Catalyst recovery control system and method based on multistage centrifugal separation

Abstract

The invention provides a catalyst recovery control system and method based on multi-stage centrifugal separation, and relates to the technical field of catalyst separation, the system comprises: a device building module for building a multi-stage centrifugal separation device; the separation and analysis module is used for determining M catalyst centrifugal process parameters; the separation treatment module is used for performing preliminary separation treatment on the target mixture and acquiring M catalyst state parameters at the same time; and the recovery control module is used for carrying out separation effect calculation and multi-stage dynamic adjustment on the M catalyst centrifugal process parameters according to the catalyst state parameters, determining multi-stage centrifugal separation strategy parameters and carrying out catalyst recovery control on the target mixture. The technical problem that the catalyst recovery efficiency is low due to the fact that the catalyst separation efficiency is insufficient and the catalyst is wasted in the prior art can be solved, the optimization of the catalyst state in the whole separation process is ensured through multi-stage centrifugal separation, and the catalyst recovery efficiency is improved.

Description

Technical Field

[0001] This application relates to the technical field of catalyst separation, and particularly to a catalyst recovery control system and method based on multi-stage centrifugal separation. Background Art

[0002] Catalyst recovery refers to the process of separating, purifying, and reusing the catalyst used in the chemical reaction process from the reaction system through a series of technical means. Although existing catalyst recovery methods (such as filtration, sedimentation, and conventional centrifugation) have been widely used, these methods usually have problems of low separation efficiency when dealing with catalysts with small particle sizes and complex morphologies. Especially when dealing with tiny particles or catalysts with strong surface adhesion, traditional separation methods often cannot achieve complete catalyst separation. Usually, only most of the catalysts can be recovered, while the fine particles or catalysts attached to the reaction system often are discarded together with the waste, resulting in catalyst waste. In addition, many existing catalyst recovery methods adopt a static operation mode, that is, the separation parameters of the catalyst (such as centrifugal rate, temperature, etc.) are set at the beginning during the recovery process and are not adjusted during the whole recovery process. Such a setting of static parameters is difficult to adapt to the changes in the mixture composition and the catalyst state, resulting in poor separation effect and low catalyst recovery rate.

[0003] In summary, there are technical problems in the prior art that due to insufficient catalyst separation efficiency, catalyst waste occurs, resulting in low catalyst recovery efficiency. Summary of the Invention

[0004] The purpose of this application is to provide a catalyst recovery control system and method based on multi-stage centrifugal separation to solve the technical problems in the prior art that due to insufficient catalyst separation efficiency, catalyst waste occurs, resulting in low catalyst recovery efficiency.

[0005] In view of the above problems, this application provides a catalyst recovery control system and method based on multi-stage centrifugal separation.

[0006] In a first aspect, the present application provides a catalyst recovery control system based on multi-stage centrifugal separation. Among them, the catalyst recovery control system based on multi-stage centrifugal separation includes: a device construction module for constructing a multi-stage centrifugal separation device, which is composed of M centrifugal separation units connected in series, and a monitoring sensor group is arranged in each separation unit of the M centrifugal separation units; a separation and analysis module for collecting and obtaining the initial composition data of the target mixture, separating and analyzing the initial composition data based on the M centrifugal separation units, and determining M catalyst centrifugation process parameters; a separation and treatment module for performing preliminary separation treatment on the target mixture through the multi-stage centrifugal separation device based on the M catalyst centrifugation process parameters, and simultaneously using the monitoring sensor group to obtain M catalyst state parameters in real time; a recovery control module for calculating the separation effect and performing multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters, determining multi-stage centrifugal separation strategy parameters, and controlling the catalyst recovery of the target mixture through the multi-stage centrifugal separation strategy parameters.

[0007] Optionally, a database acquisition unit for mining and obtaining M centrifugal process databases of the M centrifugal separation units, where the M centrifugal process databases include historical mixture composition data, centrifugal process parameters, and corresponding separation effect data; a screening and partitioning unit for screening and partitioning the M centrifugal process databases according to the separation effect data to obtain M available centrifugal process data sets; a clustering and fitting unit for performing feature clustering and fitting on the M available centrifugal process data sets based on the historical mixture composition data to obtain a multi-stage centrifugal process analysis channel; a process analysis unit for performing centrifugal process analysis on the initial composition data using the multi-stage centrifugal process analysis channel to determine the M catalyst centrifugation process parameters.

[0008] Optionally, a feature extraction sub-unit for extracting key features from the historical mixture composition data based on the separation effect data to obtain a key composition feature set; a clustering analysis sub-unit for performing a clustering operation on the M available centrifugal process data sets according to the key composition feature set to obtain M centrifugal process cluster sets; an integration processing sub-unit for integrating the M centrifugal process cluster sets according to the clustering clusters to obtain a multi-stage centrifugal process cluster set; a channel fitting sub-unit for performing analysis channel fitting on each process cluster in the multi-stage centrifugal process cluster set to construct a multi-stage centrifugal process analysis channel.

[0009] Optionally, an architecture selection component is configured to select a set of network architectures according to the characteristic information of the multi-stage centrifugation process cluster set; a centrifugation fitting component is configured to perform centrifugation process fitting on each process cluster in the multi-stage centrifugation process cluster set by using the set of network architectures to generate an initial process analysis branch network set; a verification and tuning component is configured to perform cross-verification and iterative tuning on the initial process analysis branch network set respectively to obtain a centrifugation process analysis branch network set; a parallel embedding component is configured to perform channel architecture construction and network parallel embedding based on the centrifugation process analysis branch network set to construct the multi-stage centrifugation process analysis channel.

[0010] Optionally, a normalization processing unit is configured to normalize the M catalyst state parameters according to the data application standard to obtain M standard catalyst state parameters; an index establishment unit is configured to establish M separation effect evaluation index systems according to the M catalyst recovery targets of the M centrifugal separation units; a quantification calculation unit is configured to perform effect quantification calculation on the M standard catalyst state parameters based on the M separation effect evaluation index systems to obtain M catalyst separation effect parameters; a multi-stage adjustment unit is configured to perform multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst separation effect parameters to obtain multi-stage centrifugation separation strategy parameters.

[0011] Optionally, an index extraction subunit is configured to perform multi-dimensional index extraction and calculation formula association on the M separation effect evaluation index systems to obtain M separation effect index calculation components; an index effect calculation subunit is configured to perform index effect calculation on the M standard catalyst state parameters by using the M separation effect index calculation components to obtain M separation effect index parameters; a separation effect calculation subunit is configured to use the ratio of the M separation effect index parameters to M preset separation effect standards as the M catalyst separation effect parameters.

[0012] Optionally, a strategy optimization subunit is configured to perform optimization strategy analysis on the M catalyst centrifugation process parameters by using the M catalyst separation effect parameters to determine M centrifugation parameter optimization strategies; a simulation and modeling subunit is configured to perform evaluation simulation on the M centrifugation process databases according to the M separation effect evaluation index systems to construct M separation effect simulation units; a strategy adjustment subunit is configured to perform multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M separation effect simulation units and the M centrifugation parameter optimization strategies to obtain multi-stage centrifugation separation strategy parameters.

[0013] Optionally, a feasible threshold determination component is configured to perform multi-level dynamic adjustment on the M catalyst centrifugation process parameters based on the M centrifugation parameter optimization strategies to obtain M feasible thresholds for the catalyst centrifugation process; an initialization component is configured to initialize an M catalyst centrifugation process particle space according to the M feasible thresholds for the catalyst centrifugation process; a simulation optimization component is configured to perform global simulation optimization in the M catalyst centrifugation process particle space by using the M separation effect simulation units until a preset termination condition is reached, and determine multi-level centrifugal separation strategy parameters.

[0014] Optionally, a parameter determination channel is configured to perform global simulation optimization in the M catalyst centrifugation process particle space by using the M separation effect simulation units until a preset termination condition is reached to obtain M catalyst centrifugal separation strategy parameters; an influence factor evaluation channel is configured to evaluate and obtain influence factor information of the M catalyst centrifugal separation strategy parameters; an equilibrium correction channel is configured to perform collaborative equilibrium correction on the M catalyst centrifugal separation strategy parameters based on the influence factor information to determine the multi-level centrifugal separation strategy parameters.

[0015] In a second aspect, the present application further provides a catalyst recovery control method based on multi-level centrifugal separation. The catalyst recovery control method based on multi-level centrifugal separation includes: building a multi-level centrifugal separation device, the multi-level centrifugal separation device is composed of M centrifugal separation units connected in series, and a monitoring sensor group is arranged in each separation unit of the M centrifugal separation units; collecting and obtaining initial composition data of a target mixture, performing separation analysis on the initial composition data based on the M centrifugal separation units to determine M catalyst centrifugation process parameters; performing preliminary separation processing on the target mixture by the multi-level centrifugal separation device based on the M catalyst centrifugation process parameters, and simultaneously using the monitoring sensor group to obtain M catalyst state parameters in real time; calculating the separation effect and performing multi-level dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters to determine multi-level centrifugal separation strategy parameters, and performing catalyst recovery control on the target mixture through the multi-level centrifugal separation strategy parameters.

[0016] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The device construction module is used to construct a multi-stage centrifugal separation device, which is composed of M centrifugal separation units connected in series, and a monitoring sensor group is arranged in each separation unit of the M centrifugal separation units; the separation and analysis module is used to collect and obtain the initial composition data of the target mixture, and perform separation and analysis on the initial composition data based on the M centrifugal separation units to determine M catalyst centrifugation process parameters; the separation and treatment module is used to perform preliminary separation treatment on the target mixture based on the M catalyst centrifugation process parameters through the multi-stage centrifugal separation device, and simultaneously use the monitoring sensor group to obtain M catalyst state parameters in real time; the recovery control module is used to perform separation effect calculation and multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters, determine the multi-stage centrifugal separation strategy parameters, and perform catalyst recovery control on the target mixture through the multi-stage centrifugal separation strategy parameters. That is to say, by connecting multiple centrifugal separation units in series, equipped with a monitoring sensor group, performing multi-stage centrifugal separation, monitoring the catalyst state in real time, performing separation effect calculation and multi-stage dynamic adjustment, accurately regulating the working parameters of each centrifugal separation unit, ensuring that the recovery effect reaches the best, achieving the best recovery purity of the catalyst, and improving the efficiency and accuracy of catalyst recovery.

[0017] The above description is only an overview of the technical solution 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 easy to understand, the following specific embodiments of the present application are specifically given. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. Brief Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings described below are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0019] Figure 1 It is a schematic structural diagram of the catalyst recovery control system based on multi-stage centrifugal separation of the present application; Figure 2 It is a schematic flow diagram of the catalyst recovery control method based on multi-stage centrifugal separation of the present application.

[0020] Description of the reference numerals: device construction module 11, separation and analysis module 12, separation and treatment module 13, recovery control module 14. Detailed implementation manners

[0021] By providing a catalyst recovery control system and method based on multi-stage centrifugal separation, the present application solves the technical problem in the prior art that due to insufficient catalyst separation efficiency, catalyst waste occurs, resulting in low catalyst recovery efficiency. By connecting multiple centrifugal separation units in series and equipping with a monitoring sensor group, multi-stage centrifugal separation is carried out, the catalyst state is monitored in real time, and the separation effect is calculated and multi-stage dynamic adjustment is performed to accurately control the working parameters of each centrifugal separation unit, ensuring that the recovery effect reaches the optimal, achieving the optimal recovery purity of the catalyst, and improving the efficiency and accuracy of catalyst recovery.

[0022] Next, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the scope of protection of the present application. In addition, it should be noted that for the sake of description, only the parts related to the present application are shown in the drawings rather than all.

[0023] Embodiment 1. Please refer to the attached Figure 1 drawings. The present application provides a catalyst recovery control system based on multi-stage centrifugal separation. Among them, the catalyst recovery control system based on multi-stage centrifugal separation is used to implement the steps of the catalyst recovery control method based on multi-stage centrifugal separation. The catalyst recovery control system based on multi-stage centrifugal separation includes: A device construction module 11, configured to construct a multi-stage centrifugal separation device, where the multi-stage centrifugal separation device is composed of M centrifugal separation units connected in series, and a monitoring sensor group is arranged in each separation unit of the M centrifugal separation units.

[0024] Specifically, a multi-stage centrifugal separation device composed of M centrifugal separation units connected in series is constructed. That is to say, the number of centrifugal separation units required (M) is determined, and the M centrifugal separation units are connected in series to form a continuous separation process to complete the construction of the multi-stage centrifugal separation device. Inside each centrifugal separation unit, a monitoring sensor group is arranged to monitor and record various parameters in the separation process in real time. The multi-stage centrifugal device works through the cooperation of multiple centrifugal separation units to gradually improve the separation efficiency and solve the problem that a single centrifugal separation unit cannot efficiently process complex mixtures. During the centrifugation process, particles or substances with different densities will be separated due to the action of centrifugal force. Substances with higher density will be thrown to the outside, while substances with lower density will be concentrated in the central area.

[0025] In each centrifugal separation unit, a monitoring sensor group is arranged to monitor the physicochemical state of the catalyst in real time, such as information on the density, movement speed, temperature, pressure, etc. of the catalyst particles. The sensor group can not only monitor the separation effect but also capture the change of the catalyst state in real time. For example, assume that in a certain catalyst recovery process, a multi-stage centrifugal device with 3 centrifugal separation units is used. Temperature, pressure, particle density and other sensors are installed in each centrifugal separation unit. The test data shows that through real-time monitoring and adjustment, the centrifugal speed of the first unit is set at 3000 rpm, the second unit is set at 4000 rpm, and the third unit is 5000 rpm. During the entire recovery process, the recovery rate of the catalyst is increased from 70% to 85%, and due to the adjustment of real-time monitoring, the loss of the catalyst is reduced by 10%. Through the collaborative work of the multi-stage centrifugal separation units, the separation process is more refined, effectively improving the catalyst recovery rate.

[0026] The separation and analysis module 12 is used to collect and obtain the initial composition data of the target mixture, and perform separation and analysis on the initial composition data based on the M centrifugal separation units to determine M catalyst centrifugation process parameters.

[0027] Furthermore, the separation and analysis module 12 in the catalyst recovery control system based on multi-stage centrifugal separation is further used for: The database acquisition unit is used to mine and obtain M centrifugal process databases of the M centrifugal separation units. The M centrifugal process databases include historical mixture composition data, centrifugal process parameters, and corresponding separation effect data; the screening and partitioning unit is used to screen and partition the M centrifugal process databases according to the separation effect data to obtain M available centrifugal process data sets; the clustering and fitting unit is used to perform feature clustering and fitting on the M available centrifugal process data sets based on the historical mixture composition data to obtain a multi-stage centrifugal process analysis channel; the process analysis unit is used to perform centrifugal process analysis on the initial composition data by using the multi-stage centrifugal process analysis channel to determine the M catalyst centrifugation process parameters.

[0028] The feature extraction subunit is used to extract key features from the historical mixture composition data based on the separation effect data to obtain a key composition feature set; the clustering analysis subunit is used to perform a clustering operation on the M available centrifugal process data sets according to the key composition feature set to obtain M centrifugal process cluster sets; the integration processing subunit is used to integrate and process the M centrifugal process cluster sets according to the clustering clusters to obtain a multi-stage centrifugal process cluster set; the channel fitting subunit is used to perform analysis channel fitting on each process cluster in the multi-stage centrifugal process cluster set to construct a multi-stage centrifugal process analysis channel.

[0029] Specifically, the initial composition data of the target mixture is obtained, and the detailed composition and concentration of the mixture are obtained through chemical analysis methods (such as gas chromatography, liquid chromatography, mass spectrometry, etc.). The initial composition data of the target mixture refers to the data of each component in the target mixture obtained before the catalyst recovery is carried out, which usually includes information such as the type and concentration of the catalyst, reaction products and other impurities. For example, the average particle size of the catalyst particles is 10μm, the density of the catalyst particles is 2.5g / cm³, and the solvent concentration is 50%. A process database related to catalyst recovery is extracted from the historical production logs of M centrifugal separation units, which contains different process data of multiple M centrifugal separation units, including the composition information of the historical mixture (such as the ratio of catalysts, reaction products, impurities, etc.), the centrifugal parameters used during separation (such as centrifugal speed, temperature, etc.), and the actual separation effect (such as the recovery rate and purity of the catalyst).

[0030] Historical mixture composition data refers to the composition of different mixtures from previous experiments or production processes. Centrifugation process parameters are the specific operating conditions used in previous centrifugation separation processes, such as centrifugation rate, temperature, time, and centrifugal force. Separation performance data refers to the results of previous centrifugation, including catalyst recovery rate, separation purity, recovery time, etc., to evaluate the separation performance under different process parameters.

[0031] Based on the performance data of each centrifugal separation unit (such as catalyst recovery rate and separation purity), M centrifugal process databases were screened and divided, retaining process data with good separation performance and eliminating process data with poor separation performance, resulting in M usable centrifugal process data sets. Feature clustering fitting was performed on these M usable centrifugal process data sets based on historical mixture composition data. Based on the separation performance data, the features most relevant to the separation performance (i.e., the features with the greatest impact) were extracted from the historical mixture composition data. Cluster analysis was then performed on these M usable centrifugal process data sets based on this extracted feature set, resulting in a multi-stage centrifugal process analysis channel.

[0032] Specifically, key features are extracted from historical mixture composition data based on separation performance data. This means extracting features that are most relevant to catalyst recovery and separation performance from historical mixture composition data. For example, factors such as the ratio of catalysts A and B, impurity type, and solution viscosity significantly impact separation performance, so a set of features with significant impact on separation performance is extracted. Key component feature sets are the most representative component feature data extracted from historical mixture composition data that most significantly impact separation performance. These typically refer to key factors influencing catalyst recovery and separation purity, such as the ratio of different catalysts and impurity types.

[0033] Cluster the M available centrifugation process datasets according to the key component feature set, grouping the historical centrifugation process data with similar features together to form a set of centrifugation process clusters. For example, based on the key feature set (such as the concentration of catalyst A, separation temperature, etc.), all processes are divided into several clusters: Cluster 1 is low-temperature and low-speed separation, Cluster 2 is high-temperature and high-speed separation, and Cluster 3 is medium-speed and medium-temperature separation. Each cluster represents a combination of process parameters, and within each cluster, the separation effects are relatively similar. The clustering operation is a data analysis method that helps find similar process patterns in large-scale data by grouping similar data (such as similar centrifugation process parameters and effects), thereby optimizing the centrifugation separation process in a targeted manner. For example, classify according to the catalyst particle size and material composition, clustering the mixture with catalyst particles larger than 50 μm into one category and the particles smaller than 50 μm into another category.

[0034] Each cluster in the set of M centrifugation process clusters represents a similar centrifugation process pattern, including similar process parameters and corresponding separation effects. Integrate the set of M centrifugation process clusters obtained through the clustering operation according to the clustering clusters to form a multi-level centrifugation process cluster set, that is, reasonably merge different centrifugation process clusters to construct a centrifugation process system. The purpose of integration is to merge multiple process clusters with similar features into a larger and more hierarchical process set, which not only includes different process clusters but also reflects the relationships and interactions between different process stages.

[0035] During the integration process, the cluster set may be organized according to different stages of centrifugation separation. For example, the low-speed separation in the first stage, the medium-speed separation in the second stage, and the high-speed separation in the third stage may represent different sets of process clusters. Integrating these clusters at different stages can construct a multi-level centrifugation process system that includes different process steps. For example, assume there are 3 clustering clusters: Cluster 1 (low-speed centrifugation), Cluster 2 (medium-speed centrifugation), and Cluster 3 (high-speed centrifugation), which represent different separation stages. When integrating, according to different separation conditions, merge Cluster 1, Cluster 2, and Cluster 3 into an overall multi-level centrifugation process cluster set. In this set, the relationships between different clusters are clear, ensuring that the most suitable process parameters are available for each stage. The integrated multi-level centrifugation process cluster set is usually organized according to the hierarchical relationship of process steps, ensuring that in the actual operation process, the optimal process cluster can be dynamically selected according to the characteristics of the target mixture for operation.

[0036] Select a suitable set of network architectures based on the characteristic information of the multi-stage centrifugation process cluster set, and use the set of network architectures to fit each process cluster in the multi-stage centrifugation process cluster set to generate an initial process analysis branch network set. Optimize these networks through cross-validation and iterative tuning to improve the prediction accuracy and obtain the centrifugation process analysis branch network set. Construct the channel architecture and parallelly embed the optimized centrifugation process analysis branch network set to complete the construction of the multi-stage centrifugation process analysis channel.

[0037] Use the multi-stage centrifugation process analysis channel to preliminarily separate the initial composition data of the obtained target mixture, that is, according to the initial composition data of the target mixture, use the established analysis channel to specifically analyze the separation process, and determine the catalyst centrifugation process parameters suitable for the current mixture, including rotation speed, time, temperature, pressure, etc., for preliminary separation processing of the initial composition data.

[0038] By obtaining the initial composition data of the target mixture and establishing a multi-stage centrifugation process analysis channel using historical data to determine the catalyst centrifugation process parameters, it helps to optimize the centrifugation separation process and improve the separation efficiency. Through clustering and analysis channel fitting, the most effective process patterns are summarized from a large amount of historical data, reducing the trial-and-error process in experiments and improving the catalyst recovery efficiency. By fitting the analysis channels of each process cluster, the most suitable process conditions are selected for different catalyst recovery tasks, thereby maximizing the catalyst recovery rate and separation purity.

[0039] Furthermore, the separation analysis module 12 in the catalyst recovery control system based on multi-stage centrifugation separation is also used for: an architecture selection component, which is used to select a set of network architectures according to the characteristic information of the multi-stage centrifugation process cluster set; a centrifugation fitting component, which is used to perform centrifugation process fitting on each process cluster in the multi-stage centrifugation process cluster set using the set of network architectures to generate an initial process analysis branch network set; a verification and tuning component, which is used to perform cross-validation and iterative tuning on the initial process analysis branch network set respectively to obtain the centrifugation process analysis branch network set; a parallel embedding component, which is used to build a channel architecture and perform network parallel embedding based on the centrifugation process analysis branch network set to construct the multi-stage centrifugation process analysis channel.

[0040] Specifically, according to the characteristics of the multi-stage centrifugation process cluster set (such as the composition of each cluster, separation effect, etc.), considering the data structure and characteristics of the cluster, a suitable network architecture is selected. For example, a convolutional neural network (CNN) is selected to process image data, or a recurrent neural network (RNN) is selected to process time-series data. The selection of the network architecture should optimize the processing effect according to the characteristics of different clusters. Suppose a certain centrifugation process cluster is based on time-series data (such as the separation rate at each stage changes over time), an RNN architecture is selected for fitting and prediction; while the characteristics of another cluster are based on image data (such as the operating state image of the equipment), then a CNN is selected for processing. The network architecture generally refers to the structure of the neural network, including the number of layers, the number of neurons in each layer, activation functions, etc. Different network architectures show different effects when processing different types of data.

[0041] Use the selected network architecture to fit the data of each centrifugation process cluster. According to the characteristics of each cluster (such as separation efficiency, process parameters, etc.), optimize the weights of the network so that it can effectively predict and explain the process behavior. The goal of fitting is to adjust the parameters of the network through the training process so that it can effectively describe the behavior of each cluster, and then achieve accurate separation effect prediction. For example, suppose the data characteristics of cluster 1 are based on time-series data (such as separation efficiency at different time points), an RNN architecture is selected for fitting, and the mean squared error is used as the loss function for training. By adjusting the learning rate and the number of layers, a model that can better fit the data of this cluster is finally obtained. Optimize the weights of the network by minimizing the loss function so that it can fit the data of each process cluster. After fitting, a network model is generated for each centrifugation process cluster, forming an initial process analysis branch network set. Each branch network works independently and focuses on processing the data of a specific process cluster.

[0042] Perform cross-validation on the initial process analysis branch network set. By dividing the data into multiple subsets and verifying the performance of each model on different data sets one by one, the robustness of the model is ensured. Specifically, divide the data set corresponding to each network model (that is, the data corresponding to the multi-stage centrifugation process cluster set) into several subsets, and then sequentially use each subset as the validation set, and the remaining part as the training set for model training and validation. Through iterative tuning, optimize the hyperparameters (such as learning rate, number of layers, activation functions, etc.) of each branch network to improve the generalization ability of the model, find the optimal combination of hyperparameters, and make the performance of each branch network reach the best. After cross-validation and iterative tuning, each initial process analysis branch network has undergone an optimization process, can more accurately fit the data of the centrifugation process cluster, and obtain a centrifugation process analysis branch network set.

[0043] A hierarchical and multi - level channel architecture is constructed by integrating multiple branch networks, with each network responsible for parsing a centrifugation process cluster. Each network processes data of different process clusters and finally combines the parsing results of all process clusters. Each centrifugation process parsing branch network is embedded into the whole as a sub - module, and the input and output channels of the sub - module are constructed to ensure that data of different process clusters can flow in sequence and be properly processed. The construction of the channel architecture refers to designing and building an overall data flow channel suitable for the multi - level centrifugation process on the basis of the set of centrifugation process parsing branch networks to ensure the smooth flow of data from input to output.

[0044] Parallel embedding means running multiple computing tasks or processing flows simultaneously to improve efficiency. The multiple centrifugation process parsing branch networks are processed in parallel to accelerate the computing process and consider the separation effects of multiple centrifugation processes at the same time. All the parallel - embedded networks are integrated to construct a multi - level centrifugation process parsing channel, enabling all data in the overall centrifugation process to be smoothly processed through this channel. By defining the input and output interfaces of each parsing channel, the parsing results of each process cluster can be smoothly passed to the next link. In the multi - level centrifugation process, the separation effects at different stages will affect the subsequent steps, so the output of each stage needs to be reasonably scheduled.

[0045] Through the multi - level centrifugation process parsing channel, precise centrifugation process parsing can be provided for different separation stages and process conditions, ensuring that each step of catalyst recovery reaches the optimal effect. With the continuous accumulation of process data, the model can continuously improve its adaptability through iterative tuning and cross - validation, ensuring high - efficient separation performance under different process conditions.

[0046] The separation processing module 13 is used to perform preliminary separation processing on the target mixture based on the M catalyst centrifugation process parameters through the multi - level centrifugation separation device, and at the same time, the monitoring sensor group is used to obtain M catalyst state parameters in real - time.

[0047] Specifically, according to the M catalyst centrifugation process parameters, the multi - level centrifugation separation device is controlled to perform preliminary separation on the target mixture, quickly remove most of the impurities, and recover the catalyst. The multi - level centrifugation separation unit will perform separation processing on the target mixture in sequence to separate the catalyst from other components. That is to say, according to the composition of the target mixture and the characteristics of the catalyst, M catalyst centrifugation process parameters are determined, including the centrifugation rate, temperature, and time of each centrifugation separation unit. The multi - level centrifugation separation device starts to perform preliminary separation of the target mixture according to the catalyst centrifugation process parameters (such as centrifugation rate, time, and temperature) set by the M centrifugation separation units, and separates the catalyst in the mixture from other impurities through centrifugal force.

[0048] The monitoring sensor group arranged in each centrifugal separation unit collects the catalyst state parameters in real time, detects the changes during the centrifugal separation process, and evaluates the separation effect based on the catalyst state parameters, so as to adjust the separation strategy parameters to optimize the separation effect. The catalyst state parameters are data reflecting the state of the catalyst during the centrifugal separation process, including the particle size, concentration, morphology, distribution, etc. of the catalyst, and are used to judge whether the separation process is proceeding normally.

[0049] The recovery control module 14 is used to calculate the separation effect and perform multi-level dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters, determine the multi-level centrifugal separation strategy parameters, and control the catalyst recovery of the target mixture through the multi-level centrifugal separation strategy parameters.

[0050] Furthermore, the recovery control module 14 in the catalyst recovery control system based on multi-level centrifugal separation is further used for: a normalization processing unit, which is used to perform normalization processing on the M catalyst state parameters according to the data application standard to obtain M standard catalyst state parameters; an index establishment unit, which is used to establish M separation effect evaluation index systems according to the M catalyst recovery targets of the M centrifugal separation units; a quantification calculation unit, which is used to perform effect quantification calculation on the M standard catalyst state parameters based on the M separation effect evaluation index systems to obtain M catalyst separation effect parameters; a multi-level adjustment unit, which is used to perform multi-level dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst separation effect parameters to obtain multi-level centrifugal separation strategy parameters.

[0051] The index extraction subunit is used to perform multi-dimensional index extraction and calculation formula association on the M separation effect evaluation index systems to obtain M separation effect index calculation components; the index effect calculation subunit is used to perform index effect calculation on the M standard catalyst state parameters by using the M separation effect index calculation components to obtain M separation effect index parameters; the separation effect calculation subunit is used to use the ratio of the M separation effect index parameters to M preset separation effect standards as the M catalyst separation effect parameters.

[0052] Specifically, the M monitored catalyst state parameters are normalized to eliminate the differences in different parameter dimensions, and M standard catalyst state parameters are obtained. The purpose of normalization is to convert all catalyst state parameters (such as particle concentration, particle size, etc.) into values within a standard range (such as [0,1]), so that the M catalyst state parameters can be uniformly calculated under the same dimension. The data application standard is a set of specifications used to ensure the consistency and accuracy of data, such as data format, data unit, data precision, data security, etc.

[0053] Obtain the M catalyst recovery targets for the M centrifugal separation units, that is, the specific catalyst recovery targets to be achieved by each centrifugal separation unit, such as separation efficiency, recovery rate, purity, etc. These are usually the standards set during the centrifugal separation process and are used to evaluate the separation effect. Based on the catalyst recovery targets (such as recovery rate, separation efficiency, catalyst loss, etc.) of each centrifugal separation unit, establish an evaluation index system for the separation effect to evaluate the catalyst state parameters. Effect quantification calculation refers to performing mathematical calculations on the standard catalyst state parameters according to the established evaluation index system for the separation effect to evaluate the quality of the actual separation effect, and obtaining a comprehensive evaluation value for the catalyst separation process through quantification calculation.

[0054] Analyze the M evaluation index systems for the separation effect, and perform multi-dimensional index extraction and formula association. Multi-dimensional index extraction refers to extracting important and representative features from the existing multiple separation effect evaluation indexes to reduce complexity while retaining sufficient information to ensure accurate evaluation. Formula association is to associate multiple separation effect indexes with actual data through mathematical methods and formulas, that is, associate a calculation formula for each extracted index, thereby constructing M calculation components for the separation effect indexes. The M calculation components for the separation effect indexes can process the input catalyst state parameters according to the built-in multi-dimensional indexes and calculation formulas, and calculate the values of each separation effect index. For example, for the recovery rate, it is calculated by the ratio of the mass of the recovered catalyst to the mass of the original catalyst. If the mass of the original catalyst is 100 g and the mass of the recovered catalyst is 95 g, the calculated recovery rate is 95%; for the separation purity, it is calculated by the ratio of the mass of the effective component of the separated catalyst to the total mass of the separated catalyst. If the total mass of the separated catalyst is 95 g and the effective component is 90 g, the purity is 94.74%.

[0055] According to the M calculation components for the separation effect indexes, perform index effect calculations on the M normalized standard catalyst state parameters to obtain M separation effect index parameters, which represent the performance of different indexes (such as recovery rate 0.95, purity 0.94, efficiency 0.833, etc.) in the current separation process of the catalyst separation. The preset separation effect standard is the expected value defined during design and is used as the standard for the catalyst separation effect. It usually comes from theoretical design and experimental results and represents the separation effect that should be achieved under ideal conditions. Calculate the ratio of the M separation effect index parameters to the M preset separation effect standards to obtain M catalyst separation effect parameters. The catalyst separation effect parameter is the final parameter obtained after calculating the ratio of the separation effect index parameter to the preset separation effect standard, and is used to judge the deviation between the actual separation effect and the target effect, evaluate the success or failure of the separation process, and then adjust the separation parameters.

[0056] Based on the separation effect parameters of M catalysts, determine which stages have performance deviations, and dynamically adjust the centrifugation process parameters (such as centrifugation rate, temperature, etc.) according to the specific separation effect of each centrifugation unit. For example, if the catalyst separation efficiency in a certain stage is low, optimize the separation effect by increasing the centrifugation rate or changing the temperature. Through real-time feedback and adjustment, ensure that each stage operates under the optimal operating conditions. Combine multi-stage dynamic regulation with global simulation optimization (such as particle swarm optimization algorithm) to determine the optimal centrifugation parameter threshold, and synergistically correct the strategy parameters based on the influencing factors to ensure the balance between multi-stage separation effect and economy.

[0057] Adjust the operating conditions of the centrifugation separation device according to the finally determined multi-stage centrifugation separation strategy parameters, such as setting specific centrifugation speed, separation time, temperature, pressure, etc., to control the catalyst recovery of the target mixture to achieve the best catalyst recovery effect. Through normalization processing, the consistency of the data is ensured, and an evaluation index system for separation effect is established to evaluate the separation effect according to specific recovery targets. Through effect quantification calculation, specific separation effect parameters are obtained for further optimizing and analyzing the separation parameters. Obtain multi-stage centrifugation separation strategy parameters through multi-stage dynamic regulation to guide the actual centrifugation separation process and improve the catalyst recovery efficiency.

[0058] Furthermore, the recovery control module 14 in the catalyst recovery control system based on multi-stage centrifugation separation is further configured to: a strategy optimization subunit, which is used to perform optimization strategy analysis on the M catalyst centrifugation process parameters by using the M catalyst separation effect parameters to determine M centrifugation parameter optimization strategies; a simulation and simulation subunit, which is used to evaluate and simulate the M centrifugation process databases according to the M separation effect evaluation index systems to construct M separation effect simulation units; a strategy adjustment subunit, which is used to perform multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M separation effect simulation units according to the M centrifugation parameter optimization strategies to obtain multi-stage centrifugation separation strategy parameters.

[0059] Specifically, use the M catalyst separation effect parameters to identify the influence of different process parameters on the separation effect, including recovery rate, purity, separation time, etc. Perform optimization strategy analysis on the M catalyst centrifugation process parameters according to the M catalyst separation effect parameters, that is, adjust the parameters that affect the separation effect to make them relatively better. Identify which centrifugation process parameters (such as centrifugation rate, time, temperature, etc.) have a significant impact on the catalyst recovery efficiency and separation effect through multi-dimensional data analysis, and simply adjust the M catalyst centrifugation process parameters to adjust each parameter to a more reasonable value. The M centrifugation parameter optimization strategies are not the final optimal strategies, but preliminary optimization schemes based on the current data and analysis results, which can obtain a better separation effect under the existing conditions.

[0060] Based on the historical data in M centrifugation process databases (such as historical mixture compositions, centrifugation process parameters, and separation effect data), a simulation model is constructed using a simulation tool. Specifically, based on the historical data in M centrifugation process databases, a suitable simulation tool is selected for simulation to construct the simulation model. By inputting different centrifugation process parameters (such as rate, time, temperature), the working process of each centrifugation separation unit is simulated, and separation effect parameters are output. During the simulation, the separation effect evaluation index systems of M catalysts (such as recovery rate, separation efficiency, etc.) are applied to the simulation model to evaluate the separation effects of different process parameter combinations. M separation effect simulation units are established based on the simulation model, and each unit represents an independent simulation unit responsible for evaluating the influence of specific centrifugation process settings on the catalyst separation effect.

[0061] Input the M centrifugation parameter optimization strategies into the M separation effect simulation units for simulation to obtain evaluation results, and perform multi-level dynamic adjustment on the M catalyst centrifugation process parameters to obtain multi-level centrifugation separation strategy parameters. That is to say, through the real-time feedback of each unit, parameters such as centrifugation rate and temperature are continuously optimized to ensure that the catalyst separation effect in each stage reaches the best. Multi-level centrifugation separation strategy parameters are generated through multi-level adjustment to guide the operation in the whole separation process and optimize the catalyst recovery rate, purity, and separation efficiency. By optimizing the process parameters in the catalyst separation process, each centrifugation separation unit can work under the best operating conditions, thereby improving the overall recovery rate.

[0062] Furthermore, the recovery control module 14 in the catalyst recovery control system based on multi-level centrifugation separation is further configured to: a feasible threshold determination component, which is used to perform multi-level dynamic adjustment on the M catalyst centrifugation process parameters based on the M centrifugation parameter optimization strategies to obtain the M catalyst centrifugation process feasible thresholds; an initialization component, which is used to initialize the M catalyst centrifugation process particle spaces according to the M catalyst centrifugation process feasible thresholds; a simulation optimization component, which is used to perform global simulation optimization in the M catalyst centrifugation process particle spaces by using the M separation effect simulation units until a preset termination condition is reached to determine the multi-level centrifugation separation strategy parameters.

[0063] Specifically, according to the determined M centrifugation parameter optimization strategies, multi-level dynamic adjustment is performed on the M catalyst centrifugation process parameters. Using the M centrifugation parameter optimization strategies, the process parameters of each centrifugation separation unit are adjusted, and each centrifugation separation unit is gradually adjusted until a most suitable parameter range is found, which can achieve the best separation effect and meet the preset requirements for catalyst recovery within this range. The M catalyst centrifugation process feasible thresholds include the maximum and minimum values of the process parameters, indicating that within the feasible thresholds, the catalyst separation effect is the best. The feasible threshold refers to the parameter boundary that can ensure the separation effect reaches the expected standard within the given process parameter range.

[0064] According to the M catalyst centrifugation process feasible thresholds, the search range of each particle in the particle space is determined. Each particle represents a combination of centrifugation process parameters, and each dimension of the parameter space corresponds to a specific centrifugation process parameter (such as rate, time, etc.). When initializing the particle space, the position (i.e., process parameter) of each particle is randomly generated to ensure that the particle space covers all possible combinations of process parameters. For example, for a centrifugation separation unit, there are 3 process parameters that need to be optimized: centrifugation rate, centrifugation time, and temperature. For a certain centrifugation unit, the recovery rate is the highest within a certain rate range. When the rate is between 1500 rpm and 2000 rpm, the separation effect is the best, and the recovery rate exceeds 95%. Then, the rate range between 1500 rpm and 2000 rpm is the feasible threshold. Each particle in the particle space will be composed of a 3D parameter combination (rate, time, temperature). The rate range is [1500 rpm, 2500 rpm], the time range is [5 min, 30 min], and the temperature range is [20 °C, 35 °C].

[0065] The particle swarm optimization algorithm is used to perform simulated optimization within the particle space. Each particle represents a potential combination of process parameters. By simulating and calculating the separation effect, the particle will adjust its position (process parameter) according to its separation effect (such as recovery rate, purity, etc.), so as to approach the optimal solution. Through multiple rounds of simulation and iteration until the preset termination conditions are met (such as reaching the preset maximum number of iterations or the error converges to the set threshold). During the simulation process, the particle swarm continuously adjusts the process parameters according to the effect of each centrifugation process parameter combination (such as separation recovery rate) to search for the most suitable combination. If after a certain iteration, the recovery rate reaches more than 98% and the change in particle position is small, the optimization process can end. The preset termination conditions are the stop criteria set during the optimization process, such as reaching the maximum number of iterations, the convergence error being less than a certain threshold, etc. When the preset termination conditions are met, the optimization process terminates.

[0066] Determine the final parameter combination for each centrifugal separation unit according to the best solution in the particle swarm optimization process. Apply the final parameter combination to the actual separation process to form a complete multi-stage centrifugal separation strategy, including optimized parameters such as the centrifugation rate, time, and temperature of each centrifugal separation unit, ensuring operation under optimal conditions. Through a global optimization algorithm, find the most suitable parameter combination for the catalyst centrifugation process to improve the recovery rate and purity of the catalyst. Through multi-stage dynamic adjustment and global optimization, adjust the centrifugation process parameters under different catalyst states to ensure always being in the best working state, thereby improving the separation efficiency.

[0067] Further, the recovery control module 14 in the catalyst recovery control system based on multi-stage centrifugal separation is further configured to: a parameter determination channel, which is used to perform global simulation optimization in the M catalyst centrifugation process particle spaces by using the M separation effect simulation units until a preset termination condition is reached, to obtain M catalyst centrifugal separation strategy parameters; an influence factor evaluation channel, which is used to evaluate and obtain the influence factor information of the M catalyst centrifugal separation strategy parameters; and a balance correction channel, which is used to perform collaborative balance correction on the M catalyst centrifugal separation strategy parameters based on the influence factor information to determine the multi-stage centrifugal separation strategy parameters.

[0068] Specifically, perform global simulation optimization in the M catalyst centrifugation process particle spaces by using the M separation effect simulation units. That is to say, input all the separation strategy parameters in the M catalyst centrifugation process particle spaces into the M separation effect simulation units for simulation, so as to obtain the separation effect of each parameter combination. Search the particle space through a global optimization algorithm (such as particle swarm optimization), gradually adjust the centrifugation process parameters, and find the optimal process parameter combination. During the optimization process, the optimization algorithm will iteratively update the position of each particle (i.e., the process parameter) until the preset termination condition is reached (such as reaching the maximum number of iterations or the convergence error is less than a certain threshold). The M catalyst centrifugal separation strategy parameters are based on the results of global optimization and simulation and are the parameter combinations most suitable for separating the target mixture.

[0069] For each optimized catalyst centrifugation process parameter combination, evaluate the influence of different process parameters (such as rate, temperature, etc.) on the separation effect, and identify the influence factors, including the physical properties of the catalyst (such as particle size, surface activity, etc.), process parameters (such as centrifugation rate, time, temperature, etc.), and even external factors (such as environmental temperature, pressure, etc.). The influence factor information is some key parameters or variables identified during the optimization process that have a significant impact on the separation effect, such as the physical properties of the catalyst, process conditions, operating environment, etc., and are used to further adjust the strategy during the optimization process.

[0070] Based on the impact factor information, identify which parameters have a greater impact on the separation effect and optimize and adjust them. Through collaborative correction, ensure that the interaction between multiple process parameters is optimized. For example, if the temperature of a certain centrifugal separation unit is too high, it may cause catalyst degradation, then appropriately adjust other parameters (such as centrifugation rate or time) to compensate for the negative impact of temperature on the separation effect. Collaborative equilibrium correction is to adjust M catalyst centrifugation process parameters according to the impact factor information, so that different parameters are coordinated with each other, avoiding that some process parameters have too large or too small an impact on the separation effect, and to ensure that the process parameters of each centrifugal separation unit complement each other in the overall strategy, so as to achieve the best separation effect. Equilibrium correction is used to reach an optimal process balance point.

[0071] Combined with the parameters after collaborative equilibrium correction, determine the final multi-stage centrifugal separation strategy, including the optimized process parameters of each centrifugal separation unit, such as rate, time, temperature, etc. For example, the corrected process parameter centrifugation rate is set at 1800 rpm, the time is set at 20 minutes, and the temperature is set at 25 °C. The final strategy will cover multiple centrifugal separation units to ensure that the working state of each unit in the whole is optimal. The multi-stage centrifugal separation strategy parameters are the final set of optimized parameters, including the best operating parameters of each centrifugal separation unit, such as rate, time, temperature, etc., which are obtained after collaborative equilibrium correction. Through global simulation and optimization algorithms, find the optimal combination of catalyst centrifugation process parameters, improve the catalyst recovery efficiency and purity, reduce catalyst waste by optimizing process parameters, improve resource utilization rate, and reduce production costs.

[0072] In summary, the catalyst recovery control system based on multi-stage centrifugal separation provided by the present application has the following technical effects: Through the device construction module, a multi-stage centrifugal separation device is constructed. The multi-stage centrifugal separation device is composed of M centrifugal separation units connected in series, and a monitoring sensor group is arranged in each separation unit of the M centrifugal separation units; the separation and analysis module is used to collect and obtain the initial composition data of the target mixture, and based on the M centrifugal separation units, the initial composition data is separated and analyzed to determine M catalyst centrifugation process parameters; the separation and treatment module is used to perform preliminary separation treatment on the target mixture through the multi-stage centrifugal separation device based on the M catalyst centrifugation process parameters, and at the same time use the monitoring sensor group to obtain M catalyst state parameters in real time; the recovery control module is used to calculate the separation effect and perform multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters, determine the multi-stage centrifugal separation strategy parameters, and perform catalyst recovery control on the target mixture through the multi-stage centrifugal separation strategy parameters. That is to say, by connecting multiple centrifugal separation units in series, equipped with a monitoring sensor group, performing multi-stage centrifugal separation, monitoring the catalyst state in real time, calculating the separation effect and performing multi-stage dynamic adjustment, accurately regulating the working parameters of each centrifugal separation unit, ensuring that the recovery effect reaches the optimal, achieving the optimal recovery purity of the catalyst, and improving the efficiency and accuracy of catalyst recovery.

[0073] Embodiment 2. Based on the same inventive concept as the catalyst recovery control system based on multi-stage centrifugal separation in the foregoing Embodiment 1, the present application also provides a catalyst recovery control method based on multi-stage centrifugal separation. Please refer to the attached Figure 2 , the catalyst recovery control method based on multi-stage centrifugal separation includes: S100: Construct a multi-stage centrifugal separation device. The multi-stage centrifugal separation device is composed of M centrifugal separation units connected in series, and a monitoring sensor group is arranged in each separation unit of the M centrifugal separation units; S200: Collect and obtain the initial composition data of the target mixture, and based on the M centrifugal separation units, separate and analyze the initial composition data to determine M catalyst centrifugation process parameters; S300: Perform preliminary separation treatment on the target mixture through the multi-stage centrifugal separation device based on the M catalyst centrifugation process parameters, and at the same time use the monitoring sensor group to obtain M catalyst state parameters in real time; S400: Calculate the separation effect and perform multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters, determine the multi-stage centrifugal separation strategy parameters, and perform catalyst recovery control on the target mixture through the multi-stage centrifugal separation strategy parameters.

[0074] Further, the determination of the M catalyst centrifugation process parameters includes: mining and obtaining M centrifugation process databases of the M centrifugal separation units, where the M centrifugation process databases include historical mixture composition data, centrifugation process parameters, and corresponding separation effect data; screening and dividing the M centrifugation process databases according to the separation effect data to obtain M available centrifugation process data sets; performing feature clustering fitting on the M available centrifugation process data sets based on the historical mixture composition data to obtain a multi-level centrifugation process analysis channel; and using the multi-level centrifugation process analysis channel to perform centrifugation process analysis on the initial composition data to determine the M catalyst centrifugation process parameters.

[0075] Further, the obtaining of the multi-level centrifugation process analysis channel includes: extracting key features from the historical mixture composition data based on the separation effect data to obtain a key composition feature set; performing clustering operations on the M available centrifugation process data sets according to the key composition feature set to obtain M centrifugation process cluster sets; integrating the M centrifugation process cluster sets according to the clustering clusters to obtain a multi-level centrifugation process cluster set; and fitting an analysis channel for each process cluster in the multi-level centrifugation process cluster set to construct a multi-level centrifugation process analysis channel.

[0076] Further, the construction of the multi-level centrifugation process analysis channel includes: selecting a network architecture set according to the characteristic information of the multi-level centrifugation process cluster set; using the network architecture set to perform centrifugation process fitting on each process cluster in the multi-level centrifugation process cluster set to generate an initial process analysis branch network set; respectively performing cross-validation and iterative tuning on the initial process analysis branch network set to obtain a centrifugation process analysis branch network set; and building a channel architecture and network parallel embedding based on the centrifugation process analysis branch network set to construct the multi-level centrifugation process analysis channel.

[0077] Further, the determination of the multi-level centrifugal separation strategy parameters includes: normalizing the M catalyst state parameters according to the data application standard to obtain M standard catalyst state parameters; establishing M separation effect evaluation index systems according to the M catalyst recovery targets of the M centrifugal separation units; performing effect quantification calculation on the M standard catalyst state parameters based on the M separation effect evaluation index systems to obtain M catalyst separation effect parameters; and performing multi-level dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst separation effect parameters to obtain multi-level centrifugal separation strategy parameters.

[0078] Further, obtaining the M catalyst separation effect parameters includes: performing multi-dimensional index extraction and calculation formula association on the M separation effect evaluation index systems to obtain M separation effect index calculation components; using the M separation effect index calculation components to perform index effect calculations on the M standard catalyst state parameters to obtain M separation effect index parameters; and taking the ratios of the M separation effect index parameters to M preset separation effect standards as the M catalyst separation effect parameters.

[0079] Further, obtaining the multi-stage centrifugal separation strategy parameters includes: performing optimization strategy analysis on the M catalyst centrifugation process parameters using the M catalyst separation effect parameters to determine M centrifugation parameter optimization strategies; evaluating and simulating the M centrifugation process databases according to the M separation effect evaluation index systems to construct M separation effect simulation units; and using the M separation effect simulation units to perform multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M centrifugation parameter optimization strategies to obtain the multi-stage centrifugal separation strategy parameters.

[0080] Further, obtaining the multi-stage centrifugal separation strategy parameters includes: performing multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M centrifugation parameter optimization strategies to obtain M catalyst centrifugation process feasible thresholds; initializing an M catalyst centrifugation process particle space according to the M catalyst centrifugation process feasible thresholds; and using the M separation effect simulation units to perform global simulation optimization within the M catalyst centrifugation process particle space until a preset termination condition is reached to determine the multi-stage centrifugal separation strategy parameters.

[0081] Further, determining the multi-stage centrifugal separation strategy parameters includes: using the M separation effect simulation units to perform global simulation optimization within the M catalyst centrifugation process particle space until a preset termination condition is reached to obtain M catalyst centrifugal separation strategy parameters; evaluating and obtaining the influence factor information of the M catalyst centrifugal separation strategy parameters; and performing collaborative equilibrium correction on the M catalyst centrifugal separation strategy parameters based on the influence factor information to determine the multi-stage centrifugal separation strategy parameters.

[0082] The various embodiments in this specification are described in a progressive manner, and the key point of each embodiment is the difference from other embodiments. The Figure 1 catalyst recovery control system and specific examples based on multi-stage centrifugal separation in the first foregoing embodiment are equally applicable to the catalyst recovery control method based on multi-stage centrifugal separation in this embodiment. Through the detailed description of the catalyst recovery control system based on multi-stage centrifugal separation above, those skilled in the art can clearly know the catalyst recovery control method based on multi-stage centrifugal separation in this embodiment. Therefore, for the sake of brevity of the specification, it will not be elaborated here.

[0083] The foregoing description of the disclosed embodiments enables those skilled in the art to practice or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0084] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is also intended to include these changes and modifications.

Claims

1. A catalyst recovery control system based on multistage centrifugal separation, characterized in that Including: A device construction module, configured to construct a multi-stage centrifugal separation device, where the multi-stage centrifugal separation device is composed of M centrifugal separation units connected in series, and monitoring sensor groups are arranged in each separation unit of the M centrifugal separation units; A separation and analysis module, configured to collect and obtain initial composition data of a target mixture, perform separation and analysis on the initial composition data based on the M centrifugal separation units, and determine M catalyst centrifugation process parameters; A separation and treatment module, configured to perform preliminary separation and treatment on the target mixture based on the M catalyst centrifugation process parameters through the multi-stage centrifugal separation device, and simultaneously use the monitoring sensor groups to obtain M catalyst state parameters in real time; A recovery control module, configured to perform separation effect calculation and multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters, determine multi-stage centrifugal separation strategy parameters, and perform catalyst recovery control on the target mixture through the multi-stage centrifugal separation strategy parameters.

2. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 1, wherein The separation and analysis module includes: A database acquisition unit, configured to mine and obtain M centrifugation process databases of the M centrifugal separation units, where the M centrifugation process databases include historical mixture composition data, centrifugation process parameters, and corresponding separation effect data; A screening and partitioning unit, configured to screen and partition the M centrifugation process databases according to the separation effect data to obtain M available centrifugation process data sets; A clustering and fitting unit, configured to perform feature clustering and fitting on the M available centrifugation process data sets based on the historical mixture composition data to obtain a multi-stage centrifugation process analysis channel; A process analysis unit, configured to perform centrifugation process analysis on the initial composition data by using the multi-stage centrifugation process analysis channel to determine the M catalyst centrifugation process parameters.

3. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 2, wherein The clustering and fitting unit includes: A feature extraction sub-unit, configured to perform key feature extraction on the historical mixture composition data based on the separation effect data to obtain a key composition feature set; A clustering analysis sub-unit, configured to perform a clustering operation on the M available centrifugation process data sets according to the key composition feature set to obtain M centrifugation process cluster sets; An integration processing sub-unit, configured to integrate and process the M centrifugation process cluster sets according to the clustering clusters to obtain a multi-stage centrifugation process cluster set; A channel fitting sub-unit, configured to perform analysis channel fitting on each process cluster in the multi-stage centrifugation process cluster set to construct a multi-stage centrifugation process analysis channel.

4. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 3, characterized in that, The channel fitting sub-unit includes: An architecture selection component, configured to select a network architecture set according to the characteristic information of the multi-stage centrifugation process cluster set; A centrifugation fitting component, configured to perform centrifugation process fitting on each process cluster in the multi-stage centrifugation process cluster set by using the network architecture set to generate an initial process analysis branch network set; A verification and optimization component, configured to perform cross-validation and iterative optimization on the initial process analysis branch network set respectively to obtain a centrifugation process analysis branch network set. Parallel embedding components are used to build a channel architecture and perform network parallel embedding based on the analysis of the branch network set of the centrifugation process, and construct the multi-stage centrifugation process analysis channels.

5. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 2, characterized in that, The recovery control module includes: A normalization processing unit for normalizing the M catalyst state parameters according to the data application standard to obtain M standard catalyst state parameters; An index establishment unit for establishing M separation effect evaluation index systems according to the M catalyst recovery targets of the M centrifugal separation units; A quantization calculation unit for performing effect quantization calculations on the M standard catalyst state parameters based on the M separation effect evaluation index systems to obtain M catalyst separation effect parameters; A multi-stage adjustment unit for performing multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst separation effect parameters to obtain multi-stage centrifugal separation strategy parameters.

6. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 5, wherein, The quantization calculation unit includes: An index extraction sub-unit for performing multi-dimensional index extraction and calculation formula association on the M separation effect evaluation index systems to obtain M separation effect index calculation components; An index effect calculation sub-unit for using the M separation effect index calculation components to perform index effect calculations on the M standard catalyst state parameters to obtain M separation effect index parameters; A separation effect calculation sub-unit for taking the ratio of the M separation effect index parameters to M preset separation effect standards as the M catalyst separation effect parameters.

7. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 5, characterized in that The multi-stage adjustment unit includes: A strategy optimization sub-unit for performing optimization strategy analysis on the M catalyst centrifugation process parameters using the M catalyst separation effect parameters to determine M centrifugation parameter optimization strategies; A simulation and modeling sub-unit for evaluating and simulating the M centrifugation process databases according to the M separation effect evaluation index systems to construct M separation effect simulation units; A strategy adjustment sub-unit for using the M separation effect simulation units to perform multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M centrifugation parameter optimization strategies to obtain multi-stage centrifugal separation strategy parameters.

8. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 7, characterized in that, The strategy adjustment sub-unit includes: A feasible threshold determination component for performing multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M centrifugation parameter optimization strategies to obtain M catalyst centrifugation process feasible thresholds; An initialization component for initializing the M catalyst centrifugation process particle spaces according to the M catalyst centrifugation process feasible thresholds; A simulation optimization component for using the M separation effect simulation units to perform global simulation optimization within the M catalyst centrifugation process particle spaces until a preset termination condition is reached to determine multi-stage centrifugal separation strategy parameters.

9. The catalyst recovery control system based on multi-stage centrifugal separation according to claim 8, characterized in that, The simulation optimization component includes: A parameter determination channel for using the M separation effect simulation units to perform global simulation optimization within the M catalyst centrifugation process particle spaces until a preset termination condition is reached to obtain M catalyst centrifugal separation strategy parameters; An influence factor evaluation channel for evaluating and obtaining the influence factor information of the M catalyst centrifugal separation strategy parameters. An equilibrium correction channel is used to perform collaborative equilibrium correction on the M catalyst centrifugal separation strategy parameters based on the influence factor information to determine the multi-stage centrifugal separation strategy parameters.

10. A catalyst recovery control method based on multi-stage centrifugal separation, characterized in that, Executed by the catalyst recovery control system based on multi-stage centrifugal separation according to any one of claims 1 to 9, the catalyst recovery control method based on multi-stage centrifugal separation includes: Build a multi-stage centrifugal separation device, which is composed of M centrifugal separation units connected in series, and a monitoring sensor group is arranged in each separation unit of the M centrifugal separation units; Collect and obtain the initial composition data of the target mixture, and perform separation and analysis on the initial composition data based on the M centrifugal separation units to determine M catalyst centrifugation process parameters; Perform preliminary separation processing on the target mixture based on the M catalyst centrifugation process parameters through the multi-stage centrifugal separation device, and simultaneously use the monitoring sensor group to obtain M catalyst state parameters in real time; Perform separation effect calculation and multi-stage dynamic adjustment on the M catalyst centrifugation process parameters based on the M catalyst state parameters to determine the multi-stage centrifugal separation strategy parameters, and perform catalyst recovery control on the target mixture through the multi-stage centrifugal separation strategy parameters.

Images