and raw material formulas that meet the requirements of the novel silane coupling agent to be prepared, so as to serve as a basis for subsequent optimization, and then the first preparation data is analyzed and processed to obtain second preparation data, and the basic process and necessary process for preparing the novel silane coupling agent can be obtained by processing the first preparation data, and the total preparation data can be obtained by optimizing the first preparation data and the second preparation data, so as to obtain the most suitable preparation raw materials compared with other raw materials, and targeted preparation raw materials, formulas and preparation processes can be selected according to the diversified needs of the market, and then the silane coupling agent preparation device is controlled to prepare the novel silane coupling agent according to the total preparation data, and the silane coupling agent preparation device can improve the reaction rate between the raw materials when preparing the novel silane coupling agent, improve the purity of the novel silane coupling agent, and produce the novel silane coupling agent that meets the stability requirements.
[0012] In a second aspect, the present application provides a novel silane coupling agent preparation system, comprising:
[0013] An acquisition unit, used for acquiring preparation requirement data of the new silane coupling agent to be prepared; wherein the preparation requirement data is used for reflecting the target performance that the new silane coupling agent to be prepared needs to meet;
[0014] A first analysis unit is used to perform data analysis based on the preparation demand data to obtain first preparation data; wherein the first preparation data is used to reflect the raw material formula required for preparing the new silane coupling agent;
[0015] A second analysis unit is used to perform analysis based on the first preparation data to obtain second preparation data; wherein the second preparation data is used to reflect the preparation process required for preparing the new silane coupling agent;
[0016] A result unit is used to optimize according to the first preparation data and the second preparation data to obtain total preparation data; wherein the total preparation data is used to reflect the raw material formula parameters and preparation process parameters that can meet the target performance of the new silane coupling agent;
[0017] The control unit is used for controlling the silane coupling agent preparation device to prepare a new silane coupling agent according to the total preparation data.
[0018] In a third aspect, an embodiment of the present application provides a novel silane coupling agent preparation device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the method described in any one of the first aspects above.
[0019] In a fourth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on a new silane coupling agent preparation device, the new silane coupling agent preparation device executes the new silane coupling agent preparation method described in any one of the first aspects above.
[0020] It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0022] Figure 1 It is a schematic flow chart of a method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0023] Figure 2 This is a schematic diagram of the implementation process of step S200 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0024] Figure 3 It is a schematic diagram of the implementation process of step S230 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0025] Figure 4 This is a schematic diagram of the implementation process of step S234 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0026] Figure 5 It is a schematic diagram of the implementation process of step S2344 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0027] Figure 6 It is a schematic diagram of the implementation process of step S23446 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0028] Figure 7 It is a schematic diagram of the implementation process of step S300 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0029] Figure 8 It is a schematic diagram of the implementation process of step S320 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0030] Fig. 9It is a schematic diagram of the implementation process of step S400 in the method for preparing a novel silane coupling agent provided in one embodiment of the present application;
[0031] Fig.10 Schematic diagram of the structure of a novel silane coupling agent preparation system provided in an embodiment of the present application;
[0032] Fig.11 It is a structural schematic diagram of a control device for a novel silane coupling agent preparation device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0033] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0034] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.
[0035] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0036] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.
[0037] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0038] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
[0039] Silane coupling agents are a class of organic silicon compounds with special structures and properties; coupling agents usually have two different functional groups: one functional group can react with inorganic materials to form chemical bonds; the other functional group can react with organic materials.
[0040] In the related art, when preparing silane coupling agents, the selection of raw materials is usually not accurate and the ratio is not precise enough, so that the reaction yield between the raw materials during preparation is low, the purity of the prepared silane coupling agent is low, and the stability in use is poor, resulting in unstable quality of the prepared silane coupling agent product.
[0041] In order to solve the above problems, the embodiments of the present application provide a novel method for preparing a silane coupling agent.
[0042] In the method, first, by obtaining the preparation requirement data of the new silane coupling agent to be prepared, the required performance of the new silane coupling agent can be understood, and data analysis is performed based on the preparation requirement data to obtain the first preparation data, and the preparation raw materials and raw material formula that meet the new silane coupling agent to be prepared are obtained to serve as a basis for subsequent optimization, and then the first preparation data is analyzed and processed to obtain the second preparation data. By processing the first preparation data, the basic process and necessary process for preparing the new silane coupling agent can be obtained, and the optimization is performed according to the first preparation data and the second preparation data to obtain the total preparation data, and the most suitable preparation raw materials compared with other raw materials can be obtained. Targeted preparation raw materials, formulas and preparation processes can also be selected according to the diversified needs of the market, and then according to the total preparation data, the silane coupling agent preparation device is controlled to prepare the new silane coupling agent. The silane coupling agent preparation device can improve the reaction rate between the raw materials when preparing the new silane coupling agent, improve the purity of the new silane coupling agent, and produce a new silane coupling agent that meets stability.
[0043] The novel silane coupling agent preparation method provided in the embodiment of the present application can be applied to a silane coupling agent preparation device. In this case, the silane coupling agent preparation device is the executor of the novel silane coupling agent preparation method provided in the embodiment of the present application. The embodiment of the present application does not impose any restriction on the specific type of the silane coupling agent preparation device.
[0044] For example, the silane coupling agent preparation equipment includes a preparation device and a control device; the preparation device and the control device are communicatively connected; the preparation device can be a variety of devices capable of preparing silane coupling agents, and can adopt a device for preparing silane coupling agents in the prior art, or an improved device for preparing silane coupling agents in the prior art; the control device can be a tablet computer, a laptop computer, a netbook, a desktop computer, a smart large screen, a computer, a handheld computing device, etc.
[0045] In order to better understand the novel silane coupling agent preparation method provided in the embodiments of the present application, the specific implementation process of the novel silane coupling agent preparation method provided in the embodiments of the present application is exemplarily introduced below.
[0046] Figure 1 A schematic flow chart of a method for preparing a novel silane coupling agent provided in an embodiment of the present application is shown. The method for preparing a novel silane coupling agent comprises:
[0047] S100, obtaining preparation requirement data of a new type of silane coupling agent to be prepared; wherein the preparation requirement data is used to reflect the target performance that the new type of silane coupling agent to be prepared needs to meet.
[0048] Exemplarily, obtaining the preparation demand data of the new silane coupling agent to be prepared can be directly input through the user interface on the silane coupling agent preparation device, such as a touch screen or a keyboard, by the customer or operator directly inputting the specific preparation demand. These requirements may include performance indicators of the target product (such as purity, stability, heat resistance, water resistance, etc.), expected application scenarios (such as decorative mortar, rubber reinforcing agent, coating additives, composite material interface agents, etc.) and expected output, etc. Acquiring the preparation demand data can also be database query and matching, and establishing a database containing a variety of silane coupling agent formulas and their corresponding properties. When the preparation demand is received, the system automatically queries and matches the formula or parameter combination closest to the demand in the database through an algorithm. Acquiring the preparation demand data can also be an intelligent optimization algorithm, using intelligent optimization algorithms such as particle swarm algorithms, etc., to optimize parameters such as raw material ratios and preparation processes to maximize the preparation requirements. Different acquisition methods can make the preparation demand data more accurate and provide a basis for subsequent preparation.
[0049] S200, performing data analysis based on the preparation requirement data to obtain first preparation data; wherein the first preparation data is used to reflect the raw material formula required for preparing the new silane coupling agent.
[0050] Exemplarily, the preparation demand data is first preprocessed, including steps such as data cleaning and normalization, to eliminate the influence of outliers and different dimensions on the analysis results, and then the principal component analysis method is used to extract the key features in the preparation demand, and the machine learning model is constructed using historical preparation data. This embodiment preferably uses a deep neural network or a random forest algorithm, which can capture complex nonlinear relationships and continuously optimize internal parameters through training to improve the accuracy of the prediction. The model forms a prediction rule for training through the mapping relationship between different raw material ratios and the performance of the final product. After training, the preprocessed preparation demand data is input into the trained model to obtain the optimal raw material formula, i.e., the first preparation data.
[0051] In one possible implementation, see Figure 2 , S200, performing data analysis based on the preparation demand data to obtain first preparation data, including:
[0052] S210, obtaining application data of the new silane coupling agent; wherein the application data is used to reflect the usage scenario of the new silane coupling agent to be prepared.
[0053] For example, first, a sensor network can be deployed in potential use scenarios, such as composite material manufacturing, coating treatment or rubber processing, to record key parameters such as temperature, humidity, pressure, chemical reaction rate, etc. These parameters are directly related to the performance and use effect of the new silane coupling agent. These massive real-time data are pre-processed, including data cleaning, denoising, normalization and other steps to improve data quality and analysis accuracy, and understand the specific application of the new silane coupling agent in different use scenarios. Association rule mining technology can be used. This technology can reveal the potential relationship between the variables in the data set, find out which environmental factors (such as temperature range, humidity level) have the most significant impact on the performance of the new silane coupling agent, and further verify and refine the analysis results. Advanced simulation software can be used to simulate the reaction process of the new silane coupling agent under different conditions, observe its structural changes, performance and interaction with the substrate, and obtain application data.
[0054] S220, obtaining prompt data after matching the application data and the preparation requirement data; wherein the prompt data is used to reflect the preparation formula that meets the use scenario and target performance of the new silane coupling agent.
[0055] For example, first, the key parameters extracted from the application data (such as temperature range, humidity level, required performance indicators, etc.) are used as input variables, and the objective function is set at the same time, that is, to maximize the performance of the new silane coupling agent in a specific use scenario, and these initial formulas are evaluated using a fuzzy logic system, and fuzzy sets are defined according to fuzzy logic (such as "good high temperature tolerance", "strong stability under low humidity", etc.), and the corresponding membership is assigned to each formula according to the actual data. In the iterative process of the genetic algorithm, the population is continuously evolved through operations such as selection, crossover and mutation to generate new formulas. Each generation of formulas is selected according to the results of the fuzzy logic evaluation, and gradually approaches the optimal solution. An adaptive mutation rate can also be introduced to enable the algorithm to maintain a high exploration ability in the early stage of the search, and reduce mutation when approaching the optimal solution, improve the convergence speed, and when it converges to a stable state, select the formula with the highest fitness as the final prompt data, and then obtain the prompt data. This prompt data not only meets the use scenario and target performance requirements of the new silane coupling agent, but also reflects the detailed information such as the proportion of various raw materials and reaction conditions that should be used in the preparation process.
[0056] S230, optimizing the preparation formula indicated by the prompt data to obtain first preparation data.
[0057] Exemplarily, after obtaining the prompt data, in order to enable the first preparation data to maximize the performance of the new silane coupling agent, the present embodiment adopts an optimization strategy based on deep reinforcement learning, combined with molecular dynamics simulation technology for optimization. First, the preparation formula in the prompt data is input into the reinforcement learning model as the initial state. The model is constructed by a deep neural network, which is obtained by continuous trial and error, learning and improvement through the reinforcement learning mechanism, and then by designing a specific reward function, which is based on the performance indicators (such as strength, heat resistance, weather resistance, etc.) of the prepared silane coupling agent. Quantify to guide the optimization direction of the model. Formula adjustment suggestions can be generated during the optimization process, and it is recommended to be quickly verified by molecular dynamics simulation. Molecular dynamics simulation can simulate the physical and chemical behavior of materials at the atomic level, thereby predicting the specific impact of formula adjustment on the performance of silane coupling agents, and then find the equilibrium point through continuous iteration, that is, obtain the first preparation data.
[0058] In one possible implementation, see Figure 3 , S230, optimizing the preparation formula indicated by the prompt data to obtain first preparation data, including:
[0059] S231, the silane coupling agent preparation device responds to the preparation operation of the new silane coupling agent.
[0060] Exemplarily, first, the control device receives the first preparation data from the optimization model, including parameters such as accurate raw material ratio, reaction temperature, pressure, stirring rate, etc. These data are parsed and converted into specific operating instructions by the preparation device. For example, a high-precision mass flowmeter can be used to accurately control the feeding amount of the raw material, the constant temperature environment in the reactor can be maintained by an intelligent temperature control system, and a variable frequency motor is used to drive the agitator to achieve the best stirring effect. During the preparation process, the chemical composition and physical properties of the reactants and products are monitored in real time. These real-time monitoring data are immediately fed back to the optimization model and compared and analyzed with the preset target performance parameters. If deviations are found, the optimization model will quickly adjust the formula parameters and fine-tune the preparation process through the control device to form a closed-loop control. In order to further enhance the flexibility and adaptability of the preparation process, an adaptive control algorithm can be introduced. The algorithm can automatically adjust the control strategy according to real-time monitoring data and historical preparation records to cope with the slight differences in different batches of raw materials and changes in the preparation environment, and obtain the prepared silane coupling agent.
[0061] S232, extracting the prompt data to obtain formula data; wherein the formula data is used to reflect the types of raw materials used for the new silane coupling agent to be prepared, and the types of raw materials include silicon hydride compounds and organic functional groups, and the silicon hydride compounds and organic functional groups can undergo chemical reactions.
[0062] Exemplarily, the prompt data is first parsed, including but not limited to encoding the text using a pre-trained BERT model to capture contextual semantic information, using chemical informatics tools (such as ChemSpider, PubChem, etc.) to query the detailed chemical structure, molecular formula, molecular weight and other key information of each raw material, and the information is integrated into a structured data table to provide basic data support for the subsequent optimization algorithm. By verifying whether the silicon hydrogen compound and the organic functional group have the potential for chemical reaction, the molecular docking technology is used to simulate the interaction between them. Molecular docking can predict the binding energy, binding site and binding mode between molecules, thereby evaluating the feasibility of chemical reactions. Based on the results of molecular docking, a reaction network diagram is constructed, in which nodes represent raw materials or intermediates, and edges represent possible chemical reaction paths. The optimal reaction path from raw materials to target products is searched using a graph theory algorithm (such as Dijkstra algorithm), while considering the constraints of reaction conditions (such as temperature, pressure, catalyst, etc.). According to the optimal reaction path selected in the reaction network diagram, preliminary formula data is generated, including raw material types, proportions, reaction conditions, etc. At this point, the recipe data may contain multiple alternatives, each of which corresponds to a possible reaction path. A multi-objective optimization algorithm is used to simultaneously optimize multiple performance indicators (such as yield, purity, cost, environmental impact, etc.) to obtain the recipe data.
[0063] S233, determining the raw materials based on the silicon hydride compounds and organic functional groups indicated by the formula data to obtain raw material data; wherein the raw material data is used to indicate the raw materials in the silicon hydride compounds and the raw materials in the organic functional groups, the raw materials in the silicon hydride compounds include epoxy silane, amino silane and hydroxy silane, and the raw materials in the organic functional groups include carboxylic acids, ketones and amines.
[0064] Exemplarily, the classification and regression technology in machine learning is first used to build an intelligent matching engine. The engine first learns the mapping relationship between the raw material combination and performance of a large number of known silane coupling agents to predict the most suitable raw material combination under the new formula. Ensemble learning methods such as random forests and gradient boosting trees are used to improve the accuracy and robustness of predictions. On the basis of intelligent matching, the characteristics of candidate raw materials are further evaluated. Including factors such as the purity, stability, reactivity, cost and environmental friendliness of the raw materials to screen out the optimal raw material combination and obtain raw material data. Before the raw materials are put into storage, the raw materials are tested for purity; pretreatment operations are performed according to the characteristics of the raw materials and reaction requirements. For example, for volatile epoxy silane, low-temperature storage and inert gas protection are used; for easily hydrolyzed hydroxy silane, it is necessary to seal under dry conditions. In addition, steps such as purification and activation of raw materials may be involved to improve the efficiency and yield of subsequent reactions.
[0065] S234, optimizing and analyzing the raw material data according to a genetic algorithm to obtain first preparation data.
[0066] Exemplarily, first, a coding method suitable for the characteristics of raw material data is designed. Due to the variety of raw materials and complex ratios, a hybrid strategy of real number coding combined with binary coding can be adopted. Real number coding is used to represent the ratio of raw materials (such as mass percentage), while binary coding is used to represent specific preparation conditions (such as temperature, pressure, etc.). The hybrid coding method can accurately express the raw material ratio and flexibly represent a variety of preparation conditions, and then randomly generate a certain number of individuals as the initial population. Each individual represents a possible combination of raw material ratio and preparation conditions, that is, a potential solution. In order to achieve the diversity of the population, a certain degree of randomness is introduced, and prior knowledge is added when necessary to guide the generation of the initial population. According to the experimental objectives and formula data, a fitness function is designed. This function is used to evaluate the pros and cons of each individual, that is, the comprehensive performance (such as yield, purity, cost-effectiveness, etc.) of the raw material ratio and preparation condition combination represented by it when preparing a new silane coupling agent. The fitness function should accurately reflect the experimental objectives and effectively distinguish the differences between different individuals. These excellent individuals have higher fitness values and are more likely to contain genetic information that is conducive to the achievement of the goal. The selected individuals are crossovered to generate new individuals by exchanging some genetic information. At the same time, mutation operations are introduced to change some genetic information of individuals with a certain probability, thereby increasing the diversity of the population and avoiding falling into the local optimal solution. In the crossover and mutation process, adaptive crossover rate, mutation rate and heuristic mutation strategy can be introduced to improve the performance of the algorithm. In the iterative process, the changing trend of the population and the fitness change of the optimal individual are continuously observed to adjust the algorithm parameters or strategies. When the iterative evolution is completed, the individual with the highest fitness is selected from the final population to obtain the first preparation data.
[0067] In one possible implementation, see Figure 4 , S234, optimizing and analyzing the raw material data according to the genetic algorithm to obtain first preparation data, including:
[0068] S2341, extract at least two raw materials of epoxy silane, amino silane and hydroxy silane in the silicon hydrogen compound indicated by the raw material data to obtain first raw material data.
[0069] Exemplarily, first, the mixture of silicon hydrogen compounds indicated by the raw material data is subjected to meticulous homogenization, so that each component can be evenly distributed, which is convenient for subsequent operation, and impurities such as inorganic salts and heavy metal ions in the mixture are preliminarily removed by selective adsorption resin or ion exchange resin, so as to reduce interference with subsequent extraction steps, and a low-toxic, highly selective environmentally friendly solvent (such as supercritical carbon dioxide) and a mixed system of a water-soluble solvent (such as ethanol or ethylene glycol) are used. This mixed solvent can effectively dissolve and separate the target silane compound while reducing the impact on the environment. The optimal extraction efficiency of epoxy silane, amino silane and hydroxy silane is achieved by the extraction parameters (such as temperature, pressure, solvent ratio and extraction time) optimized by the genetic algorithm. The application of the genetic algorithm at this stage can automatically adjust multiple variables, quickly find the optimal combination of conditions, and use supercritical fluid as the mobile phase, which exhibits special properties between gas and liquid under high pressure, to efficiently separate different components in the mixture. SFC combines the high separation efficiency of gas chromatography and the high selectivity of liquid chromatography, and is particularly suitable for separating heat-sensitive, volatile or polar compounds with large differences. The mixture after solvent extraction is injected into the SFC system, and the target components such as epoxy silane, amino silane and hydroxy silane are finely separated by adjusting the pressure, temperature, flow rate of the mobile phase and the possible addition of modifiers (such as methanol, ethanol, etc.) to obtain the first raw material data.
[0070] S2342, extracting at least two raw materials from carboxylic acid, ketone and amine in the organic functional group indicated by the raw material data to obtain second raw material data.
[0071] Exemplarily, first, according to the structural characteristics of the target compound (carboxylic acid, ketone, amine), a suitable template molecule is selected. In the presence of the template molecule, a functional monomer (such as methacrylic acid, acrylamide, etc.) is copolymerized with a crosslinking agent by a crosslinking polymerization reaction, and the template molecule is removed by an appropriate method (such as solvent elution), leaving a hole with a shape, size and functional group matching the template molecule. In view of the high efficiency of MAE, a solvent system that can effectively dissolve the target organic functional group and has good microwave absorption is selected, such as a non-polar solvent mixture containing a small amount of polar solvent (such as ethanol, acetone). Such a solvent combination can not only promote the dissolution of the target, but also effectively utilize microwave energy to accelerate the extraction process. Experimental design and optimization techniques (such as response surface method) are used to finely adjust parameters such as microwave power, radiation time, and solvent volume ratio to achieve the best extraction effect, and then a suitable desorption solvent (such as a weakly acidic or alkaline solution) is used to elute the target compound under mild conditions, and the desorption liquid is subjected to necessary separation operations (such as liquid-liquid extraction, solid phase extraction, etc.) to remove solvents and impurities to obtain high-purity second raw material data.
[0072] S2343, randomly and one-to-one correspondingly combine at least two raw materials among epoxy silane, amino silane and hydroxy silane indicated by the first raw material data and at least two raw materials among carboxylic acid, ketone and amine indicated by the second raw material data to obtain multiple raw material combination data; wherein one raw material combination data indicates one raw material combination, and each raw material combination includes any one raw material among epoxy silane, amino silane and hydroxy silane in silicon hydrogen compounds and any one raw material among carboxylic acid, ketone and amine in organic functional groups.
[0073] For example, all raw materials can be uniquely coded, and can be accurately identified and tracked during the automated processing. According to chemical properties (such as functional group type), the raw materials are divided into two groups: silicon hydride group (epoxy silane, amino silane, hydroxy silane) and organic functional group group (carboxylic acid, ketone, amine), and the raw materials can be selected and combined from the two groups of raw materials. In each combination, there is one silicon hydride and one organic functional group raw material, and the combination method is random and non-repetitive. During the processing, each generated raw material combination can be assigned to an independent reaction container, and a preliminary reactivity test or stability evaluation is performed. The high-throughput screening platform is used to process multiple raw material combinations, and the number, composition, processing conditions and preliminary test results of each combination are recorded at the same time to generate a detailed raw material combination data report. The generated raw material combination data is strictly verified to improve the accuracy and reliability of the data. The combination data is deeply analyzed using statistical analysis software to identify raw material data combinations with potential application value to generate multiple raw material combination data. This process not only improves the experimental efficiency, but also greatly broadens the diversity of raw material combinations, providing strong support for subsequent research or production.
[0074] S2344, performing fitness calculation based on the raw material combination data to obtain first preparation data.
[0075] For example, after obtaining the raw material combination data, the raw material combination data is first cleaned to remove outliers or erroneous records, improve data quality, and extract key feature information, such as the chemical structure of the raw materials, the type of functional groups, and the predicted value of reaction conditions, as input parameters for fitness calculation, and then the interaction of the raw material combination data at the microscopic scale is simulated and predicted using technologies such as quantum computing or molecular dynamics simulation. These technologies can capture the reaction mechanism, and then calculate the fitness of each raw material combination according to the designed fitness index and the selected calculation method. The calculation results reflect the comprehensive performance of the combination in the preparation of a specific product. The fitness scores of all raw material combinations are sorted, and several combinations with the highest fitness are selected as candidates for the first preparation data. The high-fitness raw material combinations that are initially screened are experimentally verified to confirm whether their performance in the actual preparation process meets expectations. According to the verification results, the fitness evaluation model is feedback adjusted, the algorithm parameters are optimized or the evaluation indicators are improved, and the accuracy and reliability of the evaluation are improved to obtain the first preparation data.
[0076] In one possible implementation, see Figure 5 , S2344, performing fitness calculation based on the raw material combination data to obtain first preparation data, including:
[0077] S23441, analyze each raw material combination data to obtain reaction data; wherein the reaction data is used to reflect the reactant ratio and reaction conditions between any one of the raw materials selected from epoxy silane, amino silane and hydroxy silane and any one of the raw materials selected from carboxylic acid, ketone and amine. The reactant ratio is the ratio value between any one of the raw materials selected from epoxy silane, amino silane and hydroxy silane and any one of the raw materials selected from carboxylic acid, ketone and amine. The reaction conditions include the reaction temperature value and the reaction time. Each raw material combination data corresponds to each reaction data.
[0078] Exemplarily, the chemical structure of the raw materials is first analyzed to identify key information such as functional groups and intermolecular interactions. Based on preliminary chemical structure analysis, the reaction path and product structure are predicted, and the possible reaction types and condition ranges are preliminarily screened out, and then trained by machine learning models (such as random forests, gradient boosting trees, neural networks, etc.). The chemical structure characteristics of each raw material combination data are input into the model, and the model outputs the predicted optimal reactant ratio (such as molar ratio), reaction temperature range, and reaction time and other key reaction data. For some key or complex raw material combinations, quantum chemical simulation can calculate the electron transfer between molecules, the breaking and formation process of bonds, and provide a basis for reaction conditions. By simulating the reaction path and energy changes under different reaction conditions, the prediction results of the machine learning model are verified and optimized to improve the reliability of reaction data.
[0079] S23442, respectively encode the reactant ratio and reaction conditions indicated by each reaction data to obtain encoded reaction data; wherein the encoded reaction data includes the reactant ratio encoding data and the reaction condition encoding data.
[0080] Exemplarily, the reactant ratio (such as molar ratio) is first subjected to hierarchical quantization. Unlike simple digital conversion, it is divided into multiple intervals according to the actual meaning of the ratio (such as reaction balance, efficiency optimization, etc.), and each interval is assigned a specific code. This hierarchical quantization not only reduces the complexity of the coding, but also retains the key information of the effect of the ratio change on the reaction. For example, for the reaction of epoxy silane and carboxylic acid, the molar ratio can be divided into three intervals of "low ratio", "medium ratio" and "high ratio", and represented by "L", "M" and "H" respectively. In order to further increase the complexity and security of the coding, the ratio data after hierarchical quantization is mapped to a complex network. Each ratio interval corresponds to a node in the network, and the connection between the nodes is constructed according to the law of the ratio change and possible influencing factors. A unique path code can be generated for each ratio by traversal. Finally, the path code obtained by complex network mapping is encrypted using a hash function, and the choice of the hash function should make it highly sensitive and collision-resistant so that the coded data has security and uniqueness. The encrypted reactant ratio coding data will be stored in the database for subsequent analysis. For reaction conditions (such as reaction temperature, reaction time, etc.), multi-dimensional feature extraction is first performed. In addition to basic numerical information, there are also advanced features such as the rate of temperature change and the uniformity of time distribution to fully reflect the complexity and diversity of reaction conditions. Since reaction conditions often have certain ambiguity and uncertainty in actual operation (such as temperature fluctuations, time fine-tuning, etc.), fuzzy logic processing methods can be used to quantify these features, and then by defining fuzzy sets and fuzzy rule bases, the fuzzy description of the reaction conditions can be converted into precise numerical codes, and accurate coded reaction data can be obtained.
[0081] S23443, based on the coded reaction data, experimentally operate any one of the raw materials among epoxy silane, amino silane and hydroxy silane with any one of the raw materials among carboxylic acid, ketone and amine to obtain reaction performance data; wherein the reaction performance data is used to reflect the performance index value after a simulated experimental reaction between any one of the raw materials among epoxy silane, amino silane and hydroxy silane and any one of the raw materials among carboxylic acid, ketone and amine.
[0082] Exemplarily, first, an intelligent pairing algorithm based on coded reaction data. The algorithm can parse the reactant ratio and condition information in the code, and quickly screen out theoretically feasible raw material combinations according to preset chemical reaction rules and empirical databases. For example, the algorithm can identify the combination of "epoxy silane" and "carboxylic acid", and recommend the corresponding reaction temperature range and reaction time accordingly. Then, the experiment can be performed according to the conditional coding in the coded reaction data, and the reaction temperature, stirring speed, pressure and other parameters can be adjusted to achieve the control of the reaction conditions. In order to obtain the performance data of the reaction process in real time, in-situ spectral analysis technology can be used. This technology can monitor the reaction system online without interrupting the reaction, record the spectral changes, and infer the performance indicators such as reaction rate, conversion rate, and product distribution, and then obtain the reaction performance data.
[0083] S23444, calculating and transforming the reaction performance data to obtain adaptation data; wherein the adaptation data is used to reflect the fitness value adapted to the usage scenario.
[0084] Exemplarily, firstly, an automatic data cleaning technology based on deep learning is adopted, such as using a convolutional neural network (CNN) to identify and remove outliers or noise data, which may be caused by equipment failure, measurement error, etc. Compared with the traditional threshold judgment method, deep learning can automatically learn the normal distribution pattern of data, so as to more accurately identify and handle anomalies, not only relying on basic statistical features (such as mean, standard deviation), but also introducing time series feature extraction technology (such as LSTM network) and image recognition technology (if the reaction performance data can be represented in the form of an image) to mine hidden information in the data from multiple dimensions. Ensemble learning methods, such as the combination of gradient boosting tree (GBDT) and random forest, are used to capture the complex relationship between features to find the optimal solution. Bayesian optimization technology can also be used to select parameter combinations to try to obtain adaptive data.
[0085] S23445, screening all the coded reaction data according to the requirements of the adaptation data to obtain a plurality of satisfying data; wherein the satisfying data is the coded reaction data that meets the requirements of the adaptation data, and all the coded reaction data has a plurality of coded reaction data that meets the requirements of the adaptation data.
[0086] Exemplarily, first, the requirements of the adapted data are parsed, and before screening, all the coded reaction data are preprocessed, including data cleaning, format unification and coding standardization, to build an efficient data indexing system, and an inverted index or distributed hash table can be used to accelerate the data retrieval process. In addition, graph database technology can be used to store the relationship between the coded reaction data in the form of a graph, which is convenient for complex relationship queries, and then an intelligent screening algorithm based on the adapted data requirements is used. The algorithm can be a multi-stage process, first using simple rules to filter out data that obviously does not meet the requirements, and then gradually increase the complexity and accuracy of the screening. For example, in the first stage, a Bloom filter is used to quickly exclude a large amount of irrelevant data, and in the subsequent stage, a machine learning model or a deep neural network is used to deeply analyze the data to evaluate whether it meets the specific indicators of the adapted data. Since the coded reaction data may be large, parallel processing and distributed computing technology are used to accelerate the screening process. The data is processed in blocks and distributed to multiple computing nodes for simultaneous screening. The screened data is strictly verified so that the data meets the requirements of the adapted data, and multiple satisfied data are obtained.
[0087] S23446, performing a reorganization operation on the plurality of satisfying data to obtain first prepared data.
[0088] Exemplarily, cluster analysis or principal component analysis is first used to conduct preliminary exploration of multiple satisfying data, identify natural groups or features in the data, and use association rule mining algorithms (such as Apriori algorithm or FP-Growth algorithm) to analyze the potential associations between different data sets, which may be direct (such as numerical relationships) or indirect (such as through common attributes or behavior patterns). Since multiple satisfying data may come from different data sources or have different formats, data alignment and standardization processing are required, including the unification of timestamps, conversion of data types, filling of missing values, and processing of outliers. Parallel computing or distributed processing technology can be used to accelerate the reorganization process by performing reorganization operations based on the reorganization strategy of data characteristics and analysis targets. At the same time, the efficiency and stability of the reorganization operation are improved by performance tuning. For example, the distributed reorganization of data is realized by using processing frameworks such as Apache Spark. The first prepared data after reorganization is strictly verified and evaluated, mainly including checking the integrity, consistency and accuracy of the data, and then obtaining the first prepared data.
[0089] In one possible implementation, see Figure 6 , S23446, reorganize the plurality of satisfying data to obtain first prepared data, including:
[0090] S234461, extract multiple reactant ratio coding data and reaction condition coding data that meet the data.
[0091] Exemplarily, first, the data is initially parsed to meet the requirements, and key entities (such as reactants, conditions) and their attributes (such as ratio, temperature, pressure, etc.) in the data are identified. Based on the parsing results, pattern recognition is performed on the reactant ratio and reaction conditions, and the coded data that meet the specific pattern are extracted after identification. In order to further improve the accuracy of extraction, the model output is post-processed in combination with the domain knowledge base or expert rules. The extracted coded data may contain noise or errors, so strict cleaning and verification are required. This includes checking the consistency of the data (such as whether the sum of the ratios is equal to 1), verifying the validity of the coding (such as checking whether it is within the preset coding range) and using domain knowledge to identify and correct abnormal values. The reactant ratio coded data and the reaction condition coded data after cleaning are integrated into a unified data structure, so that the data format is standardized, so that subsequent processing and analysis can be smoothly carried out, and the data after extraction and integration are subjected to quality assessment, including accuracy, recall rate and F1 score index calculation, and reactant ratio coded data and reaction condition coded data are effectively extracted from multiple satisfying data, providing strong support for subsequent data analysis and decision support.
[0092] S234462, randomly shuffle the multiple reactant ratio coding data that meet the extraction from the data, randomly shuffle the multiple reaction condition coding data that meet the extraction from the data, and then combine the shuffled multiple reactant ratio coding data and the shuffled multiple reaction condition coding data one by one according to the usage scenario.
[0093] Exemplarily, first, the reactant ratio coding data and reaction condition coding data extracted from the satisfying data are de-duplicated to ensure the uniqueness of each coding. The present embodiment is different from the traditional shuffling algorithm by designing a non-standard random scrambling algorithm; Due to the specific structure or constraints that the coding data may have (such as certain ratios must maintain a certain proportional relationship), the present embodiment can adopt a constraint-based randomization method. For example, the coded data can be divided into several groups first, and the relative proportions are kept unchanged inside each group, but random exchanges are performed between groups. Alternatively, the idea of optimization algorithms such as simulated annealing is used to perform random disturbances under the premise of maintaining certain constraints to generate a diverse data combination. In order to improve the efficiency of scrambling, parallel computing techniques, such as multi-threading or multi-process, can be used to scramble the reactant ratio coding data and the reaction condition coding data respectively. At the same time, in order to scramble the randomness, a random number generator can be introduced, which can produce different results when executed multiple times. According to specific usage scenarios (such as different experimental requirements, industrial production conditions, etc.), intelligent combination strategies are designed. Including but not limited to rule-based combinations (such as setting priorities between certain ratios and conditions), optimization-based combinations (such as using optimization algorithms to find optimal or near-optimal combinations) and simulation-based combinations (such as predicting the effects of combinations through simulation experimental processes). All generated combinations are then evaluated in multiple dimensions, including but not limited to feasibility (whether they meet actual production conditions), stability (whether the system remains stable after combination), efficiency (whether it improves production efficiency or reduces costs) and safety (whether there are potential safety hazards).
[0094] S234463, after the combination is completed, based on the preparation demand data and the usage scenario, the ratio values in multiple reactant ratio coding data are randomly changed, and the reaction parameter values in multiple reaction condition coding data are randomly changed, and the changed reactant ratio coding data and reaction condition coding data are recombined and combined to obtain reaction optimization data; wherein, the reaction optimization data is the reactant ratio coding data and reaction condition coding data after optimization and adjustment.
[0095] Exemplarily, first, the preparation demand data is analyzed in depth, including but not limited to the purity, output, cost requirements, etc. of the target product. At the same time, the specific conditions of the use scenario are clarified, such as the type of reaction vessel, temperature control range, pressure conditions, etc. Due to the complex relationship between the ratio and the reaction parameters, this embodiment can design an adjustment strategy based on a nonlinear optimization algorithm. Unlike the traditional linear adjustment method, the nonlinear optimization algorithm can more accurately simulate the complex changes in the actual reaction process, so as to find a better combination of ratios and parameters. The algorithm can be combined with a variety of factors, such as reaction kinetics, thermodynamic equilibrium, mass transfer and heat transfer limitations, etc.; while maintaining a certain probability, the ratio values in the reactant ratio coding data and the reaction parameter values in the reaction condition coding data are randomly changed. This mechanism helps to jump out of the local optimal solution and find a better global solution. Through the mutated reactant ratio and reaction conditions, the combination that is more likely to meet the preparation requirements and the use scenario requirements can be identified using a recombinant strategy, and the combination attempt is prioritized. At the same time, each group of recombined data is evaluated in multiple dimensions, including reaction efficiency, product quality, cost-effectiveness, etc. The above steps constitute an iterative loop, and the random variation, intelligent reorganization and evaluation process of the ratio and parameters are continuously repeated. Then, by setting a reasonable convergence criterion, for example, if the optimal solution is not significantly improved after multiple consecutive iterations, or if the preset upper limit of the number of iterations is reached, the iteration is stopped. The reaction optimization data obtained at this time is the reactant ratio coding data and the reaction condition coding data after optimization adjustment, that is, the reaction optimization data is obtained.
[0096] S234464, based on the reaction optimization data, obtain the first preparation data.
[0097] Exemplarily, the reaction optimization data is first deeply cleaned to remove potential noise or outliers, and a data cleaning algorithm, such as anomaly detection based on machine learning, is used to improve the purity of the data, and the data is standardized so that the ratios and parameters of different dimensions can be compared and analyzed under the same framework. According to the actual capabilities of the experimental equipment, a mapping relationship between reaction parameters and equipment operation instructions is constructed, and through a fine mapping construction, each optimized parameter can find the corresponding equipment operation instruction. Based on the optimized reactant ratio and reaction conditions, this embodiment designs a set of dynamically adjusted experimental schemes, which is different from the traditional static experimental design. The dynamic experimental planning allows parameter fine-tuning according to real-time feedback during the experiment to better adapt to changes in experimental conditions, effectively improving the success rate of the experiment, and converting the real-time data collected during the experiment into the first preparation data that can be used for actual production, effectively converting the reaction optimization data into practical preparation data, significantly improving the experimental efficiency and the reliability of the results, and laying the foundation for subsequent optimization.
[0098] S300, analyzing and processing the first preparation data to obtain second preparation data; wherein the second preparation data is used to reflect the preparation process required for preparing the new silane coupling agent.
[0099] Exemplarily, the first preparation data is first subjected to advanced filtering. Different from the traditional low-pass or high-pass filtering, the adaptive filtering technology of wavelet transform can dynamically adjust the filtering parameters according to the characteristics of the data itself to effectively remove noise interference while retaining the key feature information in the data. Then, the outliers in the data set are accurately identified and removed through the isolation forest algorithm, which can improve the accuracy and reliability of subsequent analysis. In the feature extraction stage, the pre-processed data is mined by combining the convolutional neural network and the recurrent neural network model in deep learning. The convolutional neural network can automatically learn and extract the spatial hierarchical features in the data, which is suitable for analyzing the relationship between reaction conditions (such as temperature and pressure when the silane coupling agent raw material reacts) and product characteristics; while the recurrent neural network is good at processing time series data and can capture the dynamic laws of the changes of various parameters over time in the preparation process. Molecular simulation technology, such as quantum chemical calculation and molecular dynamics simulation, can also be introduced to predict the changes in the molecular structure of the silane coupling agent from the microscopic level, providing a theoretical basis for the optimization of the preparation process. Based on the extracted features, advanced association analysis algorithms, such as the maximum information coefficient or the Pearson correlation coefficient, are preferably used to explore the intrinsic connection and potential laws between the parameters, and a prediction model is constructed using a gradient boosting tree. The model can consider the influence of multiple parameters and predict the preparation effect of new silane coupling agents. After hyperparameter tuning and cross-validation, the generalization ability and stability of the model are observed. After obtaining the prediction model, the particle swarm optimization algorithm is used to perform global search and optimization of the preparation process parameters. These algorithms find the optimal solution or approximate optimal solution by simulating the natural evolution process or group behavior, and determine the optimal preparation process of the new silane coupling agent based on the optimization results, including reaction temperature, time, raw material ratio, catalyst and other parameters.
[0100] In one possible implementation, see Figure 7 , S300, analyzing and processing the first preparation data to obtain second preparation data, including:
[0101] S310, analyzing the preparation demand data to obtain process preparation data; wherein the process preparation data is used to reflect the preparation process of the new silane coupling agent.
[0102] For example, first, the preparation demand data is deeply analyzed, including but not limited to product performance indicators (such as purity, viscosity, thermal stability), production scale, cost control and environmental protection requirements. Using knowledge graph technology, these macro requirements are mapped to specific preparation process parameters, such as the type and purity of reactants, the choice of solvents, the design of reactors and post-processing steps; the mapped preparation process is virtually simulated using process simulation software (such as Aspen Plus, ChemCAD). These software can simulate complex chemical reaction processes and predict the conversion rate of reactants, the distribution of products and the generation of by-products. At the same time, combined with risk assessment models (such as failure mode and effect analysis FMEA, Monte Carlo simulation, etc.), the risk factors that may appear in the preparation process are evaluated and corresponding preventive measures are formulated. In the actual execution of the preparation process, dynamic optimization and adaptive adjustment mechanisms can be introduced, and real-time monitoring systems can be used to collect various data in the production process, including key parameters such as temperature, pressure, and flow. By comparing the actual data with the output of the prediction model, the model is updated and optimized in real time to adapt to various changes that may occur in the production process. At the same time, the process parameters are automatically adjusted according to the optimization results, such as adjusting the reaction temperature, changing the raw material ratio or switching the catalyst, so as to achieve dynamic optimization and adaptive adjustment of the preparation process, and then obtain process preparation data.
[0103] S320, performing a preparation test on the first preparation data through process preparation data to obtain process defect data; wherein the process defect data is used to reflect the situation that the organic functional group obtained after the preparation test on the first preparation data has not reacted completely in the silicon hydride compound and impurities exist.
[0104] For example, first, microreactor technology is used to decompose traditional large-scale reactions into multiple trace parallel reactions, reduce experimental costs and time, and improve the controllability and repeatability of reaction conditions. Through high-throughput screening strategies, a variety of process conditions are quickly evaluated to identify key factors affecting the thoroughness of the reaction of organic functional groups in silicon hydrogen compounds, as well as potential sources that may lead to the generation of impurities. During the preparation and testing process, in-situ characterization techniques such as infrared spectroscopy and Raman spectroscopy are introduced to perform real-time detection of the reaction system. The breaking and formation of chemical bonds and the generation and change of products during the reaction can be observed, providing a basis for the immediate diagnosis of process defects. Combined with image processing and pattern recognition algorithms, the in-situ characterization data is identified and potential process defects are marked, and the defects are sorted to obtain process defect data.
[0105] In one possible implementation, see Figure 8 , S320, analyzing and processing the first preparation data to obtain second preparation data, including:
[0106] S321, first divide the raw material formula indicated by the first preparation data into N groups of raw material formulas according to the preparation process, and then prioritize the N groups of raw material formulas according to the preparation sequence and reaction conditions; wherein N is a positive integer.
[0107] Exemplarily, first clarify the type, quantity and order of addition of raw materials required for each step, and split the raw material formula into N independent but interrelated groups according to the natural breakpoints or key reaction steps of the preparation process. Each group of raw materials corresponds to one or more consecutive steps in the process. For each group of raw materials, evaluate the temperature, pressure, catalyst and other conditions required for its reaction, as well as the impact of these conditions on the reaction of other groups of raw materials. Identify the dependencies between groups of raw materials, for example, the generation of certain intermediates is a necessary condition for subsequent reactions. According to the actual order of addition of raw materials in the preparation process, assign initial priorities to each group of raw material formulas. Generally, the raw material formula added first has a higher priority, and the reaction rate and efficiency of each group of raw materials are evaluated in combination with reaction kinetics. For raw material formulas that react quickly and have a great impact on subsequent steps, their priority can be appropriately increased. According to the stability, toxicity and potential safety risks of the raw materials, high-risk raw material formulas are specially marked to select the best solution.
[0108] S322, inputting N groups of raw material formulas into the process preparation data in order according to priority to perform process preparation testing to obtain reaction purity data; wherein the reaction purity data is used to reflect the purity value of the new silane coupling agent obtained after the preparation test.
[0109] Exemplarily, each group of raw material formulas is added to the reaction system one by one according to the established priority order. During the reaction process, the reaction system is monitored in real time using online analysis technology, such as gas chromatography, liquid chromatography or mass spectrometry, etc., which can accurately measure the consumption of reactants and the generation of products, thereby calculating the purity of the new silane coupling agent. By sampling and analyzing, multiple groups of reaction purity data are obtained to evaluate the stability and repeatability of process preparation data. Statistical analysis of multiple batches of experimental results can identify potential outliers or unstable factors, and adjust and optimize the process preparation data accordingly, so as to clearly understand which factors have a significant impact on the purity of the new silane coupling agent, and the by-products and impurities that may occur during the reaction process. In order to reduce the impact of these adverse factors, separation and purification technologies, such as solvent extraction, distillation, crystallization, etc., can be used to post-treat the reaction products to obtain reaction purity data.
[0110] S323, calculating the conversion rate based on the reaction purity data to obtain product conversion data; wherein the product conversion data is used to reflect the product conversion status of N groups of raw material formulas into products.
[0111] Exemplarily, the mass of the target product actually obtained after each group of reactions is first calculated using the reaction purity data, which can be achieved by multiplying the total mass of the mixture after the reaction by the purity percentage of the product. For example, if the total mass of the mixture after the reaction is M grams and the purity of the target product is P%, the actual product mass is M×P% grams. The raw material conversion rate refers to the efficiency of the raw material conversion into the target product, usually expressed as a percentage. The calculation formula is: conversion rate = (consumed raw material amount / starting raw material amount) × 100%; wherein the amount of raw material consumed refers to the amount of raw material actually converted into the product in the reaction, and the amount of starting raw material refers to the total amount of raw material added at the beginning of the reaction. By summarizing the conversion rates of N groups of raw material formulas, a conversion rate data set can be obtained, and product conversion data can be obtained through the conversion rate data set. After obtaining the product conversion data, by comparing the conversion rates between different groups, it is possible to identify which raw material formulas or process conditions have a significant effect on the conversion rate. Formulas and conditions with high conversion rates should be regarded as optimization directions, while those with low conversion rates need to be further adjusted or optimized. Based on the product conversion data, analysis can be performed to understand the factors that affect the conversion rate, including reaction temperature, pressure, catalyst type and dosage, solvent selection, reaction time, etc.
[0112] S324, obtaining process defect data based on the product conversion data.
[0113] Exemplarily, first, the product conversion rate of each group of raw material formula is statistically analyzed, and the statistical quantities such as the mean value and standard deviation are calculated. By comparing the distribution of the conversion rates of each group, outliers that deviate significantly from the average level are identified. Outliers indicate that there are significant problems with certain raw material formulas or process conditions, such as incomplete reactions, increased side reactions, or raw material losses. Then, the product conversion rates of multiple batches of experiments are trend analyzed. By drawing the curve of the conversion rate changing with time, temperature, pressure or other key process parameters, it is observed whether there is an obvious downward trend, fluctuation or mutation point. The trend change reflects the decline in process stability, catalyst deactivation, equipment failure or operational errors, etc., which provides an important basis for the diagnosis of process defects. Then, the correlation analysis technology is used to check the relationship between the product conversion rate and various process parameters, and the factors that have a significant impact on the conversion rate are identified according to the correlation coefficient. For example, it is found that there is a strong positive relationship between the amount of a certain type of catalyst and the conversion rate, while a slight change in the reaction temperature has little effect on the conversion rate, which helps to determine the direction of process optimization and find the source of process defects. Based on the analysis results, combined with process principles, reaction mechanisms and practical experience, the types and causes of process defects are determined. For example, low conversion rate may be caused by insufficient raw material purity, decreased catalyst activity, improper reaction condition control (such as large temperature fluctuations, unstable pressure), or poor equipment sealing leading to raw material leakage. Finally, the obtained process defect data are sorted out, including information such as defect type, cause, impact range and severity, to obtain process defect data.
[0114] S330, inputting the process defect data into a defect analysis model for analysis to obtain defect cause data; wherein the defect cause data indicates that a preparation sequence of the new silane coupling agent is wrong.
[0115] For example, a multi-level defect analysis model can be designed; this model combines statistical learning, deep learning, and graph neural network technology to process and understand the complex correlations and nonlinear characteristics in process defect data. By training a large amount of historical process data and known defect cases, the model can identify the potential connection between different defect modes and preparation sequence errors. Before the process defect data is input into the model, data preprocessing is performed, including data cleaning (removing noise and outliers), data normalization (ensuring the comparability of data from different sources), and feature extraction (extracting features valuable for defect analysis from the original data). In the core part of the model, the long short-term memory network in the deep learning framework is used to process time series data to capture the dynamic relationship between the steps in the preparation process. The preparation process can be regarded as a complex network structure (nodes such as raw materials, intermediates, and reaction conditions are interconnected). The neural network structure is introduced to learn the information transmission rules between nodes to more accurately understand the impact of the preparation sequence on the quality of the final product and obtain defect cause data.
[0116] S340: Processing is performed based on the defect cause data to obtain second preparation data.
[0117] Exemplarily, the defect mapping technology is first used to map the specific problems in the defect cause data (such as wrong preparation order) to the corresponding links of the preparation process, which is similar to reverse engineering. Starting from the results, it is traced back to the source to clarify the specific location and scope of the problem. By constructing a detailed defect mapping diagram, it is possible to clearly understand which steps or conditions lead to the current defects, providing clear guidance for subsequent optimization work. Based on the results of defect mapping, the rule engine generates customized solutions. The rule engine is mainly responsible for screening and optimizing these suggestions, which can improve the final generated solution to be both in line with the actual situation and operational. Then, through virtual experiments and simulation verification technology, a virtual preparation environment for the new silane coupling agent is first established, in which the reaction process and results under different preparation conditions are simulated. The second preparation data is generated by continuously adjusting and optimizing the preparation parameters (such as temperature, pressure and reaction time).
[0118] In this way, the macro requirements are mapped to specific preparation process parameters, the accuracy of the preparation process is improved, and virtual simulation is performed using process simulation software. The changes in various parameters during the reaction process can be predicted, the predictability of the preparation process can be enhanced, the accuracy, predictability and safety of the preparation of new silane coupling agents can be improved, and the ability to identify and prevent process defects can be enhanced.
[0119] S400, optimizing according to the first preparation data and the second preparation data to obtain total preparation data; wherein the total preparation data is used to reflect the raw material formula parameters and preparation process parameters that can meet the target performance of the new silane coupling agent.
[0120] Exemplarily, the first and second preparation data are first subjected to strict preprocessing, including but not limited to data cleaning (removing outliers, filling missing values), normalization, and feature selection. A distribution random neighborhood embedding method can be used to retain the key information of the data while reducing the computational complexity. By contrastive analysis, the first preparation data and the second preparation data are compared in detail in terms of raw material formula parameters (e.g., silane type, catalyst concentration) and preparation process parameters (e.g., reaction temperature, stirring rate), and the key factors affecting the performance and their changing trends are identified using time series analysis or cluster analysis techniques. In view of the fact that the performance optimization of the new silane coupling agent often involves multiple objectives (e.g., strength, heat resistance, water resistance, etc.), a multi-objective optimization algorithm (e.g., NSGA-II) can be used to search for the optimal raw material formula and preparation process combination. The algorithm can simultaneously optimize multiple objective functions under given constraints to find a set of optimal solution sets. In the optimization process, molecular dynamics simulation or Monte Carlo simulation technology can be used to construct a three-dimensional model of the silane coupling agent molecule and simulate its reaction process under different conditions. By comparing the simulation results with the experimental data, the effectiveness of the optimization scheme is verified, and the model is iteratively adjusted to improve the prediction accuracy. Finally, the actual application effect of the total preparation data is verified through testing; if there is any deviation, the experimental data needs to be fed back and a new round of adjustment and optimization is carried out until the target performance requirements are met or exceeded.
[0121] In one possible implementation, see Fig. 9 S400: Analyze the first preparation data and the second preparation data to obtain total preparation data, including:
[0122] S410, performing data combination test according to the raw material formula indicated by the first preparation data and the preparation process indicated by the second preparation data to obtain complete preparation data; wherein the complete preparation data is used to reflect the raw material ratio and preparation sequence obtained after the preparation test.
[0123] Exemplarily, first, according to the raw material formula of the first preparation data, the basic ratio range is set, combined with the preparation process parameters of the second preparation data, an orthogonal test table containing key variables (such as raw material ratio, reaction temperature, stirring time, etc.) is designed, and the automated laboratory system is used to perform experimental operations to reduce human errors and improve experimental repeatability and precision. According to the customized experimental scheme, the automated laboratory system automatically weighs and mixes the raw materials according to the preset raw material ratio, and then reacts under the conditions of precise temperature, pressure, stirring rate, etc. During the whole process, the system collects key data such as reactant state, temperature, pressure, product generation in real time, provides detailed information for subsequent analysis, and then uses multidimensional data fusion analysis technology, combined with clustering, classification and regression algorithms in machine learning, to process the collected multidimensional experimental data. First, the experimental data is cleaned and normalized to eliminate dimensional differences and noise interference. Then the complex relationship between raw material ratio, preparation process parameters and product performance (such as strength, heat resistance, water resistance, etc.) is analyzed, and the key variables with the greatest impact on performance and their optimal combination are identified through gradient boosting tree, and principal component analysis is used to visualize the data distribution to assist in understanding the correlation and trend between variables. According to the data analysis results, several groups of optimized raw material formulas and preparation process combinations are selected for small-batch trial production, and comprehensive performance testing is carried out. At this time, complete preparation data is obtained.
[0124] S420, inputting the raw material ratio and preparation sequence indicated by the complete preparation data into the reaction path analysis model, and testing the reaction mechanism, intermediate products and structural performance between the raw materials through the reaction path analysis model.
[0125] It can be understood that the reaction path analysis model is a detailed mechanism for understanding how raw materials are converted into final products during complex chemical reactions. It can analyze the overall conversion of reactants to products, as well as all intermediate products that may be formed during the reaction, the stability of these intermediate products, the conversion relationship between them, and how they ultimately affect the structure and performance of the product. The reaction path analysis model is usually established by the reaction mechanism database, reaction kinetics model, quantum chemical calculation, kinetic simulation and path optimization module; the reaction mechanism database module is the basis of the model and contains the mechanism information of known chemical reactions, including the chemical bond breaking and formation rules between reactants, intermediates and products, energy changes, etc. These data can be derived from experimental data, theoretical calculations or literature reports. The reaction kinetics model module describes how the reaction rate changes with reaction conditions (such as temperature, pressure, catalyst type and concentration); it is mainly based on factors such as reactant concentration, activation energy, and pre-exponential factor, and can predict the reaction speed and product distribution under different conditions. The quantum chemical calculation module is for unknown or complex reaction mechanisms. Quantum chemical calculation methods (such as density functional theory DFT) can be used to calculate key parameters such as energy changes, transition state structures, bond energies, etc. on the reaction path to reveal the microscopic details of the reaction. The kinetic simulation module and the path optimization module combine the reaction mechanism and kinetic model, dynamically simulate the reaction process through simulation software, and optimize the reaction conditions to find the most efficient and economical reaction path. This includes adjusting the raw material ratio, reaction temperature, catalyst type, etc. to maximize the yield and selectivity of the target product.
[0126] Method of use in the present embodiment: First, the raw material ratio and preparation order in the complete preparation data are refined and converted into an input format recognizable by the model. Then, quantum chemical calculation software (such as Gaussian, VASP, etc.) is used to simulate the key reaction steps, and the electronic structure, energy state and other microscopic information of the reactants, intermediates and products are obtained. The microscopic information is used as the training data of the machine learning model, combined with the existing chemical knowledge base, to build a model that can predict the reaction path and product performance, select similar systems of some known reaction paths as verification sets, input the complete preparation data into the model for prediction, and compare and analyze with the actual results. By adjusting the model parameters (such as the number of neural network layers, learning rate, regularization coefficient, etc.), the model performance is optimized until a satisfactory prediction accuracy is reached. At the same time, parameter sensitivity analysis is carried out to identify the key parameters that have the greatest impact on the model's prediction results, provide guidance for subsequent analysis, input the complete preparation data into the reaction path analysis model, and start the dynamic simulation program. The model simulates the concentration changes, reaction rate and reaction mechanism of each species during the reaction according to the raw material ratio, temperature, pressure and other conditions. Combined with molecular dynamics simulation, it deeply explores the interactions between reactant molecules, bond breaking and formation and other microscopic processes, predicts and verifies the existence and structure of intermediates, and uses X-ray diffraction, infrared spectroscopy and other characterization methods to perform structural analysis on the actual prepared products. The analysis results are compared with the structural performance predicted by the model to obtain the test situation.
[0127] S430, obtaining total preparation data according to simulation test analysis of the reaction path analysis model.
[0128] Exemplarily, first, various data (such as reaction rate constant, intermediate product structure, product performance parameters, etc.) output by the reaction path analysis model are integrated with experimental characterization data (such as XRD spectrum, SEM image, etc.) to form a unified data format, and unsupervised learning techniques (such as cluster analysis or isolation forest, etc.) are used to detect and remove outliers or inconsistent information in the data to improve the cleanliness and accuracy of the data. Graph theory and network analysis methods are used to treat data points such as raw material ratio, preparation conditions, reaction path, intermediate products and final product performance as nodes in the network, and the interaction relationship between them is regarded as an edge. By calculating the similarity, correlation and other indicators between nodes, a multi-dimensional data association network is constructed, which not only shows the direct connection between data, but also reveals hidden, nonlinear dependencies, providing intuitive visual support for subsequent data analysis, and then combining deep learning algorithms (which can be recurrent neural networks RNN) and data mining techniques (which can be association rule mining) to analyze the integrated data, identify the complex mapping relationship between raw material ratios, preparation conditions and product performance, and organize the data to obtain total preparation data.
[0129] This setting can identify the key variables that have the greatest impact on product performance and their optimal combination, thereby improving the accuracy of product performance prediction at the design stage. By introducing the reaction path analysis model, R&D personnel can deeply explore the reaction mechanism between raw materials, the formation and transformation of intermediates, and how these processes affect the structure and performance of the final product. This deep understanding of the reaction mechanism provides a basis for further optimizing reaction conditions and improving yield and selectivity.
[0130] S500, controlling the silane coupling agent preparation device to prepare a new silane coupling agent according to the total preparation data.
[0131] For example, the key parameters in the total preparation data (such as the best raw material ratio, the best reaction temperature, the best reaction time, etc.) are input into the adaptive optimization algorithm. The algorithm optimizes the best preparation scheme through iterative calculations and considering the interaction between different parameters. The parameters are transmitted to the control device through the Internet of Things technology. During the preparation process, the silane coupling agent preparation device continuously monitors the change trend of key parameters and reports abnormal situations to the operator in a timely manner through a real-time feedback mechanism, so that the silane coupling agent preparation device can successfully prepare a new silane coupling agent.
[0132] With such a configuration, the most suitable preparation raw materials can be optimized through the above description, and different preparation schemes can be customized according to the diverse needs of the market. The preparation raw materials selected after optimization can identify the key preparation parameters that affect the performance of the silane coupling agent, improve the reaction rate between the raw materials when preparing the silane coupling agent, increase the purity of the silane coupling agent, produce a silane coupling agent with specific properties, and enhance the stability of the quality of the silane coupling agent product.
[0133] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0134] Corresponding to the novel silane coupling agent preparation method described in the above embodiment, the embodiment of the present application also provides a novel silane coupling agent preparation system, and each unit of the system can implement each step of the novel silane coupling agent preparation method. Fig.10 The structural block diagram of the novel silane coupling agent preparation system provided in the embodiment of the present application is shown. For the convenience of explanation, only the part related to the embodiment of the present application is shown.
[0135] Reference Fig.10 , the new silane coupling agent preparation system includes:
[0136] An acquisition unit, used for acquiring preparation requirement data of the new silane coupling agent to be prepared; wherein the preparation requirement data is used for reflecting the target performance that the new silane coupling agent to be prepared needs to meet;
[0137] A first analysis unit is used to perform data analysis based on the preparation demand data to obtain first preparation data; wherein the first preparation data is used to reflect the raw material formula required for preparing the new silane coupling agent;
[0138] A second analysis unit is used to perform analysis based on the first preparation data to obtain second preparation data; wherein the second preparation data is used to reflect the preparation process required for preparing the new silane coupling agent;
[0139] A result unit is used to optimize according to the first preparation data and the second preparation data to obtain total preparation data; wherein the total preparation data is used to reflect the raw material formula parameters and preparation process parameters that can meet the target performance of the new silane coupling agent;
[0140] The control unit is used for controlling the silane coupling agent preparation device to prepare a new silane coupling agent according to the total preparation data.
[0141] It should be noted that the information interaction, execution process, etc. between the above-mentioned systems / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.
[0142] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0143] The present application also provides a silane coupling agent preparation device. Fig.11 This is a schematic diagram of the structure of a control device for a silane coupling agent preparation device provided in one embodiment of the present application. Fig.11 As shown, the control device 6 of the silane coupling agent preparation equipment of this embodiment includes: at least one processor 60 ( Fig.11 Only one is shown), at least one memory 61 ( Fig.11 Only one is shown in the figure) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the control device 6 of the silane coupling agent preparation device implements the steps in any of the above-mentioned novel silane coupling agent preparation method embodiments, or the control device 6 of the silane coupling agent preparation device implements the functions of each module / unit in the above-mentioned device embodiments.
[0144] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 62 in the control device 6 of the silane coupling agent preparation device.
[0145] The control device 6 of the silane coupling agent preparation device can be a computing device such as a desktop computer, a notebook, a palm computer, etc. The control device of the silane coupling agent preparation device can include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that Fig.11 The control device 6 of the silane coupling agent preparation device is merely an example and does not constitute a limitation on the control device 6 of the silane coupling agent preparation device. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, buses, etc.
[0146] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0147] In some embodiments, the memory 61 may be an internal storage unit of the control device 6 of the silane coupling agent preparation device, such as a hard disk or memory of the control device 6 of the silane coupling agent preparation device. In other embodiments, the memory 61 may also be an external storage device of the control device 6 of the silane coupling agent preparation device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the control device 6 of the silane coupling agent preparation device. Further, the memory 61 may also include both an internal storage unit and an external storage device of the control device 6 of the silane coupling agent preparation device. The memory 61 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 61 may also be used to temporarily store data that has been output or is to be output.
[0148] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.
[0149] An embodiment of the present application provides a computer program product. When the computer program product is run on a silane coupling agent preparation device, the silane coupling agent preparation device is enabled to implement the steps in any of the above-mentioned method embodiments.
[0150] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device capable of carrying the computer program code to the silane coupling agent preparation device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, U disk, mobile hard disk, disk or optical disk, etc. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.
[0151] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0152] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0153] In the embodiments provided in the present application, it should be understood that the disclosed novel silane coupling agent preparation method, system and equipment can be implemented in other ways. For example, the above-described silane coupling agent preparation system and equipment embodiments are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0154] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0155] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A novel method for preparing a silane coupling agent, It is characterized in that include: Acquiring preparation requirement data of a new type of silane coupling agent to be prepared; wherein the preparation requirement data is used to reflect the target performance that the new type of silane coupling agent to be prepared needs to meet; Performing data analysis based on the preparation demand data to obtain first preparation data; wherein the first preparation data is used to reflect the raw material formula required for preparing the new silane coupling agent; Performing analysis and processing based on the first preparation data to obtain second preparation data; wherein the second preparation data is used to reflect the preparation process required for preparing the novel silane coupling agent; Optimizing the first preparation data and the second preparation data to obtain total preparation data; wherein the total preparation data is used to reflect the raw material formula parameters and preparation process parameters that can meet the target performance of the new silane coupling agent; According to the total preparation data, the silane coupling agent preparation device is controlled to prepare the novel silane coupling agent.
2. The method for preparing the novel silane coupling agent according to claim 1, It is characterized in that The performing data analysis based on the preparation requirement data to obtain first preparation data includes: Acquiring application data of the novel silane coupling agent; wherein the application data is used to reflect the usage scenario of the novel silane coupling agent to be prepared; Prompt data is obtained after matching the application data and the preparation requirement data; wherein the prompt data is used to reflect the preparation formula that meets the use scenario and target performance of the new silane coupling agent; The preparation formula indicated by the prompt data is optimized to obtain the first preparation data.
3. The method for preparing the novel silane coupling agent according to claim 2, It is characterized in that The step of optimizing the preparation formula indicated by the prompt data to obtain the first preparation data includes: The silane coupling agent preparation device is responsive to the preparation operation of the novel silane coupling agent; Extracting the prompt data to obtain formula data; wherein the formula data is used to reflect the types of raw materials used for the new silane coupling agent to be prepared, the types of raw materials include silicon hydride compounds and organic functional groups, and the silicon hydride compounds and the organic functional groups can undergo chemical reactions; Determine the raw materials based on the silicon hydride and the organic functional group indicated by the formula data to obtain raw material data; wherein the raw material data is used to indicate the raw materials in the silicon hydride and the raw materials in the organic functional group, the raw materials in the silicon hydride include epoxy silane, amino silane and hydroxy silane, and the raw materials in the organic functional group include carboxylic acid, ketone and amine; The raw material data is optimized according to a genetic algorithm to obtain the first preparation data.
4. The method for preparing the novel silane coupling agent according to claim 3, It is characterized in that The step of optimizing the raw material data according to a genetic algorithm to obtain the first preparation data includes: Extracting at least two raw materials of the epoxy silane, the amino silane and the hydroxy silane in the silicon hydrogen compound indicated by the raw material data to obtain first raw material data; extracting at least two raw materials of the carboxylic acid, the ketone and the amine in the organic functional group indicated by the raw material data to obtain second raw material data; At least two of the epoxy silane, the amino silane and the hydroxy silane indicated by the first raw material data are randomly and one-to-one combined with at least two of the carboxylic acid, the ketone and the amine indicated by the second raw material data to obtain a plurality of raw material combination data; wherein one raw material combination data indicates one raw material combination, and each raw material combination includes any one of the epoxy silane, the amino silane and the hydroxy silane in the silicon hydrogen compound and any one of the carboxylic acid, the ketone and the amine in the organic functional group; A fitness calculation is performed based on the raw material combination data to obtain the first preparation data.
5. The method for preparing the novel silane coupling agent according to claim 4, It is characterized in that The step of performing fitness calculation based on the raw material combination data to obtain the first preparation data includes: Analyze each of the raw material combination data to obtain reaction data; wherein the reaction data is used to reflect the reactant ratio and reaction conditions between any one of the epoxy silane, the amino silane and the hydroxy silane and any one of the carboxylic acid, the ketone and the amine, the reactant ratio is the ratio value between any one of the epoxy silane, the amino silane and the hydroxy silane and any one of the carboxylic acid, the ketone and the amine, the reaction conditions include the reaction temperature value and the reaction time, and each of the raw material combination data corresponds to each of the reaction data; Encoding the reactant ratio and the reaction condition indicated by each reaction data respectively to obtain coded reaction data; wherein the coded reaction data includes reactant ratio coded data and reaction condition coded data; According to the coded reaction data, any one of the raw materials of the epoxy silane, the amino silane and the hydroxy silane is subjected to an experimental operation with any one of the raw materials of the carboxylic acid, the ketone and the amine to obtain reaction performance data; wherein the reaction performance data is used to reflect the performance index value after the simulated experimental reaction of any one of the raw materials of the epoxy silane, the amino silane and the hydroxy silane with any one of the raw materials of the carboxylic acid, the ketone and the amine; The reaction performance data is calculated and converted to obtain adaptation data; wherein the adaptation data is used to reflect the fitness value adapted to the usage scenario; Screening all the coded reaction data according to the requirements of the adaptation data to obtain a plurality of satisfying data; wherein the satisfying data is the coded reaction data that meets the requirements of the adaptation data, and all the coded reaction data has a plurality of coded reaction data that meets the requirements of the adaptation data; A reorganization operation is performed on a plurality of the satisfying data to obtain the first prepared data.
6. The method for preparing the novel silane coupling agent according to claim 5, It is characterized in that The step of reorganizing the plurality of satisfying data to obtain the first prepared data includes: Respectively extracting the reactant ratio coding data and the reaction condition coding data in the plurality of satisfying data; Randomly shuffle the plurality of reactant ratio coding data extracted from the satisfying data, randomly shuffle the plurality of reaction condition coding data extracted from the satisfying data, and then combine the shuffled plurality of reactant ratio coding data and the shuffled plurality of reaction condition coding data one by one according to the usage scenario; After the combination is completed, based on the preparation requirement data and the usage scenario, the ratio values in the plurality of reactant ratio coding data are randomly changed, and the reaction parameter values in the plurality of reaction condition coding data are randomly changed, and the changed reactant ratio coding data and the reaction condition coding data are recombined to obtain reaction optimization data; wherein the reaction optimization data is the reactant ratio coding data and the reaction condition coding data after optimization adjustment; Based on the reaction optimization data, the first preparation data is obtained.
7. The method for preparing the novel silane coupling agent according to claim 6, It is characterized in that The analyzing and processing based on the first preparation data to obtain second preparation data includes: Analyzing the preparation demand data to obtain process preparation data; wherein the process preparation data is used to reflect the preparation process of the novel silane coupling agent; The first preparation data is subjected to preparation test through the process preparation data to obtain process defect data; wherein the process defect data is used to reflect the situation that the organic functional group obtained after the preparation test of the first preparation data has not reacted completely in the silicon hydride compound and impurities exist; Inputting the process defect data into a defect analysis model for analysis to obtain defect cause data; wherein the defect cause data indicates that the preparation sequence of the new silane coupling agent is wrong; Processing is performed based on the defect cause data to obtain second preparation data.
8. The method for preparing the novel silane coupling agent according to claim 1, It is characterized in that The analyzing the first preparation data and the second preparation data to obtain total preparation data includes: Performing a data combination test according to the raw material formula indicated by the first preparation data and the preparation process indicated by the second preparation data to obtain complete preparation data; wherein the complete preparation data is used to reflect the raw material ratio and preparation sequence obtained after the preparation test; Inputting the raw material ratio and preparation sequence indicated by the complete preparation data into a reaction path analysis model, and testing the reaction mechanism, intermediate products and structural properties between the raw materials through the reaction path analysis model; According to the simulation test analysis of the reaction path analysis model, the overall preparation data is obtained.
9. The method for preparing the novel silane coupling agent according to claim 7, It is characterized in that The step of performing a preparation test on the first preparation data through the process preparation data to obtain process defect data includes: Dividing the raw material formula indicated by the first preparation data into N groups of raw material formulas according to the preparation process, and then prioritizing the N groups of raw material formulas according to the preparation sequence and reaction conditions; wherein N is a positive integer; Inputting N groups of raw material formulas into the process preparation data in order according to priority to perform process preparation test, and obtaining reaction purity data; wherein the reaction purity data is used to reflect the purity value of the new silane coupling agent obtained after the preparation test; The conversion rate is calculated based on the reaction purity data to obtain product conversion data; wherein the product conversion data is used to reflect the product conversion status of the N groups of raw material formulas into products; Process defect data are obtained based on the product conversion data.
10. A new silane coupling agent preparation system, It is characterized in that include: An acquisition unit, used for acquiring preparation requirement data of a new type of silane coupling agent to be prepared; wherein the preparation requirement data is used for reflecting the target performance that the new type of silane coupling agent to be prepared needs to meet; A first analysis unit, configured to perform data analysis based on the preparation requirement data to obtain first preparation data; wherein the first preparation data is used to reflect the raw material formula required for preparing the novel silane coupling agent; A second analysis unit is used to perform analysis and processing based on the first preparation data to obtain second preparation data; wherein the second preparation data is used to reflect the preparation process required for preparing the new silane coupling agent; A result unit is used to optimize according to the first preparation data and the second preparation data to obtain total preparation data; wherein the total preparation data is used to reflect the raw material formula parameters and preparation process parameters that can meet the target performance of the new silane coupling agent; A control unit is used to control the silane coupling agent preparation device to prepare the novel silane coupling agent according to the total preparation data.
Citation Information
Cited By
Preparation method of metal oxide nanoparticles with controllable particle size
CN119785902A
Method for automatically optimizing continuous intelligent microreactor platform
CN120493571A
A method for automatically optimizing a continuous intelligent microreactor platform
CN120493571B
