Method and device for optimizing preparation parameters of oxygen production molecular sieve

Abstract

The invention discloses a preparation parameter optimization method and device of an oxygen production molecular sieve, and relates to the technical field of preparation parameter optimizing.The method comprises the steps that sampling detection is conducted on an interactive target oxygen production molecular sieve in a raw material library in a production cycle, and a raw material feature set is obtained; traversing process equipment for state analysis to obtain a reliable coefficient; the preparation type, the preparation amount and the quality requirement are collected, and constraint information is generated; a historical preparation database is called and evaluated, information is input into the database for retrieval, and initial preparation parameters are obtained; optimizing the parameters based on the equipment reliability coefficient to obtain target preparation parameters; and controlling the equipment to prepare according to the target parameter. The technical problems of unstable preparation parameters and inconsistent quality caused by fluctuation of equipment states and raw material characteristics in the preparation process of different types of oxygen production molecular sieves are solved, and the effects of ensuring that the preparation process of the oxygen production molecular sieves is stable and the quality meets expected requirements by optimizing the preparation parameters through comprehensive evaluation and multi-scale optimization are achieved.

Description

Technical Field

[0001] This application relates to the technical field of preparation parameter optimization, and specifically to a method and device for optimizing the preparation parameters of oxygen-making molecular sieves. Background Art

[0002] Oxygen-making molecular sieves are important materials widely used in the fields of oxygen preparation and gas separation. Their excellent adsorption properties make them important in medical oxygen concentrators, industrial gas separation, environmental protection, etc. However, different types of oxygen-making molecular sieves have different requirements for preparation methods and conditions, which means that precise control of multiple steps and parameters needs to be considered during the preparation process. With the continuous improvement of the requirements for oxygen purity and production efficiency, the preparation process of oxygen-making molecular sieves has become more complex, and it is necessary to continuously optimize the preparation process and conditions to ensure the stability of product quality and the high efficiency of production. In the traditional preparation process of oxygen-making molecular sieves, the determination of process parameters usually relies on experience accumulation and experimental adjustment. This method is not only inefficient but also difficult to cope with the uncertainties brought by fluctuations in raw material characteristics and equipment status. At the same time, due to the differences in their chemical compositions and structures, different types of oxygen-making molecular sieves have different requirements for temperature, pressure, time, material ratio, etc. during the preparation process, further increasing the complexity of process optimization. Therefore, how to quickly and accurately optimize the preparation parameters under complex and variable preparation conditions has become the key to the development of oxygen-making molecular sieve preparation technology. Summary of the Invention

[0003] This application provides a method and device for optimizing the preparation parameters of oxygen-making molecular sieves, to solve the technical problems of unstable preparation parameters and inconsistent quality caused by fluctuations in equipment status and raw material characteristics during the preparation of different types of oxygen-making molecular sieves, and to achieve the effect of optimizing the preparation parameters through comprehensive evaluation and multi-scale optimization, ensuring the stability of the preparation process of oxygen-making molecular sieves and the quality meeting the expected requirements.

[0004] The present application provides a method for optimizing the preparation parameters of an oxygen - producing molecular sieve. The method includes: interacting with the raw material library of the target oxygen - producing molecular sieve within a preset production cycle for raw material sampling and testing to obtain a raw material feature set; traversing K sets of process equipment of the target oxygen - producing molecular sieve for equipment status analysis to obtain K process equipment reliability coefficients; collecting the preparation type, preparation quantity, and preparation quality requirements of the target oxygen - producing molecular sieve to generate preparation constraint information; retrieving the historical preparation database of the target oxygen - producing molecular sieve, comprehensively evaluating the historical preparation database, and when the evaluation passes, inputting the raw material feature set, K process equipment reliability coefficients, and the preparation constraint information into the historical preparation database for retrieval to obtain an initial preparation parameter set, where the initial preparation parameter set includes K process initial preparation parameter sets; performing multi - scale optimization on the K process initial preparation parameter sets based on the K process equipment reliability coefficients, and using the preparation constraint information to constrain the optimization process until preset requirements are met to obtain a target preparation parameter set, where the target preparation parameter set includes K process target preparation parameter sets; controlling the K sets of process equipment to prepare the target oxygen - producing molecular sieve according to the K process target preparation parameter sets.

[0005] The present application also provides a device for optimizing the preparation parameters of an oxygen - producing molecular sieve, including: a raw material sampling and testing module for interacting with the raw material library of the target oxygen - producing molecular sieve within a preset production cycle for raw material sampling and testing to obtain a raw material feature set; an equipment status analysis module for traversing K sets of process equipment of the target oxygen - producing molecular sieve for equipment status analysis to obtain K process equipment reliability coefficients; a preparation constraint information generation module for collecting the preparation type, preparation quantity, and preparation quality requirements of the target oxygen - producing molecular sieve to generate preparation constraint information; a comprehensive evaluation module for retrieving the historical preparation database of the target oxygen - producing molecular sieve, comprehensively evaluating the historical preparation database, and when the evaluation passes, inputting the raw material feature set, K process equipment reliability coefficients, and the preparation constraint information into the historical preparation database for retrieval to obtain an initial preparation parameter set, where the initial preparation parameter set includes K process initial preparation parameter sets; a multi - scale optimization module for performing multi - scale optimization on the K process initial preparation parameter sets based on the K process equipment reliability coefficients, and using the preparation constraint information to constrain the optimization process until preset requirements are met to obtain a target preparation parameter set, where the target preparation parameter set includes K process target preparation parameter sets; an oxygen - producing molecular sieve preparation module for controlling the K sets of process equipment to prepare the target oxygen - producing molecular sieve according to the K process target preparation parameter sets.

[0006] It is proposed to optimize the preparation parameters of the oxygen-making molecular sieve and the device through this application. Perform raw material sampling and testing on the raw material library of the target oxygen-making molecular sieve within a preset production cycle to obtain a set of raw material characteristics; traverse the K process equipment sets of the target oxygen-making molecular sieve for equipment status analysis to obtain K process equipment reliability coefficients; collect the preparation type, preparation quantity, and preparation quality requirements of the target oxygen-making molecular sieve to generate preparation constraint information; retrieve the historical preparation database of the target oxygen-making molecular sieve, conduct a comprehensive evaluation of the historical preparation database, and when the evaluation passes, input the set of raw material characteristics, the K process equipment reliability coefficients, and the preparation constraint information into the historical preparation database for retrieval to obtain an initial set of preparation parameters, where the initial set of preparation parameters includes K sets of initial process preparation parameters; perform multi-scale optimization on the K sets of initial process preparation parameters based on the K process equipment reliability coefficients, and use the preparation constraint information to constrain the optimization process until the preset requirements are met to obtain a set of target preparation parameters, where the set of target preparation parameters includes K sets of target process preparation parameters; control the K process equipment sets to prepare the target oxygen-making molecular sieve according to the K sets of target process preparation parameters. Solve the technical problems of unstable preparation parameters and inconsistent quality caused by fluctuations in equipment status and raw material characteristics during the preparation of different types of oxygen-making molecular sieves, and achieve the effect of optimizing the preparation parameters through comprehensive evaluation and multi-scale optimization to ensure the stability of the preparation process of the oxygen-making molecular sieve and the quality meets the expected requirements. Description of the Drawings

[0007] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure 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 before or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0008] Figure 1 Schematic diagram of the process for optimizing the preparation parameters of the oxygen-making molecular sieve provided by the embodiment of the present application; Figure 2 Schematic diagram of the structure of the device for optimizing the preparation parameters of the oxygen-making molecular sieve provided by the embodiment of the present application.

[0009] Explanation of the reference numerals in the drawings: Raw material sampling and testing module 1, Equipment status analysis module 2, Preparation constraint information generation module 3, Comprehensive evaluation module 4, Multi-scale optimization module 5, Oxygen-making molecular sieve preparation module 6. Detailed Embodiments

[0010] The above description is only an overview of the technical solution of this application. In order to understand the technical means of this 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 this application more obvious and understandable, the specific embodiments of this application are specifically given below.

[0011] In order to make the purpose, technical solution and advantages of this application clearer, the following will further describe this application in detail with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

[0012] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first / second" involved 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, device, product or server comprising a series of steps or units does not necessarily limit to those steps or units clearly listed, but may include other steps or modules not clearly listed or 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 technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0013] The embodiments of this application provide a method for optimizing the preparation parameters of oxygen-making molecular sieves, as Figure 1 shown, the method includes: Interactively sampling and detecting raw materials in the raw material library of the target oxygen-making molecular sieve within a preset production cycle to obtain a set of raw material characteristics.

[0014] In the embodiment of the present application, the system terminal first determines the preset production cycle of the oxygen-making molecular sieve. This cycle is set according to factors such as production plans and market demands. During the preset production cycle, a raw material library for preparing the target oxygen-making molecular sieve is determined. This raw material library contains a variety of raw materials, and the sources, batches, and storage conditions of these materials may vary. Subsequently, according to the type of raw materials and the requirements of the preparation process, a specific sampling strategy is formulated. For example, simple random sampling, stratified sampling, systematic sampling, etc., to ensure that the extracted samples are representative. Taking simple random sampling as an example, the system terminal randomly extracts a certain number of raw material samples from the raw material library. The quantity and type of the samples are determined according to actual business needs and should be sufficient to represent the characteristics of the entire raw material library to ensure the effectiveness of sampling. After that, multi-parameter detection is performed on the extracted raw material samples, including component analysis, purity detection, particle size distribution, etc. Then, the detection results are sorted using centralized feature analysis, key parameters representing the characteristics of the raw materials are extracted, and a raw material feature set is generated. This set will include component characteristics, purity characteristics, particle size distribution characteristics, etc. of each raw material, providing data support for subsequent process parameter optimization and the preparation process of the oxygen-making molecular sieve.

[0015] Furthermore, the present application provides a method for obtaining a raw material feature set, including: Randomly extract M raw material samples from the raw material library in the manner of simple random sampling; traverse the M raw material samples for multi-parameter detection to obtain M raw material sample feature sets, where each raw material sample feature set includes component characteristics, purity characteristics, and particle size distribution characteristics; respectively use the components, purity, and particle size distribution as indexes to retrieve the M raw material sample feature sets to obtain a raw material sample component feature set, a raw material sample purity feature set, and a raw material particle size distribution feature set; traverse the raw material sample component feature set, the raw material sample purity feature set, and the raw material particle size distribution feature set for centralized feature analysis to obtain raw material component characteristics, raw material purity characteristics, and raw material particle size distribution characteristics; use the raw material component characteristics, raw material purity characteristics, and the raw material particle size distribution characteristics as the raw material feature set.

[0016] Preferably, the system terminal selects suitable raw materials according to the performance requirements of the oxygen-making molecular sieve, such as silicon source and aluminum source, etc. The chemical composition and purity of these raw materials directly affect the final performance of the oxygen-making molecular sieve. Therefore, when selecting raw materials, it is necessary to strictly control their composition and purity to ensure compliance with the requirements of the preparation process. Then, select a suitable particle size distribution of the raw materials to ensure that during the preparation process, the raw materials can be evenly mixed and react smoothly. An appropriate particle size distribution not only helps to improve the uniformity of the raw materials but also enhances the performance of the prepared oxygen-making molecular sieve. Before the formal extraction, the system terminal also pre-treats the raw materials, such as drying or grinding. These pre-treatment steps help to ensure that the raw materials can better adapt to the subsequent process flow, improving the preparation efficiency and product quality of the oxygen-making molecular sieve. Subsequently, according to the production requirements of the oxygen-making molecular sieve, M raw material samples are randomly selected from the processed raw material library in a simple random sampling manner. Random sampling can ensure the representativeness of the samples, enabling the samples to reflect the characteristics of the entire raw material library. Then, detailed multi-parameter detection is carried out on the M selected raw material samples. The detection contents include component analysis, purity detection, and particle size distribution detection. The detection results of each raw material sample will form a raw material sample feature set, including component features, purity features, and particle size distribution features. These M raw material sample feature sets will be used as the basis for subsequent analysis. After that, according to the generated M raw material sample feature sets, the M raw material sample feature sets are retrieved respectively with components, purity, and particle size distribution as indexes, generating a raw material sample component feature set, a raw material sample purity feature set, and a raw material particle size distribution feature set, that is, the same features of the M raw material sample feature sets are classified in one set. This step ensures the orderly arrangement of each feature set, facilitating subsequent in-depth analysis. Then, traverse the obtained raw material sample component feature set, raw material sample purity feature set, and raw material particle size distribution feature set respectively, and conduct centralized feature analysis on each traversed feature set to obtain raw material component features, raw material purity features, and raw material particle size distribution features. These features are considered to be the most representative features because they conform to the overall trend of the samples and are within the fluctuation range. Taking the raw material sample component feature set as an example, the system terminal will construct a D-dimensional centralized analysis space based on the results of traversing the raw material sample component feature set, and then use the pre-constructed center update formula to update the centroid of the D-dimensional centralized analysis space, repeating this process until the preset iteration requirements are met. When the iteration ends, the system terminal takes the raw material sample component feature corresponding to the current clustering center as the raw material component feature. For the raw material sample purity feature set and the raw material particle size distribution feature set, the system terminal performs the same operations to obtain the raw material purity feature and the raw material particle size distribution feature. Finally, the extracted raw material component features, raw material purity features, and raw material particle size distribution features are summarized to form the final raw material feature set.This set of raw material characteristics provides necessary data support for the optimization of the preparation parameters of the oxygen production molecular sieve, ensuring the stability of the preparation process and the high quality of the final product.

[0017] Furthermore, the present application provides a method for obtaining the raw material component characteristics, including: Constructing a D-dimensional centralized analysis space based on the set of raw material sample component characteristics, where D is the number of sub-characteristics in each raw material sample component characteristic, D is a positive integer greater than or equal to 2, and the D-dimensional centralized analysis space includes M spatial points, each spatial point corresponding to a raw material sample component characteristic; taking the centroid of the D-dimensional centralized analysis space as the first clustering center, and determining the first clustering neighborhood of the first clustering center according to a preset distance bandwidth; inputting the first clustering center and the first clustering neighborhood into a center update formula to update the clustering center and obtain a second clustering center; after multiple iterations of clustering center update until a preset iteration requirement is met, taking the raw material sample component characteristic corresponding to the clustering center obtained in the last iteration as the raw material component characteristic.

[0018] Optionally, the system terminal traverses the set of raw material sample composition features and constructs a D-dimensional centralized analysis space using the traversed raw material sample composition features, where D represents the number of sub-features in each raw material sample composition feature, for example, the type of chemical component, which is a positive integer greater than or equal to 2. In this D-dimensional space, each raw material sample composition feature can be represented as a D-dimensional vector. Therefore, the space will contain M spatial points, and each spatial point corresponds to the composition feature of a raw material sample. Subsequently, the centroid of the M spatial points in the D-dimensional centralized analysis space is calculated. The centroid is calculated as the coordinate mean of all spatial points. This centroid will be used as the first clustering center, representing the central position of the composition features within this space. Then, according to the preset distance bandwidth, the first clustering neighborhood is determined. The first clustering neighborhood is a set composed of all spatial points whose distances from the first clustering center are within the preset distance bandwidth range. Specifically, the system terminal traverses the M spatial points in the D-dimensional space and calculates the Euclidean distance between each point and the first clustering center. If the distance of a certain point is less than or equal to the preset distance bandwidth, then this point belongs to the first clustering neighborhood. After that, the first clustering center and the first clustering neighborhood are input into a pre-constructed center update formula for clustering center update to calculate a new clustering center. The new clustering center will be closer to the sample points in the neighborhood, reflecting the average position of these points. Then, the clustering center update process is iterated multiple times. After each update, the clustering center is recalculated, and the clustering neighborhood is redefined. The iteration process continues until the preset iteration requirement is met, such as the change amount between the new and old clustering centers is less than the center change threshold. When the change amount of the clustering center is less than the center change threshold, the clustering center obtained in the last iteration is used as the final clustering center. The raw material sample composition feature corresponding to this final clustering center is then determined as the raw material composition feature. This raw material composition feature is regarded as the most representative feature and is used for subsequent process optimization and parameter adjustment. Through this process, the most representative raw material composition feature can be accurately extracted in the multi-dimensional feature space, ensuring that the preparation process of the oxygen production molecular sieve is based on stable and consistent raw material characteristics.

[0019] Furthermore, the present application provides a center update formula, including: The center update formula is: 。

[0020] Optionally, during the preparation parameter optimization process of the oxygen production molecular sieve, a clustering method is used to determine the raw material sample composition features. During the clustering process, the position of the clustering center plays a crucial role in the final feature extraction. Therefore, it is necessary to update the position of the clustering center through multiple iterations to make it more accurately reflect the central tendency of the samples. To update the clustering center, the system terminal constructs a center update formula. This formula is used to calculate the position of the new clustering center, specifically as follows: ; where is the second clustering center, i.e., the position of the new clustering center to be calculated. is the th spatial point in the first clustering neighborhood, and these points are a set of points whose distances from the current clustering center are within the preset bandwidth. is the first clustering center, i.e., the position of the current clustering center. is the first clustering neighborhood, representing a set of all points whose distances from the first clustering center are within the preset bandwidth.

[0021] Traverse the K sets of process equipment of the target oxygen - producing molecular sieve to perform equipment status analysis, and obtain the reliability coefficients of the K sets of process equipment; In one embodiment, the system terminal classifies the process equipment for preparing the target oxygen - producing molecular sieve into multiple categories. These equipment cover the entire production process of the oxygen - producing molecular sieve, including raw material processing equipment, reaction synthesis equipment, post - processing equipment, quality control equipment, and auxiliary equipment. Each category of equipment undertakes a specific function during the preparation process. For each category of equipment, the system terminal collects its operating status data to form K sets of process equipment. These data include the operating time, failure times, maintenance records, working environmental conditions, etc. of the equipment. By monitoring and recording these data, the current status of the equipment can be comprehensively understood. Subsequently, the system terminal traverses the obtained K sets of process equipment. In each traversal, the system terminal analyzes the maintenance records of each process equipment, conducts reliability analysis from two dimensions of the number of maintenance times and equipment loss amount, and calculates the reliability coefficients of the K sets of process equipment. The reliability coefficient of each equipment reflects its stability and reliability under the current production conditions, ensuring the operating stability and production efficiency of the equipment during the preparation of the oxygen - producing molecular sieve.

[0022] Furthermore, the present application provides obtaining the reliability coefficients of the K sets of process equipment, including: Traverse the K sets of process equipment to search for maintenance records, and obtain K clusters of historical maintenance records of process equipment. Each cluster of historical maintenance records of process equipment includes multiple sets of historical maintenance records of process equipment, and each set of historical maintenance records of process equipment corresponds to a process equipment; conduct reliability analysis on the K clusters of historical maintenance records of process equipment from two dimensions of the number of maintenance times and equipment loss amount, and obtain the reliability coefficients of the K sets of process equipment.

[0023] Preferably, after obtaining the K sets of process equipment, the system terminal traverses these sets of process equipment. In each traversal, for each piece of equipment, its historical maintenance records of the process equipment are obtained. These historical maintenance records of the process equipment cover all maintenance data since the process equipment was put into use, including failure time, maintenance content, maintenance duration, spare parts used, the status of the equipment after maintenance, etc. Then, all the maintenance records of each piece of process equipment are grouped into a set of historical maintenance records of the process equipment. Each record set corresponds to a specific piece of process equipment and details all the maintenance histories of this piece of process equipment. Subsequently, the sets of maintenance records of the K pieces of process equipment obtained are combined together to form a cluster of historical maintenance records of the process equipment. This cluster contains the maintenance record information of all key process equipment in the preparation process of the oxygen production molecular sieve. After that, the system terminal counts the number of maintenance times of each piece of equipment from the obtained cluster of historical maintenance records of the process equipment, which is one of the important indicators to measure the reliability of the process equipment. The more maintenance times, the higher the failure frequency of the process equipment and the lower the reliability. According to the actual business requirements, indicators of equipment losses are defined, such as downtime, production losses, maintenance costs, etc. Then, reliable analysis is carried out based on the counted number of maintenance times and the defined indicators of equipment losses. Specifically, for the number of maintenance times, the system terminal traverses the historical maintenance records of each piece of process equipment, counts its total number of maintenance times within a certain period, and then obtains the maximum and minimum values of the number of maintenance times among all process equipment. Subsequently, the difference between the number of maintenance times of each piece of process equipment and the minimum value of the number of maintenance times is divided by the difference between the maximum and minimum values of the number of maintenance times to obtain the maintenance times coefficient of each piece of process equipment. For the indicators of equipment losses, taking downtime, production losses, and maintenance costs as examples, the system terminal measures the downtime according to the ratio of the total downtime of each piece of process equipment to the production time, calculates the production losses by multiplying the production loss per unit time by the downtime, and determines the maintenance costs based on the spare parts and labor costs of the process equipment. Then, the dimensions between the obtained downtime, production losses, and maintenance costs are eliminated through existing standardized formulas, and the equipment loss amount of each piece of process equipment is calculated by using the method of weighted summation, where the weight of each equipment loss indicator is determined according to the actual business requirements. After that, using the same method as calculating the maintenance times coefficient above, the equipment loss amount coefficient of each piece of process equipment is calculated, and then the weighted summation of the calculated maintenance times coefficient and equipment loss amount coefficient is carried out to calculate the reliability coefficient of each piece of process equipment, where the weights of the maintenance times coefficient and equipment loss amount coefficient depend on the priority of the equipment maintenance strategy. The K reliability coefficients of the process equipment obtained by the system terminal will be used for the selection and allocation of equipment in the subsequent process parameter optimization process to ensure the reliability of the equipment in the preparation process of the oxygen production molecular sieve.

[0024] Collect the preparation type, preparation quantity, and preparation quality requirements of the target oxygen production molecular sieve to generate preparation constraint information.

[0025] In one embodiment, before preparing the target oxygen - producing molecular sieve, the system terminal collects the preparation type, preparation quantity, and preparation quality requirements of the target oxygen - producing molecular sieve. These information will jointly constitute the constraint information in the preparation process, which is used to guide and optimize the entire preparation process. Among them, the preparation type refers to the specific type of oxygen - producing molecular sieve to be prepared. Different types of molecular sieves may have different chemical compositions, crystal structures, and performance requirements. Therefore, clarifying the preparation type helps to determine the appropriate process parameters and equipment configurations to ensure that the required molecular sieve characteristics can be achieved during the production process. The preparation quantity refers to the quantity of the oxygen - producing molecular sieve planned to be produced. The preparation quantity determines the production scale and the configuration requirements of resources, such as the usage amount of raw materials, the running time of equipment, etc. Accurate preparation quantity information can help optimize the production plan, reasonably arrange equipment resources, and avoid waste or shortage. The preparation quality requirements refer to the specific requirements for the performance and quality standards of the final product, such as purity, adsorption capacity, particle size distribution, etc. These quality requirements directly affect the setting and control of process parameters and are the key to ensuring that the final product meets the expected performance indicators.

[0026] Retrieve the historical preparation database of the target oxygen - producing molecular sieve, conduct a comprehensive evaluation of the historical preparation database. When the evaluation passes, input the raw material feature set, the reliable coefficients of K process equipment, and the preparation constraint information into the historical preparation database for retrieval to obtain an initial set of preparation parameters, where the initial set of preparation parameters includes K initial sets of process preparation parameters.

[0027] In one embodiment, after obtaining the preparation constraint information, the system terminal retrieves the historical preparation database storing the historical preparation data of the oxygen-making molecular sieve. These data include past preparation records, relevant process parameters, production environment conditions, equipment status, and quality inspection results of the final product, etc. Subsequently, a comprehensive evaluation is carried out on the retrieved historical preparation database to ensure the integrity, accuracy, and reliability of the information in the database. This evaluation process includes data integrity check, data accuracy verification, and correlation analysis. The data integrity check can ensure that the key data in the database, such as raw material characteristics, equipment status, process parameters, quality inspection results, etc., are complete without omission. The data accuracy verification can check whether the historical data is accurate, which involves comparing and verifying with the actual production records, or using data cleaning methods to exclude outliers and incorrect data. The correlation analysis can evaluate the correlation between the historical data and the current preparation conditions to ensure that the historical records stored in the database have sufficient reference value for the type of molecular sieve to be prepared currently. When the comprehensive evaluation passes, that is, when it is determined that the data in the historical database is complete, accurate, and highly relevant, subsequent operations can be continued. If the evaluation fails, data needs to be supplemented or the content of the database needs to be adjusted. After the historical preparation database passes the evaluation, the system terminal inputs the raw material feature set of the current target oxygen-making molecular sieve, the reliable coefficients of K process equipment, and the preparation constraint information into the comprehensively evaluated historical preparation database for retrieval. In the historical preparation database, the system terminal retrieves according to the input conditions to find the historical preparation record that best matches the current preparation task. The process parameters included in these records will be used as a reference to form the initial preparation parameter set. This initial preparation parameter set includes the process parameters initially recommended for the K process equipment under the current conditions. These parameters include setting values related to equipment operation such as temperature, pressure, time, stirring speed, etc., and serve as the basis for the subsequent preparation process. These parameters will be further optimized in the subsequent multi-scale optimization steps to finally determine the optimal preparation plan.

[0028] Perform multi-scale optimization on the K process initial preparation parameter sets based on the reliable coefficients of the K process equipment, and use the preparation constraint information to constrain the optimization process until the preset requirements are met to obtain the target preparation parameter set, where the target preparation parameter set includes K process target preparation parameter sets.

[0029] In one embodiment, during the multi-scale optimization process, the system terminal determines K adjustment steps for K sets of initial process preparation parameters according to the reliability coefficients of K process devices. Process devices with lower reliability allow a larger range of parameter adjustment, while devices with higher reliability may require more conservative adjustments to avoid production failures caused by exceeding their capabilities. Subsequently, based on the determined K adjustment steps, the system terminal adjusts the K sets of initial process preparation parameters and evaluates the preparation quality of the adjustment results to determine whether the preparation constraint information is satisfied. If the requirements of the preparation constraint information are met, the system terminal evaluates the adjustment results from two dimensions: overall preparation quality and individual process quality, to obtain the preparation fitness of the adjustment results. Then, the adjustment results are adjusted again using the K adjustment steps, and the above process is repeated to obtain the secondary adjustment results and the corresponding preparation fitness. Then, the magnitude relationship between the preparation fitness of the secondary adjustment results and the preparation fitness of the adjustment results is judged to determine the adjustment object for the next stage. This process is repeated until the calculated preparation fitness meets the preset requirements, and a set of target preparation parameters is generated. This set of target preparation parameters includes K sets of process target preparation parameters, with each set corresponding to a process device, and is used to optimize the production process of oxygen production molecular sieve to ensure product quality and production efficiency.

[0030] Further, the present application provides a method for obtaining a set of target preparation parameters, including: Calculate the ratio of the reliability coefficients of the K process devices to the sum of the reliability coefficients of the K process devices respectively, and take the difference between 1 and the ratio as the K adjustment steps; randomly adjust the K sets of initial process preparation parameters based on the K adjustment steps to obtain K sets of fine-tuned process preparation parameters; evaluate the preparation quality of the K sets of fine-tuned process preparation parameters to determine whether the preparation constraint information is satisfied. If so, perform random adjustment again and add the random adjustment method for obtaining the K sets of fine-tuned process preparation parameters to the taboo list, where the taboo list is used to prohibit the random adjustment method from being selected in the next preset taboo times; evaluate the K sets of fine-tuned process preparation parameters from two dimensions: overall preparation quality and individual process quality, to obtain the first preparation fitness.

[0031] Preferably, for K process devices, the system terminal adds up the reliability coefficients of the K process devices to calculate the total sum of the reliability coefficients of the K process devices. Then, it calculates the ratio of the reliability coefficient of each process device to the total sum of the reliability coefficients of the process devices, and subtracts the calculated ratio from 1 to obtain the adjustment step size for each process device, that is, K adjustment step sizes. For each adjustment step size, the larger the corresponding reliability coefficient of the process device, the smaller the adjustment step size, indicating that a smaller adjustment ratio is required during fine-tuning and the fine-tuning amplitude is smaller to maintain the stability of the process device. Subsequently, based on the calculated K adjustment step sizes, the initial preparation parameter sets of each process device are randomly adjusted to obtain K process fine-tuning preparation parameter sets, that is, adjusted according to the corresponding adjustment step sizes on the basis of the initial parameters. The adjustment amplitude can be positive or negative to maintain randomness. After obtaining the K process fine-tuning preparation parameter sets, the system terminal evaluates the preparation quality of these process fine-tuning preparation parameter sets. The evaluation criterion is to determine whether the adjusted parameter set meets the preparation constraint information. If the evaluation result shows that the preparation constraint information is met, subsequent random adjustments are performed, and the current random adjustment method is recorded in the taboo list. The taboo list is used to record the random adjustment results that have been used to avoid repeating the same adjustment results in several subsequent adjustments, thereby preventing falling into a local optimal solution. And the random adjustment results in the taboo list will be unlocked after a set number of taboo times and are allowed to be reused. Then, the K process fine-tuning preparation parameter sets are evaluated from two dimensions: overall preparation quality and individual process quality to obtain the overall first preparation fitness. Specifically, the system terminal controls the parameters of the process device by using the K process fine-tuning preparation parameter sets, conducts small-batch tests, and collects the chemical purity of the molecular sieve and the size distribution of the molecular sieve particles. Then, it calculates the deviation degree between the collected chemical purity and particle distribution and the corresponding standard values, and performs weighted summation on the calculated deviation degree by using the weights assigned according to the actual business needs to generate an overall preparation quality score. For each process device, the system terminal statistics the corresponding index data of the device. For example, the stability of the reaction temperature, pressure control accuracy, reaction time, etc. of the reaction kettle, the raw material mixing uniformity, stability of the stirring speed, etc. of the mixing device, the control accuracy of the drying temperature, humidity change situation, etc. of the drying device. Then, using the same method, the process quality score of each device is calculated and the mean value is calculated to obtain the individual process quality score. Then, the system terminal performs mean processing on the obtained overall preparation quality score and individual process quality score to generate the first preparation fitness for further optimization and adjustment.

[0032] Perform multi-scale optimization on the K sets of process fine-tuning preparation parameters based on the K adjustment step sizes to obtain K sets of process fine-tuning preparation parameters for different stages, where the K sets of process fine-tuning preparation parameters for different stages include the second preparation fitness. Determine whether the second preparation fitness is greater than or equal to the first preparation fitness. If so, use the K sets of process fine-tuning preparation parameters for different stages as the adjustment object; if not, use the K sets of process fine-tuning preparation parameters as the adjustment object. Perform multi-scale optimization on the adjustment object until the preset requirements are met, and use the K sets of process fine-tuning preparation parameters for different stages corresponding to the maximum preparation fitness during the optimization process as the target preparation parameter set.

[0033] Preferably, after obtaining the first preparation fitness, the system terminal uses the K adjustment step sizes to fine-tune the obtained K sets of process fine-tuning preparation parameters to obtain K sets of process fine-tuning preparation parameters for different stages, and then uses the same method as above to calculate the second preparation fitness of the K sets of process fine-tuning preparation parameters for different stages. Subsequently, compare the second preparation fitness with the first preparation fitness. If the second preparation fitness is greater than or equal to the first preparation fitness, it means that the current multi-scale optimization result is better than or equal to the previous result. At this time, the system terminal uses the current K sets of process fine-tuning preparation parameters for different stages as the new adjustment object for the next round of optimization. Otherwise, it indicates that the current optimization result is worse than the previous result. At this time, the system terminal retains the previous K sets of process fine-tuning preparation parameters as the adjustment object. After that, continue to perform multi-scale optimization on the selected adjustment object, that is, the K sets of process fine-tuning preparation parameters for different stages or the K sets of process fine-tuning preparation parameters. Continuously iterate, adjust different parameters, and continue to generate new sets of phased process parameters. As the multi-scale optimization progresses, the adjustment range of the parameters gradually narrows, making the result gradually approach the global optimal solution until the preset requirements are met, and retain the K sets of process fine-tuning preparation parameters for different stages corresponding to the maximum fitness as the final target preparation parameter set. This target preparation parameter set can maximize the quality and production efficiency of the target oxygen-producing molecular sieve under the existing conditions, improve the stability of the production process and the product quality, and achieve the optimal production effect.

[0034] Furthermore, the present application provides preset requirements, including: The preset requirement is that the number of optimization adjustment times meets the preset number or the difference in preparation fitness obtained from two adjacent adjustments is less than or equal to the preset preparation fitness difference.

[0035] Optionally, during the multi-scale optimization process, in order to determine when to stop the adjustment, the system terminal sets a preset requirement. This preset requirement includes two cases, namely the number of adjustments and the difference in preparation fitness. The number of adjustments means that after multiple iterative adjustments, if the preset maximum number of adjustments has been reached, it indicates that the optimization process has made enough attempts and further adjustments can be stopped. The difference in preparation fitness means that if the difference in preparation fitness between two adjacent adjustments is less than or equal to the preset difference in preparation fitness, it indicates that the process parameters have been adjusted to a relatively excellent state, which is equivalent to approaching or reaching the peak of optimization. At this time, the improvement that further adjustments may bring is very limited, so it can be considered that the optimization process is basically completed and the adjustment can be stopped.

[0036] Control the K process equipment sets to prepare the target oxygen generation molecular sieve according to the K process target preparation parameter sets.

[0037] In one embodiment, after finally determining the K process target preparation parameter sets, the system terminal uses these optimized parameters to control the K process equipment, that is, according to the best operating parameters of each device, set and adjust the operating conditions of these devices one by one to ensure that the entire preparation process of the oxygen generation molecular sieve proceeds according to the optimized plan. Through this precise control, it is possible to ensure that the preparation quality of the oxygen generation molecular sieve reaches the expected target and achieve efficient and stable production.

[0038] In the above text, with reference to Figure 1 The method for optimizing the preparation parameters of the oxygen generation molecular sieve according to the embodiments of the present invention has been described in detail. Next, with reference to Figure 2 The device for optimizing the preparation parameters of the oxygen generation molecular sieve according to the embodiments of the present invention will be described.

[0039] The device for optimizing the preparation parameters of the oxygen generation molecular sieve according to the embodiments of the present invention is used to solve the technical problems of unstable preparation parameters and inconsistent quality caused by fluctuations in equipment status and raw material characteristics during the preparation of different types of oxygen generation molecular sieves, and achieve the effect of optimizing the preparation parameters through comprehensive evaluation and multi-scale optimization to ensure the stability of the preparation process of the oxygen generation molecular sieve and the quality meets the expected requirements. The device for optimizing the preparation parameters of the oxygen generation molecular sieve includes: a raw material sampling and detection module 1, an equipment status analysis module 2, a preparation constraint information generation module 3, a comprehensive evaluation module 4, a multi-scale optimization module 5, and an oxygen generation molecular sieve preparation module 6.

[0040] Raw material sampling and testing module 1: The raw material sampling and testing module 1 is used to interact with the raw material library of the target oxygen-making molecular sieve within a preset production cycle to conduct raw material sampling and testing, and obtain a raw material feature set; Equipment status analysis module 2: The equipment status analysis module 2 is used to traverse the K process equipment sets of the target oxygen-making molecular sieve to conduct equipment status analysis, and obtain K process equipment reliability coefficients; Preparation constraint information generation module 3: The preparation constraint information generation module 3 is used to collect the preparation type, preparation quantity, and preparation quality requirements of the target oxygen-making molecular sieve, and generate preparation constraint information; Comprehensive evaluation module 4: The comprehensive evaluation module 4 is used to retrieve the historical preparation database of the target oxygen-making molecular sieve, conduct a comprehensive evaluation on the historical preparation database. When the evaluation passes, the raw material feature set, the K process equipment reliability coefficients, and the preparation constraint information are input into the historical preparation database for retrieval, and an initial preparation parameter set is obtained. Among them, the initial preparation parameter set includes K process initial preparation parameter sets; Multi-scale optimization module 5: The multi-scale optimization module 5 is used to perform multi-scale optimization on the K process initial preparation parameter sets based on the K process equipment reliability coefficients, and use the preparation constraint information to constrain the optimization process until the preset requirements are met, and obtain a target preparation parameter set. Among them, the target preparation parameter set includes K process target preparation parameter sets; Oxygen-making molecular sieve preparation module 6: The oxygen-making molecular sieve preparation module 6 is used to control the K process equipment sets to prepare the target oxygen-making molecular sieve according to the K process target preparation parameter sets.

[0041] Further, the raw material sampling and testing module 1 further includes: Randomly select M raw material samples from the raw material library in accordance with the simple random sampling method; Traverse the M raw material samples to conduct multi-parameter testing, and obtain M raw material sample feature sets. Among them, each raw material sample feature set includes composition features, purity features, and particle size distribution features; Respectively, using composition, purity, and particle size distribution as indexes, retrieve the M raw material sample feature sets to obtain a raw material sample composition feature set, a raw material sample purity feature set, and a raw material particle size distribution feature set; Traverse the raw material sample composition feature set, the raw material sample purity feature set, and the raw material particle size distribution feature set to conduct centralized feature analysis, and obtain raw material composition features, raw material purity features, and raw material particle size distribution features; Use the raw material composition features, raw material purity features, and the raw material particle size distribution features as the raw material feature set.

[0042] Further, the raw material sampling and testing module 1 further includes: Construct a D-dimensional centralized analysis space based on the set of raw material sample component features, where D is the number of sub-features in each raw material sample component feature, D is a positive integer greater than or equal to 2, and the D-dimensional centralized analysis space includes M spatial points, each spatial point corresponding to a raw material sample component feature; take the centroid of the D-dimensional centralized analysis space as the first clustering center, and determine the first clustering neighborhood of the first clustering center according to a preset distance bandwidth; input the first clustering center and the first clustering neighborhood into the center update formula for clustering center update to obtain the second clustering center; after multiple clustering center update iterations until the preset iteration requirement is met, take the raw material sample component features corresponding to the clustering center obtained in the last iteration as the raw material component features.

[0043] Further, the raw material sampling and detection module 1 further includes: The center update formula is: ; where is the second clustering center, is the th spatial point in the first clustering neighborhood, is the first clustering center, is the first clustering neighborhood.

[0044] Further, the equipment status analysis module 2 further includes: Traverse the K process equipment sets to search for maintenance records, and obtain K clusters of process equipment historical maintenance records. Each cluster of process equipment historical maintenance records includes multiple sets of process equipment historical maintenance records, and each set of process equipment historical maintenance records corresponds to a process equipment; perform reliability analysis on the K clusters of process equipment historical maintenance records from two dimensions of the number of maintenance times and equipment loss amount to obtain K process equipment reliability coefficients.

[0045] Further, the multi-scale optimization module 5 further includes: Calculate the ratios of the K process equipment reliability coefficients to the sum of the K process equipment reliability coefficients respectively, and take the difference between 1 and the ratios as the K adjustment steps; Randomly adjust the K sets of initial process preparation parameters based on the K adjustment step sizes to obtain K sets of fine-tuned process preparation parameters; evaluate the preparation quality of the K sets of fine-tuned process preparation parameters to determine whether the preparation constraint information is satisfied. If so, perform random adjustment again and add the random adjustment method for obtaining the K sets of fine-tuned process preparation parameters to the taboo list, where the taboo list is used to prohibit the selection of the random adjustment method in the next preset taboo times; evaluate the K sets of fine-tuned process preparation parameters from two dimensions of overall preparation quality and single-process quality to obtain the first preparation fitness; perform multi-scale optimization on the K sets of fine-tuned process preparation parameters based on the K adjustment step sizes to obtain K sets of stage fine-tuned process preparation parameters, where the K sets of stage fine-tuned process preparation parameters include the second preparation fitness; determine whether the second preparation fitness is greater than or equal to the first preparation fitness. If so, use the K sets of stage fine-tuned process preparation parameters as the adjustment object; if not, use the K sets of fine-tuned process preparation parameters as the adjustment object; perform multi-scale optimization on the adjustment object until the preset requirements are met, and use the K sets of stage fine-tuned process preparation parameters corresponding to the maximum preparation fitness during the optimization process as the target preparation parameters.

[0046] Further, the multi-scale optimization module 5 further includes: The preset requirement is that the number of optimization adjustment times meets the preset number or the difference in preparation fitness obtained from two adjacent adjustments is less than or equal to the preset preparation fitness difference.

[0047] The device for optimizing the preparation parameters of the oxygen-producing molecular sieve provided by the embodiments of the present invention can execute the method for optimizing the preparation parameters of the oxygen-producing molecular sieve provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0048] Although various references are made to certain modules in the device according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The included units and modules 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.

[0049] The above specific embodiments do not constitute a limitation to 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.

Claims

1. A method for optimizing the preparation parameters of an oxygen-producing molecular sieve, characterized in that: The method comprises: The interactive target oxygen-producing molecular sieve performs raw material sampling and testing in the raw material warehouse within the preset production cycle to obtain a raw material feature set; Traversing the K process equipment sets of the target oxygen-producing molecular sieve to perform equipment status analysis and obtain K process equipment reliability coefficients; Collecting the preparation type, preparation quantity and preparation quality requirements of the target oxygen-producing molecular sieve to generate preparation constraint information; Retrieving the historical preparation database of the target oxygen-producing molecular sieve, and comprehensively evaluating the historical preparation database. When the evaluation is passed, inputting the raw material feature set, K process equipment reliability coefficients and the preparation constraint information into the historical preparation database for retrieval to obtain an initial preparation parameter set, wherein the initial preparation parameter set includes K process initial preparation parameter sets; Based on the K process equipment reliability coefficients, multi-scale optimization is performed on the K process initial preparation parameter sets, and the optimization process is constrained by using the preparation constraint information until the preset requirements are met, so as to obtain a target preparation parameter set, wherein the target preparation parameter set includes K process target preparation parameter sets; The K process equipment sets are controlled according to the K process target preparation parameter sets to prepare the target oxygen-producing molecular sieve.

2. The method for optimizing the preparation parameters of the oxygen-producing molecular sieve according to claim 1, characterized in that: include: Randomly select M raw material samples from the raw material library in a simple random sampling manner; Traversing the M raw material samples to perform multi-parameter detection to obtain M raw material sample feature sets, wherein each raw material sample feature set includes a component feature, a purity feature, and a particle size distribution feature; Retrieve the M raw material sample feature sets using composition, purity and particle size distribution as indexes respectively, and obtain a raw material sample composition feature set, a raw material sample purity feature set and a raw material particle size distribution feature set; Traversing the raw material sample component feature set, the raw material sample purity feature set and the raw material particle size distribution feature set to perform centralized feature analysis to obtain raw material component features, raw material purity features and raw material particle size distribution features; The raw material component characteristics, raw material purity characteristics and raw material particle size distribution characteristics are taken as the raw material characteristic set.

3. The method for optimizing the preparation parameters of the oxygen-producing molecular sieve according to claim 2, characterized in that: include: Constructing a D-dimensional centralized analysis space based on the raw material sample component feature set, wherein D is the number of sub-features in each raw material sample component feature, D is a positive integer greater than or equal to 2, and the D-dimensional centralized analysis space includes M space points, each space point corresponding to a raw material sample component feature; Taking the center of gravity of the D-dimensional concentrated analysis space as a first cluster center, and determining a first cluster neighborhood of the first cluster center according to a preset distance bandwidth; Inputting the first cluster center and the first cluster neighborhood into a center update formula to update the cluster center, thereby obtaining a second cluster center; After multiple cluster center update iterations until the preset iteration requirements are met, the raw material sample component characteristics corresponding to the cluster center obtained in the last iteration are used as the raw material component characteristics.

4. The method for optimizing the preparation parameters of the oxygen-producing molecular sieve according to claim 3, characterized in that: The center update formula is: ； in, is the second cluster center, is the first cluster neighborhood A point in space, is the first cluster center, is the first cluster neighborhood.

5. The method for optimizing the preparation parameters of the oxygen-producing molecular sieve according to claim 1, characterized in that: include: Traversing the K process equipment sets to search for maintenance records, and obtaining K process equipment historical maintenance record clusters, each process equipment historical maintenance record cluster including multiple process equipment historical maintenance record sets, and each process equipment historical maintenance record set corresponding to one process equipment; The K process equipment historical maintenance record clusters are subjected to reliability analysis from two dimensions, namely, the number of maintenance times and the amount of equipment loss, to obtain K process equipment reliability coefficients.

6. The method for optimizing the preparation parameters of the oxygen-producing molecular sieve according to claim 1, characterized in that: include: Calculate the ratio of the K process equipment reliability coefficients to the sum of the K process equipment reliability coefficients respectively, and use the difference between 1 and the ratio as K adjustment step lengths; Randomly adjusting the K initial process preparation parameter sets based on the K adjustment step sizes to obtain K process fine-tuning preparation parameter sets; Performing a preparation quality assessment on the K process fine-tuning preparation parameter sets to determine whether the preparation constraint information is satisfied, and if so, re-performing a random adjustment, and adding the random adjustment method of the K process fine-tuning preparation parameter sets to a taboo list, wherein the taboo list is used to prohibit the random adjustment method from being selected in the next preset taboo number of times; Evaluate the K process fine-tuning preparation parameter sets from two dimensions: overall preparation quality and individual process quality, to obtain a first preparation fitness; Performing multi-scale optimization on the K process fine-tuning preparation parameter sets based on the K adjustment step sizes to obtain K stage process fine-tuning preparation parameter sets, wherein the K stage process fine-tuning preparation parameter sets include a second preparation fitness; Determine whether the second preparation fitness is greater than or equal to the first preparation fitness, and if so, take the K stage process fine-tuning preparation parameter sets as adjustment objects; If not, the K process fine-tuning preparation parameter sets are taken as adjustment objects; Multi-scale optimization is performed on the adjustment object until the preset requirements are met, and the K-stage process fine-tuning preparation parameter sets corresponding to the maximum preparation fitness in the optimization process are taken as the target preparation parameter set.

7. The method for optimizing the preparation parameters of the oxygen-producing molecular sieve according to claim 6, characterized in that: The preset requirement is that the number of optimization adjustments meets the preset number or the preparation fitness difference obtained from two adjacent adjustments is less than or equal to the preset preparation fitness difference.

8. A device for optimizing the preparation parameters of oxygen-producing molecular sieves, characterized in that: The device is used to implement the preparation parameter optimization method of the oxygen-producing molecular sieve according to any one of claims 1 to 7, comprising: Raw material sampling and detection module: The interactive target oxygen-producing molecular sieve performs raw material sampling and detection in the raw material warehouse within the preset production cycle to obtain a raw material feature set; Equipment status analysis module: traverses the K process equipment sets of the target oxygen-producing molecular sieve to perform equipment status analysis and obtain K process equipment reliability coefficients; Preparation constraint information generation module: collects the preparation type, preparation amount and preparation quality requirements of the target oxygen-producing molecular sieve, and generates preparation constraint information; Comprehensiveness evaluation module: retrieve the historical preparation database of the target oxygen-producing molecular sieve, conduct a comprehensive evaluation on the historical preparation database, and when the evaluation passes, input the raw material feature set, K process equipment reliability coefficients and the preparation constraint information into the historical preparation database for retrieval to obtain an initial preparation parameter set, wherein the initial preparation parameter set includes K process initial preparation parameter sets; Multi-scale optimization module: based on the K process equipment reliability coefficients, multi-scale optimization is performed on the K process initial preparation parameter sets, and the preparation constraint information is used to constrain the optimization process until the preset requirements are met to obtain a target preparation parameter set, wherein the target preparation parameter set includes K process target preparation parameter sets; Oxygen-producing molecular sieve preparation module: controls the K process equipment sets to prepare the target oxygen-producing molecular sieve according to the K process target preparation parameter sets.

Images