Intelligent control method for traditional Chinese medicine concentration process

Abstract

The invention discloses an intelligent control method for a traditional Chinese medicine concentration process, and relates to the technical field of traditional Chinese medicine production and processing. Gas-phase and liquid-phase multi-dimensional data in the whole traditional Chinese medicine concentration process are synchronously acquired and structurally processed, and the activity degree of volatile gas behaviors and the volatilization trend of aromatic substances are quantitatively analyzed; according to the method, heat-sensitive aromatic components can be effectively reserved, meanwhile, component proportion imbalance caused by premature or delayed release is avoided, the drug effect consistency of a finished product is improved, a comprehensive convergence score SCF is introduced, end point judgment better conforms to the actual concentration state change, the batch-to-batch difference is reduced, and meanwhile, the product quality is improved. Early warning and operation adjustment can be automatically triggered when a liquid medicine concentration system does not reach a stable convergence state, so that excessive heating or insufficient concentration is avoided, and the intelligence, refinement and self-adaptive capabilities of a traditional Chinese medicine concentration process are integrally improved.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine production and processing technology, specifically to an intelligent control method for a traditional Chinese medicine concentration process. Background Technology

[0002] Currently, in the production of traditional Chinese medicine (TCM) decoction pieces, granules, oral liquids, and other TCM preparations, vacuum concentration technology has become an important process for improving extraction efficiency, reducing the degradation of heat-sensitive components, and improving flavor and taste. However, the temperature, pressure, flow rate, and heat transfer conditions involved in this process exhibit significant dynamic fluctuations, which not only affect the final quality of the concentrate but also lead to changes in the content of important components such as aromatics, trace active substances, and dissolved oxygen in the liquid, thereby affecting the efficacy and stability of the drug. As the production of TCM preparations places higher demands on the content of effective components and flavor stability, simply relying on fixed time or temperature thresholds to determine the concentration endpoint is no longer sufficient to meet the requirements of production precision and consistency. This is especially true in TCM formulations containing multiple heat-sensitive aromatic components, such as volatile oils, aromatic esters, and trace amounts of aldehydes and ketones, where the release time window of these heat-sensitive aromatic components is extremely narrow, and a lack of refined control can easily lead to component loss.

[0003] In existing Chinese medicine concentration processes, especially in the preparation of health supplement decoction pieces, fixed decompression temperature control curves or constant vacuum concentration strategies are often used. These traditional control methods rely on empirically set parameters and do not provide real-time perception and feedback on the gas phase composition and its changing trends during the actual concentration process. Since the volatilization behavior of some aromatic or heat-sensitive components has a strong time correlation and stage, this static control strategy will cause the effective components of the liquid to evaporate in large quantities in advance, resulting in problems such as a bland flavor, loss of effective components, and large quality fluctuations. Especially in batch continuous production, this type of loss has a cumulative effect, which seriously affects the consistency and taste stability of the final product. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent control method for traditional Chinese medicine concentration processes, which solves the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent control method for a traditional Chinese medicine concentration process, comprising the following steps: S1. Real-time acquisition of drug concentration-related data through intelligent sensor group, and preprocessing of drug concentration-related data to construct structured parameter set S, wherein drug concentration-related data includes gas phase data and liquid phase data; S2. Based on the structured parameter set S, quantitative analysis is performed on the activity level of volatile gas behavior and the volatilization trend of aromatic substances to obtain the gas phase disturbance accumulation index. (t) and the aromatic volatile tendency index AVT(t); S3, Accumulation index of gas phase disturbance The aromatic volatile tendency index AVT(t) is dimensionlessly processed and coupled with the norm synthesis method to obtain the gas phase volatile endpoint determination index MVT(t), which is used to evaluate the release status of effective volatile components. When the evaluation result indicates that the release of effective volatile components is complete, the concentration and enrichment stage is entered. S4. Based on the structured parameter set S, analyze the changes in drug concentration to obtain the liquid phase concentration endpoint determination index LTI(t), and couple the liquid phase concentration endpoint determination index LTI(t) with the gas phase volatilization endpoint determination index MVT(t) to obtain the comprehensive convergence score SCF and evaluate the convergence status of drug concentration in the concentration and enrichment stage. S5. Based on the evaluation results of the convergence state of the concentrated medicine solution during the concentration and enrichment stage, when the concentration of the medicine solution has not reached a stable convergence state, the early warning system is automatically triggered to control the intelligent operation of the medicine solution concentration equipment.

[0006] Preferably, step S1 specifically includes: S11. Install an intelligent sensor group in the liquid concentration equipment and use the intelligent sensor group to acquire relevant data of the traditional Chinese medicine concentration process in real time, namely, liquid concentration related data. Perform data preprocessing on the acquired liquid concentration related data and construct a structured parameter set S based on the preprocessed liquid concentration related data. The liquid concentration related data includes gas phase data and liquid phase data. Data preprocessing includes processing the original sensor signals using the median filtering method, synchronizing the sampling frequency, normalizing the data, and removing outliers. The traditional Chinese medicine concentration process includes a vacuum heating evaporation stage, a concentration and enrichment stage, and a steady-state maintenance stage. The liquid concentration equipment refers to a vacuum decompression concentration device used for the concentration treatment of traditional Chinese medicine liquid. The vacuum heating evaporation stage involves controlled heating of the liquid medicine under vacuum conditions, which causes the solvent in the liquid phase to evaporate rapidly at low temperature, thereby avoiding the destruction of the active ingredients by high temperature. The main task of this stage is to use vacuum to lower the boiling point, and with the help of a stable heat source input, to achieve a large amount of initial removal of water from the liquid medicine, and to form a vapor stream that can be condensed and recovered on the gas phase side. The concentration and enrichment stage is based on the vacuum heating and evaporation stage, further reducing the water content of the liquid medicine and increasing the concentration value of the liquid medicine to the set target concentration value. The steady-state maintenance stage is the operational stage that maintains the stable state of the drug solution after it has reached the set target concentration value. During this stage, low-power heat sources and constant vacuum conditions are used to suppress moisture reabsorption or continued evaporation, so as to maintain the temperature and pressure inside the tank within the set range and prevent the active ingredients from continuing to decompose under high temperature or high vacuum.

[0007] Preferably, step S1 further includes, S12. The intelligent sensor group includes a thermal anemometer, a capacitive negative pressure sensor, a non-contact infrared temperature sensor, a gas chromatograph, an online refractive index sensor, and a high-temperature platinum resistance temperature sensor. S13, Gas phase data includes gas phase flow rate Vapor phase pressure Vapor phase temperature and gas phase signal intensity ; gas phase flow rate The instantaneous flow rate of the mixed airflow discharged through the steam exhaust pipe inside the concentration pot is obtained by using a thermal anemometer. The mixed airflow refers to the gas mixture in the steam exhaust pipe of the drug concentration equipment that simultaneously contains water vapor and gaseous components volatilized from the drug solution, mainly solvent vapor and some volatile active ingredients. Vapor pressure The pressure of the gas phase space inside the concentrator under vacuum is obtained by using a capacitive negative pressure sensor. Vapor temperature The steam temperature in the steam exhaust pipe inside the concentrator is obtained by using a non-contact infrared temperature sensor. Gas phase signal strength This indicates the chromatographic response amplitudes of various volatile aromatic components detected during the drug concentration process, which were acquired using a gas chromatograph. S14. Liquid phase data includes drug concentration value C and liquid phase temperature. ; The concentration value C of the drug solution represents the degree of solute enrichment during the concentration process, that is, the total amount of effective ingredients contained in a unit volume of drug solution, which is obtained by using an online refractive index sensor. Liquid phase temperature The temperature of the liquid medicine is indicated by a high-temperature resistant platinum resistance temperature sensor.

[0008] Preferably, step S2 specifically includes: S21. Based on the structured parameter set S, during the vacuum heating and evaporation stage of the drug solution, the activity level of volatile gases in the gas phase environment above the drug solution is quantitatively analyzed to obtain the gas phase disturbance accumulation index. Among them, the cumulative index of gas phase disturbance The specific method of obtaining it is as follows: ; In the formula, This indicates the starting point of data acquisition for the vacuum heating and evaporation stage of the medicinal liquid. Indicates the current data collection time point. Represents the integral variable. Indicates the data collection time point Gas flow rate at that time Indicates the time point of data collection The gas phase pressure at that time, Indicates the time point of data collection The liquid phase temperature at that time Indicates the time point of data collection The gas phase temperature at that time.

[0009] Preferably, step S2 further includes, S22. Based on the structured parameter set S, calculate the gas phase signal intensity for each type of volatile aromatic component. First derivative calculations were performed to obtain the gas-phase signal intensity of volatile aromatic components. The release rate over time, i.e., the rate of change of gas phase signal intensity. ,in, This represents the gas-phase signal intensity of the i-th type of volatile aromatic component; S23. Calculate the rate of change of gas phase signal intensity at the current data acquisition time point t. With gas phase signal intensity Coupled calculations are performed to construct a trend adjustment factor, the specific form of which is... ; S24. The rate of change of gas phase signal intensity of all n types of volatile aromatic components Gas phase signal intensity at the current data acquisition time point t Coupled calculations are performed and averaged to obtain the aromatic substance volatility trend index AVT(t).

[0010] Preferably, step S3 specifically includes: S31, Accumulation index of gas phase disturbance The volatile tendency index of aromatic substances, AVT(t), was dimensionlessly processed, and the cumulative index of gas-phase disturbances after dimensionless processing was used as the basis for the calculation. The aromatic volatility tendency index AVT(t) was coupled with the norm synthesis method to obtain the gas phase volatilization endpoint determination index MVT(t). S32. Collect relevant data on the concentration of qualified pharmaceutical products over a historical period through the pharmaceutical concentration system database to obtain the vapor phase volatilization endpoint determination index (MVT) of qualified pharmaceutical products. Use the MVT of qualified pharmaceutical products as a sample for statistical analysis, and select the 95th percentile of the MVT(t) as the determination threshold δ. The pharmaceutical concentration system database refers to an electronic information system used to store, manage and analyze various process parameters and detection data during the pharmaceutical concentration process. S33. Compare and analyze the gas-phase volatilization endpoint determination index MVT(t) with the determination threshold δ to evaluate the release status of effective volatile components during the vacuum heating evaporation stage. The specific evaluation content is as follows: If the gas phase evaporation endpoint determination index MVT(t) is greater than the determination threshold δ, it is determined that the release of effective volatile components has not been completed, and the current vacuum concentration state is maintained and operation continues. If the gas phase volatilization endpoint determination index MVT(t) is less than or equal to the determination threshold δ, it is determined that the effective volatile components have been released. At this time, the vacuum decompression heating operation is immediately terminated, and the concentration and enrichment stage begins.

[0011] Preferably, step S4 specifically includes: S41. Based on the structured parameter set S, after the effective volatile components are released, the difference between the drug concentration C(t) at the current data acquisition time point t and the drug concentration C(t-1) at the previous data acquisition time point t-1 is calculated based on the liquid phase data in the structured parameter set S. The difference is then divided by the data acquisition time interval Δt to obtain the instantaneous change rate of the drug concentration at the current data acquisition time point t. The instantaneous change rate of the drug concentration for a period of time before the current data acquisition time point t is also calculated and the average value is calculated to obtain the drug concentration change rate LCR(t). S42. Calculate the ratio of the drug concentration value C(t) at the current data acquisition time t to the target drug concentration value to obtain the drug concentration completion index CCI(t). Then, perform a composite operation with the drug concentration change rate LCR(t) and the drug concentration completion index CCI(t) to obtain the liquid phase concentration endpoint determination index LTI(t). The specific method for obtaining the liquid phase concentration endpoint determination index LTI(t) is as follows: ; In the formula, This represents the drug concentration completion index at the current data collection time point t. This indicates the rate of change of the drug concentration at the current data acquisition time point t. This indicates the allowable relative deviation of the drug concentration completion index. This represents the maximum acceptable range of the rate of change of drug concentration (LCR(t)) under stable conditions at the endpoint of liquid phase concentration. By taking the qualified state of the process as a reference, the absolute magnitude of LCR(t) is normalized to a uniform scale, thereby constructing a comparable and reproducible judgment benchmark for the "change tends to stabilize" attribute.

[0012] Preferably, step S4 further includes, S43. The Liquid Phase Concentration Endpoint Determination Index (LTI) Gas-phase volatilization endpoint determination index (MVT) Coupled calculations are performed to obtain the overall convergence score (SCF). The specific method for obtaining the overall convergence score (SCF) is as follows: ; In the formula, Represents the natural logarithm function. This represents the maximum value of the determination index for the gas-phase volatilization endpoint, where the maximum value of the determination index for the gas-phase volatilization endpoint is... Extracted from the drug concentration system database.

[0013] Preferably, step S4 further includes, S44. A preset comprehensive convergence threshold ф is established, and the comprehensive convergence threshold ф is compared and analyzed with the comprehensive convergence score SCF to evaluate the convergence status of the drug solution concentration during the concentration and enrichment stage. The comprehensive convergence threshold ф is obtained by combining the comprehensive convergence score SCF of multiple batches of historical qualified drug solution concentration processes using a sliding window stability judgment method. Specifically, the comprehensive convergence score SCF of each batch of qualified drug solution when it entered a stable state is identified from historical drug solution concentration data and used as a sample for statistical analysis. The sample mean is selected as the comprehensive convergence threshold ф. Specific evaluation content is as follows: If the overall convergence score SCF is less than or equal to the overall convergence threshold ф, i.e. SCF≤ф, then it is determined that the concentration of the medicine solution in the concentration and enrichment stage has not reached a stable convergence state. If the overall convergence score SCF is greater than or equal to the overall convergence threshold ф, i.e. SCF > ф, it is determined that the concentration of the drug solution in the concentration and enrichment stage has reached a stable convergence state. At this time, it enters the steady-state maintenance stage and marks the drug solution concentration system as the endpoint state. At the same time, the overall convergence score SCF value and the corresponding timestamp are recorded as the basis for the completion of the concentration of the current batch of drug solution and archived.

[0014] Preferably, step S5 specifically includes: S51. Based on the convergence state evaluation results of the concentrated medicine solution during the concentration and enrichment stage, if the concentrated medicine solution fails to reach a stable convergence state during the concentration and enrichment stage, the early warning system is automatically triggered and enters the alarm state. At this time, the concentrated medicine solution system is immediately controlled to maintain the current heating power and vacuum degree, and the heat preservation time of the medicine solution is extended. The relevant data of the concentrated medicine solution over a period of time is continuously collected to calculate the comprehensive convergence score (SCF) for the future period of time. If the comprehensive convergence score (SCF) for the future period of time is still less than or equal to the comprehensive convergence threshold (ф), a dual alarm mechanism is triggered, generating an endpoint lag prompt and a thermal coupling abnormality prompt, and starting a self-test program to perform equipment calibration. If the comprehensive convergence score (SCF) for the future period of time is greater than the comprehensive convergence threshold (ф), the alarm state is lifted and the steady-state maintenance stage is entered. The equipment calibration includes heat source stability calibration, vacuum system airtightness calibration, and thermocouple sensor calibration.

[0015] This invention provides an intelligent control method for a traditional Chinese medicine concentration process, which has the following beneficial effects: (1) By deploying a smart sensor group in the drug concentration equipment, the synchronous acquisition of gas phase data and liquid phase data is realized. Data preprocessing methods such as median filtering, sampling frequency synchronization, normalization and outlier removal are used to construct a structured parameter set S integrating gas and liquid phases, and the gas phase disturbance accumulation index is introduced. The quantitative analysis of the aromatic substance volatilization trend index AVT(t) and the dynamic determination of the volatilization endpoint and concentration endpoint under the coupling of gas-liquid two-phase information can effectively avoid the premature loss of low-boiling-point components and the delayed release of high-boiling-point components, thereby significantly improving the retention rate of effective ingredients and the consistency of the flavor of the medicine.

[0016] (2) By unweighting the liquid phase concentration endpoint determination index LTI(t) and the gas phase volatilization endpoint determination index MVT(t), a comprehensive convergence score SCF is constructed to achieve synchronous monitoring of the convergence of the gas and liquid phases and the thermal conductivity stability of the equipment. Compared with the method that relies on human experience weights, the comprehensive convergence score SCF of the present invention can suppress the endpoint determination in time when either phase has not reached a stable state, avoid misjudgment caused by local fluctuations or parameter deviations, and ensure that the convergence evaluation results have high consistency and repeatability under different batches of raw materials, different equipment loads and different operating conditions, thereby significantly improving the stability and objectivity of the endpoint identification of the Chinese medicine concentration process.

[0017] (3) By introducing quantile threshold judgment and dual alarm mechanism, when the comprehensive convergence score SCF is less than or equal to the comprehensive convergence threshold ф, the strategy of extending the heat preservation, maintaining the heating power and vacuum degree can be automatically triggered, and subsequent data can be continuously collected for secondary evaluation. If the continuous evaluation still fails to meet the standard, the equipment self-test and the calibration process of heat source, vacuum system and sensing unit can be automatically started. This closed-loop dynamic control mechanism can not only compensate for process fluctuations and extend the extraction time of the released components, but also perform rapid diagnosis and correction under equipment performance degradation or abnormal working conditions, significantly improving the system's adaptability to changing working conditions, raw material differences and external disturbances, and ensuring the accuracy of concentration endpoint judgment and the stability of the production process. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the intelligent control method for a traditional Chinese medicine concentration process according to the present invention; Figure 2 This is a schematic diagram of the data structure related to the concentration of the drug solution in this invention; Figure 3 This is a schematic diagram of the effective volatile component release status assessment process of the present invention; Figure 4 This is a schematic diagram comparing the overall convergence score (SCF) and the overall convergence threshold (ф) of this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1 Please see Figure 1 This invention provides an intelligent control method for a traditional Chinese medicine concentration process, comprising the following steps: S1. Real-time acquisition of drug concentration-related data through intelligent sensor group, and preprocessing of drug concentration-related data to construct structured parameter set S, wherein drug concentration-related data includes gas phase data and liquid phase data; S2. Based on the structured parameter set S, quantitative analysis is performed on the activity level of volatile gas behavior and the volatilization trend of aromatic substances to obtain the gas phase disturbance accumulation index. (t) and the aromatic volatile tendency index AVT(t); S3, Accumulation index of gas phase disturbance The aromatic volatile tendency index AVT(t) is dimensionlessly processed and coupled with the norm synthesis method to obtain the gas phase volatile endpoint determination index MVT(t), which is used to evaluate the release status of effective volatile components. When the evaluation result indicates that the release of effective volatile components is complete, the concentration and enrichment stage is entered. S4. Based on the structured parameter set S, analyze the changes in drug concentration to obtain the liquid phase concentration endpoint determination index LTI(t), and couple the liquid phase concentration endpoint determination index LTI(t) with the gas phase volatilization endpoint determination index MVT(t) to obtain the comprehensive convergence score SCF and evaluate the convergence status of drug concentration in the concentration and enrichment stage. S5. Based on the evaluation results of the convergence state of the concentrated medicine solution during the concentration and enrichment stage, when the concentration of the medicine solution has not reached a stable convergence state, the early warning system is automatically triggered to control the intelligent operation of the medicine solution concentration equipment.

[0021] In this embodiment, by simultaneously acquiring and structuring multi-dimensional gas-phase and liquid-phase data throughout the entire process of traditional Chinese medicine concentration, the limitations of relying on a single time or temperature threshold for endpoint determination in existing technologies are overcome. This significantly improves the accuracy and stability of endpoint determination. Furthermore, the quantitative analysis of the activity level of volatile gases and the volatilization trend of aromatic substances can effectively retain heat-sensitive aromatic components while avoiding imbalances in component ratios caused by premature or delayed release, thus improving the consistency of the finished product's efficacy. The endpoint determination results of the gas-phase and liquid-phase phases are dynamically coupled through a comprehensive convergence score (SCF), making the endpoint determination more consistent with actual concentration state changes and reducing batch-to-batch differences. At the same time, it can automatically trigger warnings and operational adjustments when the drug concentration system has not reached a stable convergence state, avoiding overheating or insufficient concentration, reducing energy consumption, and extending equipment lifespan. Overall, this enhances the intelligence, precision, and adaptability of the traditional Chinese medicine concentration process, meeting the dual requirements of modern traditional Chinese medicine preparations for effective component retention rate and batch stability.

[0022] Example 2 Please refer to Figure 1 , Figure 2 and Figure 3 Specifically: S1 includes the following steps: S11. Install an intelligent sensor group in the liquid concentration equipment and use the intelligent sensor group to acquire relevant data of the traditional Chinese medicine concentration process in real time, namely, liquid concentration related data. Perform data preprocessing on the acquired liquid concentration related data and construct a structured parameter set S based on the preprocessed liquid concentration related data. The liquid concentration related data includes gas phase data and liquid phase data. Data preprocessing includes processing the original sensor signals using the median filtering method, synchronizing the sampling frequency, normalizing the data, and removing outliers. The traditional Chinese medicine concentration process includes a vacuum heating evaporation stage, a concentration and enrichment stage, and a steady-state maintenance stage. The liquid concentration equipment refers to a vacuum decompression concentration device used for the concentration treatment of traditional Chinese medicine liquid. The vacuum heating evaporation stage involves controlled heating of the liquid medicine under vacuum conditions, which causes the solvent in the liquid phase to evaporate rapidly at low temperature, thereby avoiding the destruction of the active ingredients by high temperature. The main task of this stage is to use vacuum to lower the boiling point, and with the help of a stable heat source input, to achieve a large amount of initial removal of water from the liquid medicine, and to form a vapor stream that can be condensed and recovered on the gas phase side. The concentration and enrichment stage is based on the vacuum heating and evaporation stage, further reducing the water content of the liquid medicine and increasing the concentration value of the liquid medicine to the set target concentration value. The steady-state maintenance stage is the operational stage that maintains the stable state of the drug solution after it has reached the set target concentration value. During this stage, low-power heat sources and constant vacuum conditions are used to suppress moisture reabsorption or continued evaporation, so as to maintain the temperature and pressure inside the tank within the set range and prevent the active ingredients from continuing to decompose under high temperature or high vacuum.

[0023] S1 specific steps also include, S12. The intelligent sensor group includes a thermal anemometer, a capacitive negative pressure sensor, a non-contact infrared temperature sensor, a gas chromatograph, an online refractive index sensor, and a high-temperature platinum resistance temperature sensor. S13, Gas phase data includes gas phase flow rate Vapor phase pressure Vapor phase temperature and gas phase signal intensity ; gas phase flow rate The instantaneous flow rate of the mixed airflow discharged through the steam exhaust pipe inside the concentration pot is obtained by using a thermal anemometer. The mixed airflow refers to the gas mixture in the steam exhaust pipe of the drug concentration equipment that simultaneously contains water vapor and gaseous components volatilized from the drug solution, mainly solvent vapor and some volatile active ingredients. Vapor pressure The pressure of the gas phase space inside the concentrator under vacuum is obtained by using a capacitive negative pressure sensor. Vapor temperature The steam temperature in the steam exhaust pipe inside the concentrator is obtained by using a non-contact infrared temperature sensor. It reflects the energy state of the gas produced by evaporation and is a direct manifestation of the "boiling effect" and "volatilization drive". Gas phase signal strength This indicates the chromatographic response amplitudes of various volatile aromatic components detected during the drug concentration process, which were acquired using a gas chromatograph. S14. Liquid phase data includes drug concentration value C and liquid phase temperature. ; The concentration value C of the drug solution represents the degree of solute enrichment during the concentration process, that is, the total amount of effective ingredients contained in a unit volume of drug solution, which is obtained by using an online refractive index sensor. Liquid phase temperature The temperature of the liquid medicine is indicated by a high-temperature resistant platinum resistance temperature sensor.

[0024] In this embodiment, by deploying multiple types of intelligent sensors within the drug concentration equipment, real-time acquisition and synchronous processing of gas and liquid phase parameters are achieved throughout the entire process. This breaks through the traditional extensive management mode that relies on a single temperature, pressure, or time as the basis for control. The introduction of gas phase parameters enables the system to dynamically capture the degree of volatility and the release state of aromatic substances, significantly improving the ability to protect heat-sensitive and aromatic components. Synchronous monitoring of liquid phase parameters ensures accurate control of the solute enrichment degree and heat transfer process, helping to avoid the loss of medicinal components due to uneven heat transfer or over-concentration. The collected drug concentration-related data are subjected to median filtering, sampling synchronization, normalization, and outlier removal to construct a structured parameter set S, providing high-quality data support for subsequent intelligent judgment and stage control. This method can precisely control the different physical characteristics of the three stages of vacuum heating evaporation, concentration and enrichment, and steady-state maintenance, reducing batch-to-batch differences and improving the consistency and repeatability of the concentration endpoint, thereby comprehensively improving the refinement and intelligence level of the traditional Chinese medicine concentration process.

[0025] Example 3 Please refer to Figure 1 and Figure 3 Specifically: The specific steps of S2 include, S21. Based on the structured parameter set S, during the vacuum heating and evaporation stage of the drug solution, the activity level of volatile gases in the gas phase environment above the drug solution is quantitatively analyzed to obtain the gas phase disturbance accumulation index. Among them, the cumulative index of gas phase disturbance The specific method of obtaining it is as follows: ; In the formula, This indicates the starting point of data acquisition for the vacuum heating and evaporation stage of the medicinal liquid. Indicates the current data collection time point. Represents the integral variable. Indicates the time point of data collection Gas flow rate at that time Indicates the time point of data collection The gas phase pressure at that time, Indicates the time point of data collection The liquid phase temperature at that time Indicates the time point of data collection The gas phase temperature at that time.

[0026] Cumulative index of gas phase disturbance This is a comprehensive parameter used to quantify the activity of gas disturbances in the upper gas phase environment during the vacuum heating and evaporation of pharmaceutical liquids. It is calculated by integrating the instantaneous gas velocity change rate, gas phase pressure change rate, and gas-liquid temperature difference. It reflects the strength and accumulation of disturbances in the gas phase caused by factors such as liquid surface evaporation and temperature difference during the vacuum heating and evaporation stage. In the vacuum heating and evaporation stage, the strength of gas phase disturbances directly affects the stability of the evaporation rate, heat and mass transfer efficiency, as well as energy consumption and process efficiency. Therefore, it is necessary to introduce a gas phase disturbance accumulation index. By monitoring and quantifying gas phase disturbances, abnormal conditions in the vacuum heating evaporation stage can be detected in a timely manner during the operation of the traditional Chinese medicine concentration process. This can also prevent contamination or loss of concentrate caused by excessive gas phase disturbances, solve the problems of decreased evaporation rate and prolonged concentration time when gas phase disturbances are too weak, improve the controllability and repeatability of the vacuum heating evaporation process, and ensure the consistency of the quality of the concentrated medicine.

[0027] Formula derivation and physical meaning: The formula introduces a disturbance response mechanism, in which, It is the gas-phase disturbance excitation term, representing the second derivative of the gas-phase pressure change, and denoted as the wave acceleration. It is used to characterize the instantaneous response intensity of the system under disturbance, and to capture abnormal oscillations or points of violent component release. This is a reverse velocity amplification term, representing a reverse amplification of the disturbance in the direction of airflow velocity to counteract the masking effect of high-speed flow. Based on the turbulence suppression effect, the faster the gas flow velocity, the harder it is for the disturbance signal to be detected. Therefore, the reciprocal of the gas flow velocity is introduced to improve the sensitivity to low-speed disturbances and avoid misjudging a steady state. In addition, the formula includes a liquid-temperature temperature difference correction term. , To correct for perturbation sensitivity in the presence of temperature gradients, the formula consists of three parts: gas phase perturbation excitation term × reverse velocity amplification term × liquid phase temperature difference correction term. A temperature gradient ratio is constructed in the formula, introducing the temperature difference value and correcting the thermal field through the square of the gas phase temperature, thus adapting to the situation of uneven temperature distribution in real multiphase systems. Finally, to avoid the random interference of instantaneous abrupt changes, time integration is used to cumulatively sum all perturbation behaviors throughout the entire evaporation stage, yielding the gas phase perturbation cumulative index. .

[0028] A specific example is as follows: Suppose that during the concentration process of a certain traditional Chinese medicine, a smart sensor array is used to collect relevant data on the concentration of the medicinal liquid during the vacuum heating evaporation stage in real time, as shown in Table 1: Based on Table 1, obtain the gas phase disturbance excitation term. First, calculate the first derivative of the gas phase pressure. The calculation formula is: ; When = 0, forward difference is used. =0.38; When =10, central difference is used. =0.46; When the value is 20, central difference is used. =0.61; When the value is 30, central difference is used. =0.75; When the value is 40, backward difference is used. =0.82; Subsequently, the first derivative Based on this, continue calculating the second derivative. The second derivative is obtained. The values ​​are 0.0080, 0.0116, 0.0145, 0.0101, and 0.0065, respectively. Based on the obtained second derivative values, for each data acquisition time point of Calculations were performed to obtain the data collection time points 0, 10, 20, 30, and 40. They are respectively , , , , The gas phase disturbance cumulative index is obtained by performing time integration and summarization. for ; S22. Based on the structured parameter set S, calculate the gas phase signal intensity for each type of volatile aromatic component. First derivative calculations were performed to obtain the gas-phase signal intensity of volatile aromatic components. The release rate over time, i.e., the rate of change of gas phase signal intensity. ,in, This represents the gas-phase signal intensity of the i-th type of volatile aromatic component; Volatile aromatic components refer to different types of aromatic compounds that are released into the gas phase during the concentration process of traditional Chinese medicine. These compounds usually come from volatile oils or other small molecules with aromatic odors in Chinese medicinal materials, including monoterpenes, sesquiterpenes, aromatic aldehydes, aromatic ketones, aromatic alcohols, as well as esters and ethers. S23. Calculate the rate of change of gas phase signal intensity at the current data acquisition time point t. With gas phase signal intensity Coupled calculations are performed to construct a trend adjustment factor, the specific form of which is... ; Among them, the gas phase signal intensity at the current data acquisition time point t The method of adding one and then taking the logarithm aims to map the gas phase signal intensity to a logarithmic scale. This compresses the amplitude of change under high-amplitude signal conditions, thus avoiding nonlinear amplification of the trend adjustment factor due to excessively large instantaneous peak values. Simultaneously, it retains high sensitivity to change under low-amplitude signal conditions, enabling the trend adjustment factor to amplify weak changes in the gas phase signal in the low-amplitude range. This achieves smoothing and robust adjustment of the gas phase disturbance trend. In a specific example, assuming the gas phase signal intensities at the current data acquisition time point t and the previous data acquisition time point t-1 are 0.63 and 0.70 respectively, calculate the rate of change of the gas phase signal intensity at the current data acquisition time point t. Logarithmic terms =0.53, at this point the trend adjustment factor is derived as = ×0.53=0.0037.

[0029] S24. The rate of change of gas phase signal intensity of all n types of volatile aromatic components Gas phase signal intensity at the current data acquisition time point t Coupled calculations are performed and averaged to obtain the aromatic substance volatility trend index AVT(t).

[0030] The Aromatic Volatility Trend Index (AVT(t)) is a comprehensive index obtained by trend correction and cumulative calculation of the rate of change of the gas phase signal intensity of volatile aromatic components during processes such as concentration or heating. It not only reflects the release rate of aromatic components over a period of time, but also avoids the interference of instantaneous fluctuations after smoothing by a trend adjustment factor, thus more accurately reflecting the overall trend of aromatic component volatilization. Through the AVT(t) index, the aroma quality of the product can be controlled, process conditions can be prevented from going out of control, and a basis for endpoint judgment can be provided. It can reduce the concentrated loss of aromatic components caused by instantaneous high temperature or vacuum fluctuations, solve the problem that traditional endpoint detection alone cannot reflect the process trend, and improve the controllability and consistency of aroma quality during the concentration process, reducing batch-to-batch differences.

[0031] In this embodiment, by introducing dual quantitative analysis of the gas-phase disturbance accumulation index and the aromatic substance volatilization trend index during the vacuum heating evaporation stage, a fine characterization of the dynamic features of the gas-liquid two phases is achieved. The gas-phase disturbance accumulation index is constructed based on multiple factors such as the second derivative of pressure, the reciprocal of flow velocity, and liquid-temperature temperature difference correction. It can accurately capture the sensitive transition from stability to instability and the point of violent release of components, avoiding the masking or misjudgment of disturbance signals caused by high flow rates or local temperature differences. The aromatic substance volatilization trend index, by monitoring the signal intensity change rate of various volatile aromatic components in real time and introducing a trend adjustment factor, achieves an amplified response to the volatilization trend under low signal intensity, thereby effectively tracking the release process of heat-sensitive and aromatic components. This method breaks through the limitations of relying on a single endpoint parameter, enabling the monitoring of volatilization behavior to shift from static endpoint determination to dynamic evaluation of the entire process. It not only improves the process's ability to grasp the release timing of key components, but also significantly reduces the risk of loss of effective components, thereby ensuring the flavor, efficacy, and stability of traditional Chinese medicine products.

[0032] Example 4 Please refer to Figure 1 and Figure 3 Specifically: The specific steps of S3 include, S31, Accumulation index of gas phase disturbance The volatile tendency index of aromatic substances, AVT(t), was dimensionlessly processed, and the cumulative index of gas-phase disturbances after dimensionless processing was used as the basis for the calculation. The aromatic volatility tendency index AVT(t) was coupled with the norm synthesis method to obtain the gas phase volatilization endpoint determination index MVT(t). During vacuum compression, the release of volatile components is a combined manifestation of gas-phase disturbance and the release of aromatic components, with the gas-phase disturbance cumulative index being [missing information]. The fluctuation amplitude and cumulative effect of gas phase disturbances were captured, reflecting the activity of drug liquid evaporation. The aromatic substance volatilization trend index AVT(t) describes the release rate and trend of aromatic substances, which directly corresponds to the volatilization of effective components. As the effective volatile components are gradually released, the degree of disturbance and the release trend of aromatic components will tend to stabilize or decrease, resulting in a decrease in the gas phase volatilization endpoint determination index MVT(t). When the gas phase volatilization endpoint determination index MVT(t) decreases to below the threshold, it indicates that the gas volatilization is close to completion and the release state of aromatic components tends to stabilize.

[0033] Norm synthesis is a mathematical method that unifies multiple indicators into a single comprehensive evaluation value. It is often used for the comprehensive analysis of multidimensional data. In this scheme, the dimensionless cumulative index of gas phase disturbance is used. The volatile tendency index of aromatic substances AVT(t) is regarded as two components of a two-dimensional vector. The norm synthesis method is to regard these two components as two coordinates of a two-dimensional vector and calculate their Euclidean norm. S32. Collect relevant data on the concentration of qualified pharmaceutical products over a historical period through the pharmaceutical concentration system database to obtain the vapor phase volatilization endpoint determination index (MVT) of qualified pharmaceutical products. Use the MVT of qualified pharmaceutical products as a sample for statistical analysis, and select the 95th percentile of the MVT(t) as the determination threshold δ. The pharmaceutical concentration system database refers to an electronic information system used to store, manage and analyze various process parameters and detection data during the pharmaceutical concentration process. S33. Compare and analyze the gas phase evaporation endpoint determination index MVT(t) with the determination threshold δ to evaluate the release status of effective volatile components during the vacuum heating evaporation stage. Effective volatile components refer to the set of volatile compounds in the traditional Chinese medicine formula that have a substantial impact on the product's flavor consistency or efficacy-related quality indicators, and that can be stably detected by the gas phase monitoring channel during the vacuum heating evaporation stage. Specific evaluation content is as follows: If the gas phase evaporation endpoint determination index MVT(t) is greater than the determination threshold δ, it is determined that the release of effective volatile components has not been completed, and the current vacuum concentration state is maintained and operation continues. If the gas phase volatilization endpoint determination index MVT(t) is less than or equal to the determination threshold δ, it is determined that the effective volatile components have been released. At this time, the vacuum decompression heating operation is immediately terminated, and the concentration and enrichment stage begins.

[0034] In this embodiment, by introducing a comparative analysis of the vapor phase evaporation endpoint determination index MVT(t) and the determination threshold δ, dynamic endpoint determination of the vacuum heating evaporation stage is achieved, overcoming the traditional crude method that relies on fixed time or temperature control. A 95th percentile threshold is constructed using historical data, giving the endpoint determination a statistical basis and enabling it to adapt to process fluctuations under different batches and raw material characteristics, thereby improving the accuracy and stability of the determination. Continuing heating when MVT(t) is higher than the threshold ensures the full release of effective volatile components, avoiding insufficient residue of medicinal components. Immediately stopping heating when MVT(t) reaches or falls below the threshold effectively prevents excessive loss of aromatic and heat-sensitive components, maintaining the flavor and efficacy of the medicinal liquid. This method makes the control of the evaporation stage more precise and controllable, significantly reducing the risk of quality fluctuations caused by premature or delayed termination, ensuring the stability of effective components and batch-to-batch consistency during the concentration of traditional Chinese medicine, and meeting the needs of modern traditional Chinese medicine production for precise quality control.

[0035] Example 5 Please refer to Figure 1 and Figure 4 Specifically: The specific steps of S4 include, S41. Based on the structured parameter set S, after the effective volatile components are released, the difference between the drug concentration C(t) at the current data acquisition time point t and the drug concentration C(t-1) at the previous data acquisition time point t-1 is calculated based on the liquid phase data in the structured parameter set S. The difference is then divided by the data acquisition time interval Δt to obtain the instantaneous change rate of the drug concentration at the current data acquisition time point t. The instantaneous change rate of the drug concentration for a period of time before the current data acquisition time point t is also calculated and the average value is calculated to obtain the drug concentration change rate LCR(t). S42. Calculate the ratio of the drug concentration value C(t) at the current data acquisition time t to the target drug concentration value to obtain the drug concentration completion index CCI(t). Then, perform a composite operation with the drug concentration change rate LCR(t) and the drug concentration completion index CCI(t) to obtain the liquid phase concentration endpoint determination index LTI(t). The specific method for obtaining the liquid phase concentration endpoint determination index LTI(t) is as follows: ; In the formula, This represents the drug concentration completion index at the current data collection time point t. This indicates the rate of change of the drug concentration at the current data acquisition time point t. This indicates the allowable relative deviation of the drug concentration completion index. This represents the maximum acceptable range of the rate of change of drug concentration (LCR(t)) under stable conditions at the endpoint of liquid phase concentration. By taking the qualified state of the process as a reference, the absolute magnitude of LCR(t) is normalized to a uniform scale, thereby constructing a comparable and reproducible judgment benchmark for the "change tends to stabilize" attribute.

[0036] Permissible relative deviation of the drug concentration completion index and the allowable rate of change of drug concentration It is set through process standards and specifications.

[0037] The target concentration value of the medicinal liquid represents a pre-set target concentration in the Chinese medicine concentration process, used to determine when the medicinal liquid reaches the expected degree of concentration, and is set through process standards and specifications.

[0038] The determination of the liquid phase endpoint should meet two conditions simultaneously: first, the drug concentration value is close to the target drug concentration value; second, the drug concentration change tends to be stable. The calculation formula of the liquid phase concentration endpoint determination index LTI(t) maps the two conditions to the drug concentration completion score and the drug concentration change stability score, respectively, and uses a product form to achieve "a high value is given only when both conditions are met at the same time". Only when the drug concentration completion index is high and the drug concentration change tends to be stable will the liquid phase concentration endpoint determination index LTI(t) obtain a high value, thereby avoiding the problem of extracting the endpoint when only one condition is met.

[0039] Derivation of the formula and its physical meaning: The score indicates the degree of completion of drug concentration. The score represents the stability of the drug concentration change, among which and This involves normalization, dividing the numerical value by a reference value to map parameters of different dimensions and magnitudes onto a unified, comparable scale, i.e., between 0 and 1. The multiplicative coupling of the drug concentration completion score and the drug concentration change stability score ensures that a high LTI(t) value is only given when both conditions are met simultaneously. High but If the concentration is large, then the LTI(t) is low. At this point, the drug concentration is complete, but the concentration of the drug solution is unstable. Low but If the value is small, then LTI(t) is low. At this time, the concentration of the drug solution changes stably, but the drug solution concentration is not complete.

[0040] Specific examples are as follows: Assuming that the effective volatile components have been released, based on the structured parameter set S, the target concentration of the drug solution is set to 100.00, the data acquisition time interval Δt is 1 min, and the acquired data is shown in Table 2: In Table 2, at t=6, the drug concentration completion index is high, but the rate of change of drug concentration is large, so the obtained liquid phase concentration endpoint determination index (LTI) is low. The initial concentration index is low, while at t=8 or t=9, the concentration completion index is high, but the rate of change of drug concentration decreases significantly. At this time, the change of drug concentration tends to be stable, so the LTI (Liquid Phase Concentration Endpoint Judgment Index) obtained is low. Larger.

[0041] The specific steps in S4 also include, S43. The Liquid Phase Concentration Endpoint Determination Index (LTI) Gas-phase volatilization endpoint determination index (MVT) Coupled calculations are performed to obtain the overall convergence score (SCF). The specific method for obtaining the overall convergence score (SCF) is as follows: ; In the formula, ln represents the natural logarithm function. This represents the maximum value of the determination index for the gas-phase volatilization endpoint, where the maximum value of the determination index for the gas-phase volatilization endpoint is... Extracted from the drug concentration system database; Formula derivation and physical meaning: The formula uses the liquid concentration endpoint determination index (LTI) to determine the endpoint of liquid phase concentration. Gas-phase volatilization endpoint determination index (MVT) The summation of squares reflects the nonlinear superposition of the gas-phase and liquid-phase concentration states on the overall drug solution convergence. The use of "+1" avoids the input of the natural logarithm function being 0, and the final formula uses the natural logarithm function ln to compress the large liquid-phase concentration endpoint determination index LTI. Gas-phase volatilization endpoint determination index (MVT) The variation range of the gas phase volatilization endpoint determination index (MVT) is improved to enhance numerical stability and assessment resolution. The gas phase evaporation endpoint determination index (MVT) is negatively correlated with the overall convergence score (SCF). The lower the value, the closer it is to the end of gas phase volatilization. Therefore, the formula uses the gas phase volatilization endpoint determination index (MVT). Normalization mapping is performed to convert it into an index positively correlated with the overall convergence score (SCF), thus making the vapor phase evaporation endpoint determination index (MVT)... Liquid Concentration Endpoint Judgment Index (LTI) Changes in the same trend direction; The Comprehensive Convergence Score (SCF) is a comprehensive numerical value that measures whether the liquid and gas phases simultaneously approach and stably reach the endpoint. It uses a single value to reflect whether the degree of drug concentration completion and the stability of changes converge synchronously. This ensures that the score is significantly suppressed when either the liquid or gas phase fails to meet the target or experiences significant fluctuations. Only when the Liquid Concentration Endpoint Judgment Index (LTI) is reached... Gas-phase volatilization endpoint determination index (MVT) At the same time, SCF values ​​are only achieved when they are high and stable, which can be used for anomaly diagnosis, such as insufficient condensation, leakage, limited heat transfer, and batch consistency assessment.

[0042] Specific examples are as follows: Suppose that during the vacuum heating evaporation stage of a batch of traditional Chinese medicine concentration production, the operator collects the process parameters of the liquid and gas phases in real time through the intelligent control system. The fluctuations in both the liquid and gas phases of the current batch of medicine are slowing down, and the endpoint determination is being carried out to obtain the Liquid Phase Concentration Endpoint Determination Index (LTI). Gas-phase volatilization endpoint determination index (MVT) The maximum values ​​of the gas-phase volatilization endpoint determination index are 0.864 and 0.812, respectively. Given a value of 1.000, calculate the overall convergence score (SCF): ln =0.026.

[0043] The specific steps in S4 also include, S44. A preset comprehensive convergence threshold ф is established, and the comprehensive convergence threshold ф is compared and analyzed with the comprehensive convergence score SCF to evaluate the convergence status of the drug solution concentration during the concentration and enrichment stage. The comprehensive convergence threshold ф is obtained by combining the comprehensive convergence score SCF of multiple batches of historical qualified drug solution concentration processes using a sliding window stability judgment method. Specifically, the comprehensive convergence score SCF of each batch of qualified drug solution when it entered a stable state is identified from historical drug solution concentration data and used as a sample for statistical analysis. The sample mean is selected as the comprehensive convergence threshold ф. Specific evaluation content is as follows: If the overall convergence score SCF is less than or equal to the overall convergence threshold ф, i.e. SCF≤ф, then it is determined that the concentration of the medicine solution in the concentration and enrichment stage has not reached a stable convergence state. If the overall convergence score SCF is greater than or equal to the overall convergence threshold ф, i.e. SCF > ф, it is determined that the concentration of the drug solution in the concentration and enrichment stage has reached a stable convergence state. At this time, it enters the steady-state maintenance stage and marks the drug solution concentration system as the endpoint state. At the same time, the overall convergence score SCF value and the corresponding timestamp are recorded as the basis for the completion of the concentration of the current batch of drug solution and archived.

[0044] In this embodiment, the liquid phase concentration endpoint determination index (LTI) is introduced. Gas-phase volatilization endpoint determination index (MVT) By employing bidirectional coupling, a comprehensive convergence score (SCF) is constructed, enabling multi-dimensional synchronous convergence evaluation of the entire process of traditional Chinese medicine concentration. This method not only reflects the trend and stability of the concentration of the medicinal liquid simultaneously, avoiding incomplete concentration or component loss due to insufficient or excessive heating, but also eliminates the bias caused by human parameter settings through unweighted objective fusion. This improves the adaptability and consistency of judgment under different batches and equipment conditions. When the comprehensive convergence score reaches the threshold, it can accurately determine that the concentration is complete and switch to the steady-state maintenance stage, significantly reducing energy consumption and equipment load, while ensuring the stability of the medicinal liquid components and the preservation of flavor, thereby improving the overall quality and production efficiency of traditional Chinese medicine concentration.

[0045] Example 6 Please refer to Figure 1 Specifically: The S5 steps include, S51. Based on the convergence state evaluation results of the concentrated medicine solution during the concentration and enrichment stage, if the concentrated medicine solution fails to reach a stable convergence state during the concentration and enrichment stage, the early warning system is automatically triggered and enters the alarm state. At this time, the concentrated medicine solution system is immediately controlled to maintain the current heating power and vacuum degree, and the heat preservation time of the medicine solution is extended. The relevant data of the concentrated medicine solution over a period of time is continuously collected to calculate the comprehensive convergence score (SCF) for the future period of time. If the comprehensive convergence score (SCF) for the future period of time is still less than or equal to the comprehensive convergence threshold (ф), a dual alarm mechanism is triggered, generating an endpoint lag prompt and a thermal coupling abnormality prompt, and starting a self-test program to perform equipment calibration. If the comprehensive convergence score (SCF) for the future period of time is greater than the comprehensive convergence threshold (ф), the alarm state is lifted and the steady-state maintenance stage is entered. The equipment calibration includes heat source stability calibration, vacuum system airtightness calibration, and thermocouple sensor calibration.

[0046] In this embodiment, by automatically triggering an early warning system when the drug concentration system fails to reach a stable convergence state, real-time safety control and process assurance of the concentration process are achieved. When the comprehensive convergence score is detected to be below the threshold, the system can automatically maintain the current heating power and vacuum level, and extend the holding time to ensure that the trace effective components released later are fully volatilized and concentrated, avoiding incomplete components or quality degradation due to premature termination. At the same time, subsequent operating data is continuously collected and a second comprehensive convergence evaluation is performed. If the threshold is still not reached, a dual alarm is triggered, indicating endpoint lag and abnormal heat transfer coupling, and initiating equipment self-check and calibration, covering comprehensive detection of heat source, vacuum system and sensors. This mechanism can effectively prevent concentration abnormalities caused by equipment status deviation, improve the reliability and consistency of endpoint determination, ensure the stability of the concentrated quality of traditional Chinese medicine, reduce batch-to-batch differences, and improve the controllability and safety of the production process.

[0047] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent control of a traditional Chinese medicine concentration process, characterized in that: Includes the following steps, S1. Real-time acquisition of drug concentration-related data through intelligent sensor group, and preprocessing of drug concentration-related data to construct structured parameter set S, wherein drug concentration-related data includes gas phase data and liquid phase data; S2. Based on the structured parameter set S, quantitative analysis is performed on the activity level of volatile gas behavior and the volatilization trend of aromatic substances to obtain the gas phase disturbance accumulation index. (t) and the aromatic volatile tendency index AVT(t); S3, Accumulation index of gas phase disturbance The aromatic volatile tendency index AVT(t) is dimensionlessly processed and coupled with the norm synthesis method to obtain the gas phase volatile endpoint determination index MVT(t), which is used to evaluate the release status of effective volatile components. When the evaluation result indicates that the release of effective volatile components is complete, the concentration and enrichment stage is entered. S4. Based on the structured parameter set S, analyze the changes in drug concentration to obtain the liquid phase concentration endpoint determination index LTI(t), and couple the liquid phase concentration endpoint determination index LTI(t) with the gas phase volatilization endpoint determination index MVT(t) to obtain the comprehensive convergence score SCF and evaluate the convergence status of drug concentration in the concentration and enrichment stage. S5. Based on the evaluation results of the convergence state of the concentrated medicine solution during the concentration and enrichment stage, when the concentration of the medicine solution has not reached a stable convergence state, the early warning system is automatically triggered to control the intelligent operation of the medicine solution concentration equipment.

2. The intelligent control method for a traditional Chinese medicine concentration process according to claim 1, characterized in that: The specific steps in S1 include: S11. Install an intelligent sensor group in the liquid concentration equipment and use the intelligent sensor group to acquire relevant data of the traditional Chinese medicine concentration process in real time, namely, liquid concentration related data. Perform data preprocessing on the acquired liquid concentration related data and construct a structured parameter set S based on the preprocessed liquid concentration related data. The liquid concentration related data includes gas phase data and liquid phase data. Data preprocessing includes processing the original sensor signals using the median filtering method, synchronizing the sampling frequency, normalizing the data, and removing outliers. The traditional Chinese medicine concentration process includes a vacuum heating evaporation stage, a concentration and enrichment stage, and a steady-state maintenance stage. The liquid concentration equipment refers to a vacuum decompression concentration device used for the concentration treatment of traditional Chinese medicine liquid. The vacuum heating evaporation stage involves controlled heating of the liquid medicine under vacuum conditions, which causes the solvent in the liquid phase to evaporate rapidly at low temperature, thereby avoiding the destruction of the active ingredients by high temperature. The main task of this stage is to use vacuum to lower the boiling point, and with the help of a stable heat source input, to achieve a large amount of initial removal of water from the liquid medicine, and to form a vapor stream that can be condensed and recovered on the gas phase side. The concentration and enrichment stage is based on the vacuum heating and evaporation stage, further reducing the water content of the liquid medicine and increasing the concentration value of the liquid medicine to the set target concentration value. The steady-state maintenance stage is the operational stage that maintains the stable state of the drug solution after it has reached the set target concentration value. During this stage, low-power heat sources and constant vacuum conditions are used to suppress moisture reabsorption or continued evaporation, so as to maintain the temperature and pressure inside the tank within the set range and prevent the active ingredients from continuing to decompose under high temperature or high vacuum.

3. The intelligent control method for a traditional Chinese medicine concentration process according to claim 2, characterized in that: S1 specific steps also include, S12. The intelligent sensor group includes a thermal anemometer, a capacitive negative pressure sensor, a non-contact infrared temperature sensor, a gas chromatograph, an online refractive index sensor, and a high-temperature platinum resistance temperature sensor. S13, Gas phase data includes gas phase flow rate Vapor phase pressure Vapor phase temperature and gas phase signal intensity ; gas phase flow rate The instantaneous flow rate of the mixed airflow discharged through the steam exhaust pipe inside the concentration pot is obtained by using a thermal anemometer. The mixed airflow refers to the gas mixture in the steam exhaust pipe of the drug concentration equipment that simultaneously contains water vapor and gaseous components volatilized from the drug solution, mainly solvent vapor and some volatile active ingredients. Vapor pressure The pressure of the gas phase space inside the concentrator under vacuum is obtained by using a capacitive negative pressure sensor. Vapor phase temperature The steam temperature in the steam exhaust pipe inside the concentrator is obtained by using a non-contact infrared temperature sensor. Gas phase signal strength This indicates the chromatographic response amplitudes of various volatile aromatic components detected during the drug concentration process, which were acquired using a gas chromatograph. S14. Liquid phase data includes drug concentration value C and liquid phase temperature. ; The concentration value C of the drug solution represents the degree of solute enrichment during the concentration process, that is, the total amount of effective ingredients contained in a unit volume of drug solution, which is obtained by using an online refractive index sensor. Liquid phase temperature The temperature of the liquid medicine is indicated by a high-temperature resistant platinum resistance temperature sensor.

4. The intelligent control method for a traditional Chinese medicine concentration process according to claim 3, characterized in that: The specific steps of S2 include, S21. Based on the structured parameter set S, during the vacuum heating and evaporation stage of the drug solution, the activity level of volatile gases in the gas phase environment above the drug solution is quantitatively analyzed to obtain the gas phase disturbance accumulation index. Among them, the cumulative index of gas phase disturbance The specific method of obtaining it is as follows: ； In the formula, This indicates the starting point of data acquisition for the vacuum heating and evaporation stage of the medicinal liquid. Indicates the current data collection time point. Represents the integral variable. Indicates the time point of data collection Gas flow rate at that time Indicates the time point of data collection The gas phase pressure at that time, Indicates the time point of data collection The liquid phase temperature at that time Indicates the time point of data collection The gas phase temperature at that time.

5. The intelligent control method for a traditional Chinese medicine concentration process according to claim 4, characterized in that: S2 specific steps also include, S22. Based on the structured parameter set S, calculate the gas phase signal intensity for each type of volatile aromatic component. First derivative calculations were performed to obtain the gas-phase signal intensity of volatile aromatic components. The release rate over time, i.e., the rate of change of gas phase signal intensity. ,in, This represents the gas-phase signal intensity of the i-th type of volatile aromatic component; S23. Calculate the rate of change of gas phase signal intensity at the current data acquisition time point t. With gas phase signal intensity Coupled calculations are performed to construct a trend adjustment factor, the specific form of which is... ; S24. The rate of change of gas phase signal intensity of all n types of volatile aromatic components Gas phase signal intensity at the current data acquisition time point t Coupled calculations are performed and averaged to obtain the aromatic substance volatility trend index AVT(t).

6. The intelligent control method for a traditional Chinese medicine concentration process according to claim 5, characterized in that: The specific steps of S3 include, S31, Accumulation index of gas phase disturbance The volatile tendency index of aromatic substances, AVT(t), was dimensionlessly processed, and the cumulative index of gas-phase disturbances after dimensionless processing was used as the basis for the calculation. The aromatic volatility tendency index AVT(t) was coupled with the norm synthesis method to obtain the gas phase volatilization endpoint determination index MVT(t). S32. Collect relevant data on the concentration of qualified pharmaceutical products over a historical period through the pharmaceutical concentration system database to obtain the vapor phase volatilization endpoint determination index (MVT) of qualified pharmaceutical products. Use the MVT of qualified pharmaceutical products as a sample for statistical analysis, and select the 95th percentile of the MVT(t) as the determination threshold δ. The pharmaceutical concentration system database refers to an electronic information system used to store, manage and analyze various process parameters and detection data during the pharmaceutical concentration process. S33. Compare and analyze the gas-phase volatilization endpoint determination index MVT(t) with the determination threshold δ to evaluate the release status of effective volatile components during the vacuum heating evaporation stage. The specific evaluation content is as follows: If the gas phase evaporation endpoint determination index MVT(t) is greater than the determination threshold δ, it is determined that the release of effective volatile components has not been completed, and the current vacuum concentration state is maintained and operation continues. If the gas phase volatilization endpoint determination index MVT(t) is less than or equal to the determination threshold δ, it is determined that the effective volatile components have been released. At this time, the vacuum decompression heating operation is immediately terminated, and the concentration and enrichment stage begins.

7. The intelligent control method for a traditional Chinese medicine concentration process according to claim 6, characterized in that: The specific steps of S4 include, S41. Based on the structured parameter set S, after the effective volatile components are released, the difference between the drug concentration C(t) at the current data acquisition time point t and the drug concentration C(t-1) at the previous data acquisition time point t-1 is calculated based on the liquid phase data in the structured parameter set S. The difference is then divided by the data acquisition time interval Δt to obtain the instantaneous change rate of the drug concentration at the current data acquisition time point t. The instantaneous change rate of the drug concentration for a period of time before the current data acquisition time point t is also calculated and the average value is calculated to obtain the drug concentration change rate LCR(t). S42. Calculate the ratio of the drug concentration value C(t) at the current data acquisition time t to the target drug concentration value to obtain the drug concentration completion index CCI(t). Then, perform a composite operation with the drug concentration change rate LCR(t) and the drug concentration completion index CCI(t) to obtain the liquid phase concentration endpoint determination index LTI(t). The specific method for obtaining the liquid phase concentration endpoint determination index LTI(t) is as follows: ； In the formula, This represents the drug concentration completion index at the current data collection time point t. This indicates the rate of change of the drug concentration at the current data acquisition time point t. This indicates the allowable relative deviation of the drug concentration completion index. This represents the maximum acceptable range of the rate of change of drug concentration (LCR(t)) under stable conditions at the endpoint of liquid phase concentration. By taking the qualified state of the process as a reference, the absolute magnitude of LCR(t) is normalized to a uniform scale, thereby constructing a comparable and reproducible judgment benchmark for the "change tends to stabilize" attribute.

8. The intelligent control method for a traditional Chinese medicine concentration process according to claim 7, characterized in that: The specific steps in S4 also include, S43. The Liquid Phase Concentration Endpoint Determination Index (LTI) Gas-phase volatilization endpoint determination index (MVT) Coupled calculations are performed to obtain the overall convergence score (SCF). The specific method for obtaining the overall convergence score (SCF) is as follows: ； In the formula, Represents the natural logarithm function. This represents the maximum value of the determination index for the gas-phase volatilization endpoint, where the maximum value of the determination index for the gas-phase volatilization endpoint is... Extracted from the drug concentration system database.

9. The intelligent control method for a traditional Chinese medicine concentration process according to claim 8, characterized in that: The specific steps in S4 also include, S44. A preset comprehensive convergence threshold ф is established, and the comprehensive convergence threshold ф is compared and analyzed with the comprehensive convergence score SCF to evaluate the convergence status of the drug solution concentration during the concentration and enrichment stage. The comprehensive convergence threshold ф is obtained by combining the comprehensive convergence score SCF of multiple batches of historical qualified drug solution concentration processes using a sliding window stability judgment method. Specifically, the comprehensive convergence score SCF of each batch of qualified drug solution when it entered a stable state is identified from historical drug solution concentration data and used as a sample for statistical analysis. The sample mean is selected as the comprehensive convergence threshold ф. Specific evaluation content is as follows: If the overall convergence score SCF is less than or equal to the overall convergence threshold ф, i.e. SCF≤ф, then it is determined that the concentration of the medicine solution in the concentration and enrichment stage has not reached a stable convergence state. If the overall convergence score SCF is greater than or equal to the overall convergence threshold ф, i.e. SCF > ф, it is determined that the concentration of the drug solution in the concentration and enrichment stage has reached a stable convergence state. At this time, it enters the steady-state maintenance stage and marks the drug solution concentration system as the endpoint state. At the same time, the overall convergence score SCF value and the corresponding timestamp are recorded as the basis for the completion of the concentration of the current batch of drug solution and archived.

10. The intelligent control method for a traditional Chinese medicine concentration process according to claim 9, characterized in that: The specific steps of S5 include, S51. Based on the convergence state evaluation results of the concentrated medicine solution during the concentration and enrichment stage, if the concentrated medicine solution fails to reach a stable convergence state during the concentration and enrichment stage, the early warning system is automatically triggered and enters the alarm state. At this time, the concentrated medicine solution system is immediately controlled to maintain the current heating power and vacuum degree, and the heat preservation time of the medicine solution is extended. The relevant data of the concentrated medicine solution over a period of time is continuously collected to calculate the comprehensive convergence score (SCF) for the future period of time. If the comprehensive convergence score (SCF) for the future period of time is still less than or equal to the comprehensive convergence threshold (ф), a dual alarm mechanism is triggered, generating an endpoint lag prompt and a thermal coupling abnormality prompt, and starting a self-test program to perform equipment calibration. If the comprehensive convergence score (SCF) for the future period of time is greater than the comprehensive convergence threshold (ф), the alarm state is lifted and the steady-state maintenance stage is entered. The equipment calibration includes heat source stability calibration, vacuum system airtightness calibration, and thermocouple sensor calibration.

Images