Separation method and system for sampling of pharmaceutical equipment

Through chemical analysis and component identification channels, the drug mixture components are identified, combined with separation methods and control model monitoring, the problem of low separation adaptability of drug mixtures in the prior art is solved, and efficient drug separation and resource conservation is achieved.

CN119064108AInactive Publication Date: 2024-12-03JIANGSU SAIQI SEPARATION TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411562413.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-05
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing drug mixture separation methods used for sampling in pharmaceutical equipment are poor in separation control schemes and drug mixtures, resulting in poor separation quality and low efficiency, resulting in waste of resources.

Method used

By constructing chemical analysis channels and component identification channels, identify the target drug mixture, obtain the mixture component information, combine the separation method based on the preset separation data table and evaluation function, generate an optimized separation control scheme, and control twin models through separation control Perform synchronous monitoring and feedback control.

Benefits of technology

It improves the accuracy and efficiency of drug mixture separation control, improves the separation quality, and saves separation resources.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119064108A_ABST
    Figure CN119064108A_ABST
Patent Text Reader

Abstract

The invention provides a pharmaceutical equipment sampling separation method and system, and relates to the technical field of pharmaceutical separation control, and the method comprises the following steps: based on a first separation evaluation function and an expected separation index, carrying out combination optimization by using a separation mode set, and determining an optimized separation combination scheme which comprises separation stage and expected stage separation indexes; and based on a stage separation evaluation function, optimizing separation control parameters of each separation stage, generating an optimized separation control scheme according to an optimization result and the optimized separation combination scheme for separation processing, and executing synchronous monitoring and feedback control through a separation control twinborn model. The technical problems that according to an existing medicine mixture separation method, due to the fact that the matching degree of a separation control scheme and the medicine mixture is low, the separation quality of the medicine mixture is poor, the separation efficiency is low, and meanwhile separation resources are wasted can be solved, and the purposes of improving the separation quality and the separation efficiency of the medicine mixture and improving the separation efficiency of the medicine mixture can be achieved. And separation resources are saved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of pharmaceutical separation control technology, and in particular to a separation method and system for sampling pharmaceutical equipment. Background Art

[0002] Drug mixture separation technology is an indispensable part of drug research and development and production. It is used to extract and purify target drug components from complex mixtures. With the continuous development of artificial intelligence technology, new separation technologies and intelligent separation control methods are also emerging, providing more options and possibilities for drug mixture separation.

[0003] Currently, the existing drug mixture separation methods used for pharmaceutical equipment sampling have technical problems such as poor drug mixture separation quality and low separation efficiency due to the low compatibility of the separation control scheme with the drug mixture, and also cause waste of separation resources. Summary of the invention

[0004] The purpose of the present application is to provide a separation method and system for sampling pharmaceutical equipment, so as to solve the technical problem that the existing drug mixture separation method used for sampling pharmaceutical equipment has poor drug mixture separation quality and low separation efficiency due to the low compatibility of the separation control scheme with the drug mixture, and also causes waste of separation resources.

[0005] In view of the above problems, the present application provides a separation method and system for sampling pharmaceutical equipment.

[0006] In a first aspect, the present application provides a separation method for sampling pharmaceutical equipment, the method is implemented by a separation system for sampling pharmaceutical equipment, wherein the method comprises: S100: obtaining a target drug mixture, parsing the target drug mixture through a chemical parsing channel to obtain a chemical parsing data set, then identifying the chemical parsing data set through a component identification channel, and outputting mixture component information, wherein the mixture component information includes component type and mixing ratio; S200: inputting the component type into a preset separation data table, matching to obtain a separation mode set, based on a first separation evaluation function and an expected separation index, using the separation mode set for combination optimization, and determining an optimized separation combination scheme, wherein the optimized separation combination scheme includes a separation stage and an expected stage separation index, and each separation stage corresponds to a separation mode; S300: based on a stage separation evaluation function, with the expected stage separation index as a constraint, optimizing the separation control parameters of each separation stage, and generating an optimized separation control scheme according to the optimization result and the optimized separation combination scheme; S400: separating and processing the target drug mixture according to the optimized separation control scheme, and performing synchronous monitoring and feedback control through a separation control twin model.

[0007] In a second aspect, the present application further provides a separation system for sampling pharmaceutical equipment, which is used to execute a separation method for sampling pharmaceutical equipment as described in the first aspect, wherein the system comprises: a mixture component information output module, which is used to obtain a target drug mixture, parse the target drug mixture through a chemical analysis channel to obtain a chemical analysis data set, and then identify the chemical analysis data set through a component identification channel to output mixture component information, wherein the mixture component information includes component type and mixing ratio; an optimization separation combination scheme determination module, which is used to input the component type into a preset separation data table, match to obtain a separation method set, and select a separation method based on a first separation evaluation function and a preset separation ratio. The separation index of the expected stage is used, and the combination optimization is performed by using the separation mode set to determine the optimized separation combination scheme, wherein the optimized separation combination scheme includes separation stage and expected stage separation indexes, and each separation stage corresponds to a separation mode; an optimized separation control scheme generation module is used to optimize the separation control parameters of each separation stage based on the stage separation evaluation function, with the expected stage separation index as a constraint, and generate an optimized separation control scheme according to the optimization result and the optimized separation combination scheme; a separation monitoring and control module is used to separate and process the target drug mixture according to the optimized separation control scheme, and perform synchronous monitoring and feedback control through the separation control twin model.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: A target drug mixture is obtained, the target drug mixture is analyzed through a chemical analysis channel to obtain a chemical analysis data set, and then the chemical analysis data set is identified through a component identification channel to output mixture component information, wherein the mixture component information includes component type and mixing ratio; the component type is input into a preset separation data table, a separation mode set is obtained by matching, and based on a first separation evaluation function and an expected separation index, the separation mode set is used to perform combination optimization to determine an optimized separation combination scheme, wherein the optimized separation combination scheme includes separation stages and expected stage separation indicators, and each separation stage corresponds to a separation mode; based on a stage separation evaluation function, the separation control parameters of each separation stage are optimized with the expected stage separation index as a constraint, and an optimized separation control scheme is generated according to the optimization result and the optimized separation combination scheme; the target drug mixture is separated and processed according to the optimized separation control scheme, and synchronous monitoring and feedback control are performed through a separation control twin model. That is to say, by constructing a chemical analysis channel and a component identification channel to identify the components of the target drug mixture and obtain the mixture component information, the accuracy and efficiency of the mixture component information can be improved; then, based on the mixture component information matching, a set of suitable separation methods is obtained, and the separation method set is used to perform combination optimization to meet the expected separation index, and the optimized separation combination scheme is obtained, wherein the optimized separation combination scheme includes the separation stage and the expected stage separation index; then, with the expected stage separation index as a constraint, the separation control parameters of each separation stage are optimized, and the optimized separation control scheme is generated according to the optimization results and the optimized separation combination scheme; finally, the target drug mixture is separated and processed according to the optimized separation control scheme, and synchronous monitoring and feedback control are performed through the separation control twin model during the separation process. The fineness and accuracy of the drug mixture separation control scheme setting can be improved, thereby improving the accuracy of the drug mixture separation control, achieving the technical effect of improving the drug mixture separation quality and separation efficiency, and saving separation resources.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents 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 specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended 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

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 A schematic diagram of a separation method for sampling pharmaceutical equipment in this application; Figure 2 A schematic diagram of a process for constructing a component identification channel in a separation method for sampling pharmaceutical equipment in the present application; Figure 3 This is a schematic diagram of the structure of a separation system for sampling pharmaceutical equipment in this application.

[0012] Description of reference numerals: Mixture composition information output module 11, optimized separation combination scheme determination module 12, optimized separation control scheme generation module 13, separation monitoring control module 14. DETAILED DESCRIPTION

[0013] The present application provides a separation method and system for sampling pharmaceutical equipment, which solves the technical problem that the existing separation method for pharmaceutical equipment sampling has poor separation quality and low separation efficiency, and wastes separation resources due to the low adaptability of the separation control scheme to the pharmaceutical mixture. The fineness and accuracy of the setting of the separation control scheme for the pharmaceutical mixture can be improved, thereby improving the accuracy of the separation control of the pharmaceutical mixture, achieving the technical effect of improving the separation quality and separation efficiency of the pharmaceutical mixture and saving separation resources.

[0014] Below, 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 part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0015] Embodiment 1 Please see attached Figure 1 The present application provides a separation method for sampling pharmaceutical equipment, wherein the method is applied to a separation system for sampling pharmaceutical equipment, and the method specifically comprises the following steps: S100: Acquire a target drug mixture, parse the target drug mixture through a chemical analysis channel to obtain a chemical analysis data set, then identify the chemical analysis data set through a component identification channel, and output mixture component information, wherein the mixture component information includes component type and mixing ratio.

[0016] Specifically, first, a target drug mixture is obtained, which refers to a complex mixture composed of multiple chemical components in the pharmaceutical process, which contains the required target drug components and other unnecessary impurities or auxiliary components, and can be set according to actual conditions. Then, the target drug mixture is analyzed through a chemical analysis channel, wherein the chemical analysis channel includes multiple component analysis techniques, such as chromatography, mass spectrometry, and infrared spectroscopy; then, a chemical analysis data set is obtained based on the analysis results, and the chemical analysis data set includes analysis data of multiple analysis techniques, such as chromatography analysis data, mass spectrometry analysis data, etc.

[0017] The chemical analysis data set is further input into a component identification channel for component information identification, wherein the component identification channel comprises a plurality of component identification sub-channels, and each component identification sub-channel corresponds to a machine learning algorithm; firstly, the chemical analysis data set is respectively identified through the plurality of component identification sub-channels, and a plurality of identification results are outputted, and then a weighted calculation is performed on the plurality of identification results to obtain mixture component information, wherein the mixture component information comprises a component type and a mixing ratio, wherein the component type refers to the type of compound contained in the mixture; and the mixing ratio refers to the relative content or proportion of each compound in the mixture.

[0018] By constructing component identification channels, since different identification sub-channels are constructed based on different algorithms, the advantages of multiple algorithms can be integrated to reduce the limitations of single algorithm identification and improve the accuracy and reliability of component identification.

[0019] S200: Input the component type into a preset separation data table, match and obtain a set of separation methods, and based on a first separation evaluation function and an expected separation index, use the set of separation methods to perform combination optimization to determine an optimized separation combination scheme, wherein the optimized separation combination scheme includes separation stages and expected stage separation indicators, and each separation stage corresponds to a separation method.

[0020] Specifically, first, a preset separation data table is constructed, wherein the preset separation data table is constructed based on the tree-based principle, including multiple sample mixture component information and multiple separation method sets with mapping relationships, which can be obtained through training with sample data, wherein the separation method is a separation technology used to separate the mixture, such as: chromatography separation technology, extraction separation technology, crystallization separation technology, distillation separation technology, centrifugal separation technology, etc. Then, the component type is input into the preset separation data table for separation method matching, and a separation method set suitable for the component type is obtained.

[0021] Obtain the expected separation index, which refers to the expected separation requirement and can be set according to actual needs, including purity index, separation index and recovery index. Construct a first separation evaluation function, which is used to evaluate the separation effect of the separation method combination. Then, based on the separation method set, an optimization space is constructed, and based on the first separation evaluation function and the expected separation index, the separation method combination scheme is optimized in the optimization space to obtain an optimized separation combination scheme, wherein the optimized separation combination scheme includes a separation stage and an expected stage separation index, and the separation stage refers to the separation time corresponding to each separation method, and the separation stage has a sequential connection relationship, and each separation stage corresponds to a separation method, such as: using a combination of multiple separation technologies to separate complex mixtures, such as first performing preliminary separation by chromatography, and then further purifying by solid phase extraction, etc.; the expected stage separation index refers to the expected separation index corresponding to each separation stage after the separation is completed, including a stage purity index, a stage separation index and a stage recovery index.

[0022] S300: Based on the stage separation evaluation function, the separation control parameters of each separation stage are optimized with the expected stage separation index being satisfied as a constraint, and an optimized separation control scheme is generated according to the optimization result and the optimized separation combination scheme.

[0023] S400: Separate the target drug mixture according to the optimized separation control scheme, and perform synchronous monitoring and feedback control through the separation control twin model.

[0024] Specifically, a stage separation evaluation function is constructed, and the stage separation evaluation function is used to evaluate the separation quality of the separation control parameters of each separation stage. Then, based on the stage separation evaluation function, the separation control parameters of each separation stage are optimized by using an optimization algorithm to meet the expected stage separation index as a constraint, and the optimized separation control parameters of each separation stage are obtained; further, the optimized separation control parameters of each separation stage are mapped and embedded into multiple separation stages corresponding to the optimized separation combination scheme to generate an optimized separation control scheme.

[0025] Then, based on the optimized separation control scheme, the target drug mixture is separated and processed according to the order of the separation stages and the optimized separation control parameters corresponding to the separation stages, and during the separation process, the separation control twin model corresponding to the separation method of each separation stage is called to synchronously monitor the stage separation process, and timely feedback and optimization processing are performed on the control deviations that occur during the monitoring process. The above method can improve the precision and accuracy of the drug mixture separation control scheme setting, thereby improving the accuracy of drug mixture separation control, achieving the technical effect of improving the drug mixture separation quality and separation efficiency, and saving separation resources.

[0026] The separation method for sampling pharmaceutical equipment is applied to a separation system for sampling pharmaceutical equipment, which can solve the existing separation method for pharmaceutical equipment sampling. Due to the low adaptability of the separation control scheme to the pharmaceutical mixture, the separation quality of the pharmaceutical mixture is poor and the separation efficiency is low, and the separation resources are wasted. Technical problems. First, the target drug mixture is obtained, the target drug mixture is parsed through the chemical parsing channel to obtain a chemical parsing data set, and then the chemical parsing data set is identified through the component identification channel, and the mixture component information is output, wherein the mixture component information includes the component type and the mixing ratio. Then, the component type is input into the preset separation data table, and the separation mode set is matched to obtain the separation mode set. Based on the first separation evaluation function and the expected separation index, the separation mode set is used to perform combination optimization to determine the optimized separation combination scheme, wherein the optimized separation combination scheme includes the separation stage and the expected stage separation index, and each separation stage corresponds to a separation mode. Then, based on the stage separation evaluation function, the separation control parameters of each separation stage are optimized with the expected stage separation index as a constraint, and the optimized separation control scheme is generated according to the optimization result and the optimized separation combination scheme. Finally, the target drug mixture is separated and processed according to the optimized separation control scheme, and synchronous monitoring and feedback control are performed through the separation control twin model. By constructing a chemical analysis channel and a component identification channel to identify the components of the target drug mixture, the mixture component information is obtained, which can improve the accuracy and efficiency of the mixture component information; then, based on the mixture component information matching, an adaptive separation mode set is obtained, and the separation mode set is used to perform combination optimization to obtain an optimized separation combination scheme, wherein the optimized separation combination scheme includes separation stage and expected stage separation index; then, with the expected stage separation index as a constraint, the separation control parameters of each separation stage are optimized, and an optimized separation control scheme is generated according to the optimization results and the optimized separation combination scheme; finally, the target drug mixture is separated and processed according to the optimized separation control scheme, and synchronous monitoring and feedback control are performed through the separation control twin model during the separation process. The fineness and accuracy of the drug mixture separation control scheme setting can be improved, thereby improving the accuracy of drug mixture separation control, achieving the technical effect of improving drug mixture separation quality and separation efficiency, and saving separation resources.

[0027] Further, step S100 of the present application includes: The chemical analysis channel includes a chromatographic analysis unit, a mass spectrometry analysis unit and an infrared spectrum analysis unit.

[0028] Specifically, the chemical analysis channel includes a chromatographic analysis unit, a mass spectrometry analysis unit and an infrared spectrum analysis unit. The chromatographic analysis unit is constructed based on the principle of chromatographic analysis. Chromatography is a physical and chemical separation and analysis method that uses a specific chromatographic system (such as thin layer chromatography, high performance liquid chromatography or gas chromatography, etc.) to separate and analyze the components in a mixture; the mass spectrometry analysis unit analyzes the composition and structure of a substance based on mass spectrometry analysis technology. Mass spectrometry analysis technology is based on the mass-to-charge ratio of ions. Under the action of electric and magnetic fields, the moving ions are separated and then detected. The mass spectrometer is the main technical instrument used in the mass spectrometry analysis unit. It can draw ion spectra, and obtain relevant information including molecular weight, fragmentation rules, chemical structure formula, etc. by observing and analyzing the intensity of the ion spectrum peak and the compound content, thereby realizing chemical analysis; the infrared spectrum analysis unit mainly uses infrared spectrum analysis technology to analyze substances. Infrared spectrum analysis refers to the analysis and identification of substance molecules using infrared spectra, and its working principle is based on molecular vibration and rotational energy level transitions.

[0029] By constructing a chemical analysis channel based on a chromatographic analysis unit, a mass spectrometry analysis unit, and an infrared spectroscopy analysis unit, the advantages of multiple analysis techniques can be combined to improve the accuracy of the analysis of the target drug mixture.

[0030] Further, construct a component identification channel, such as the attached Figure 2 As shown, the present application S100 includes: Acquire a preset recognition operator, wherein the preset recognition operator includes at least a BP network, a support vector machine and a random decision forest; select a first recognition operator to build a first recognition channel, and call N sample training sets based on the analysis unit of the chemical analysis channel, wherein N is the number of analysis units; perform supervised training and feedback verification on the first recognition channel through the N sample training sets respectively, and obtain N first convergent recognition sub-channels, wherein each first convergent recognition sub-channel has a convergence degree mark; obtain N convergent recognition sub-channel sets through iterative training, cluster the N convergent recognition sub-channel sets based on the analysis unit, and generate N convergent recognition units; set a trusted weight based on the attribute information of the analysis unit, embed the trusted weight into the N convergent recognition units, and establish the component recognition channel after fusion, wherein the attribute information includes the accuracy of the analysis data and the reliability of the analysis data.

[0031] Specifically, the method for constructing a component recognition channel is as follows: first, a preset recognition operator is obtained, wherein the preset recognition operator includes at least a BP network, a support vector machine, and a random decision forest. Additional recognition operators can also be added according to actual conditions to further improve the accuracy and reliability of recognition; wherein the BP network is a feedforward neural network that can be iteratively optimized in machine learning, and automatically optimizes the model based on the error back propagation algorithm. It has a strong pattern recognition ability and can learn and recognize specific patterns from a large amount of data; the support vector machine is a type of generalized linear classifier that performs binary classification of data in a supervised learning manner, and can be used for feature recognition; the random decision forest is an ensemble learning method based on decision trees, which constructs multiple decision trees by randomly selecting features and samples in a training set, and integrates the prediction results of these decision trees (such as voting or averaging) to obtain a more stable and accurate prediction model.

[0032] Then, a first recognition operator is randomly selected from the preset recognition operators, and the first recognition operator is any one of the preset recognition operators, such as: BP network; then a first recognition channel is built based on the first recognition operator, such as: BP neural network model; next, based on the analysis unit of the chemical analysis channel, N sample training sets are called, where N is the number of analysis units, for example: assuming that the analysis unit is a chromatographic analysis unit, the sample training data corresponding to the chromatographic analysis unit includes sample chromatographic analysis data and sample component identification results; then the N sample training sets are divided into N training sets and N verification sets; supervised training is performed on the first recognition channel through the N training sets, and after reaching the preset number of training times, feedback verification is performed on the first recognition channel through the N verification sets to obtain N first convergent recognition sub-channels, wherein each first convergent recognition sub-channel has a convergence degree mark, and the convergence degree is the output accuracy of the first convergent recognition sub-channel.

[0033] Using the same method for obtaining the N first convergent identification sub-channels, a sample training set is obtained to iteratively train other preset identification operators to obtain a set of N convergent identification sub-channels; then, the set of N convergent identification sub-channels is clustered based on the analysis unit, that is, the convergent identification sub-channels with the same analysis properties are clustered into one category to generate N convergent identification units, wherein the output result of each convergent identification unit is a weighted calculation result of multiple convergent identification sub-channels with weights set based on the convergence degree, for example, multiple convergent identification sub-channels corresponding to the chromatographic analysis unit are clustered into one category to generate a chromatographic convergence identification unit.

[0034] The trusted weight is set based on the attribute information of the analysis unit, and the attribute information includes the accuracy of the analysis data and the reliability of the analysis data, wherein the greater the accuracy of the analysis data and the reliability of the analysis data, the greater the corresponding trusted weight, and the weight can be set using the existing coefficient of variation method; then the trusted weight is embedded in the N convergent identification units, and the component identification channel is established after fusion, wherein the output result of the component identification channel is the weighted calculation result of the N convergent identification units based on the trusted weight. By constructing the component identification channel, since different identification sub-channels are constructed based on different algorithms, the advantages of multiple algorithms can be integrated to reduce the limitations of single algorithm identification and improve the accuracy and reliability of component identification.

[0035] Further, based on the first separation evaluation function and the expected separation index, the separation method set is used to perform combination optimization to determine an optimized separation combination scheme. S200 of the present application includes: Obtaining expected separation indicators, wherein the expected separation indicators include purity indicators, separation degree indicators, and recovery rate indicators; taking the expected separation indicators as optimization constraints, based on the first separation evaluation function and the separation effect prediction model, using the separation mode set to perform combination optimization to obtain the optimized separation combination scheme; the expression of the first separation evaluation function is: ;in, Characterize separation and evaluate fitness, is the weight coefficient of separation cost, is the weight coefficient of separation efficiency, M is the number of separation methods, is the first weight, is the second weight, is the third weight, is the material loss of the ith separation mode, is the equipment loss of the ith separation mode, is the energy consumption of the ith separation mode, is the operation duration of the ith separation method.

[0036] Specifically, first, the expected separation index is obtained, wherein the expected separation index includes a purity index, a separation index, and a recovery index; then a first separation evaluation function and a separation effect prediction model are constructed. The expression of the first separation evaluation function is: ; In the first separation evaluation function, Characterize the separation evaluation fitness, where the higher the separation evaluation fitness, the better the characterization separation evaluation effect; is the weight coefficient of separation cost, is the weight coefficient of separation efficiency, where and The sum is 1, and the influence of separation cost and separation efficiency on the separation evaluation result can be set. The higher the influence, the greater the corresponding weight coefficient. M is the number of separation methods in the separation combination scheme. is the first weight, is the second weight, is the third weight, wherein the sum of the first weight, the second weight and the third weight is 1, and the separation cost influence is set based on the indicator; is the material loss of the ith separation method, where the ith separation method is any one of the M separation methods in the separation combination scheme; is the equipment loss of the ith separation mode, is the energy consumption of the ith separation mode, is the operation duration of the ith separation method.

[0037] The separation effect prediction model is constructed based on a BP neural network and is used to evaluate the separation effect of a separation combination scheme, wherein the input data of the separation effect prediction model is the separation combination scheme, and the output data is the separation effect evaluation result, including the purity evaluation result, the separation degree evaluation result and the recovery rate evaluation result, and sample training data is obtained to train the separation effect prediction model to a convergence state. Then, the expected separation index is taken as an optimization constraint condition, based on the first separation evaluation function and the separation effect prediction model, an optimization space is constructed with the separation mode set, and the separation combination scheme is optimized using an optimization algorithm to obtain an optimized separation combination scheme.

[0038] Furthermore, taking satisfying the expected separation index as an optimization constraint, based on the first separation evaluation function and the separation effect prediction model, and using the separation method set to perform combined optimization, the present application further includes the following steps: Based on the separation method set, a plurality of initial separation combination schemes are randomly generated, wherein the separation time and the order of the separation methods in each initial separation combination scheme are different, and each initial separation combination scheme includes at least one separation method; the separation effect of the plurality of initial separation combination schemes is evaluated by the separation effect prediction model, and the initial separation combination scheme whose separation effect evaluation result meets the expected separation index is selected as the initial solution to obtain a plurality of initial solutions; the plurality of initial solutions are evaluated by the first separation evaluation function to obtain a plurality of evaluation fitnesses, and the plurality of initial solutions are divided from large to small according to the evaluation fitnesses to obtain a plurality of head solutions and a plurality of tail solutions, wherein the number of tail solutions is much greater than the number of head solutions; the plurality of tail solutions are randomly clustered with the head solution as the center to generate a plurality of solutions Threshold, and within the multiple solution thresholds, take the head solution as the optimization direction, mutate the tail solutions in the multiple solution thresholds based on the preset mutation strategy to obtain multiple updated solution thresholds, wherein the separation effect of the mutated tail solution is evaluated by the separation effect prediction model, and if the evaluation result does not meet the expected separation index, the tail solution is not mutated; identify the multiple updated solution thresholds, if within the updated solution threshold, there is a situation where the evaluation fitness of the tail solution is greater than the evaluation fitness of the head solution, then use the tail solution to update and replace the head solution; continuously perform iterative optimization until the current number of optimization times reaches the preset number threshold, then stop the optimization and output the current multiple solution thresholds; evaluate the current multiple solution thresholds, select the solution threshold with the largest evaluation fitness mean as the optimal solution threshold, and extract the head solution in the optimal solution threshold as the optimized separation combination scheme.

[0039] Specifically, first, a plurality of initial separation combination schemes are randomly generated based on the separation mode set, wherein the separation time and the order of the separation modes in each initial separation combination scheme are different, and each initial separation combination scheme includes at least one separation mode. Then, the separation effect prediction model is used to evaluate the separation effect of the plurality of initial separation combination schemes to obtain a plurality of separation effect evaluation results, and the initial separation combination schemes whose separation effect evaluation results are greater than the expected separation index are further selected as initial solutions to obtain a plurality of initial solutions; then, the first separation evaluation function is used to evaluate the plurality of initial solutions to obtain a plurality of evaluation fitnesses, and the plurality of initial solutions are divided from large to small according to the evaluation fitnesses, and a plurality of initial solutions with larger fitnesses are set as head solutions, and a plurality of solutions with smaller fitnesses are set as tail solutions to obtain a plurality of head solutions and a plurality of tail solutions, wherein the number of tail solutions is much greater than the number of head solutions.

[0040] Then, taking the head solution as the center, the multiple tail solutions are randomly clustered to generate multiple solution thresholds, wherein the number of tail solutions in each solution threshold is the same; then, within the multiple solution thresholds, the head solution is used as the optimization direction, and the tail solutions in the multiple solution thresholds are mutated based on the preset mutation strategy, wherein the preset mutation strategy includes the order adjustment step length and the separation time adjustment step length, which can be set according to actual needs, wherein the higher the optimization requirement accuracy, the shorter the adjustment step length, and multiple updated solution thresholds are obtained, wherein the separation effect of the mutated tail solution is evaluated by the separation effect prediction model, and if the evaluation result does not meet the expected separation index, the tail solution is not mutated. Then, the multiple updated solution thresholds are identified, and the mutated tail solutions within the multiple updated solution thresholds are evaluated by the first separation evaluation function to obtain the fitness of multiple mutated tail solutions, and then within the updated solution threshold, if there is a situation where the tail solution evaluation fitness is greater than the head solution evaluation fitness, the head solution is updated and replaced by the tail solution. Iterate the optimization until the current number of optimization times reaches a preset number threshold, which can be set according to actual conditions. Then stop the optimization and output the current multiple solution thresholds; then evaluate the current multiple solution thresholds, select the solution threshold with the largest evaluation fitness mean and set it as the optimal solution threshold, and at the same time extract the head solution in the optimal solution threshold and set it as the optimized separation combination scheme.

[0041] By using the optimization algorithm to optimize multiple initial separation combination schemes, due to the strong global search ability of the algorithm, the accuracy and comprehensiveness of the initial separation combination scheme optimization can be improved, thereby improving the accuracy and rationality of the optimized separation combination scheme.

[0042] Further, based on the stage separation evaluation function, the separation control parameters of each separation stage are optimized with the expected stage separation index being satisfied as a constraint. S300 of the present application includes: Select the first separation stage in the optimized separation combination scheme, and obtain the first expected stage separation index of the first separation stage, wherein the first separation stage is the separation stage with the highest ranking, and the first expected stage separation index includes a first expected purity index, a first expected separation index, and a first expected recovery rate index; construct a separation quality evaluation model, wherein the separation quality evaluation model includes a plurality of separation quality evaluation modules, and the separation quality evaluation modules correspond one to one to the separation mode; construct a stage separation evaluation function: ;in, is the fitness of the j-th separation control parameter, is the weight of the purity index, is the weight of the separation index, is the weight of the recovery rate indicator, is the predicted purity of the jth separation control parameter, is the first expected purity indicator, The predicted separation degree of the jth separation control parameter, is the first expected separation index, is the predicted recovery rate of the jth separation control parameter, is the first expected recovery rate indicator; based on the separation quality evaluation model and the stage separation evaluation function, the separation control parameters of the first separation mode of the first separation stage are optimized to obtain the first stage optimized control parameters, which are added to the multiple stage optimized control parameters; the multiple stage optimized control parameters and the optimized separation combination scheme are integrated to generate the optimized separation control scheme.

[0043] Specifically, first, the first separation stage in the optimized separation combination scheme is selected, and the first expected stage separation index of the first separation stage is obtained, wherein the first separation stage is the most highly ranked separation stage, and the first expected stage separation index includes a first expected purity index, a first expected separation index, and a first expected recovery index. Then, a separation quality evaluation model is constructed, wherein the separation quality evaluation model includes a plurality of separation quality evaluation modules, and the separation quality evaluation modules correspond one to one with the separation mode, and the separation quality evaluation modules are constructed based on a BP neural network, and are used to predict the separation quality under different separation control parameters, and obtain predicted purity, predicted separation degree, and predicted recovery rate.

[0044] Construct a phase separation evaluation function, the phase separation evaluation function expression is: ; In the stage separation evaluation function, is the fitness of the jth separation control parameter, where the larger the fitness, the better the separation quality; is the weight of the purity index, is the weight of the separation index, is the weight of the recovery rate indicator, is the predicted purity of the jth separation control parameter, is the first expected purity indicator, The predicted separation degree of the jth separation control parameter, is the first expected separation index, is the predicted recovery rate of the jth separation control parameter, It is the first expected recovery rate indicator.

[0045] First, the rated threshold of the separation control parameter of the first separation mode of the first separation stage is obtained, the first separation control parameter optimization space is constructed, and a plurality of first initial separation control parameters are randomly generated based on the first separation control parameter optimization space; then, based on the separation quality evaluation model and the stage separation evaluation function, the plurality of first initial separation control parameters are optimized to obtain the first stage optimized control parameters, wherein the optimization method is the same as the above-mentioned patrol method for obtaining the optimized separation combination scheme, and in order to reflect the simplicity of the specification, it will not be described in detail here, and the first stage optimized control parameters are obtained and added to the multiple stage optimized control parameters, wherein each separation stage corresponds to a stage optimized control parameter. Finally, the multiple stage optimized control parameters and the optimized separation combination scheme are integrated to obtain the optimized separation control scheme.

[0046] Further, by performing synchronous monitoring and feedback control through the separate control twin model, step S400 of the present application includes: Based on digital twin technology, multiple separation methods are simulated to construct a separation control twin space, wherein the separation control twin space includes multiple separation control twin models, and the separation control twin models correspond to the separation methods one by one; based on the optimized separation control scheme, the matching separation control twin model in the separation control twin space is called to synchronously monitor the separation process, and when the monitoring results deviate from the preset threshold, the subsequent separation control scheme is optimized and adjusted.

[0047] Specifically, digital twin is a technology that integrates multi-physics, multi-scale and multi-disciplinary attributes, and has the characteristics of real-time synchronization, faithful mapping and high fidelity. Digital twin integrates multiple technologies such as collaborative modeling, data mining, and human-computer interaction, and realizes collaborative interaction between the physical world and the digital world. Compared with traditional fault prediction methods, fault prediction methods based on digital twins have more advantages in improving prediction accuracy and accuracy. Based on digital twin technology, the separation process of multiple separation methods is simulated to construct a separation control twin space, wherein the separation control twin space includes multiple separation control twin models, and the separation control twin models correspond to the separation methods one by one.

[0048] Then, when separation control is performed according to the optimized separation control scheme, the matching separation control twin model in the separation control twin space is called based on the separation mode in the optimized separation control scheme, and the separation process is synchronously monitored by the matching separation control twin model, and when the monitoring result deviates from the preset threshold, the subsequent separation control scheme is optimized and adjusted. The preset threshold is the control error range, which can be set according to the simulation accuracy of the separation control twin model. By constructing a separation control twin model to synchronously monitor and feedback control the separation process, the accuracy of separation control of the target drug mixture can be further improved, thereby improving the separation quality and separation efficiency of the drug mixture.

[0049] In summary, the separation method for sampling pharmaceutical equipment provided in this application has the following technical effects: 1. By constructing a chemical analysis channel and a component identification channel to identify the components of the target drug mixture, the mixture component information can be obtained, which can improve the accuracy and efficiency of the mixture component information; then, based on the mixture component information matching, a set of suitable separation methods is obtained, and the separation method set is used to perform combination optimization to obtain an optimized separation combination scheme, wherein the optimized separation combination scheme includes separation stage and expected stage separation indicators; then, with the expected stage separation indicator as a constraint, the separation control parameters of each separation stage are optimized, and an optimized separation control scheme is generated based on the optimization results and the optimized separation combination scheme; finally, the target drug mixture is separated and processed according to the optimized separation control scheme, and synchronous monitoring and feedback control are performed through the separation control twin model during the separation process. The precision and accuracy of the drug mixture separation control scheme setting can be improved, thereby improving the accuracy of drug mixture separation control, achieving the technical effect of improving the separation quality and efficiency of drug mixtures and saving separation resources.

[0050] 2. By constructing component identification channels, since different identification sub-channels are constructed based on different algorithms, the advantages of multiple algorithms can be integrated to reduce the limitations of single algorithm identification and improve the accuracy and reliability of component identification.

[0051] 3. By using the optimization algorithm to optimize multiple initial separation combination schemes, due to the strong global search capability of the algorithm, the accuracy and comprehensiveness of the initial separation combination scheme optimization can be improved, thereby improving the accuracy and rationality of the optimized separation combination scheme.

[0052] 4. By constructing a separation control twin model to synchronously monitor and feedback control the separation process, the accuracy of separation control of the target drug mixture can be further improved, thereby improving the separation quality and efficiency of the drug mixture.

[0053] Embodiment 2 Based on the same inventive concept as the separation method for sampling pharmaceutical equipment in the aforementioned embodiment, the present application also provides a separation system for sampling pharmaceutical equipment, see the attached Figure 3 , the system comprising: The mixture component information output module 11 is used to obtain a target drug mixture, analyze the target drug mixture through a chemical analysis channel to obtain a chemical analysis data set, and then identify the chemical analysis data set through a component identification channel to output the mixture component information, wherein the mixture component information includes component type and mixing ratio.

[0054] The optimized separation combination scheme determination module 12 is used to input the component type into a preset separation data table, match and obtain a set of separation methods, and based on the first separation evaluation function and the expected separation index, use the set of separation methods to perform combination optimization to determine the optimized separation combination scheme, wherein the optimized separation combination scheme includes separation stages and expected stage separation indicators, and each separation stage corresponds to a separation method.

[0055] The optimized separation control scheme generating module 13 is used to optimize the separation control parameters of each separation stage based on the stage separation evaluation function and to satisfy the expected stage separation index as a constraint, and to generate an optimized separation control scheme according to the optimization result and the optimized separation combination scheme.

[0056] The separation monitoring and control module 14 is used to separate the target drug mixture according to the optimized separation control scheme, and perform synchronous monitoring and feedback control through the separation control twin model.

[0057] Furthermore, the mixture composition information output module 11 in the system is also used for: the chemical analysis channel includes a chromatographic analysis unit, a mass spectrometry analysis unit and an infrared spectrum analysis unit.

[0058] Furthermore, the mixture component information output module 11 in the system is also used to: obtain a preset recognition operator, wherein the preset recognition operator includes at least a BP network, a support vector machine and a random decision forest; select a first recognition operator to build a first recognition channel, and based on the analysis unit of the chemical analysis channel, call N sample training sets, wherein N is the number of analysis units; perform supervised training and feedback verification on the first recognition channel through the N sample training sets respectively, to obtain N first convergent recognition sub-channels, wherein each first convergent recognition sub-channel has a convergence degree mark; obtain N convergent recognition sub-channel sets through iterative training, cluster the N convergent recognition sub-channel sets based on the analysis unit, and generate N convergent recognition units; set a trusted weight based on the attribute information of the analysis unit, embed the trusted weight into the N convergent recognition units, and establish the component recognition channel after fusion, wherein the attribute information includes the accuracy of the analysis data and the reliability of the analysis data.

[0059] Further, the optimized separation combination scheme determination module 12 in the system is also used to: obtain expected separation indicators, wherein the expected separation indicators include purity indicators, separation degree indicators and recovery rate indicators; based on the first separation evaluation function and the separation effect prediction model, the separation mode set is used to perform combination optimization to obtain the optimized separation combination scheme, with the expected separation index being taken as the optimization constraint; the expression of the first separation evaluation function is: ;in, Characterize separation and evaluate fitness, is the weight coefficient of separation cost, is the weight coefficient of separation efficiency, M is the number of separation methods, is the first weight, is the second weight, is the third weight, is the material loss of the ith separation mode, is the equipment loss of the ith separation mode, is the energy consumption of the ith separation mode, is the operation duration of the ith separation method.

[0060] Furthermore, the optimized separation combination scheme determination module 12 in the system is also used to: randomly generate multiple initial separation combination schemes based on the separation method set, wherein the separation time and sequence of the separation methods in each initial separation combination scheme are different, and each initial separation combination scheme includes at least one separation method; evaluate the separation effect of the multiple initial separation combination schemes through the separation effect prediction model, select the initial separation combination scheme whose separation effect evaluation result meets the expected separation index and set it as the initial solution to obtain multiple initial solutions; evaluate the multiple initial solutions using the first separation evaluation function to obtain multiple evaluation fitness, and divide the multiple initial solutions from large to small according to the evaluation fitness to obtain multiple head solutions and multiple tail solutions, wherein the number of tail solutions is much larger than the number of head solutions; and evaluate the multiple initial solutions with the head solution as the center. The tail solutions are randomly clustered to generate multiple solution thresholds, and within the multiple solution thresholds, the head solution is taken as the optimization direction, and the tail solutions in the multiple solution thresholds are mutated based on a preset mutation strategy to obtain multiple updated solution thresholds, wherein the separation effect of the mutated tail solution is evaluated by the separation effect prediction model, and if the evaluation result does not meet the expected separation index, the tail solution is not mutated; the multiple updated solution thresholds are identified, and if within the updated solution threshold, the tail solution is used to update and replace the head solution; iterative optimization is continuously performed until the current number of optimization searches reaches a preset number threshold, then the optimization is stopped and the current multiple solution thresholds are output; the current multiple solution thresholds are evaluated, and the solution threshold with the largest mean evaluation fitness is selected as the optimal solution threshold, and the head solution in the optimal solution threshold is extracted and set as the optimized separation combination scheme.

[0061] Further, the optimized separation control scheme generation module 13 in the system is also used to: select the first separation stage in the optimized separation combination scheme, and obtain the first expected stage separation index of the first separation stage, wherein the first separation stage is the separation stage with the highest ranking, and the first expected stage separation index includes a first expected purity index, a first expected separation degree index and a first expected recovery rate index; construct a separation quality evaluation model, wherein the separation quality evaluation model includes a plurality of separation quality evaluation modules, and the separation quality evaluation modules correspond one to one to the separation mode; construct a stage separation evaluation function: ;in, is the fitness of the j-th separation control parameter, is the weight of the purity index, is the weight of the separation index, is the weight of the recovery rate indicator, is the predicted purity of the jth separation control parameter, is the first expected purity indicator, The predicted separation degree of the jth separation control parameter, is the first expected separation index, is the predicted recovery rate of the jth separation control parameter, is the first expected recovery rate indicator; based on the separation quality evaluation model and the stage separation evaluation function, the separation control parameters of the first separation mode of the first separation stage are optimized to obtain the first stage optimized control parameters, which are added to the multiple stage optimized control parameters; the multiple stage optimized control parameters and the optimized separation combination scheme are integrated to generate the optimized separation control scheme.

[0062] Furthermore, the separation monitoring and control module 14 in the system is also used to: simulate a variety of separation methods based on digital twin technology, and construct a separation control twin space, wherein the separation control twin space includes multiple separation control twin models, and the separation control twin models correspond one to one to the separation methods; based on the optimized separation control scheme, call the matching separation control twin model in the separation control twin space to synchronously monitor the separation process, and when the monitoring results deviate from the preset threshold, optimize and adjust the subsequent separation control scheme.

[0063] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The separation method and specific examples of a pharmaceutical equipment sampling in the aforementioned embodiment 1 are also applicable to a pharmaceutical equipment sampling separation system in this embodiment. Through the aforementioned detailed description of a pharmaceutical equipment sampling separation method, those skilled in the art can clearly understand the pharmaceutical equipment sampling separation system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0064] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be 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. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0065] 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 belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A separation method for sampling pharmaceutical equipment, characterized in that: The method comprises: S100: Acquire a target drug mixture, parse the target drug mixture through a chemical parsing channel to obtain a chemical parsing data set, then identify the chemical parsing data set through a component identification channel, and output mixture component information, wherein the mixture component information includes component type and mixing ratio; S200: Input the component type into a preset separation data table, match to obtain a set of separation methods, perform combination optimization using the set of separation methods based on a first separation evaluation function and an expected separation index, and determine an optimized separation combination scheme, wherein the optimized separation combination scheme includes a separation stage and an expected stage separation index, and each separation stage corresponds to a separation method; S300: Based on the stage separation evaluation function, the separation control parameters of each separation stage are optimized with the expected stage separation index being satisfied as a constraint, and an optimized separation control scheme is generated according to the optimization result and the optimized separation combination scheme; S400: Separate the target drug mixture according to the optimized separation control scheme, and perform synchronous monitoring and feedback control through the separation control twin model.

2. The method according to claim 1, characterized in that The chemical analysis channel includes a chromatographic analysis unit, a mass spectrometry analysis unit and an infrared spectrum analysis unit.

3. The method according to claim 2, characterized in that Build a component identification pipeline, including: Obtaining a preset recognition operator, wherein the preset recognition operator includes at least a BP network, a support vector machine, and a random decision forest; Selecting a first recognition operator to build a first recognition channel, and calling N sample training sets based on the analysis unit of the chemical analysis channel, where N is the number of analysis units; Performing supervised training and feedback verification on the first recognition channel through N sample training sets to obtain N first convergent recognition sub-channels, wherein each first convergent recognition sub-channel has a convergence degree mark; Obtaining N convergent identification sub-channel sets through iterative training, clustering the N convergent identification sub-channel sets based on the parsing unit to generate N convergent identification units; A trusted weight is set based on the attribute information of the analysis unit, and the trusted weight is embedded in the N convergence identification units, and the component identification channel is established after fusion, wherein the attribute information includes the accuracy and reliability of the analysis data.

4. The method according to claim 1, characterized in that: Based on the first separation evaluation function and the expected separation index, the separation method set is used to perform combination optimization to determine an optimized separation combination scheme, including: Obtaining expected separation indicators, wherein the expected separation indicators include a purity indicator, a separation index, and a recovery rate index; Taking satisfying the expected separation index as an optimization constraint, based on the first separation evaluation function and the separation effect prediction model, using the separation method set to perform combined optimization to obtain the optimized separation combination scheme; The expression of the first separation evaluation function is: ; in, Characterize separation and evaluate fitness, is the weight coefficient of separation cost, is the weight coefficient of separation efficiency, M is the number of separation methods, is the first weight, is the second weight, is the third weight, is the material loss of the ith separation mode, is the equipment loss of the ith separation mode, is the energy consumption of the ith separation mode, is the operation duration of the ith separation method.

5. The method according to claim 4, characterized in that Taking satisfying the expected separation index as an optimization constraint, based on the first separation evaluation function and the separation effect prediction model, using the separation mode set to perform combined optimization, including: Randomly generate a plurality of initial separation combination schemes based on the separation method set, wherein the separation time and the order of the separation methods in each initial separation combination scheme are different, and each initial separation combination scheme includes at least one separation method; Performing separation effect evaluation on the multiple initial separation combination schemes through the separation effect prediction model, selecting the initial separation combination scheme whose separation effect evaluation result meets the expected separation index as the initial solution, and obtaining multiple initial solutions; Using the first separation evaluation function to evaluate the multiple initial solutions to obtain multiple evaluation fitnesses, and dividing the multiple initial solutions according to the evaluation fitnesses from large to small to obtain multiple head solutions and multiple tail solutions, wherein the number of tail solutions is much greater than the number of head solutions; Taking the head solution as the center, a plurality of tail solutions are randomly clustered to generate a plurality of solution thresholds, and within the plurality of solution thresholds, the head solution is taken as the optimization direction, and the tail solutions in the plurality of solution thresholds are mutated based on a preset mutation strategy to obtain a plurality of updated solution thresholds, wherein the separation effect of the mutated tail solutions is evaluated by the separation effect prediction model, and if the evaluation result does not meet the expected separation index, the tail solution is not mutated; Identify the multiple update solution thresholds, and if within the update solution threshold, there is a situation where the evaluation fitness of the tail solution is greater than the evaluation fitness of the head solution, then use the tail solution to update and replace the head solution; Continue to iterate and optimize until the current number of optimizations reaches the preset number threshold, then stop the optimization and output the current multiple solution thresholds; The current multiple solution thresholds are evaluated, and the solution threshold with the largest evaluation fitness mean is selected as the optimal solution threshold, and the head solution in the optimal solution threshold is extracted and set as the optimized separation combination scheme.

6. The method according to claim 5, characterized in that Based on the stage separation evaluation function, the separation control parameters of each separation stage are optimized with the expected stage separation index as a constraint, including: Selecting the first separation stage in the optimized separation combination scheme, and obtaining a first expected stage separation index of the first separation stage, wherein the first separation stage is the separation stage with the highest ranking, and the first expected stage separation index includes a first expected purity index, a first expected separation index, and a first expected recovery rate index; Constructing a separation quality evaluation model, wherein the separation quality evaluation model includes a plurality of separation quality evaluation modules, and the separation quality evaluation modules correspond one to one to the separation modes; Separate evaluation function during construction phase: ; in, is the fitness of the j-th separation control parameter, is the weight of the purity index, is the weight of the separation index, is the weight of the recovery rate indicator, is the predicted purity of the jth separation control parameter, is the first expected purity indicator, The predicted separation degree of the jth separation control parameter, is the first expected separation index, is the predicted recovery rate of the jth separation control parameter, is the first expected recovery rate indicator; Based on the separation quality evaluation model and the stage separation evaluation function, optimizing the separation control parameters of the first separation mode of the first separation stage, obtaining the first stage optimized control parameters, and adding them to the multiple stage optimized control parameters; The optimized control parameters of the multiple stages and the optimized separation combination scheme are integrated to generate the optimized separation control scheme.

7. The method according to claim 6, characterized in that Perform simultaneous monitoring and feedback control via a separate control twin, including: Based on the digital twin technology, multiple separation modes are simulated to construct a separation control twin space, wherein the separation control twin space includes multiple separation control twin models, and the separation control twin models correspond to the separation modes one by one; Based on the optimized separation control scheme, the matching separation control twin model in the separation control twin space is called to synchronously monitor the separation process, and when the monitoring results deviate from the preset threshold, the subsequent separation control scheme is optimized and adjusted.

8. A separation system for sampling pharmaceutical equipment, characterized in that: For implementing the steps of the method according to any one of claims 1 to 7, the system comprises: A mixture component information output module is used to obtain a target drug mixture, parse the target drug mixture through a chemical analysis channel to obtain a chemical analysis data set, then identify the chemical analysis data set through a component identification channel, and output mixture component information, wherein the mixture component information includes component type and mixing ratio; An optimized separation combination scheme determination module is used to input the component type into a preset separation data table, match to obtain a set of separation methods, and perform combination optimization using the set of separation methods based on a first separation evaluation function and an expected separation index to determine an optimized separation combination scheme, wherein the optimized separation combination scheme includes a separation stage and an expected stage separation index, and each separation stage corresponds to a separation method; An optimized separation control scheme generation module is used to optimize the separation control parameters of each separation stage based on the stage separation evaluation function and to satisfy the expected stage separation index as a constraint, and to generate an optimized separation control scheme according to the optimization result and the optimized separation combination scheme; The separation monitoring and control module is used to separate the target drug mixture according to the optimized separation control scheme, and perform synchronous monitoring and feedback control through the separation control twin model.

Citation Information

Cited By

  • Self-adaptive collaborative optimization method and system based on separation simulation control

    CN121979128A