Method and control system for extracting active ingredients of artemisia apiacea through enzymolysis-gradient combination

Abstract

The invention discloses a method and a control system for extracting active ingredients of sweet wormwood by enzymolysis-gradient combination, and particularly relates to the technical field of plant extraction, a high-sensitivity sensor is used for collecting key parameters in enzymolysis reaction, key process parameter characteristics are extracted from high-dimensional data through principal component analysis and a clustering algorithm, and the active ingredients of the sweet wormwood are extracted; establishing a reaction mode library of different extraction risk scores, and identifying an abnormal reaction mode causing artemisinin extraction; evaluating and extracting a risk score in real time by combining real-time and historical data based on a machine learning model, and triggering early warning when the risk exceeds a preset threshold value; the intelligent feedback and prevention and control module applies a control algorithm, automatically adjusts reaction conditions according to risk scores, and forms closed-loop feedback; the concentration and efficiency of artemisinin are monitored in real time through an online high performance liquid chromatography or mass spectrometry system, a verification result is fed back to the extraction risk prediction model, and the system performance is continuously optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of plant extraction, and more specifically to a method for extracting active ingredients of Artemisia annua by enzymatic hydrolysis-gradient combination and a control system. Background Art

[0002] The extraction process of artemisinin usually adopts the method of enzymatic hydrolysis-gradient combination to improve the extraction efficiency and product efficiency. However, the key process parameters in the enzymatic hydrolysis reaction, such as temperature, pH value, enzyme concentration and reaction time, have a significant impact on the extraction effect of artemisinin. Traditional extraction methods rely on manual monitoring and manual adjustment of process parameters, which makes it difficult to achieve precise control of complex reaction conditions, resulting in instability in artemisinin yield and efficiency.

[0003] In the actual production process, slight fluctuations in the enzymatic reaction conditions may trigger the enzymatic hydrolysis of artemisinin, thereby affecting the quality and efficacy of the final product. In the existing technology, there is a lack of effective real-time monitoring and dynamic optimization methods, which makes it difficult to timely identify and adjust harmful combinations of reaction conditions, resulting in extraction risks that are difficult to predict and control. In addition, traditional methods have lags in data processing and risk assessment, and cannot meet the needs of efficient extraction and stable production. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and control system for extracting active ingredients of Artemisia annua by enzymatic hydrolysis-gradient combination, which realizes accurate monitoring and dynamic adjustment of enzymatic hydrolysis reaction conditions through real-time data collection, risk prediction and early warning, as well as intelligent feedback and prevention and control, so as to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination, comprising the following steps: Step 1: Pretreatment step: preparing clean and dry Artemisia annua into an Artemisia annua sample with uniform particle size; Step 2, enzymolysis step: mixing the pretreated Artemisia annua sample with the enzyme solution to obtain a mixture reaction system, adjusting the pH of the mixture reaction system to 4.5-4.7, controlling the temperature to 40-45° C., performing enzymolysis reaction for 2 hours under stirring conditions, and outputting an enzymolysis mixture; Step 3, gradient extraction step: after the enzymatic hydrolysis reaction is completed, the enzymatic hydrolysis mixture is extracted using a gradient solvent system to output a final extract, wherein the final extract includes a first crude extract, a second crude extract and a third crude extract; Step 4: Concentration and purification: After the final extract is initially treated by centrifugation and filtration, the extract is concentrated by vacuum concentration or reduced pressure evaporation technology, and then purified by column chromatography or liquid chromatography.

[0006] Preferably, the enzymatic hydrolysis step comprises: adding 100 kg of the Artemisia annua sample to an acetic acid-sodium acetate aqueous solution with a pH value of 5.5 according to a raw material weight ratio of 1:4, and adding the enzyme solution, wherein the amount of the enzyme solution added is 1% of the weight of the Artemisia annua sample; the enzyme solution is a combination of cellulase, hemicellulase and pectinase.

[0007] Preferably, the enzymatic hydrolysis mixture flows into the first extraction chamber from the top and contacts with the solvent A flowing in the countercurrent direction, and outputs the first crude extract and the remaining enzymatic hydrolysis mixture; the remaining enzymatic hydrolysis mixture flows from the first extraction chamber into the second extraction chamber and contacts with the solvent B, and outputs the second crude extract and the remaining enzymatic hydrolysis mixture; the remaining enzymatic hydrolysis mixture continues to flow into the third extraction chamber and contacts with the solvent C, and outputs the third crude extract.

[0008] Preferably, during the enzymatic hydrolysis reaction, the reaction progress is monitored in real time by a pH sensor and a temperature sensor, and the amount of acid and alkali solution added and the heating / cooling power are automatically adjusted by a PID controller to maintain the set enzymatic hydrolysis conditions.

[0009] Preferably, the solvent A is n-hexane or petroleum ether, the solvent B is ethyl acetate, and the solvent C is n-butanol; the solvents A, B, and C are combined in a volume ratio of 1:1:1 to form a gradient solvent system.

[0010] To achieve the above object, the present invention provides the following technical solution: a control system for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination, comprising: Real-time data acquisition and preprocessing module: install temperature sensor, pH sensor, dissolved oxygen sensor and online artemisinin concentration detection instrument in the enzymatic hydrolysis reactor; perform denoising, smoothing and standardization on the collected data; The feature extraction and pattern recognition module uses principal component analysis to extract key process parameter features that affect the enzymatic hydrolysis reaction; uses clustering algorithms to perform pattern recognition on the data during the enzymatic hydrolysis reaction, establishes a reaction pattern library with different risk levels, and identifies abnormal reaction patterns that lead to abnormal enzymatic hydrolysis yields; The extraction risk prediction and early warning module builds an extraction risk prediction model based on machine learning algorithms, predicts the extraction risk score under current reaction conditions based on real-time data and historical data, and triggers an early warning mechanism when the extraction risk score exceeds the preset threshold; The intelligent feedback and prevention and control module applies control algorithms to automatically adjust response conditions based on the extracted risk scores, executes adjustment instructions, and forms a closed-loop feedback loop; The result verification module monitors the yield of artemisinin in real time, verifies the effect of process parameter adjustment, and feeds back the verification results to the extraction risk prediction model, continuously optimizes the extraction risk prediction model parameters, and forms an iterative optimization cycle.

[0011] Preferably, the abnormal reaction mode includes: characteristic combinations of key process parameters, rapid fluctuations in reaction conditions, continuous deviations from long-term reaction conditions, and significant differences in raw material properties.

[0012] Preferably, the construction of the extraction risk prediction model includes the following steps: Acquire key process parameter data, raw material characteristic data and corresponding artemisinin yield data in the enzymatic hydrolysis reaction from the real-time data acquisition and preprocessing module to form a training data set; According to the artemisinin yield and enzymatic hydrolysis conditions, the collected key process parameter data were annotated to distinguish normal reaction data from high-risk enzymatic hydrolysis data; Use statistical analysis methods or feature selection algorithms to screen out process parameter features related to extraction risk prediction; Using the labeled training data set to train and extract the risk prediction model; The performance of the trained extraction risk prediction model is evaluated through the cross-validation method, and the parameters of the extraction risk prediction model are adjusted or a more appropriate algorithm is selected based on the evaluation results to optimize the prediction accuracy and generalization ability of the extraction risk prediction model.

[0013] Preferably, the extraction risk score under the current reaction conditions is predicted based on the real-time data and historical data, and the extraction risk score is obtained in the following manner: The current process parameter characteristics and raw material characteristic data output by the real-time data acquisition and preprocessing module are input into the extraction risk prediction model; the historical data pattern trained in the extraction risk prediction model is used in combination with the current real-time data to calculate the risk score; the input data is analyzed in multiple dimensions by the extraction risk prediction model, and the entropy weight method and Bayesian network are used to output the extraction risk score, which reflects the risk of abnormal artemisinin extraction under the current reaction conditions; The calculated extraction risk score is used as the basis for judging whether it is necessary to automatically adjust the process parameters through the intelligent feedback and prevention and control module to prevent abnormal artemisinin extraction, which includes abnormal extraction ratio and abnormal extraction efficiency.

[0014] Preferably, the system further comprises: The dynamic optimization module optimizes the reaction conditions in real time through an optimization algorithm, including the following steps: The optimization objective function is defined, and the factors of artemisinin yield, efficiency and energy consumption are comprehensively considered. The optimization objective function F is expressed as: ; Wherein, Y is the artemisinin yield, P is the artemisinin extraction efficiency, and E is the energy consumption. , , is the weight coefficient, which is set according to actual needs; The optimization algorithm is used to search for the best reaction parameter combination, and the optimization results are passed to the intelligent feedback and prevention and control module to implement the best reaction parameter combination.

[0015] Technical effects and advantages of the present invention:

[0016] (1) The present invention provides a method for extracting active ingredients from Artemisia annua by combining enzymatic hydrolysis and gradient extraction, wherein the enzymatic hydrolyzate is transported to a gradient extraction unit for gradient extraction, and different extraction solvents are used for graded extraction according to the polarity differences of the three components, namely, artemisinins, polyphenols and flavonoids; extracts rich in artemisinins, polyphenols and flavonoids are obtained respectively, thereby improving the comprehensive utilization rate of Artemisia annua resources; the two steps of enzymatic hydrolysis and gradient extraction are integrated together to realize continuous operation, shorten the process flow and improve production efficiency.

[0017] (2) The present invention provides a method for extracting active ingredients from Artemisia annua by combining enzymatic hydrolysis and gradient extraction. The method achieves efficient extraction of artemisinin through four steps of pretreatment, enzymatic hydrolysis, gradient extraction and concentration and purification. During the enzymatic hydrolysis reaction, real-time monitoring by a pH sensor and a temperature sensor is combined with automatic adjustment by a PID controller to maintain stable enzymatic hydrolysis conditions, thereby improving the controllability and repeatability of the process, effectively improving the production efficiency and economic benefits of artemisinin, and solving the problem of low enzymatic hydrolysis and extraction efficiency of artemisinin caused by fluctuations in enzymatic hydrolysis conditions in traditional extraction methods.

[0018] (3) The present invention provides a control system for extracting the active ingredients of Artemisia annua by enzymatic hydrolysis-gradient extraction. The extraction risk prediction model constructed by machine learning algorithm can evaluate the risk level under the current reaction conditions in real time. When the risk score exceeds the preset threshold, the system automatically triggers an early warning and adjusts the process parameters to avoid enzymatic hydrolysis of the active ingredients. The intelligent feedback and prevention and control module uses advanced control algorithms such as fuzzy logic control and adaptive control to automatically optimize the reaction conditions, ensure the high yield and efficiency of artemisinin, realize automated control of the enzymatic hydrolysis and gradient extraction process, and improve production efficiency and product quality stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The present invention is a flow chart of the method for extracting active ingredients from Artemisia annua by using enzymatic hydrolysis-gradient combination.

[0020] Figure 2 The present invention is a structural block diagram of a control system for extracting active ingredients from Artemisia annua. DETAILED DESCRIPTION

[0021] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0022] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0023] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present application, its application, or uses.

[0024] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0025] Embodiment 1, the present invention provides as Figure 1 The method for extracting active ingredients of Artemisia annua by enzymatic hydrolysis-gradient combination is shown, comprising the following steps: Step 1: Pretreatment step: preparing clean and dry Artemisia annua into an Artemisia annua sample with uniform particle size; Explanation: Select freshly picked Artemisia annua and wash it quickly to remove dirt, impurities and other undesirable parts; chop the washed Artemisia annua at room temperature, and then dry it under mild conditions below 50°C until the moisture content drops below 10%; chopping and drying not only help reduce moisture and extend the shelf life of the raw materials, but also destroy some cell structures, provide a larger surface area for the subsequent enzymatic hydrolysis step, and improve the efficiency of enzymatic hydrolysis; Step 2, enzymatic hydrolysis step: mixing the pretreated Artemisia annua sample with the enzyme solution to obtain a mixture reaction system, adjusting the pH of the mixture reaction system to 4.5-4.7, controlling the temperature to 40-45° C., performing enzymatic hydrolysis reaction for 2 hours under stirring conditions, and outputting an enzymatic hydrolysis mixture; continuing for a certain period of time to decompose the cell wall and release the active ingredients; It needs to be further explained in the embodiment of the present invention that the enzymatic hydrolysis step includes: adding 100 kg of the Artemisia annua sample to an acetic acid-sodium acetate aqueous solution with a pH value of 5.5 according to a raw material weight ratio of 1:4, and adding the enzyme solution, wherein the amount of the enzyme solution added is 1% of the weight of the Artemisia annua sample; It is necessary to further explain in the embodiments of the present invention that during the enzymatic hydrolysis reaction, the reaction progress is monitored in real time by a pH sensor and a temperature sensor, and the amount of acid and alkali solution added and the heating / cooling power are automatically adjusted by a PID controller to maintain the set enzymatic hydrolysis conditions; It is necessary to further explain in the embodiments of the present invention that after the enzymatic hydrolysis reaction is completed, heating, pH adjustment, and adding inhibitors are adopted to inactivate the enzyme; for example, the temperature is raised to 65° C. and maintained for 15 minutes to inactivate the enzyme; It needs to be further explained in the embodiments of the present invention that the enzyme solution is a combination of cellulase, hemicellulase and pectinase; the enzyme solution is a mixture of cellulase, hemicellulase and pectinase in a specific ratio, wherein the specific ratio may be cellulase: hemicellulase: pectinase = 1:2:1; Step 3, gradient extraction step: after the enzymatic hydrolysis reaction is completed, the enzymatic hydrolysis mixture is extracted using a gradient solvent system, a final extract is output, and the solvent polarity is gradually changed to efficiently separate different active ingredients in Artemisia annua; the final extract includes a first crude extract, a second crude extract and a third crude extract; It is necessary to further explain in the embodiments of the present invention that, according to the polarity differences of the three components of artemisinins, polyphenols and flavonoids, different extraction solvents are used for graded extraction; extracts at different extraction stages are collected to obtain extracts rich in artemisinins, polyphenols and flavonoids respectively; In the gradient solvent system, the enzymatic hydrolysis mixture flows into the first extraction chamber from the top and contacts with the solvent A (n-hexane) flowing in the countercurrent direction. The n-hexane is used to extract the fat-soluble components of artemisinin, and the first crude extract and the remaining enzymatic hydrolysis mixture are output; the remaining enzymatic hydrolysis mixture flows from the first extraction chamber into the second extraction chamber and contacts with the solvent B (ethyl acetate). The ethyl acetate is used to extract polyphenols and some flavonoids, and the second crude extract and the remaining enzymatic hydrolysis mixture are output; the remaining enzymatic hydrolysis mixture continues to flow into the third extraction chamber and contacts with the solvent C (n-butanol), and the third crude extract is output, thereby realizing the gradient extraction of different types of components.

[0026] It is explained that each extraction chamber is equipped with a stirring device to ensure that the enzymatic hydrolysis mixture is fully mixed with solvents A, B, and C to improve the extraction efficiency; after multi-stage countercurrent extraction, the final extract flows out from the top of the unit, and the final extract contains the target ingredients of artemisinin, polyphenols, and flavonoids.

[0027] In a possible embodiment, petroleum ether is used instead of the n-hexane, and petroleum ether, ethyl acetate and n-butanol are combined in a volume ratio of 1:1:1 to form a gradient solvent system; wherein petroleum ether is used to extract artemisinins, ethyl acetate is used to extract polyphenols, and n-butanol is used to extract flavonoids; Step 4, concentration and purification step: After the final extract is subjected to preliminary treatment by centrifugation and filtration, the extract is concentrated by vacuum concentration or reduced pressure evaporation technology, and then purified by column chromatography and liquid chromatography to obtain high-efficiency Artemisia annua active ingredients, polyphenols and flavonoids active ingredients.

[0028] Explanation: The crystallization and purification of the crude extract in step 4 is based on the current advanced solution configuration in the industry. For ease of understanding, the concentration and purification steps include: The final extract is pumped into the concentration tank, and the vacuum pump is started to reduce the pressure in the concentration tank, so that the solvent boils and evaporates at a lower temperature; the evaporated solvent is condensed into liquid through the condenser and collected in the collection tank; the concentration process is monitored in real time by the temperature sensor and the pressure sensor, and the heating / cooling power and vacuum degree are automatically adjusted by the PID controller to maintain the optimal concentration conditions; when the concentrated liquid reaches the preset concentration, the heating / cooling and vacuum pump are stopped, and the concentrated liquid is discharged.

[0029] What needs to be further explained in the embodiments of the present invention is that the concentrated liquid is passed through the adsorption column, the target component is adsorbed on the resin, the impurities flow out with the water, and the impurities on the adsorption column are washed away with the eluent; the target component is eluted from the resin with the analytical solution and collected in the collection tank.

[0030] Example 2, see Figure 2 The structural block diagram of the control system for extracting the active ingredients of Artemisia annua; In the process of extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination, slight fluctuations in the enzymatic hydrolysis reaction conditions may lead to the enzymatic hydrolysis of active ingredients such as artemisinin, thereby affecting the yield of the final product. Traditional control systems are difficult to monitor and control these minor enzymatic hydrolysis reactions in real time, resulting in instability in product quality; Specific scenarios: In actual production, even in dried Artemisia annua raw materials, slight fluctuations in parameters such as temperature and pH during the enzymatic hydrolysis process may still trigger enzymatic hydrolysis of active ingredients. For example, when the temperature is slightly higher than the optimal range, artemisinin may undergo thermal enzymatic hydrolysis; when the pH value deviates from the optimal value, the activity and selectivity of the enzyme may be affected, leading to unnecessary side reactions. Therefore, a control system that can monitor and control the enzymatic hydrolysis of active ingredients in real time is needed to ensure the high efficiency and activity of artemisinin; A control system for extracting active ingredients of Artemisia annua by enzymatic hydrolysis-gradient combination, comprising: The real-time data acquisition and preprocessing module installs a temperature sensor, a pH sensor, a dissolved oxygen sensor and an online artemisinin concentration detection instrument (such as an online UV-Vis spectrometer or a near-infrared spectrometer) in the enzymatic hydrolysis reactor, performs preprocessing operations on the collected sensor data, and outputs the preprocessed data; Explanation: Ensure real-time collection and high-frequency updating of sensor data; collect key parameter data in enzymatic hydrolysis reactions in real time through high-sensitivity sensors; the collected raw data needs to be preprocessed, including denoising, smoothing and standardization, to improve data accuracy and consistency, and provide a reliable data basis for subsequent extraction risk prediction.

[0031] In the process of extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination, the conditions of the enzymatic hydrolysis reaction (such as temperature, pH value, enzyme concentration, reaction time, etc.) have an important influence on the extraction efficiency and product quality; however, certain combinations of conditions may lead to enzymatic hydrolysis of active ingredients (such as artemisinin), affecting the efficiency and efficacy of the final product; therefore, identifying and avoiding these harmful reaction condition patterns is crucial for optimizing the extraction process; The feature extraction and pattern recognition module uses principal component analysis or time series analysis methods to extract key process parameter features that affect enzymatic hydrolysis reactions from high-dimensional data, such as temperature change rate, pH fluctuation range, dissolved oxygen content change, etc.; clustering algorithms are used to perform pattern recognition on the data during the enzymatic hydrolysis process, establish a reaction pattern library with different extraction risk scores, and identify abnormal reaction patterns that lead to enzymatic hydrolysis; The pattern recognition refers to using a clustering algorithm to divide all reaction monitoring data into several clusters, each cluster representing a group of similar reaction condition combinations; analyzing the center point of each cluster to obtain the representative typical reaction conditions; evaluating the similarity of samples within the cluster to ensure the rationality of the clustering effect; Each cluster is associated with the reaction results (such as artemisinin yield, extraction efficiency) to identify abnormal reaction patterns that lead to enzymatic hydrolysis; the abnormal reaction pattern refers to the combination of process parameters that leads to a decrease in the yield of active ingredients (such as artemisinin); the extracted key features and identified abnormal reaction patterns are input into the enzymatic hydrolysis reaction prediction model for real-time risk assessment; Extraction risk prediction and early warning module, build extraction risk prediction model; extraction risk prediction model is based on real-time data and historical data to predict the extraction risk level under current reaction conditions; The explanation is that the constructed machine learning prediction model is used to conduct a real-time assessment of the extraction risk in the current enzymatic hydrolysis reaction based on the key feature data collected in real time. When the extraction risk is predicted to exceed the preset threshold, the early warning mechanism will be triggered, indicating the need for process adjustment.

[0032] Data exchange and connection: The extracted risk prediction results will be transmitted to the intelligent feedback and prevention and control module to guide the adjustment of real-time process parameters; Intelligent feedback and prevention and control module, which is used to adjust process parameters, apply control algorithms (such as fuzzy logic control, adaptive control or PID control algorithm), and automatically adjust reaction conditions (such as adjusting reaction parameters such as temperature, pH value, dissolved oxygen content, etc.) according to the extraction risk prediction results; Explanation: When the extraction risk prediction model detects a high risk, the intelligent control algorithm will automatically adjust the key parameters of the enzymatic reaction according to the preset control strategy. For example, the temperature can be appropriately lowered, the pH value can be adjusted, or the dissolved oxygen content can be increased to reduce the extraction risk of the active ingredient. The adjusted parameters will be immediately fed back to the enzymatic reactor to ensure the optimization of the reaction conditions.

[0033] It needs to be further explained in the embodiments of the present invention that the adjusted reaction conditions are again monitored in real time by sensors to form a closed-loop feedback, thereby ensuring the effectiveness of the process parameter adjustment and providing updated data for the next step of extraction risk assessment.

[0034] The result verification module monitors the concentration and efficiency of artemisinin in real time through online detection equipment (such as high-performance liquid chromatography or mass spectrometry) to verify the effect of process parameter adjustment; the verification results are fed back to the prediction model to continuously optimize the model parameters and improve the prediction accuracy; The verification results are used as new data input to further enrich and optimize the extraction risk prediction model, forming an optimization cycle of continuous iteration and improvement to ensure the stability and efficiency of the extraction process.

[0035] It needs to be further explained in the embodiment of the present invention that the construction of the extraction risk prediction model includes the following steps: The data collection step is to obtain the key process parameter data, raw material characteristic data and corresponding artemisinin yield data in the enzymatic hydrolysis reaction from the real-time data acquisition and preprocessing module to form a training data set; Data annotation step: annotate the collected key process parameter data according to artemisinin yield and enzymolysis conditions to distinguish normal reaction data from high-risk enzymolysis data; The feature selection step uses statistical analysis methods or feature selection algorithms to select process parameter features that are relevant to extraction risk prediction; The model training step uses the selected machine learning algorithm (such as support vector machine, random forest or long short-term memory network) to train and extract the risk prediction model using the labeled training data set; In the model validation and optimization step, the performance of the trained extraction risk prediction model is evaluated through the cross-validation method, and the parameters of the extraction risk prediction model are adjusted or a more appropriate algorithm is selected based on the evaluation results to optimize the prediction accuracy and generalization ability of the extraction risk prediction model.

[0036] The extraction risk prediction model includes an adaptive learning unit, which uses an online learning algorithm to continuously update and optimize the extraction risk prediction model so that the extraction risk prediction model can adapt to changes in the characteristics of different batches of raw materials and reaction conditions; the extraction risk prediction model is updated using an online learning algorithm (such as incremental learning or reinforcement learning); taking incremental learning as an example, the update formula is as follows: ; in, To extract risk prediction model parameters, is the learning rate, L is the loss function, , are the newly input data and labels respectively; Extract risk prediction model parameters for history, Extract risk prediction model parameters for the updated ones; Regularly evaluate the performance of the updated extraction risk prediction model to ensure that the prediction ability does not decline, and use the increase in information entropy or mutual information to measure the prediction improvement of the extraction risk prediction model.

[0037] It is necessary to further explain in the embodiment of the present invention that the extraction risk score under the current reaction conditions is predicted based on the real-time data and the historical data, and the extraction risk score is obtained in the following manner: Input the current process parameter characteristics and raw material characteristic data output by the real-time data acquisition and preprocessing module into the extraction risk prediction model; Historical data reference: using historical data patterns trained in the risk prediction model and combining them with current real-time data to calculate risk scores; Risk score calculation: Through the extraction risk prediction model, the input data is analyzed in multiple dimensions, and a quantitative extraction risk score is output using the entropy weight method and Bayesian network. The extraction risk score reflects the probability of abnormal artemisinin extraction under the current reaction conditions; The calculated extraction risk score is used as the basis for judging whether it is necessary to automatically adjust the process parameters through the intelligent feedback and prevention and control module to prevent abnormal artemisinin extraction and ensure the stability and efficiency of the extraction process.

[0038] It is necessary to further explain in the embodiment of the present invention that the control system for extracting the active components of Artemisia annua by enzymatic hydrolysis-gradient combination also includes: The dynamic optimization module optimizes the reaction conditions in real time through an optimization algorithm, including the following steps: The optimization objective function is defined, and the factors of artemisinin yield, efficiency and energy consumption are comprehensively considered. The optimization objective function F is expressed as: ; Wherein, Y is the artemisinin yield, P is the artemisinin extraction efficiency, and E is the energy consumption. , , is the weight coefficient, which is set according to actual needs; Use optimization algorithms to search for the best reaction parameter combination based on the current extraction risk score and reaction conditions, pass the optimization results to the intelligent feedback and prevention and control module, and implement the best reaction parameter combination; Taking particle swarm optimization as an example, each particle represents a specific combination of process parameters, i.e., a set of possible reaction conditions; Assume that there are three key process parameters that need to be optimized: temperature (T), pH value (P), and dissolved oxygen (S); Then, the position of each particle can be represented as a three-dimensional vector, such as (T, P, S); each dimension corresponds to a process parameter that needs to be optimized; The particle's speed in each dimension determines how much and in what direction it moves in that dimension; For the temperature (T) dimension, the velocity vi1 represents the moving speed of particle i in the temperature dimension, a positive value indicates an increase in temperature, and a negative value indicates a decrease in temperature; The position of the particle in each dimension represents the current process parameter value, and its update formula is as follows: ; ; in, is the velocity of particle i in dimension j, is the position of particle i in dimension j, is the best historical position of particle i in dimension j, is the global optimal position, w is the inertia weight, , is the learning factor, , Is a random number.

[0039] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination, characterized in that: The following steps are involved: Step 1: Pretreatment step: preparing clean and dry Artemisia annua into an Artemisia annua sample with uniform particle size; Step 2, enzymolysis step: mixing the pretreated Artemisia annua sample with the enzyme solution to obtain a mixture reaction system, adjusting the pH of the mixture reaction system to 4.5-4.7, controlling the temperature to 40-45° C., performing enzymolysis reaction for 2 hours under stirring conditions, and outputting an enzymolysis mixture; Step 3, gradient extraction step: after the enzymatic hydrolysis reaction is completed, the enzymatic hydrolysis mixture is extracted using a gradient solvent system to output a final extract, wherein the final extract includes a first crude extract, a second crude extract and a third crude extract; Step 4: Concentration and purification: After the final extract is initially treated by centrifugation and filtration, the extract is concentrated by vacuum concentration or reduced pressure evaporation technology, and then purified by column chromatography or liquid chromatography.

2. The method for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 1, characterized in that: The enzymatic hydrolysis step includes: adding 100 kg of the Artemisia annua sample to an acetic acid-sodium acetate aqueous solution with a pH value of 5.5 according to a raw material weight ratio of 1:4, and adding the enzyme solution, wherein the added amount of the enzyme solution is 1% of the weight of the Artemisia annua sample; the enzyme solution is a combination of cellulase, hemicellulase and pectinase.

3. The method for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 1, characterized in that: The enzymolysis mixed solution flows into the first extraction chamber from the top and contacts with the solvent A flowing in the countercurrent, and outputs the first crude extract and the remaining enzymolysis mixed solution; the remaining enzymolysis mixed solution flows from the first extraction chamber into the second extraction chamber and contacts with the solvent B, and outputs the second crude extract and the remaining enzymolysis mixed solution; the remaining enzymolysis mixed solution continues to flow into the third extraction chamber and contacts with the solvent C, and outputs the third crude extract.

4. The method for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 1, characterized in that: During the enzymatic hydrolysis reaction, the pH sensor and temperature sensor are used to monitor the reaction progress in real time, and the PID controller is used to automatically adjust the amount of acid and alkali solution added and the heating / cooling power to maintain the set enzymatic hydrolysis conditions.

5. The method for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 3, characterized in that: The solvent A is n-hexane or petroleum ether, the solvent B is ethyl acetate, and the solvent C is n-butanol; the solvents A, B, and C are combined in a volume ratio of 1:1:1 to form a gradient solvent system.

6. A control system for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination, suitable for the method described in any one of claims 1 to 5, characterized in that: include: Real-time data acquisition and preprocessing module, installing temperature sensor, pH sensor, dissolved oxygen sensor and online artemisinin concentration detection instrument in the enzymatic hydrolysis reactor; The feature extraction and pattern recognition module uses principal component analysis to extract the key process parameter characteristics that affect the enzymatic hydrolysis reaction; Clustering algorithms are used to perform pattern recognition on the data during the enzymatic hydrolysis reaction, and reaction pattern libraries with different risk levels are established to identify abnormal reaction patterns that lead to abnormal enzymatic hydrolysis yields. The extraction risk prediction and early warning module builds an extraction risk prediction model based on machine learning algorithms, predicts the extraction risk score under current reaction conditions based on real-time data and historical data, and triggers an early warning mechanism when the extraction risk score exceeds the preset threshold; Intelligent feedback and prevention and control module, which automatically adjusts the response conditions according to the extracted risk score, executes the adjustment instructions, and forms a closed-loop feedback; The result verification module monitors the yield of artemisinin in real time, verifies the effect of process parameter adjustment, and feeds back the verification results to the extraction risk prediction model, continuously optimizes the extraction risk prediction model parameters, and forms an iterative optimization cycle.

7. A control system for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 6, characterized in that: The abnormal reaction modes include: characteristic combinations of key process parameters, rapid fluctuations in reaction conditions, continuous deviations from long-term reaction conditions, and significant differences in raw material properties.

8. A control system for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 6, characterized in that: The construction of the extraction risk prediction model includes the following steps: Acquire key process parameter data, raw material characteristic data and corresponding artemisinin yield data in the enzymatic hydrolysis reaction from the real-time data acquisition and preprocessing module to form a training data set; According to the artemisinin yield and enzymatic hydrolysis conditions, the collected key process parameter data were annotated to distinguish normal reaction data from high-risk enzymatic hydrolysis data; Use statistical analysis methods or feature selection algorithms to screen out process parameter features related to extraction risk prediction; Using the labeled training data set to train and extract the risk prediction model; The performance of the trained extraction risk prediction model is evaluated through the cross-validation method, and the parameters of the extraction risk prediction model are adjusted or a more appropriate algorithm is selected based on the evaluation results to optimize the prediction accuracy and generalization ability of the extraction risk prediction model.

9. A control system for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 6, characterized in that: The extraction risk score under the current reaction conditions is predicted based on real-time data and historical data. The extraction risk score is obtained as follows: The current process parameter characteristics and raw material characteristic data output by the real-time data acquisition and preprocessing module are input into the extraction risk prediction model; the historical data pattern trained in the extraction risk prediction model is used in combination with the current real-time data to calculate the risk score; the input data is analyzed in multiple dimensions by the extraction risk prediction model, and the entropy weight method and Bayesian network are used to output the extraction risk score, which reflects the risk of abnormal artemisinin extraction under the current reaction conditions; The calculated extraction risk score is used as the basis for determining whether it is necessary to automatically adjust the process parameters through the intelligent feedback and prevention and control module to prevent abnormal extraction of artemisinin.

10. A control system for extracting active ingredients from Artemisia annua by enzymatic hydrolysis-gradient combination according to claim 6, characterized in that: Also includes: The dynamic optimization module optimizes the reaction conditions in real time through an optimization algorithm, including the following steps: The optimization objective function is defined, and the factors of artemisinin yield, efficiency and energy consumption are comprehensively considered. The optimization objective function F is expressed as: ； Wherein, Y is the artemisinin yield, P is the artemisinin extraction efficiency, and E is the energy consumption. , , is the weight coefficient, which is set according to actual needs; The optimization algorithm is used to search for the best reaction parameter combination, and the optimization results are passed to the intelligent feedback and prevention and control module to implement the best reaction parameter combination.

Images