A control optimization method and system for extraction and separation equipment for pharmaceutical use

Abstract

The present invention discloses a control optimization method and system for pharmaceutical extraction and separation equipment, which relates to data processing related fields. The method includes: performing component detection on a target mixture to obtain component characteristic information; setting extraction constraints according to extraction requirements; selecting an extractant, optimizing extraction parameters, and obtaining optimal extraction parameters; using optimal extraction parameters and extractants, controlling the extraction and separation equipment to extract, obtaining extraction results, and obtaining actual extraction purity information and actual impurity content information; setting impurity constraints, optimizing washing parameters, and obtaining optimal washing parameters; controlling the extraction and separation equipment to wash the extraction results to obtain the final extraction results. The method solves the technical problems of the lack of pertinence in the control of existing extraction and separation equipment, and the difficulty in simultaneously meeting the requirements of high extraction purity and low impurity content, and achieves the technical effect of optimizing operations according to specific component characteristics and simultaneously achieving the requirements of high extraction purity and low impurity content.

Description

Technical Field

[0001] The present application relates to data processing related fields, and in particular to a control optimization method and system for pharmaceutical extraction and separation equipment. Background Art

[0002] In the pharmaceutical industry, the extraction and purification of drug ingredients is a crucial link, which is directly related to the quality and efficacy of drugs. Traditional pharmaceutical extraction and separation equipment often relies on experience or fixed parameter settings for extraction and washing during operation, lacking pertinence and flexibility. Due to the complexity of drug ingredients and separation components, traditional operation methods often cannot meet the requirements of high extraction purity and low impurity content at the same time, resulting in unstable drug quality and affecting drug efficacy.

[0003] In the current relevant technologies, the control of extraction and separation equipment lacks specificity and is difficult to simultaneously meet the requirements of high extraction purity and low impurity content, resulting in technical problems such as poor extraction effects. Summary of the invention

[0004] The present application provides a control optimization method and system for pharmaceutical extraction and separation equipment, which uses technical means such as component detection of a target mixture to be extracted and separated, setting extraction constraints, optimizing extraction parameters, using optimal extraction parameters and extractants for extraction operations, setting impurity constraints, and optimizing washing parameters. The method achieves the technical effect of optimizing operations according to the specific component characteristics of the target mixture, while achieving the requirements of high extraction purity and low impurity content, thereby improving the quality of drugs.

[0005] The present application provides a control optimization method for a pharmaceutical extraction and separation device, comprising:

[0006] The target mixture to be extracted and separated is subjected to component detection to obtain component characteristic information, wherein the target mixture includes drug components and separation components; extraction constraints are set according to the extraction requirements of the drug components, wherein the extraction constraints include purity constraints; an extractant is selected according to the component characteristic information, and extraction parameters are optimized in combination with the extraction constraints to obtain optimal extraction parameters, and optimal extraction purity information and optimal impurity content information for extraction under the optimal extraction parameters are predicted and obtained, wherein the extraction parameters are optimized for the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant; the optimal extraction is adopted Take parameters and extractant, control the extraction and separation equipment to extract the target mixture, obtain the extraction result, and detect and obtain the actual extraction purity information and the actual impurity content information; set impurity constraints, and optimize the washing parameters of the extraction result washed with the original solvent of the target mixture according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraints, to obtain the optimal washing parameters, wherein the optimization is performed with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result; use the optimal washing parameters to control the extraction and separation equipment to wash the extraction result to obtain the final extraction result.

[0007] In a possible implementation, the target mixture to be extracted and separated is subjected to component detection and the following processing is performed:

[0008] The target mixture to be extracted and separated is randomly sampled; the randomly sampled samples are tested, and the drug component content and the separation component content are calculated to obtain component characteristic information.

[0009] In a possible implementation, according to the extraction requirements of the drug components, extraction constraints are set and the following processing is performed:

[0010] The effective purity threshold of the drug component is obtained; the purity of the drug component after extraction is greater than or equal to the effective purity threshold, which is used as a purity constraint to construct the extraction constraint.

[0011] In a possible implementation, based on the component characteristic information, an extractant is selected, and extraction parameters are optimized in combination with the extraction constraints to obtain optimal extraction parameters, and the following processing is performed:

[0012] According to the component characteristic information, the extractant is selected and the extraction parameter range is obtained; in order to improve the extraction purity of the drug components and reduce the impurity content in the extractant, an extraction optimization function is constructed as follows: ; where EXF is the extraction fitness, and The sum of is 1, which are the purity weight and the impurity weight respectively, C is the extraction purity of the drug component in the extraction result after extraction according to the extraction parameters, and K is the impurity content in the extraction result after extraction according to the extraction parameters; within the extraction parameter range, a plurality of initial extraction parameters are randomly generated, and extraction prediction is performed in combination with the component characteristic information to obtain a plurality of initial extraction purity information and a plurality of initial impurity content information; it is determined whether the plurality of initial extraction purity information meets the extraction constraints, and if not, the corresponding initial random parameters are deleted and the initial extraction parameters that meet the extraction constraints are randomly generated again for completion; according to the extraction priority The method comprises the following steps: a first step is to transform the extraction parameters into a single function, and a plurality of initial extraction fitnesses are calculated and obtained; an optimized extraction parameter, a plurality of common extraction parameters and a worst extraction parameter are selected according to the plurality of initial extraction fitnesses; the worst extraction parameter is deleted, and the extraction parameters satisfying the extraction constraints are randomly generated again for completion, and the plurality of common extraction parameters are adjusted with the optimized extraction parameter as the adjustment direction to obtain a plurality of updated extraction parameters; a plurality of updated extraction fitnesses are calculated and the optimized extraction parameter, a plurality of common extraction parameters and the worst extraction parameter are updated; the optimization is continued until convergence, and the final optimized extraction parameter is output to obtain the optimal extraction parameter.

[0013] In a possible implementation, the component feature information is combined to perform extraction prediction and execute the following processing:

[0014] According to the historical extraction data of the target mixture, a sample extraction parameter set, a sample extraction component characteristic information set, and a sample extraction purity information set and a sample impurity content information set obtained by testing after extraction are collected; the sample extraction parameter set and the sample extraction component characteristic information set are used as input, and the sample extraction purity information set and the sample impurity content information set are used as output to train an extraction predictor; the extraction predictor is used to perform extraction prediction on the component characteristic information in combination with multiple initial extraction parameters to obtain multiple initial extraction purity information and multiple initial impurity content information.

[0015] In a possible implementation, an impurity constraint is set, and according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraint, the washing parameters for washing the extraction result using the original solvent of the target mixture are optimized to obtain the optimal washing parameters, and the following processing is performed:

[0016] Obtain the effective impurity content threshold of the drug component, and set the impurity content after washing to be less than the effective impurity content threshold as the impurity constraint; with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result, construct a washing function as follows: ; Where WAF is washing fitness, and is the weight, S is the drug purity loss in the extraction result after washing, and Z is the remaining impurity content in the extraction result after washing; according to the original solvent of the target mixture, the washing parameter range is obtained, and a plurality of initial washing parameters are randomly generated within the washing parameter range; based on the plurality of initial washing parameters, washing prediction is performed in combination with the actual extraction purity information and the actual impurity content information to obtain a plurality of predicted initial loss amounts and a plurality of predicted initial remaining impurity contents; according to the plurality of predicted initial loss amounts and the actual extraction purity information, a plurality of washing purity information is calculated and compensated to obtain a plurality of compensated washing purity information, and it is judged whether the extraction constraint is satisfied, and the plurality of predicted initial remaining impurity contents are compensated and it is judged whether the impurity constraint is satisfied, and the plurality of initial washing parameters are judged to be deleted and supplemented; based on the washing function, a plurality of initial washing fitness of the plurality of initial washing parameters after supplementation is calculated, and they are selected and optimized until convergence, and the optimal washing parameters are output.

[0017] In a possible implementation, based on the multiple predicted initial loss amounts and the actual extraction purity information, multiple washing purity information is calculated and compensated, and the following processing is performed:

[0018] Based on the multiple predicted initial loss amounts and the actual extraction purity information, multiple washing purity information are calculated; based on the actual extraction purity information and the optimal extraction purity information, a purity compensation coefficient is calculated to compensate for the multiple washing purity information to obtain multiple compensated washing purity information; based on the actual impurity content information and the optimal impurity content information, an impurity content compensation coefficient is calculated to compensate for the multiple predicted initial residual impurity contents to obtain multiple compensated predicted initial residual impurity contents; it is determined whether the multiple compensated washing purity information meets the extraction constraints, and it is determined whether the multiple compensated predicted initial residual impurity contents meet the impurity constraints.

[0019] The present application also provides a control optimization system for a pharmaceutical extraction and separation device, comprising:

[0020] a component detection module, the component detection module is used to detect the components of the target mixture to be extracted and separated, and obtain component characteristic information, wherein the target mixture includes drug components and separation components; an extraction constraint setting module, the extraction constraint setting module is used to set extraction constraints according to the extraction requirements of the drug components, wherein the extraction constraints include purity constraints; an extraction parameter optimization module, the extraction parameter optimization module is used to select an extractant according to the component characteristic information, optimize the extraction parameters in combination with the extraction constraints, obtain optimal extraction parameters, and predict and obtain optimal extraction purity information and optimal impurity content information for extraction under the optimal extraction parameters, wherein the extraction parameters are optimized for the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant; an extraction module, wherein the extraction module is used to use the optimal extraction parameters and the extractant to control the extraction and separation equipment to extract the target mixture, obtain the extraction result, and detect and obtain the actual extraction purity information and the actual impurity content information; a washing parameter optimization module, wherein the washing parameter optimization module is used to set impurity constraints, and according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraints, optimize the washing parameters for washing the extraction result using the original solvent of the target mixture to obtain the optimal washing parameters, wherein the optimization is performed with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result; and a washing module, wherein the washing module is used to use the optimal washing parameters to control the extraction and separation equipment to wash the extraction result to obtain the final extraction result.

[0021] The control optimization method and system of a pharmaceutical extraction and separation equipment proposed in the present application first detects the components of a target mixture to be extracted and separated to obtain component characteristic information, wherein the target mixture includes drug components and separation components, and then sets extraction constraints according to the extraction requirements of the drug components, wherein the extraction constraints include purity constraints, and then selects an extractant according to the component characteristic information, and optimizes the extraction parameters in combination with the extraction constraints to obtain the optimal extraction parameters, and predicts and obtains the optimal extraction purity information and the optimal impurity content information for extraction under the optimal extraction parameters, wherein the extraction parameters are optimized for the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant, and then the optimal extraction parameters and extractant are used to control the extraction The separation equipment extracts the target mixture to obtain the extraction result, and detects and obtains the actual extraction purity information and the actual impurity content information, then sets the impurity constraints, and optimizes the washing parameters for washing the extraction result with the original solvent of the target mixture according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraints, to obtain the optimal washing parameters, wherein the optimization is performed with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result, and finally uses the optimal washing parameters to control the extraction separation equipment to wash the extraction result to obtain the final extraction result, achieving the technical effect of optimizing the operation according to the specific component characteristics of the target mixture, while achieving the requirements of high extraction purity and low impurity content, thereby improving the quality of the drug. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0023] Figure 1 A schematic flow chart of a control optimization method for a pharmaceutical extraction and separation device provided in an embodiment of the present application.

[0024] Figure 2 A schematic diagram of the structure of a control optimization system for a pharmaceutical extraction and separation device provided in an embodiment of the present application.

[0025] Explanation of the reference numerals: component detection module 10 , extraction constraint setting module 20 , extraction parameter optimization module 30 , extraction module 40 , washing parameter optimization module 50 , washing module 60 . DETAILED DESCRIPTION

[0026] 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 in accordance with the contents of the specification. 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 listed below.

[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0028] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0029] The present application embodiment provides a control optimization method for a pharmaceutical extraction and separation device, such as Figure 1 As shown, the method includes:

[0030] Step S100, performing component detection on the target mixture to be extracted and separated to obtain component characteristic information, wherein the target mixture includes drug components and separation components. Specifically, the target mixture refers to a mixture that needs to be subjected to extraction and separation operations, and contains drug components that need to be extracted and separation components that are not needed, wherein drug components refer to active ingredients that need to be extracted from the target mixture and have specific pharmacological effects; separation components refer to other ingredients in the target mixture other than drug components, which have a negative impact on the purity or effect of the drug and need to be removed by extraction and separation technology. Specifically, a certain amount of sample of the target mixture to be extracted is taken out from the production line or storage container to ensure that the sample is representative and can reflect the average composition of the entire batch or a large number of target mixtures. According to the properties of the target mixture (such as polarity, volatility, thermal stability, etc.), select appropriate detection technology, including high performance liquid chromatography (HPLC), gas chromatography (GC), mass spectrometry (MS), nuclear magnetic resonance (NMR), etc., pre-treat the sample, such as dissolution, dilution, filtration, derivatization, etc., so as to more accurately detect the target component. According to the selected detection technology, set appropriate detection conditions, such as mobile phase composition, flow rate, column temperature, etc., inject the prepared target mixture sample into the detection instrument, perform component analysis according to the set conditions, output the retention time, peak area, peak height and other information of each component, process the data output by the instrument, such as integration, correction, normalization, etc., to obtain the relative content or absolute content of each component. The processed data is organized into component characteristic information, including the type, concentration, ratio of the drug component, and the type and content of the separated component.

[0031] In a possible implementation, the target mixture to be extracted and separated is subjected to component detection, and step S100 further includes step S110, randomly sampling the target mixture to be extracted and separated. Specifically, according to the total amount of the target mixture and the expected uniformity, the number and position of sampling are determined, sampling points are randomly selected in the target mixture, and samples are obtained using a sampling tool. During the sampling process, the randomness of sampling is ensured so that the sample can represent the characteristics of the entire target mixture. Step S120, the randomly sampled samples are tested, and the drug component content and the separation component content are calculated as component characteristic information. Specifically, according to the nature of the sample and the detection requirements, the sample is appropriately pre-treated, such as dilution, filtration, etc., the treated sample is placed in a detection instrument, and the detection is performed according to a predetermined detection method and parameters, the data output by the detection instrument is analyzed, the content of the drug component in the sample and the content of the component to be separated are calculated, and the results of the detection and analysis are recorded as component characteristic information. This implementation achieves the technical effect of improving the accuracy of the detection of the target mixture components through random sampling and sample detection.

[0032] Step S200, according to the extraction requirements of the drug components, the extraction constraints are set, wherein the extraction constraints include purity constraints. Specifically, the extraction requirements of the drug components, that is, the specific requirements and expectations for the extraction process of the drug components, including the required purity, output, cost and other factors are analyzed, and according to the extraction requirements, the minimum purity standard that the drug components should reach after extraction is determined, and one or more purity constraint values ​​are set. The purity constraint value can be a specific purity percentage, that is, the proportion of the drug components extracted from the target mixture, or it can be a purity range. For example, the purity of the drug components is set to be no less than 90%, or the purity should be between 89% and 91%. In addition to purity constraints, other extraction constraints can also be set according to the specific extraction process and equipment performance, such as extraction time, temperature, pressure, solvent dosage, etc.

[0033] In a possible implementation, according to the extraction requirements of the drug component, an extraction constraint is set, and step S200 further includes step S210, obtaining the effective purity threshold of the drug component. Specifically, the relevant drug standards, literature or experimental data are consulted to determine the effective purity threshold of the drug component. The effective purity threshold is determined based on factors such as the drug's efficacy, safety and market requirements. The drug component needs to reach the minimum purity requirement after extraction to ensure the quality and effect of the final product. Step S220, the purity of the drug component after extraction is greater than or equal to the effective purity threshold, as a purity constraint, to construct the extraction constraint. Specifically, according to the obtained effective purity threshold, the purity requirement of the drug component after extraction is determined, and the purity requirement of the drug component after extraction is converted into a specific constraint condition, that is, the purity after extraction is greater than or equal to the effective purity threshold. This implementation method can better guide the parameter adjustment and optimization in the extraction process by setting an effective purity constraint, and achieve the technical effect of improving the accuracy of extraction parameter optimization.

[0034] Step S300, according to the component characteristic information, select the extractant, combine the extraction constraints, optimize the extraction parameters, obtain the optimal extraction parameters, and predict the optimal extraction purity information and the optimal impurity content information for extraction under the optimal extraction parameters, wherein the extraction parameters are optimized for the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant. Specifically, according to the component characteristic information of the target mixture (such as the polarity and solubility of the drug components and the separation components), select a suitable extractant (a solvent or solution for extracting the drug components from the target mixture). The selection of the extractant should ensure that the drug components can be effectively separated from the target mixture and the dissolution of the impurity components can be minimized. According to experience or equipment performance, set the initial extraction parameters, such as extraction time, temperature, solvent dosage, stirring speed, etc. The initial extraction parameters are the starting point for subsequent optimization. Establish a mathematical model or simulation model to describe the relationship between the extraction process and the extraction parameters. The model can predict the extraction purity, impurity content and other possible performance indicators of the drug components under different extraction parameters. Integrate extraction constraints such as purity constraints into the optimization model, and use optimization algorithms (such as genetic algorithms, simulated annealing algorithms, particle swarm optimization, etc.) to optimize the extraction parameters. During the optimization process, predict the extraction purity and impurity content under different extraction parameters, and calculate the fitness value. Adjust the parameters according to the fitness value, and iterate the optimization until the optimal extraction parameters are found. Use the optimal extraction parameters obtained by optimization to predict the optimal extraction purity information and optimal impurity content information after extraction through the model, compare the extraction parameters and prediction results before and after optimization, and evaluate whether the optimization effect has achieved the expected goal. If the optimization effect is not ideal, adjust the optimization algorithm or model parameters and re-optimize.

[0035] In one possible implementation, based on the component characteristic information, an extractant is selected, and in combination with the extraction constraints, extraction parameters are optimized to obtain optimal extraction parameters. Step S300 further includes step S310, in which an extractant is selected based on the component characteristic information, and an extraction parameter range is obtained. Specifically, the component characteristic information is analyzed to determine the properties and interactions of each component in the target mixture, and a suitable extractant is selected based on the properties of the components. According to the characteristics of the extractant and the experimental conditions, a reasonable value range of the extraction parameters (such as temperature, time, solvent ratio, etc.) is determined. Step S320, with the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant, an extraction optimization function is constructed as follows:

[0036] ;

[0037] Among them, EXF is the extraction fitness, and The sum of is 1, which are the purity weight and the impurity weight respectively, C is the extraction purity of the drug components in the extraction result after extraction according to the extraction parameters, and K is the impurity content in the extraction result after extraction according to the extraction parameters. Specifically, the extraction fitness is an indicator for evaluating the extraction effect, which is jointly determined by the extraction purity and the impurity content. The purity weight and the impurity weight are used to adjust the importance of the extraction purity and the impurity content in the optimization process. Step S330, within the extraction parameter range, randomly generate multiple initial extraction parameters, combine the component feature information, perform extraction prediction, and obtain multiple initial extraction purity information and multiple initial impurity content information. Specifically, within the extraction parameter range, randomly generate multiple initial extraction parameter combinations, combine the component feature information, and perform extraction prediction on each initial extraction parameter combination through simulation experiments, empirical formulas or machine learning models to obtain the corresponding initial extraction purity information and initial impurity content information. Step S340, determine whether the multiple initial extraction purity information meets the extraction constraints. If not, delete the corresponding initial random parameters and randomly generate the initial extraction parameters that meet the extraction constraints again for completion. Specifically, check whether the initial extraction purity corresponding to each initial extraction parameter satisfies the extraction purity constraint. For the initial extraction parameters that do not meet the extraction purity constraint, delete and randomly generate new initial extraction parameters that meet the extraction purity constraint, and complete the deleted initial extraction parameters. Step S350, calculate and obtain multiple initial extraction fitnesses according to the extraction optimization function. Specifically, use the extraction optimization function to calculate the extraction fitness value of each initial extraction parameter that meets the extraction constraint. Step S360, select and obtain optimized extraction parameters, multiple common extraction parameters and the worst extraction parameters according to the multiple initial extraction fitnesses. Specifically, sort the initial extraction parameters according to the initial extraction fitness value, select the parameter with the highest initial extraction fitness value as the optimized extraction parameter, select the parameter with the lowest initial extraction fitness value as the worst extraction parameter, and use other parameters with medium initial extraction fitness values ​​as the common extraction parameters. Step S370, delete the worst extraction parameter, and randomly generate and complete the extraction parameters that meet the extraction constraint, adjust the multiple common extraction parameters with the optimized extraction parameter as the adjustment direction, and obtain multiple updated extraction parameters. Specifically, the worst extraction parameters are deleted, and the optimized extraction parameters are adjusted in the direction. Based on certain rules or strategies, such as gradient descent method, genetic algorithm, etc., the common extraction parameters are fine-tuned or regenerated to obtain multiple updated extraction parameters that meet the extraction constraints, and the deleted worst extraction parameters are supplemented. Step S380, multiple updated extraction fitnesses are calculated, and the optimized extraction parameters, multiple common extraction parameters and the worst extraction parameters are updated. Specifically, the updated extraction parameters are used for extraction prediction, and the corresponding extraction fitness values ​​are calculated. According to the new extraction fitness values, the optimized extraction parameters, the common extraction parameters and the worst extraction parameters are updated.Step S390, continue to optimize until convergence, output the final optimized extraction parameters, and obtain the optimal extraction parameters. Specifically, repeat steps S370 and S380 until the preset convergence conditions (such as the number of iterations, the change in the extraction fitness value, etc.) are reached, and output the final optimized extraction parameter combination. During the optimization process, this implementation method continuously deletes the worst extraction parameters, adjusts the common extraction parameters, and introduces new parameter combinations to ensure that the optimization process is always moving towards the optimal solution. This dynamic adjustment method achieves the technical effect of making the optimization process more flexible and efficient.

[0038] In a possible implementation, extraction prediction is performed in combination with the component characteristic information, and step S330 further includes step S331, collecting a sample extraction parameter set, a sample extraction component characteristic information set, and a sample extraction purity information set and a sample impurity content information set obtained by post-extraction testing according to the historical extraction data of the target mixture. Specifically, the historical extraction data of the target mixture is retrieved from the database, and the relevant sample extraction parameter set (extraction parameters used in historical experiments, such as temperature, pressure, time, solvent type, etc.), the sample extraction component characteristic information set (information describing the characteristics of drug components, such as the type, content, molecular structure, etc. of drug components), and the corresponding sample extraction purity information set (purity data of the extract measured in historical experiments) and sample impurity content information set (content data of impurities in the extract measured in historical experiments) obtained by post-extraction testing are screened out from these data. Step S332, using the sample extraction parameter set and the sample extraction component characteristic information set as input, and using the sample extraction purity information set and the sample impurity content information set as output, to train an extraction predictor. Specifically, the collected sample data is cleaned, normalized or standardized, etc., to ensure the quality and consistency of the data. According to the characteristics and requirements of the problem, a suitable machine learning or deep learning model is selected as an extraction predictor. For example, support vector machine, random forest, neural network, etc. can be selected. The preprocessed sample extraction parameter set and the sample extraction component feature information set are used as input, and the sample extraction purity information set and the sample impurity content information set are used as output. The selected model is trained, and the prediction performance of the model is optimized by adjusting the parameters and hyperparameters of the model. Step S333, using the extraction predictor, the component feature information is combined with multiple initial extraction parameters to perform extraction prediction, and multiple initial extraction purity information and multiple initial impurity content information are obtained. Specifically, the feature information of the drug component that needs to be optimized is used as input, combined with multiple randomly generated initial extraction parameters, to form multiple groups of input data, and each group of input data is passed to the trained extraction predictor to obtain the corresponding predicted values ​​of the initial extraction purity information and the initial impurity content information. The prediction method based on historical data in this implementation makes full use of the existing experimental data, avoids experimental design from scratch, and achieves the technical effect of improving the accuracy and efficiency of extraction prediction.

[0039] Step S400, using the optimal extraction parameters and extractant, controlling the extraction and separation equipment to extract the target mixture, obtaining the extraction result, and detecting and obtaining the actual extraction purity information and the actual impurity content information. Specifically, according to the optimal extraction parameters, the control parameters of the extraction and separation equipment (special equipment for performing extraction operations, including components such as reactors, stirrers, and temperature control systems) are adjusted, including setting the extraction time, temperature, solvent dosage, stirring speed, etc., adding the extractant to the extraction and separation equipment, adding the target mixture to the extractant in a set proportion and manner, ensuring sufficient mixing, starting the extraction and separation equipment, performing the extraction operation according to the set optimal extraction parameters, monitoring the extraction process, and ensuring that all parameters operate stably within the set range. After the extraction is completed, the extract (the solution or solid substance containing the drug ingredients obtained after the extraction operation) is collected and transferred to an appropriate container after appropriate post-processing steps (such as filtration, concentration, etc.). The extraction results are tested for purity and impurity analysis using appropriate detection methods (such as HPLC, GC, MS, etc.), and the actual extraction purity information (purity information of the drug ingredients in the extract) and actual impurity content information (type and content information of the impurity components in the extract) detected are recorded.

[0040] Step S500, setting impurity constraints, according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraints, optimizing the washing parameters of the extraction results using the original solvent of the target mixture, and obtaining the optimal washing parameters, wherein the optimization is performed for the purpose of reducing the impurity content in the extraction results and reducing the loss of drug purity in the extraction results. Specifically, during the washing process, an upper limit of the impurity content, i.e., the impurity constraint, is set based on factors such as product quality requirements, safety or regulatory requirements to ensure product quality and safety. According to the actual extraction purity information and the actual impurity content information obtained, the distribution of drug components and impurities in the extraction results is analyzed to determine the main types and contents of impurities in the extraction results. The goal of washing is to reduce the impurity content in the extraction results while minimizing the loss of drug purity. Therefore, a washing goal is set, i.e., to minimize the loss of drug purity under the premise of meeting the impurity constraints. The original solvent of the target mixture is selected as the washing solvent because the original solvent has a good solubility for impurities. The washing parameters include the number of washes, washing time, washing temperature, the amount of washing solvent, etc. According to the actual extraction results and washing targets, use appropriate optimization algorithms (such as genetic algorithms, simulated annealing algorithms, etc.) to optimize the washing parameters. During the optimization process, it is necessary to ensure that the optimal washing parameters (the optimized washing parameter combination, including the amount of washing solvent, washing time, washing temperature, etc.) are obtained under the premise of meeting the extraction constraints (such as purity constraints) and washing targets. Use the optimized washing parameters to predict the extraction results, and obtain the purity and impurity content of the drug after washing. Compare the predicted results with the washing targets to evaluate whether the optimization results meet the requirements. If they do not meet the requirements, the optimization algorithm can be adjusted or the washing parameters can be modified to optimize again.

[0041] In one possible implementation, an impurity constraint is set, and the washing parameters for washing the extraction result using the original solvent of the target mixture are optimized according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraint, to obtain the optimal washing parameters. Step S500 further includes step S510, obtaining the effective impurity content threshold of the drug component, and taking the impurity content after washing to be less than the effective impurity content threshold as the impurity constraint. Specifically, the effective impurity content threshold for the drug component is obtained from a preset database or user input, that is, the upper limit of the impurity content allowed in the drug component. Exceeding this effective impurity content threshold may affect the efficacy or safety of the drug, and a condition is set that the impurity content after washing must be less than the effective impurity content threshold to ensure the quality and safety of the product. Step S520, with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result, a washing function is constructed as follows:

[0042] ;

[0043] Among them, WAF is washing fitness, and is the weight, S is the drug purity loss in the extraction result after washing, and Z is the residual impurity content in the extraction result after washing. Specifically, the washing fitness is an indicator for evaluating the effect of the washing process, and a weighted calculation is performed based on the drug purity loss and the residual impurity content. Step S530, according to the original solvent of the target mixture, obtain the washing parameter range, and randomly generate multiple initial washing parameters within the washing parameter range. Specifically, according to the original solvent characteristics of the target mixture, determine the reasonable range of washing parameters (such as washing time, temperature, solvent dosage, etc.), and randomly generate multiple initial washing parameter combinations within this range. Step S540, based on the multiple initial washing parameters, combined with the actual extraction purity information and the actual impurity content information, wash prediction is performed to obtain multiple predicted initial loss amounts and multiple predicted initial residual impurity contents. Specifically, using methods such as simulation experiments, the washing process is predicted based on the initial washing parameters, and the drug purity loss and the residual impurity content after washing are calculated according to the prediction results. Step S550, based on the multiple predicted initial loss amounts and the actual extraction purity information, multiple washing purity information is calculated and compensated, multiple compensated washing purity information is obtained, whether the extraction constraint is satisfied, and the multiple predicted initial residual impurity contents are compensated and whether the impurity constraint is satisfied, and multiple initial washing parameters are judged, deleted and supplemented. Specifically, based on the obtained multiple predicted initial loss amounts (i.e., the loss amount of drug purity after washing), combined with the obtained actual extraction purity information, the drug purity information after washing is calculated by subtracting the predicted initial loss amount from the actual extraction purity. Since there may be errors in the prediction, the actual purity after washing is different from the predicted value. Therefore, according to historical data or experience, the calculated washing purity information is appropriately compensated or adjusted to more accurately reflect the actual situation. The compensated washing purity information is compared with the preset extraction constraint. If the purity after washing is higher than or equal to the minimum purity requirement specified in the extraction constraint, it is considered that the washing parameter combination meets the extraction constraint; otherwise, subsequent processing is required. Similar to purity information, the actual impurity content after washing may be different from the predicted value due to prediction errors or unforeseen factors in actual operation. Therefore, the multiple predicted initial residual impurity contents are compensated or adjusted to more accurately reflect the actual situation. The compensated residual impurity content is compared with the effective impurity content threshold. If the impurity content after washing is lower than or equal to the threshold, it is considered that the washing parameter combination meets the impurity constraint; otherwise, subsequent processing is required.For the initial washing parameter combination that does not meet the extraction constraint or impurity constraint, it is deleted from the candidate parameter set. After deleting the initial washing parameters that do not meet the conditions, the parameter space will be sparse or incomplete. At this time, new parameter combinations are added in the parameter space by interpolation, extrapolation or other reasonable methods to ensure the integrity and diversity of the parameter space and provide sufficient search space for subsequent optimization. Step S560, based on the washing function, multiple initial washing fitness of multiple initial washing parameters after completion are calculated, and selected and optimized until convergence, and the optimal washing parameters are output. Specifically, according to the washing function, the initial washing fitness of each initial washing parameter after completion is calculated, and the initial washing parameters are optimized using an optimization algorithm (such as genetic algorithm, particle swarm optimization, etc.) until convergence, and the optimal washing parameter combination is output. This implementation method, under the premise of meeting the product quality requirements, iteratively optimizes the washing parameters through data-driven and predictive methods, and achieves the technical effect of being able to find the optimal solution that can simultaneously reduce the impurity content and reduce the purity loss.

[0044] In a possible implementation, multiple washing purity information is calculated and compensated according to the multiple predicted initial loss amounts and the actual extraction purity information. Step S550 further includes step S551, calculating multiple washing purity information according to the multiple predicted initial loss amounts and the actual extraction purity information. Specifically, according to the obtained multiple predicted initial loss amounts (i.e., the loss amount of drug purity after washing), the corresponding predicted initial loss amount is subtracted from the actual extraction purity information to obtain the washing purity information corresponding to each initial washing parameter combination. Step S552, according to the actual extraction purity information and the optimal extraction purity information, the purity compensation coefficient is calculated, the multiple washing purity information is compensated, and multiple compensated washing purity information is obtained. Specifically, the difference between the actual extraction purity information and the optimal extraction purity information is compared, and the purity compensation coefficient is calculated. The purity compensation coefficient reflects the gap between the actual extraction purity and the optimal extraction purity, and is used to compensate the washing purity information. The purity compensation coefficient is used to adjust the multiple washing purity information to obtain multiple compensated washing purity information. Step S553, according to the actual impurity content information and the optimal impurity content information, calculate the impurity content compensation coefficient, compensate the multiple predicted initial residual impurity contents, and obtain multiple compensated predicted initial residual impurity contents. Specifically, compare the difference between the actual impurity content information and the optimal impurity content information, calculate the impurity content compensation coefficient, the impurity content compensation coefficient reflects the gap between the actual impurity content and the optimal impurity content, and is used to compensate for the predicted initial residual impurity content, and use the impurity content compensation coefficient to adjust the multiple predicted initial residual impurity contents to obtain multiple compensated predicted initial residual impurity contents. Step S554, determine whether the multiple compensated washing purity information meets the extraction constraint, and determine whether the multiple compensated predicted initial residual impurity contents meet the impurity constraint. Specifically, compare the multiple compensated washing purity information obtained in step S552 with the extraction constraint to determine whether it meets the preset purity requirement; compare the multiple compensated predicted initial residual impurity contents obtained in step S553 with the impurity constraint to determine whether it meets the preset impurity content requirement. According to the comparison result, the initial washing parameter combination that does not meet the constraint is deleted. This implementation method calculates the purity compensation coefficient based on the actual extraction purity information and the optimal extraction purity information, and calculates the impurity content compensation coefficient based on the actual impurity content information and the optimal impurity content information, thereby achieving the technical effect of accurately compensating for the washing purity information and the predicted initial residual impurity content, achieving the technical effect of making the compensation result closer to the actual situation, thereby improving the accuracy of the optimization process.

[0045] Step S600, using the optimal washing parameters, control the extraction and separation equipment to wash the extraction results to obtain the final extraction results. Specifically, confirm that the extraction and separation equipment is ready for the washing operation and has been set accordingly according to the optimal washing parameters, prepare sufficient washing solvent (i.e., the original solvent of the target mixture) to ensure continuous supply during the washing process, gradually add the washing solvent to the extraction and separation equipment according to the amount set in the optimal washing parameters, ensure that the washing solvent is in full contact with the extraction results, so as to effectively remove the impurities therein, control the operation of the extraction and separation equipment according to the washing time and temperature requirements in the optimal washing parameters, and regularly detect the changes in the impurity content and drug purity in the washing solvent during the washing process. If it is found that the impurity content is not significantly reduced or the drug purity loss is too large, adjust the washing parameters in time or take other remedial measures. When the washing process is completed, separate the washing liquid from the extraction result, and perform purity detection and impurity analysis on the final extraction result (the final product obtained after extraction and washing, with high purity and low impurity content) to ensure that the product quality requirements are met. The embodiments of the present application adopt technical means such as component detection of the target mixture to be extracted and separated, setting extraction constraints, optimizing extraction parameters, using optimal extraction parameters and extractants for extraction operations, setting impurity constraints, optimizing washing parameters, etc., to achieve the technical effect of optimizing operations according to the specific component characteristics of the target mixture, while achieving the requirements of high extraction purity and low impurity content, thereby improving the quality of drugs.

[0046] In the above, refer to Figure 1 A control optimization method for a pharmaceutical extraction and separation device according to an embodiment of the present invention is described in detail. Figure 2 A control optimization system for a pharmaceutical extraction and separation device according to an embodiment of the present invention is described.

[0047] A control optimization system for a pharmaceutical extraction and separation device according to an embodiment of the present invention is used to solve the technical problem that the existing extraction and separation device control lacks pertinence, is difficult to simultaneously meet the requirements of high extraction purity and low impurity content, and leads to poor extraction effect, so as to achieve the technical effect of optimizing the operation according to the specific component characteristics of the target mixture, and simultaneously achieve the requirements of high extraction purity and low impurity content, thereby improving the quality of the drug. A control optimization system for a pharmaceutical extraction and separation device includes: a component detection module 10, an extraction constraint setting module 20, an extraction parameter optimization module 30, an extraction module 40, a washing parameter optimization module 50, and a washing module 60.

[0048] The component detection module 10 is used to perform component detection on the target mixture to be extracted and separated to obtain component characteristic information, wherein the target mixture includes drug components and separation components; the extraction constraint setting module 20 is used to set extraction constraints according to the extraction requirements of the drug components, wherein the extraction constraints include purity constraints; the extraction parameter optimization module 30 is used to select an extractant according to the component characteristic information, optimize the extraction parameters in combination with the extraction constraints, obtain the optimal extraction parameters, and predict the optimal extraction purity information and the optimal impurity content information for extraction under the optimal extraction parameters, wherein the extraction parameters are optimized for the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant; the extraction module 40 is used to use the optimal extraction parameters and extractant to control the extraction and separation equipment to extract the target mixture, obtain the extraction result, and detect and obtain the actual extraction purity information and the actual impurity content information; the washing parameter optimization module 50 is used to set impurity constraints, and according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraints, optimize the washing parameters for washing the extraction result using the original solvent of the target mixture to obtain the optimal washing parameters, wherein the optimization is performed with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result; the washing module 60 is used to use the optimal washing parameters to control the extraction and separation equipment to wash the extraction result to obtain the final extraction result.

[0049] The specific configuration of the component detection module 10 will be described in detail below. As described above, the component detection module 10 may further include: a random sampling unit for randomly sampling the target mixture to be extracted and separated; and a sample detection unit for detecting the randomly sampled samples and calculating the drug component content and the separation component content as component characteristic information.

[0050] The specific configuration of the extraction constraint setting module 20 will be described in detail below. As described above, according to the extraction requirements of the drug component, the extraction constraint is set, and the extraction constraint setting module 20 may further include: an effective purity threshold acquisition unit for acquiring an effective purity threshold of the drug component; a purity constraint construction unit for taking the purity of the drug component after extraction as greater than or equal to the effective purity threshold as a purity constraint to construct the extraction constraint.

[0051] The specific configuration of the extraction parameter optimization module 30 will be described in detail below. As described above, according to the component characteristic information, the extraction agent is selected, and the extraction parameters are optimized in combination with the extraction constraints to obtain the optimal extraction parameters. The extraction parameter optimization module 30 may further include: an extraction agent selection unit for selecting an extraction agent according to the component characteristic information and obtaining an extraction parameter range; an extraction optimization function construction unit for constructing an extraction optimization function for the purpose of improving the extraction purity of the drug component and reducing the impurity content in the extraction agent, as shown in the following formula: ; Among them, EXF is the extraction fitness, and The sum of is 1, which are the purity weight and the impurity weight respectively, C is the extraction purity of the drug component in the extraction result after extraction according to the extraction parameters, and K is the impurity content in the extraction result after extraction according to the extraction parameters; the extraction prediction unit is used to randomly generate multiple initial extraction parameters within the extraction parameter range, and perform extraction prediction in combination with the component feature information to obtain multiple initial extraction purity information and multiple initial impurity content information; the judgment unit is used to judge whether the multiple initial extraction purity information meets the extraction constraints, if not, delete the corresponding initial random parameters and randomly generate the initial extraction parameters that meet the extraction constraints again for completion; the initial extraction fitness acquisition unit is used to obtain the initial extraction fitness information according to the extraction optimization function number, and calculate to obtain multiple initial extraction fitnesses; the selection unit is used to select and obtain optimized extraction parameters, multiple common extraction parameters and the worst extraction parameters according to the multiple initial extraction fitnesses; the common extraction parameter adjustment unit is used to delete the worst extraction parameters, and randomly generate extraction parameters that meet the extraction constraints to complete them, and adjust the multiple common extraction parameters with the optimized extraction parameters as the adjustment direction to obtain multiple updated extraction parameters; the updating unit is used to calculate to obtain multiple updated extraction fitnesses, and update the optimized extraction parameters, multiple common extraction parameters and the worst extraction parameters; the optimization convergence unit is used to continue to optimize until convergence, output the final optimized extraction parameters, and obtain the optimal extraction parameters.

[0052] Wherein, extraction prediction is performed in combination with the component characteristic information, and the extraction prediction unit may further include: a sample set acquisition subunit is used to collect a sample extraction parameter set, a sample extraction component characteristic information set, and a sample extraction purity information set and a sample impurity content information set obtained by testing after extraction according to historical extraction data of the target mixture; an extraction predictor training subunit is used to use the sample extraction parameter set and the sample extraction component characteristic information set as input, and use the sample extraction purity information set and the sample impurity content information set as output to train an extraction predictor; an extraction prediction subunit is used to use the extraction predictor to perform extraction prediction on the component characteristic information in combination with multiple initial extraction parameters, and obtain multiple initial extraction purity information and multiple initial impurity content information.

[0053] The specific configuration of the washing parameter optimization module 50 will be described in detail below. As described above, the impurity constraint is set, and the washing parameters for washing the extraction result using the original solvent of the target mixture are optimized according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraint, to obtain the optimal washing parameters. The washing parameter optimization module 50 may further include: an impurity constraint setting unit for obtaining the effective impurity content threshold of the drug component, and taking the impurity content after washing less than the effective impurity content threshold as the impurity constraint; a washing function construction unit for constructing a washing function for the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result, as shown in the following formula: ; Where WAF is washing fitness, and is the weight, S is the drug purity loss in the extraction result after washing, and Z is the residual impurity content in the extraction result after washing; the initial washing parameter random generation unit is used to obtain the washing parameter range according to the original solvent of the target mixture, and randomly generate multiple initial washing parameters within the washing parameter range; the washing prediction unit is used to perform washing prediction based on the multiple initial washing parameters in combination with the actual extraction purity information and the actual impurity content information, and obtain multiple predicted initial loss amounts and multiple predicted initial residual impurity contents; the discrimination unit is used to calculate and compensate for multiple washing purity information based on the multiple predicted initial loss amounts and the actual extraction purity information, obtain multiple compensated washing purity information, judge whether the extraction constraint is met, and compensate for the multiple predicted initial residual impurity contents and judge whether the impurity constraint is met, and judge, delete and complete multiple initial washing parameters; the optimal washing parameter output unit is used to calculate multiple initial washing fitness of the multiple initial washing parameters after completion based on the washing function, select and optimize until convergence, and output the optimal washing parameters.

[0054] Among them, according to the multiple predicted initial loss amounts and the actual extraction purity information, multiple washing purity information is calculated and compensated. The judgment unit may further include: a washing purity information acquisition subunit is used to calculate and obtain multiple washing purity information according to the multiple predicted initial loss amounts and the actual extraction purity information; a compensation subunit is used to calculate and obtain a purity compensation coefficient according to the actual extraction purity information and the optimal extraction purity information, compensate for the multiple washing purity information, and obtain multiple compensated washing purity information; according to the actual impurity content information and the optimal impurity content information, calculate and obtain an impurity content compensation coefficient, compensate for the multiple predicted initial residual impurity contents, and obtain multiple compensated predicted initial residual impurity contents; the judgment subunit is used to judge whether the multiple compensated washing purity information meets the extraction constraints, and judge whether the multiple compensated predicted initial residual impurity contents meet the impurity constraints.

[0055] A control optimization system for a pharmaceutical extraction and separation device provided in an embodiment of the present invention can execute a control optimization method for a pharmaceutical extraction and separation device provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

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

[0057] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A control optimization method for pharmaceutical extraction and separation equipment, characterized in that: The method comprises: Performing component detection on a target mixture to be extracted and separated to obtain component characteristic information, wherein the target mixture includes drug components and separation components; According to the extraction requirements of the drug components, extraction constraints are set, wherein the extraction constraints include purity constraints; According to the component characteristic information, an extractant is selected, and in combination with the extraction constraints, extraction parameters are optimized to obtain optimal extraction parameters, and optimal extraction purity information and optimal impurity content information for extraction under the optimal extraction parameters are predicted and obtained, wherein the extraction parameters are optimized for the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant; Using the optimal extraction parameters and extractant, controlling the extraction and separation equipment to extract the target mixture, obtaining extraction results, and detecting and obtaining actual extraction purity information and actual impurity content information; Setting impurity constraints, optimizing washing parameters for washing the extraction result using the original solvent of the target mixture according to the actual extraction purity information and the actual impurity content information, and the extraction constraints, to obtain optimal washing parameters, wherein the optimization is performed with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result; Using the optimal washing parameters, controlling the extraction and separation equipment to wash the extraction result to obtain the final extraction result; According to the extraction requirements of the drug components, extraction constraints are set, including: Obtaining an effective purity threshold of the drug component; The purity of the drug component after extraction is greater than or equal to the effective purity threshold as a purity constraint to construct the extraction constraint; According to the component characteristic information, an extractant is selected, and in combination with the extraction constraints, extraction parameters are optimized to obtain optimal extraction parameters, including: According to the component characteristic information, an extraction agent is selected and an extraction parameter range is obtained; In order to improve the extraction purity of drug components and reduce the impurity content in the extractant, an extraction optimization function is constructed as follows: ； Among them, EXF is the extraction fitness, and The sum of is 1, which are the purity weight and impurity weight respectively, C is the extraction purity of the drug component in the extraction result after extraction according to the extraction parameters, and K is the impurity content in the extraction result after extraction according to the extraction parameters; In the extraction parameter range, a plurality of initial extraction parameters are randomly generated, and extraction prediction is performed in combination with the component characteristic information to obtain a plurality of initial extraction purity information and a plurality of initial impurity content information; Determine whether the multiple initial extraction purity information meets the extraction constraints, and if not, delete the corresponding initial random parameters and randomly generate the initial extraction parameters that meet the extraction constraints to complete the task; According to the extraction optimization function, multiple initial extraction fitness values ​​are calculated; According to a plurality of initial extraction fitnesses, an optimized extraction parameter, a plurality of common extraction parameters and a worst extraction parameter are selected; The worst extraction parameter is deleted, and an extraction parameter satisfying the extraction constraint is randomly generated to complete the extraction parameter, and the plurality of common extraction parameters are adjusted with the optimized extraction parameter as the adjustment direction to obtain a plurality of updated extraction parameters; Calculate and obtain multiple updated extraction fitnesses, and update the optimized extraction parameters, multiple common extraction parameters, and the worst extraction parameters; Continue to optimize until convergence, output the final optimized extraction parameters, and obtain the optimal extraction parameters; Combined with the component characteristic information, extraction prediction is performed, including: According to the historical extraction data of the target mixture, a sample extraction parameter set, a sample extraction component characteristic information set, and a sample extraction purity information set and a sample impurity content information set obtained by post-extraction testing are collected; Using the sample extraction parameter set and the sample extraction component characteristic information set as input, and using the sample extraction purity information set and the sample impurity content information set as output, to train an extraction predictor; Using the extraction predictor, the component characteristic information is combined with a plurality of initial extraction parameters to perform extraction prediction, and a plurality of initial extraction purity information and a plurality of initial impurity content information are obtained; Setting impurity constraints, optimizing washing parameters for washing the extraction result with the original solvent of the target mixture according to the actual extraction purity information and the actual impurity content information, and the extraction constraints, and obtaining optimal washing parameters, including: Obtaining an effective impurity content threshold of the drug component, and taking the impurity content after washing to be less than the effective impurity content threshold as the impurity constraint; In order to reduce the impurity content in the extraction result and reduce the loss of drug purity in the extraction result, a washing function is constructed as follows: ； Among them, WAF is washing fitness, and is the weight, S is the drug purity loss in the extraction result after washing, and Z is the remaining impurity content in the extraction result after washing; According to the original solvent of the target mixture, a washing parameter range is obtained, and a plurality of initial washing parameters are randomly generated within the washing parameter range; Based on the multiple initial washing parameters, combined with the actual extraction purity information and the actual impurity content information, washing prediction is performed to obtain multiple predicted initial loss amounts and multiple predicted initial residual impurity contents; According to the multiple predicted initial loss amounts and the actual extraction purity information, multiple washing purity information is calculated and compensated to obtain multiple compensated washing purity information, and whether the extraction constraint is satisfied is determined; and the multiple predicted initial residual impurity contents are compensated and whether the impurity constraint is satisfied, and multiple initial washing parameters are judged, deleted, and supplemented; Based on the washing function, multiple initial washing fitness of multiple initial washing parameters after completion are calculated, and selected and optimized until convergence, and the optimal washing parameters are output.

2. The control optimization method of a pharmaceutical extraction and separation equipment according to claim 1, characterized in that: Component testing of the target mixture to be extracted and separated includes: Random sampling of the target mixture to be extracted and separated; The randomly sampled samples are tested, and the drug component content and the separated component content are calculated as the component characteristic information.

3. The control optimization method of a pharmaceutical extraction and separation equipment according to claim 1, characterized in that: According to the plurality of predicted initial loss amounts and the actual extraction purity information, a plurality of washing purity information is calculated and compensated, including: Calculating and obtaining a plurality of washing purity information according to the plurality of predicted initial loss amounts and the actual extraction purity information; According to the actual extraction purity information and the optimal extraction purity information, a purity compensation coefficient is calculated, and the plurality of washing purity information are compensated to obtain a plurality of compensated washing purity information; According to the actual impurity content information and the optimal impurity content information, an impurity content compensation coefficient is calculated to compensate the multiple predicted initial residual impurity contents to obtain multiple compensated predicted initial residual impurity contents; It is determined whether the plurality of compensated washing purity information satisfies the extraction constraint, and it is determined whether the plurality of compensated predicted initial remaining impurity contents satisfies the impurity constraint.

4. A control optimization system for pharmaceutical extraction and separation equipment, characterized in that: The system is used to implement the control optimization method of a pharmaceutical extraction and separation device according to any one of claims 1 to 3, and the system comprises: A component detection module, which is used to detect components of a target mixture to be extracted and separated to obtain component characteristic information, wherein the target mixture includes drug components and separation components; An extraction constraint setting module, the extraction constraint setting module is used to set extraction constraints according to the extraction requirements of the drug components, wherein the extraction constraints include purity constraints; An extraction parameter optimization module, wherein the extraction parameter optimization module is used to select an extractant according to the component characteristic information, optimize the extraction parameters in combination with the extraction constraints, obtain optimal extraction parameters, and predict and obtain optimal extraction purity information and optimal impurity content information for extraction under the optimal extraction parameters, wherein the extraction parameters are optimized for the purpose of improving the extraction purity of the drug components and reducing the impurity content in the extractant; An extraction module, the extraction module is used to use the optimal extraction parameters and extraction agent to control the extraction and separation equipment to extract the target mixture, obtain the extraction result, and detect and obtain actual extraction purity information and actual impurity content information; A washing parameter optimization module, the washing parameter optimization module is used to set impurity constraints, and optimize the washing parameters of washing the extraction result using the original solvent of the target mixture according to the actual extraction purity information and the actual impurity content information, as well as the extraction constraints, to obtain the optimal washing parameters, wherein the optimization is performed with the purpose of reducing the impurity content in the extraction result and reducing the loss of drug purity in the extraction result; A washing module is used to adopt the optimal washing parameters to control the extraction and separation equipment to wash the extraction result to obtain a final extraction result.

Images