Oxygen production molecular sieve structure optimization method and device for improving thermal stability

Abstract

The invention discloses an oxygen generation molecular sieve structure optimization method and device for improving thermal stability, and relates to the technical field of intelligent optimizing.The method comprises the steps that preset structure features are read, multi-dimensional features of an oxygen generation molecular sieve are collected, physical structure features are screened out, a three-dimensional simulation model is constructed, and an optimal physical structure is obtained; and establishing a database and carrying out supervised learning to obtain an oxygen production fitness prediction model, determining an optimal structure through the model, and forming an optimal structure scheme of the oxygen production molecular sieve by the optimal physical structure and the optimal structure. The technical problem of poor oxygen production efficiency caused by poor thermal stability of the existing oxygen production molecular sieve is solved, and the technical effects of improving the thermal stability and further improving the oxygen production efficiency and quality are achieved by bilateral optimization of the physical and chemical structures of the molecular sieve and optimization of the physical structure to prevent impurities from entering oxygen due to structural collapse to influence the quality.

Description

Technical Field

[0001] The present application relates to the field of intelligent optimization technologies, and particularly to an optimization method and device for the structure of oxygen - producing molecular sieves for improving thermal stability. Background Art

[0002] In the application scenario of oxygen - producing molecular sieves, the demand for high - quality oxygen is increasing continuously. The problem that the thermal stability of oxygen - producing molecular sieves is poor, which in turn leads to poor oxygen - production efficiency, is particularly prominent. Efficiently optimizing the structure of oxygen - producing molecular sieves to improve thermal stability and oxygen - production efficiency has become a crucial link in solving the key problems of oxygen production. Traditional optimization methods for oxygen - producing molecular sieves are often relatively single, only focusing on one aspect of the physical structure or chemical structure, lacking sufficient control over the overall structure of the molecular sieve, and lacking systematic analysis and utilization. It is difficult to comprehensively and accurately improve the thermal stability and oxygen - production efficiency of the molecular sieve, and there are unreasonable situations in structure optimization, resulting in the inability to well meet the demand for high - quality oxygen in different application scenarios.

[0003] In the current related technologies, there are technical problems such as poor thermal stability of oxygen - producing molecular sieves, which in turn leads to poor oxygen - production efficiency. Summary of the Invention

[0004] The present application provides an optimization method and device for the structure of oxygen - producing molecular sieves for improving thermal stability. By reading predetermined structural features to collect multi - dimensional features of oxygen - producing molecular sieves, screening out physical - structure features, constructing a three - dimensional simulation model to obtain the optimal physical structure, forming a database with this and performing supervised learning to obtain an oxygen - production fitness prediction model, determining the optimal chemical structure through this model, and the optimal physical structure and the optimal chemical structure form the optimal structure scheme of the oxygen - producing molecular sieve, realizing the bilateral optimization of the physical and chemical structures of the molecular sieve. At the same time, the optimization of the physical structure avoids the entry of impurities into oxygen due to structure collapse, affecting the quality, achieving the technical effect of improving thermal stability, and thus improving oxygen - production efficiency and quality.

[0005] The present application provides an optimization method for the structure of oxygen - producing molecular sieves for improving thermal stability, including: Read the predetermined structural features, collect multi-dimensional features of the oxygen-producing molecular sieve based on the predetermined structural features, and obtain the molecular sieve structure feature information; traverse and screen the predetermined physical structure features in the molecular sieve structure feature information to obtain the physical structure feature information, where the physical structure feature information includes pore channel feature information and crystal grain feature information; construct a three-dimensional simulation model of the oxygen-producing molecular sieve according to the pore channel feature information and the crystal grain feature information, and perform simulation analysis on the three-dimensional simulation model in combination with the thermal simulation optimization plan to obtain the optimal physical structure; form a molecular sieve structure database with the optimal physical structure as a constraint, and extract the first structure data group from the molecular sieve structure database; perform supervised learning and testing on the first training data group formed based on the first chemical structure feature information and the first oxygen production fitness in the first structure data group to obtain an oxygen production fitness prediction model; perform prediction analysis on any chemical structure feature information through the oxygen production fitness prediction model to obtain any oxygen production fitness; when the any oxygen production fitness reaches a predetermined fitness threshold, use the any chemical structure feature information at that time as the optimal chemical structure; the optimal physical structure and the optimal chemical structure form the optimal structure plan of the oxygen-producing molecular sieve.

[0006] The present application also provides an oxygen-producing molecular sieve structure optimization device for improving thermal stability, including: A structural feature information acquisition module, which is used to read predetermined structural features and collect multi-dimensional features of the oxygen-making molecular sieve based on the predetermined structural features to obtain molecular sieve structural feature information; a physical structural feature information acquisition module, which is used to traverse and screen the predetermined physical structural features in the molecular sieve structural feature information to obtain physical structural feature information, and the physical structural feature information includes pore channel feature information and crystal grain feature information; a simulation analysis module, which is used to construct a three-dimensional simulation model of the oxygen-making molecular sieve according to the pore channel feature information and the crystal grain feature information, and perform simulation analysis on the three-dimensional simulation model in combination with a thermal simulation optimization plan to obtain an optimal physical structure; a molecular sieve structure database construction module, which is used to form a molecular sieve structure database with the optimal physical structure as a constraint and extract the first structure data group from the molecular sieve structure database; a prediction model acquisition module, which is used to perform supervised learning and verification on a first training data group formed based on the first chemical structural feature information and the first oxygen-making fitness in the first structure data group to obtain an oxygen-making fitness prediction model; a prediction analysis module, which is used to perform prediction analysis on any chemical structural feature information through the oxygen-making fitness prediction model to obtain any oxygen-making fitness; an optimal chemical structure acquisition module, which is used to use the any chemical structural feature information at that time as the optimal chemical structure when the any oxygen-making fitness reaches a predetermined fitness threshold; an optimal structure scheme acquisition module, which is used to form the optimal structure scheme of the oxygen-making molecular sieve with the optimal physical structure and the optimal chemical structure.

[0007] It is intended to propose a method and device for optimizing the structure of an oxygen-making molecular sieve for improving thermal stability through this application. First, read the predetermined structural features to collect multi-dimensional features of the oxygen-making molecular sieve, screen out the physical structural features, construct a three-dimensional simulation model to obtain the optimal physical structure, form a database with this and perform supervised learning to obtain an oxygen-making fitness prediction model, determine the optimal chemical structure through this model, and the optimal physical structure and the optimal chemical structure form the optimal structure scheme of the oxygen-making molecular sieve. By optimizing the physical and chemical structures of the molecular sieve bilaterally, and at the same time, the optimization of the physical structure avoids the structure collapse and impurities entering the oxygen, which affects the quality, achieving the technical effect of improving the thermal stability, and further improving the oxygen-making efficiency and quality. Brief Description of the Drawings

[0008] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the devices according to the embodiments of the present application. It should be understood that the operations above or below do not necessarily need to be performed precisely in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0009] Figure 1 It is a schematic flowchart of the method for optimizing the structure of an oxygen-producing molecular sieve for improving thermal stability provided by the embodiments of the present application; Figure 2 It is a schematic structural diagram of the device for optimizing the structure of an oxygen-producing molecular sieve for improving thermal stability provided by the embodiments of the present application.

[0010] Explanation of reference numerals: Structure feature information acquisition module 10, Physical structure feature information acquisition module 20, Simulation analysis module 30, Molecular sieve structure database construction module 40, Prediction model acquisition module 50, Prediction analysis module 60, Optimal chemical structure acquisition module 70, Optimal structure scheme acquisition module 80. Detailed implementation manners

[0011] The above description is only an overview of the technical solutions of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0012] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0013] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict. The terms "first" and "second" are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product or server that includes a series of steps or units need not be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0014] The embodiments of this application provide a method for optimizing the structure of oxygen-making molecular sieves to improve thermal stability, as Figure 1 shown. The method includes: Step S100, reading predetermined structural features and collecting multi-dimensional features of the oxygen-making molecular sieve based on the predetermined structural features to obtain molecular sieve structure feature information. Specifically, in the oxygen-making molecular sieve structure optimization scheme, the predetermined structural features are first determined, and the features cover physical properties, chemical properties and performance indicators. A variety of technologies are used for multi-dimensional feature collection, including observing the microstructure using an electron microscope, determining the crystal structure by X-ray diffraction, measuring the pore conditions by nitrogen adsorption-desorption testing, determining the chemical composition by elemental analysis, identifying functional groups by infrared spectroscopy, studying the thermal stability by thermogravimetric analysis, and performing thermal stability and oxygen-making performance tests, etc. A large amount of information is collected, and after sorting and analysis, a molecular sieve structure feature information set is formed, including physical structure, chemical structure and performance characteristics, etc., providing comprehensive data support for subsequent screening and optimization.

[0015] Step S200, traversing and screening the predetermined physical structure features in the molecular sieve structure feature information to obtain physical structure feature information, where the physical structure feature information includes pore channel feature information and crystal grain feature information. Specifically, in the oxygen-making molecular sieve structure optimization, the predetermined physical structure features are traversed and screened in the molecular sieve structure feature information. The predetermined physical structure features are determined based on performance expectations, theories and experiences, including a specific pore channel size range, structure, and crystal grain size range and morphology, etc. The crystal structure data, pore distribution data and electron microscope image data, etc. in the molecular sieve structure feature information are analyzed, and the required pore channel feature information and crystal grain feature information are screened out, providing an important basis for subsequent construction of a three-dimensional simulation model and further optimization of the molecular sieve structure.

[0016] In a possible implementation, the predetermined physical structure features are traversed and screened in the molecular sieve structure feature information to obtain physical structure feature information, where the physical structure feature information includes pore feature information and crystal grain feature information. Step S200 further includes step S210, and the pore feature information includes pore size and pore structure. Specifically, in the optimized molecular sieve structure for oxygen production, the pore feature information includes pore size and pore structure. Appropriate pore size has an important impact on oxygen production efficiency and oxygen purity. The crystal structure can be inferred by X-ray diffraction analysis to estimate the size range, the pore size can be inferred by nitrogen adsorption-desorption test to obtain the pore size distribution, and the molecular simulation technology can be used to predict the diffusion behavior of oxygen molecules to determine the size range. Different pore structures affect the diffusion path and adsorption sites of gas molecules. Electron microscopy technology can be used to observe the pore morphology and connection mode, and nuclear magnetic resonance technology can be used to study the internal pore structure. Theoretical calculation methods such as density functional theory can be used to simulate the adsorption and diffusion behavior of oxygen molecules to optimize the pore structure. Pore features are crucial for the performance of the oxygen-producing molecular sieve, and a variety of technologies need to be comprehensively used to accurately determine and optimize them.

[0017] In a possible implementation, the predetermined physical structure features are traversed and screened in the molecular sieve structure feature information to obtain physical structure feature information, where the physical structure feature information includes pore feature information and crystal grain feature information. Step S200 further includes step S220, and the crystal grain feature information includes crystal grain size and crystal grain morphology. Specifically, in the optimized molecular sieve structure for oxygen production, the crystal grain feature information includes crystal grain size and crystal grain morphology. Smaller crystal grains usually have a larger specific surface area, which can increase the contact opportunity with gas molecules, improve the oxygen production efficiency, and may have better thermal stability. To determine the crystal grain size, electron microscopy technology can be used to observe and measure the sizes of multiple crystal grains to obtain the distribution, or the average size can be calculated by the Scherrer formula according to the peak width of the X-ray diffraction pattern. The crystal grain morphology affects the packing mode and pore structure of the molecular sieve. Spherical crystal grains may be easily packed tightly to form a uniform pore structure, which is conducive to the diffusion and adsorption of gas molecules. Irregularly shaped crystal grains will increase the complexity of the pore structure and affect the transmission efficiency. Electron microscopy technology can be used to observe the crystal grain shape, and image processing technology can be used to extract the morphological characteristic parameters. The influence of the crystal grain morphology on the pore structure can also be indirectly inferred through physical property tests. Crystal grain features are important for the performance of the oxygen-producing molecular sieve. Accurately determining the crystal grain size and analyzing the crystal grain morphology can provide a basis for optimizing the structure, improving the thermal stability and oxygen production efficiency.

[0018] Step S300: Construct a three-dimensional simulation model of the oxygen-producing molecular sieve based on the pore channel characteristic information and the crystal grain characteristic information, and perform simulation analysis on the three-dimensional simulation model in combination with the thermal simulation optimization plan to obtain the optimal physical structure. Specifically, in the oxygen-producing molecular sieve structure optimization plan, first, based on the pore channel characteristic information (including pore channel size and structure) and the crystal grain characteristic information (including crystal grain size and morphology), use computer modeling software to construct a three-dimensional simulation model of the oxygen-producing molecular sieve to ensure that the model truly reflects the actual physical structure. Design a thermal simulation optimization plan, including an extremely high temperature simulation plan and a cyclic temperature simulation plan. Perform simulation analysis on the three-dimensional simulation model for a predetermined duration at the first extremely high temperature and the first cyclic temperature respectively, obtain the corresponding simulation records and extract the simulation physical structure characteristic information, calculate the simulation structure collapse index by comparing with the initial physical structure characteristic information. When both indices meet the predetermined collapse index threshold, use the physical structure characteristic information at this time as the optimal physical structure; if not, issue an adjustment instruction, adjust the pore channel and crystal grain characteristics, then construct the first optimized three-dimensional simulation model and perform iterative optimization until the required optimal physical structure is found, laying a foundation for improving the thermal stability and oxygen production efficiency of the molecular sieve.

[0019] In a possible implementation manner, to construct a three-dimensional simulation model of the oxygen-producing molecular sieve based on the pore channel characteristic information and the crystal grain characteristic information, and perform simulation analysis on the three-dimensional simulation model in combination with the thermal simulation optimization plan to obtain the optimal physical structure, step S300 further includes step S310, and the thermal simulation optimization plan includes an extremely high temperature simulation plan and a cyclic temperature simulation plan. Specifically, the purpose of the thermal simulation optimization plan is to comprehensively evaluate the structural stability and performance of the oxygen-producing molecular sieve under different extreme temperature conditions. The extremely high temperature simulation plan aims at the extremely high temperature situation that the molecular sieve may face during the oxygen production process, simulating the structural changes and performance degradation risks under extreme conditions. The cyclic temperature simulation plan focuses on the temperature changes brought about by the repeated adsorption and desorption cycles during the oxygen production process to evaluate the thermal stability and service life of the molecular sieve under such cyclic conditions.

[0020] Step S320: According to the extremely high temperature simulation plan, perform simulation analysis on the three-dimensional simulation model for a predetermined duration at the first extremely high temperature to obtain the first extremely high temperature simulation record. Specifically, use professional thermal simulation software to place the constructed three-dimensional simulation model of the oxygen-producing molecular sieve in the set first extremely high temperature environment. The first extremely high temperature is determined based on the highest temperature that may occur during the oxygen production process to simulate the most severe thermal environment. Perform simulation analysis on the model for a predetermined duration through the software, simulating physical processes such as heat conduction, molecular movement, and structural changes of the molecular sieve at high temperatures. During the simulation process, the software will record information such as the temperature distribution, stress distribution, and structural deformation of the molecular sieve at different time points, and finally obtain the first extremely high temperature simulation record.

[0021] Step S330: Obtain the first simulated physical structure characteristic information corresponding to the predetermined duration in the first extremely high temperature simulation record. Specifically, from the first extremely high temperature simulation record, extract the first simulated physical structure characteristic information of the molecular sieve at the end of the predetermined duration. The characteristic information includes, but is not limited to, changes in pore size, degree of pore structure distortion, changes in crystal grain size, changes in crystal grain morphology, etc. Through a specific data extraction algorithm, accurately extract this physical structure characteristic information from the large amount of data in the simulation record.

[0022] Step S340: Compare the physical structure characteristic information with the first simulated physical structure characteristic information to obtain the first simulated structure collapse index. Specifically, use a structure similarity evaluation algorithm, such as the root mean square error algorithm or the structural similarity index algorithm, to compare the initial physical structure characteristic information with the first simulated physical structure characteristic information. By calculating the degree of difference between the two, obtain the first simulated structure collapse index. The index reflects the degree of structural change of the molecular sieve at extremely high temperatures relative to the initial state. The larger the index value, the higher the risk of structural collapse.

[0023] Step S350: According to the cyclic temperature simulation scheme, perform a simulation analysis of the three-dimensional simulation model at the first cyclic temperature for the predetermined duration to obtain the first cyclic temperature simulation record. Specifically, also use thermal simulation software to set a representative cyclic temperature change range to simulate the repeated temperature changes during the oxygen production process. Place the three-dimensional simulation model in this cyclic temperature environment for a simulation analysis of the predetermined duration. The software will record various physical parameter changes of the molecular sieve at different temperature stages to obtain the first cyclic temperature simulation record.

[0024] Step S360: Obtain the second simulated physical structure characteristic information corresponding to the predetermined duration in the first cyclic temperature simulation record. Specifically, similar to the steps of the extremely high temperature simulation, extract the second simulated physical structure characteristic information of the molecular sieve at the end of the predetermined duration from the first cyclic temperature simulation record, including the structural changes of the pores and crystal grains under the cyclic temperature.

[0025] Step S370: Compare the physical structure characteristic information with the second simulated physical structure characteristic information to obtain the second simulated structure collapse index. Specifically, apply a structure similarity evaluation algorithm to compare the initial physical structure characteristic information with the second simulated physical structure characteristic information and calculate the second simulated structure collapse index to measure the structural stability of the molecular sieve under the cyclic temperature.

[0026] Step S380: When both the first simulation structure collapse index and the second simulation structure collapse index meet the predetermined collapse index threshold, use the physical structure feature information corresponding to the three-dimensional simulation model as the optimal physical structure. Specifically, based on past experimental data and experience, determine a reasonable predetermined collapse index threshold. When the calculated first simulation structure collapse index and second simulation structure collapse index are both less than this threshold, it indicates that the molecular sieve has good structural stability at extremely high temperatures and cyclic temperatures. At this time, determine the physical structure feature information corresponding to the three-dimensional simulation model as the optimal physical structure. The optimal physical structure performs excellently in terms of thermal stability and can meet the requirements for the thermal stability of the molecular sieve during the oxygen production process, thereby improving the oxygen production efficiency and the service life of the molecular sieve.

[0027] In a possible implementation, when both the first simulation structure collapse index and the second simulation structure collapse index meet the predetermined collapse index threshold, use the physical structure feature information corresponding to the three-dimensional simulation model as the optimal physical structure. Step S380 further includes step S381: When either the first simulation structure collapse index or the second simulation structure collapse index does not meet the predetermined collapse index threshold, issue a physical structure adjustment instruction. Specifically, after obtaining the first simulation structure collapse index and the second simulation structure collapse index, compare them with the predetermined collapse index threshold. If any one of the indices does not meet the threshold requirement, it means that the current three-dimensional simulation model of the oxygen production molecular sieve has insufficient structural stability under extremely high temperature or cyclic temperature conditions. The system will automatically issue a physical structure adjustment instruction, which is a trigger signal for the subsequent optimization process, indicating that the physical structure of the molecular sieve needs to be adjusted to improve its thermal stability.

[0028] Step S382: Based on the physical structure adjustment instruction, adjust the pore channel feature information and the crystal grain feature information, and construct a first optimized three-dimensional simulation model according to the adjustment results. Specifically, once the physical structure adjustment instruction is received, the system will adjust the pore channel feature information and the crystal grain feature information according to the instruction. For the pore channel feature information, it can be considered to adjust the pore channel size, such as increasing or decreasing the pore diameter to change the gas molecule transport efficiency and the thermal stability of the molecular sieve; it can also change the pore channel structure, such as adjusting the straight pore channel to a curved pore channel or adding pore channel branches, etc. For the crystal grain feature information, the crystal grain size can be adjusted, for example, by controlling the synthesis conditions to prepare smaller or larger crystal grains; the crystal grain morphology can also be changed, such as making it closer to a spherical or cubic shape, etc. According to the adjusted pore channel feature information and crystal grain feature information, reconstruct a new three-dimensional simulation model, called the first optimized three-dimensional simulation model, which will be used as the basis for the next iterative optimization.

[0029] Step S383: Iteratively optimize the first optimized three-dimensional simulation model to obtain the optimal physical structure. Specifically, repeat the previous thermal simulation analysis process for the first optimized three-dimensional simulation model, that is, perform extremely high-temperature simulation and cyclic temperature simulation again to obtain new simulation records and structure collapse indices. If the new structure collapse index still does not meet the predetermined threshold, continue to adjust the physical structure according to the results, construct a new three-dimensional simulation model for the next round of iteration, and repeat continuously until the obtained structure collapse index meets the predetermined collapse index threshold. The physical structure feature information of the corresponding three-dimensional simulation model is determined as the optimal physical structure. By means of iterative optimization, the optimal physical structure of the oxygen-producing molecular sieve can be gradually approximated, improving its thermal stability and oxygen production efficiency.

[0030] Step S400: Build a molecular sieve structure database with the optimal physical structure as a constraint and extract the first structure data group from the molecular sieve structure database. Specifically, in the oxygen-producing molecular sieve structure optimization scheme, after determining the optimal physical structure, build a molecular sieve structure database based on this. Collect a large amount of molecular sieve structure data through various channels, record their physical and chemical structure characteristics, and use the similarity calculation method and set threshold for screening and sorting with the optimal physical structure as a constraint to ensure that the molecular sieve structures in the database have a certain relevance and similarity to the optimal physical structure. Then, further extract the first structure data group that meets specific conditions in the database, that is, the similarity of its chemical structure characteristics to the structure characteristics of the optimal physical structure is very high and reaches a certain similarity threshold. This data group will serve as an important data basis for subsequent supervised learning and establishing an oxygen production fitness prediction model, providing a basis for determining the optimal chemical structure.

[0031] Step S500: Perform supervised learning and verification on the first training data group formed based on the first chemical structure feature information and the first oxygen production fitness in the first structure data group to obtain an oxygen production fitness prediction model. Specifically, in the oxygen-producing molecular sieve structure optimization scheme, extract the first chemical structure feature information (including framework structure, surface chemistry, synthesis method characteristics, etc.) from the first structure data group and determine the first oxygen production fitness (which can be calculated from the oxygen production output and quality within a predetermined period), and combine the two to form the first training data group. Select the random forest supervised learning algorithm, divide the first training data group into a training set and a verification set for training. The algorithm calculates the predicted oxygen production fitness based on the input chemical structure feature information and compares it with the actual value to continuously adjust the parameters to reduce the error. Then, use the verification set to test the model. After inputting the chemical structure feature information into the model to obtain the predicted oxygen production fitness and comparing it with the actual value to calculate the evaluation index, if the performance does not meet the requirements, adjust the model parameters, increase the training data volume, or improve the data preprocessing method, etc., and train and test again. Finally, obtain a model that can accurately predict the oxygen production fitness based on the chemical structure feature information, providing a basis for determining the optimal chemical structure.

[0032] In a possible implementation, supervised learning and testing are performed on a first training data set formed based on the first chemical structure feature information and the first oxygen generation fitness in the first structure data set to obtain an oxygen generation fitness prediction model. Step S500 further includes step S510. The first structure data set has an identifier of the first oxygen generation record, and the first oxygen generation record includes the first oxygen generation output and the first oxygen generation quality within a predetermined period. Specifically, the first structure data set is a specific data set extracted from a molecular sieve structure database. The data set having the identifier of the first oxygen generation record means that it is associated with a specific oxygen generation process. The first oxygen generation record includes two important aspects: the first oxygen generation output within a predetermined period and the first oxygen generation quality. The oxygen generation output reflects the total amount of oxygen that the molecular sieve can generate within a specific time period, which is crucial for evaluating the oxygen generation efficiency of the molecular sieve. The oxygen generation quality mainly relates to the purity of oxygen and the impurity situation. The higher the purity and the fewer the impurities, the better the oxygen generation quality. By recording the information of these two aspects, the oxygen generation performance of the molecular sieve under specific conditions can be comprehensively understood.

[0033] Step S520, normalizing and weighting the first oxygen generation output and the first oxygen generation quality to obtain the first oxygen generation fitness. Specifically, in order to comprehensively evaluate the oxygen generation performance of the molecular sieve, it is necessary to integrate the oxygen generation output and the oxygen generation quality to obtain a single index, that is, the first oxygen generation fitness. The first oxygen generation output and the first oxygen generation quality are normalized. The purpose of normalization is to convert data with different dimensions into values within the same range for weighted calculation. For the oxygen generation output, a suitable normalization method can be determined according to the actual situation. For example, it can be converted into a value between 0 and 1, representing the ratio relative to a certain standard output. For the oxygen generation quality, since the purity and impurity situation are mainly concerned, the purity can be represented as a value between 0 and 1, and the impurity situation can be converted by the reciprocal of the impurity content, etc. Then, weighted calculation is performed. According to the importance of the oxygen generation output and the oxygen generation quality in practical applications, the corresponding weight coefficients are determined. If more importance is attached to the oxygen generation efficiency, a higher weight may be assigned to the oxygen generation output; if strict requirements are imposed on the oxygen quality, the weight of the oxygen generation quality may be greater. Through weighted calculation, the normalized oxygen generation output and oxygen generation quality are combined to obtain the first oxygen generation fitness. The fitness value can more comprehensively reflect the oxygen generation performance of the molecular sieve under specific conditions and provide an important basis for subsequent supervised learning and determining the optimal chemical structure.

[0034] In a possible implementation, a supervised learning and verification are performed on a first training data set formed based on the first chemical structure feature information and the first oxygen generation fitness in the first structure data set to obtain an oxygen generation fitness prediction model. Step S500 further includes step S530. The first chemical structure feature information includes framework structure feature information, surface chemical feature information, and synthesis method feature information. Among them, the framework structure feature information refers to the framework silica-alumina ratio, the surface chemical feature information refers to the acid site and hydroxyl content, and the synthesis method feature information refers to the template agent information and synthesis conditions. Specifically, the first chemical structure feature information includes framework structure feature information, surface chemical feature information, and synthesis method feature information. The framework structure feature information refers to the framework silica-alumina ratio, which determines properties such as the thermal stability, hydrophobicity, and acidity of the molecular sieve and can be determined by X-ray fluorescence spectroscopy and nuclear magnetic resonance techniques. The surface chemical feature information includes the acid site and hydroxyl content. The acid site plays a key role in catalysis and adsorption, and the type, quantity, and intensity can be determined by pyridine adsorption infrared spectroscopy and temperature-programmed desorption techniques. The hydroxyl content affects the adsorption performance and chemical reactivity and can be determined by infrared spectroscopy and nuclear magnetic resonance techniques. The synthesis method feature information includes template agent information and synthesis conditions. The template agent plays a structure-directing role, and the type, dosage, and action mechanism can be determined by analyzing the synthesized product and comparing it with known template agents, such as inferring by combining scanning electron microscopy, transmission electron microscopy, and X-ray diffraction analysis. The synthesis conditions include temperature, time, and reactant ratio, etc., and can be determined by experimental design and optimization methods such as the response surface method to study their influence on the performance of the molecular sieve. The analysis of the feature information helps to deeply understand the relationship between the molecular sieve structure and performance and provides a basis for optimizing the oxygen generation performance.

[0035] Step S600, perform a predictive analysis on any chemical structure feature information through the oxygen generation fitness prediction model to obtain any oxygen generation fitness. Specifically, the oxygen generation fitness prediction model is obtained after performing supervised learning and verification on a first training data set formed based on the first chemical structure feature information and the first oxygen generation fitness in the first structure data set. It can be used to perform a predictive analysis on any chemical structure feature information, determine the chemical structure feature information of the molecular sieve to be predicted, including framework structure feature information, surface chemical feature information, and synthesis method feature information, extract and preprocess it through analysis techniques and experimental methods to make its numerical range consistent with the training data set, input it into the model, and the model processes the information according to the internal calculation mechanism and infers the corresponding oxygen generation fitness. After the output result is evaluated and interpreted, compare it with the known oxygen generation fitness range to understand the position, analyze the influence of different features on the fitness, and then make a decision based on the prediction result. If the fitness is high, the corresponding structure can be considered for adoption; if it is low, adjust the feature information and re-predict, providing a basis and support for determining the optimal chemical structure.

[0036] Step S700: When the any oxygen generation fitness reaches a predetermined fitness threshold, the any chemical structure feature information at that time is taken as the optimal chemical structure. Specifically, in this solution, first, according to the actual application requirements and performance goals of the oxygen generation molecular sieve, factors such as oxygen generation efficiency, oxygen purity, and stability are comprehensively considered. By analyzing historical data, experimental results, and industry standards, a reasonable predetermined fitness threshold is determined. Then, the chemical structure feature information of any molecular sieve is extracted, including framework structure, surface chemistry, and synthesis method features. After preprocessing, it is input into the oxygen generation fitness prediction model. The model analyzes and calculates the input information based on the learned relationships to obtain the corresponding any oxygen generation fitness, which reflects the potential performance of the molecular sieve in oxygen generation. The any oxygen generation fitness is compared with the predetermined fitness threshold. When it reaches or exceeds the threshold, the corresponding any chemical structure feature information at this time is taken as the optimal chemical structure, representing the best chemical composition and structure features that can meet the oxygen generation performance requirements. It can be used as the target for molecular sieve synthesis or optimization, and is further verified and improved through actual synthesis experiments and performance tests to improve the performance and practicality of the oxygen generation molecular sieve.

[0037] Step S800: The optimal physical structure and the optimal chemical structure form the optimal structure scheme of the oxygen generation molecular sieve. Specifically, in the process of determining the optimal structure scheme of the oxygen generation molecular sieve, first, the optimal physical structure is determined through a thermal simulation optimization plan, including extremely high temperature and cyclic temperature simulation plans. The three-dimensional simulation model is analyzed to obtain simulation records and a structure collapse index. If it meets the threshold, it is determined as the optimal physical structure; if not, it is adjusted and iteratively optimized. A molecular sieve structure database is constructed with the optimal physical structure as a constraint, and the first structure data group is extracted, which has an oxygen generation record identifier. The oxygen generation fitness is obtained by normalizing and weighting the oxygen generation yield and quality. Based on this, a training data group is formed for supervised learning and verification to obtain an oxygen generation fitness prediction model. The any chemical structure feature information is predicted and analyzed. When the any oxygen generation fitness reaches the predetermined threshold, it is determined as the optimal chemical structure. Finally, the optimal physical structure and the optimal chemical structure are combined to jointly form the optimal structure scheme of the oxygen generation molecular sieve, providing a direction and goal for the design, synthesis, and optimization of the molecular sieve, and improving the oxygen generation performance and stability to meet the actual application requirements.

[0038] The embodiment of this application reads the predetermined structure features to collect multi-dimensional features of the oxygen generation molecular sieve, screens out the physical structure features, constructs a three-dimensional simulation model to obtain the optimal physical structure, forms a database based on this and conducts supervised learning to obtain an oxygen generation fitness prediction model. Through this model, the optimal chemical structure is determined. The optimal physical structure and the optimal chemical structure form the optimal structure scheme of the oxygen generation molecular sieve. By optimizing both the physical and chemical structures of the molecular sieve, and at the same time, the optimization of the physical structure avoids the entry of impurities into oxygen due to structure collapse, affecting the quality, achieving the technical effects of improving thermal stability, and further improving the oxygen generation efficiency and quality.

[0039] In the foregoing, with reference to Figure 1 a method for optimizing the structure of an oxygen-producing molecular sieve for improving thermal stability according to an embodiment of the present invention was described in detail. Next, with reference to Figure 2 an apparatus for optimizing the structure of an oxygen-producing molecular sieve for improving thermal stability according to an embodiment of the present invention will be described.

[0040] The apparatus for optimizing the structure of an oxygen-producing molecular sieve for improving thermal stability according to an embodiment of the present invention is used to solve the technical problem that the existing oxygen-producing molecular sieve has poor thermal stability, which in turn leads to poor oxygen production efficiency. By optimizing both the physical and chemical structures of the molecular sieve, and at the same time, the optimization of the physical structure avoids the collapse of the structure and the entry of impurities into the oxygen, affecting the quality. The technical effect of improving the thermal stability and thus improving the oxygen production efficiency and quality is achieved. The apparatus for optimizing the structure of an oxygen-producing molecular sieve for improving thermal stability includes: a structural feature information acquisition module 10, a physical structural feature information acquisition module 20, a simulation analysis module 30, a molecular sieve structure database construction module 40, a prediction model acquisition module 50, a prediction analysis module 60, and an optimal chemical structure acquisition module 70.

[0041] The structural feature information acquisition module 10 is used to read the predetermined structural features and collect multi-dimensional features of the oxygen-producing molecular sieve based on the predetermined structural features to obtain the molecular sieve structural feature information.

[0042] The physical structural feature information acquisition module 20 is used to traverse and screen the predetermined physical structural features in the molecular sieve structural feature information to obtain the physical structural feature information, and the physical structural feature information includes pore feature information and crystal grain feature information.

[0043] The simulation analysis module 30 is used to construct a three-dimensional simulation model of the oxygen-producing molecular sieve according to the pore feature information and the crystal grain feature information, and perform simulation analysis on the three-dimensional simulation model in combination with a thermal simulation optimization plan to obtain the optimal physical structure.

[0044] The molecular sieve structure database construction module 40 is used to form a molecular sieve structure database with the optimal physical structure as a constraint and extract the first structure data group from the molecular sieve structure database.

[0045] The prediction model acquisition module 50 is used to perform supervised learning and testing on a first training data group formed based on the first chemical structure feature information and the first oxygen production fitness in the first structure data group to obtain an oxygen production fitness prediction model.

[0046] The prediction analysis module 60 is used to perform prediction analysis on any chemical structure feature information through the oxygen production fitness prediction model to obtain any oxygen production fitness.

[0047] The optimal chemical structure acquisition module 70 is configured to use the arbitrary chemical structure feature information at that time as the optimal chemical structure when the arbitrary oxygen generation fitness reaches a predetermined fitness threshold.

[0048] The optimal structure scheme acquisition module 80 is configured to form the optimal structure scheme of the oxygen generation molecular sieve with the optimal physical structure and the optimal chemical structure.

[0049] Next, the specific configuration of the physical structure feature information acquisition module 20 will be described in detail. As described above, the predetermined physical structure features are traversed and screened in the molecular sieve structure feature information to obtain the physical structure feature information, which includes pore feature information and crystal grain feature information. The physical structure feature information acquisition module 20 further includes: a pore feature information composition unit, and the pore feature information composition unit is configured to the pore feature information includes pore size and pore structure.

[0050] Among them, the physical structure feature information acquisition module 20 further includes: a crystal grain feature information composition unit, and the crystal grain feature information composition unit is configured to the crystal grain feature information includes crystal grain size and crystal grain morphology.

[0051] Next, the specific configuration of the simulation analysis module 30 will be described in detail. As described above, a three-dimensional simulation model of the oxygen-producing molecular sieve is constructed based on the pore feature information and the crystal grain feature information, and the three-dimensional simulation model is subjected to simulation analysis in combination with the thermal simulation optimization plan to obtain the optimal physical structure. The simulation analysis module 30 further includes: a thermal simulation optimization plan composition unit, where the thermal simulation optimization plan composition unit is used for the thermal simulation optimization plan including an extremely high temperature simulation plan and a cyclic temperature simulation plan; a first extremely high temperature simulation record acquisition unit, where the first extremely high temperature simulation record acquisition unit is used for performing simulation analysis on the three-dimensional simulation model for a predetermined duration at a first extremely high temperature according to the extremely high temperature simulation plan to obtain a first extremely high temperature simulation record; a first simulation physical structure feature information acquisition unit, where the first simulation physical structure feature information acquisition unit is used for acquiring first simulation physical structure feature information corresponding to the predetermined duration in the first extremely high temperature simulation record; a first simulation structure collapse index acquisition unit, where the first simulation structure collapse index acquisition unit is used for comparing the physical structure feature information with the first simulation physical structure feature information to obtain a first simulation structure collapse index; a first cyclic temperature simulation record acquisition unit, where the first cyclic temperature simulation record acquisition unit is used for performing simulation analysis on the three-dimensional simulation model for the predetermined duration at a first cyclic temperature according to the cyclic temperature simulation plan to obtain a first cyclic temperature simulation record; a second simulation physical structure feature information acquisition unit, where the second simulation physical structure feature information acquisition unit is used for acquiring second simulation physical structure feature information corresponding to the predetermined duration in the first cyclic temperature simulation record; a second simulation structure collapse index acquisition unit, where the second simulation structure collapse index acquisition unit is used for comparing the physical structure feature information with the second simulation physical structure feature information to obtain a second simulation structure collapse index; an optimal physical structure acquisition unit, where the optimal physical structure acquisition unit is used for, when both the first simulation structure collapse index and the second simulation structure collapse index meet a predetermined collapse index threshold, taking the physical structure feature information corresponding to the three-dimensional simulation model as the optimal physical structure.

[0052] Among them, when both the first simulation structure collapse index and the second simulation structure collapse index meet the predetermined collapse index threshold, the physical structure feature information corresponding to the three-dimensional simulation model is used as the optimal physical structure. The optimal physical structure acquisition unit further includes: a physical structure adjustment instruction subunit, which is configured to issue a physical structure adjustment instruction when the first simulation structure collapse index or the second simulation structure collapse index does not meet the predetermined collapse index threshold; a first optimized three-dimensional simulation model acquisition subunit, which is configured to adjust the pore feature information and the crystal grain feature information based on the physical structure adjustment instruction, and construct a first optimized three-dimensional simulation model according to the adjustment result; an iterative optimization subunit, which is configured to perform iterative optimization on the first optimized three-dimensional simulation model to obtain the optimal physical structure.

[0053] Next, the specific configuration of the prediction model acquisition module 50 will be described in detail. As described above, supervised learning and verification are performed on the first training data set composed of the first chemical structure feature information and the first oxygen production fitness in the first structure data set to obtain an oxygen production fitness prediction model. The prediction model acquisition module 50 further includes: a first oxygen production record identification unit, which is configured to identify that the first structure data set has a first oxygen production record, and the first oxygen production record includes the first oxygen production amount and the first oxygen production quality within a predetermined period; a first oxygen production fitness acquisition unit, which is configured to normalize and weight the first oxygen production amount and the first oxygen production quality to obtain the first oxygen production fitness.

[0054] Among them, the prediction model acquisition module 50 further includes: a first chemical structure feature information composition unit, which is configured to include the first chemical structure feature information including framework structure feature information, surface chemical feature information, and synthesis method feature information. Among them, the framework structure feature information refers to the framework silicon-aluminum ratio, the surface chemical feature information refers to the acid site and hydroxyl content, and the synthesis method feature information refers to the template agent information and synthesis conditions.

[0055] The oxygen production molecular sieve structure optimization device for improving thermal stability provided by the embodiments of the present invention can execute the oxygen production molecular sieve structure optimization method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0056] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server. The various units and modules included are only divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0057] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application. In some cases, the actions or steps recited in the present application can be executed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for optimizing the structure of an oxygen-producing molecular sieve for improving thermal stability, characterized in that: include: Reading predetermined structural features, and collecting multi-dimensional features of the oxygen-producing molecular sieve based on the predetermined structural features to obtain molecular sieve structural feature information; The predetermined physical structure feature is traversed and screened in the molecular sieve structure feature information to obtain the physical structure feature information, wherein the physical structure feature information includes pore feature information and grain feature information; Constructing a three-dimensional simulation model of the oxygen-producing molecular sieve according to the pore characteristic information and the grain characteristic information, and performing simulation analysis on the three-dimensional simulation model in combination with a thermal simulation optimization plan to obtain an optimal physical structure; Building a molecular sieve structure database with the optimal physical structure as a constraint, and extracting a first structure data group from the molecular sieve structure database; Performing supervised learning and testing on a first training data set formed based on the first chemical structure feature information and the first oxygen production fitness in the first structure data set to obtain an oxygen production fitness prediction model; Using the oxygen production fitness prediction model, any chemical structure characteristic information is predicted and analyzed to obtain any oxygen production fitness; When the fitness of any oxygen production reaches a predetermined fitness threshold, the characteristic information of any chemical structure at that time is used as the optimal chemical structure; The optimal physical structure and the optimal chemical structure constitute the optimal structural scheme of the oxygen-producing molecular sieve.

2. The method for optimizing the structure of oxygen-producing molecular sieves for improving thermal stability according to claim 1, characterized in that: The pore characteristic information includes pore size and pore structure.

3. The method for optimizing the structure of oxygen-producing molecular sieves for improving thermal stability according to claim 1, characterized in that: The grain characteristic information includes grain size and grain morphology.

4. The method for optimizing the structure of oxygen-producing molecular sieves for improving thermal stability according to claim 1, characterized in that: include: The thermal simulation optimization plan includes an extremely high temperature simulation plan and a circulating temperature simulation plan; According to the extremely high temperature simulation scheme, a simulation analysis is performed on the three-dimensional simulation model at a first extremely high temperature for a predetermined time to obtain a first extremely high temperature simulation record; Acquire first simulation physical structure feature information corresponding to the predetermined time length in the first extremely high temperature simulation record; Comparing the physical structure feature information with the first simulated physical structure feature information to obtain a first simulated structure collapse index; According to the cycle temperature simulation scheme, performing simulation analysis on the three-dimensional simulation model at a first cycle temperature for the predetermined time period to obtain a first cycle temperature simulation record; Acquire the second simulation physical structure feature information corresponding to the predetermined time length in the first cycle temperature simulation record; Comparing the physical structure feature information with the second simulated physical structure feature information to obtain a second simulated structure collapse index; When the collapse index of the first simulation structure and the collapse index of the second simulation structure both meet a predetermined collapse index threshold, the physical structure feature information corresponding to the three-dimensional simulation model is used as the optimal physical structure.

5. The method for optimizing the structure of oxygen-producing molecular sieves for improving thermal stability according to claim 4, characterized in that: include: When the collapse index of the first simulation structure or the collapse index of the second simulation structure does not meet the predetermined collapse index threshold, issuing a physical structure adjustment instruction; Adjusting the channel characteristic information and the grain characteristic information based on the physical structure adjustment instruction, and constructing a first optimization three-dimensional simulation model according to the adjustment result; The first optimizing three-dimensional simulation model is iteratively optimized to obtain the optimal physical structure.

6. The method for optimizing the structure of oxygen-producing molecular sieves for improving thermal stability according to claim 1, characterized in that: include: The first structure data group has an identifier of a first oxygen production record, and the first oxygen production record includes a first oxygen production output and a first oxygen production quality within a predetermined period; The first oxygen production fitness is obtained by normalizing and weighting the first oxygen production output and the first oxygen production quality.

7. The method for optimizing the structure of oxygen-producing molecular sieves for improving thermal stability according to claim 1, characterized in that: The first chemical structure characteristic information includes skeleton structure characteristic information, surface chemical characteristic information and synthesis method characteristic information, wherein the skeleton structure characteristic information refers to the skeleton silicon-aluminum ratio, the surface chemical characteristic information refers to the acidic site and hydroxyl content, and the synthesis method characteristic information refers to the template information and synthesis conditions.

8. An oxygen-producing molecular sieve structure optimization device for improving thermal stability, characterized in that: The device is used to implement the oxygen production molecular sieve structure optimization method for improving thermal stability according to any one of claims 1 to 7, and the device comprises: A structural feature information acquisition module, the structural feature information acquisition module is used to read predetermined structural features, and based on the predetermined structural features, perform multi-dimensional feature collection on the oxygen-producing molecular sieve to obtain molecular sieve structural feature information; A physical structure characteristic information acquisition module, the physical structure characteristic information acquisition module is used to traverse and screen the predetermined physical structure characteristic in the molecular sieve structure characteristic information to obtain physical structure characteristic information, the physical structure characteristic information including pore characteristic information and grain characteristic information; A simulation analysis module, wherein the simulation analysis module is used to construct a three-dimensional simulation model of the oxygen-producing molecular sieve according to the pore characteristic information and the grain characteristic information, and to perform simulation analysis on the three-dimensional simulation model in combination with a thermal simulation optimization plan to obtain an optimal physical structure; A molecular sieve structure database construction module, wherein the molecular sieve structure database construction module is used to construct a molecular sieve structure database with the optimal physical structure as a constraint, and to extract a first structure data group from the molecular sieve structure database; A prediction model acquisition module, the prediction model acquisition module is used to perform supervised learning and verification on a first training data group formed based on the first chemical structure feature information and the first oxygen production fitness in the first structure data group, to obtain an oxygen production fitness prediction model; A prediction and analysis module, which is used to perform prediction and analysis on any chemical structure feature information through the oxygen production fitness prediction model to obtain any oxygen production fitness; An optimal chemical structure acquisition module, wherein when the fitness of any oxygen production reaches a predetermined fitness threshold, the feature information of any chemical structure at that time is used as the optimal chemical structure; An optimal structural scheme acquisition module is used for the optimal structural scheme of the oxygen-producing molecular sieve composed of the optimal physical structure and the optimal chemical structure.

Images