Traditional Chinese medicine extraction tank temperature gradient intelligent control method and system

Abstract

The application discloses a traditional Chinese medicine extraction tank temperature gradient intelligent control method and system, the method adopts multiple longitudinal temperature zones constructed by multiple independent electric heating jackets distributed longitudinally along the tank wall of the extraction tank, and multiple layer and multiple point detection units of temperature and effective component concentration are arranged corresponding to each temperature zone; a double closed loop control architecture coupled by the concentration outer ring and the temperature inner ring is constructed, the deviation between the concentration measured value and the target dissolution curve is outputted by fuzzy reasoning operation to output the temperature correction amount, and then the output power of the corresponding heating jacket is adjusted by the inner ring PID; meanwhile, the temperature difference between adjacent temperature zones is dynamically limited by the gradient smoothing constraint algorithm to maintain the continuous and smooth transition of the temperature gradient. The system constructs a multiple input and multiple output control architecture, improves the synchronous dissolution efficiency of fat-soluble and water-soluble components, the application can adaptively match the dissolution temperature requirements of different properties of effective components, significantly improves the comprehensive extraction rate of multiple components, reduces the production energy consumption, and improves the stability of the extraction process and the batch quality consistency.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for pharmaceutical production processes, specifically to an intelligent control method and system for temperature gradient in a traditional Chinese medicine extraction tank. Background Technology

[0002] Extraction of traditional Chinese medicine (TCM) is a core process in TCM pharmaceutical production. The dissolution efficiency and retention rate of its active ingredients directly determine the utilization rate of medicinal materials, the efficacy of the finished product, and the production cost. During the extraction process, temperature is a key control parameter affecting the dissolution kinetics of active ingredients. Different types of active ingredients respond significantly differently to temperature: fat-soluble components usually require higher temperatures to dissolve fully, while water-soluble and heat-sensitive components are prone to decomposition and inactivation at high temperatures.

[0003] Traditional constant temperature extraction mode has a single temperature field inside the tank, which cannot simultaneously meet the dissolution requirements of different components. It is prone to problems such as insufficient dissolution of fat-soluble components or degradation of heat-sensitive components, resulting in a low overall extraction rate. In addition, the energy consumption is high due to the method of simultaneous heating of the whole tank.

[0004] Existing improved zone heating schemes still have significant drawbacks: First, they mostly employ open-loop or single-temperature closed-loop control, without introducing real-time feedback on the concentration of effective components, making it impossible to dynamically adjust parameters based on batch differences in medicinal materials and actual dissolution progress, resulting in poor batch quality stability; second, they lack gradient smoothing constraint mechanisms between adjacent temperature zones, and excessive local temperature differences can easily trigger abnormal convection of the medicinal liquid, disrupting the stability of the temperature gradient field; third, conventional PID algorithms are difficult to adapt to the nonlinear, large-lag, and multi-coupling process characteristics of traditional Chinese medicine extraction, making it difficult to balance control accuracy and dynamic response.

[0005] In summary, existing temperature control technologies for the extraction of traditional Chinese medicine cannot guarantee process stability and energy economy while simultaneously improving the extraction rate of multiple effective components. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention discloses an intelligent control method and system for temperature gradient in a traditional Chinese medicine extraction tank, in order to solve the problems mentioned in the background.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligent control of temperature gradient in a traditional Chinese medicine extraction tank, applied to a traditional Chinese medicine extraction tank with a layered electric heating jacket on the tank wall, comprising the following steps:

[0008] S1. Obtain the characteristic parameters of the Chinese medicinal materials to be extracted, and match them to generate the initial temperature gradient curves of multiple independent temperature zones distributed longitudinally in the extraction tank. Each temperature zone corresponds to each layer of electric heating jacket.

[0009] S2. During the extraction process, the actual temperature value of the medicine liquid and the measured value of the effective ingredient concentration in each temperature zone are collected in real time according to the preset sampling cycle through multi-layer and multi-point detection units arranged for each temperature zone.

[0010] S3. Construct a dual closed-loop control architecture that couples the concentration outer loop and the temperature inner loop. Use the measured concentration of the effective component in each temperature zone and the concentration deviation of the preset target dissolution curve as the outer loop input, and output the temperature setting correction amount of the corresponding temperature zone through fuzzy inference calculation.

[0011] S4. Input the temperature deviation between the target temperature value after the superimposed temperature setting correction and the actual temperature value of the liquid medicine into the inner loop PID controller, and output the power adjustment signal of the corresponding layered electric heating jacket to realize independent closed-loop tracking control of the temperature of each temperature zone.

[0012] S5. Calculate the actual temperature difference between adjacent temperature zones in real time, and dynamically adjust the heating rate of each temperature zone through the gradient smoothing constraint algorithm to maintain a continuous and smooth transition of the longitudinal temperature gradient in the extraction tank.

[0013] Preferably, step S1 specifically includes:

[0014] A pre-constructed database of extraction process parameters for traditional Chinese medicinal materials is used. The database stores temperature zone configuration parameters, initial temperature parameters for each temperature zone, and dissolution characteristic parameters of the target active ingredients for various traditional Chinese medicinal materials. Based on the input particle size range, moisture content range, and dissolution properties of the target active ingredients of the traditional Chinese medicinal materials to be extracted, the corresponding process parameters are retrieved from the database to generate an initial temperature gradient curve distributed along the longitudinal height of the extraction tank. The dissolution properties of the target active ingredients include water solubility and lipid solubility.

[0015] Preferably, in step S2, the multi-layer multi-point detection unit includes temperature detection elements and near-infrared spectral detection probes arranged at intervals along the longitudinal height of the extraction tank. At least one detection unit consisting of temperature detection elements and concentration detection probes is configured in each temperature zone. The detection points are located in the mainstream area of ​​the drug solution in the corresponding temperature zone. The collected temperature and concentration values ​​are digitally filtered and then input into the control unit. The preset sampling period is within a preset time interval.

[0016] Preferably, in step S3, the fuzzy inference operations are performed sequentially according to the following hierarchy, with each operational relation corresponding to a control step:

[0017] The first step is to calculate the concentration deviation at the current sampling time. This deviation characterizes the degree of discrepancy between the actual dissolution state of the active ingredient and the preset dissolution target, serving as the basic deviation input for outer-loop control. The concentration deviation calculation satisfies the following:

[0018] ,

[0019] The second step is to calculate the concentration deviation change rate, which characterizes the dynamic trend of dissolution deviation, provides an advance prediction basis for temperature regulation, and suppresses control lag. The concentration deviation change rate calculation satisfies the following:

[0020] ,

[0021] The third step involves using concentration deviation and the rate of change of concentration deviation as two-dimensional input variables. A fuzzy inference mapping operation is then performed to obtain the temperature setting correction value. This correction is used to dynamically adjust the temperature setting value of the temperature zone based on the actual dissolution progress, achieving adaptive optimization of dissolution efficiency. The temperature setting correction value calculation satisfies the following:

[0022] ,

[0023] The fourth step involves superimposing the initial temperature setpoint with the temperature setpoint correction to obtain the corrected target temperature value. This corrected target temperature value serves as the input for the inner-loop temperature control, completing the control transfer from the concentration outer loop to the temperature inner loop. The target temperature value calculation satisfies the following:

[0024] ,

[0025] Where k is the sampling time number; The concentration deviation at the k-th sampling time; The target concentration value on the target dissolution curve is preset for the k-th sampling time; This represents the measured concentration of the effective component at the k-th sampling time. Let be the rate of change of concentration deviation at the k-th sampling time; For the first Concentration deviation at the sampling time; The preset sampling period; Set a correction amount for the temperature output at the k-th sampling time; This is a fuzzy inference mapping function established based on the dissolution kinetics rules of traditional Chinese medicinal materials; This is the target temperature value corrected at the k-th sampling time; This is the initial temperature setting value at the k-th sampling time.

[0026] Preferably, in step S4, the power regulation signal output of the inner loop PID controller satisfies the following relationship:

[0027] ,

[0028] in, This represents the heating power control quantity output at the k-th sampling time. This is the proportionality coefficient; The integral coefficient; These are the differential coefficients; The temperature deviation at the k-th sampling time is equal to the difference between the corrected target temperature value and the actual temperature value of the liquid medicine. For the first Temperature deviation at the sampling time; The preset sampling period is consistent with the sampling period used in the concentration outer loop calculation;

[0029] The heating power control quantity is converted from digital to analog and output as a power adjustment signal, which drives the corresponding layered electric heating jacket to adjust the output power, thereby realizing closed-loop tracking control of the temperature zone.

[0030] Preferably, in step S5, the gradient smoothing constraint algorithm satisfies the following temperature difference constraint relationship:

[0031] ,

[0032] in, This represents the actual temperature value of the i-th temperature zone at the k-th sampling time. The k-th sampling time is the longitudinally adjacent temperature zone i. The actual temperature value of each temperature zone; This is the preset maximum allowable temperature difference threshold;

[0033] When the actual temperature difference between adjacent temperature zones exceeds the maximum allowable temperature difference threshold, a limiting adjustment is applied to the heating rate of the high-temperature zone. The limited heating rate satisfies the following:

[0034] ,

[0035] in, The heating rate of the i-th temperature zone before limiting; This represents the heating rate of the i-th temperature zone after limiting. The heating rate limiting coefficient has a value range of greater than 0 and less than or equal to 1. Through the above constraints and adjustments, the continuous transition of the longitudinal temperature gradient in the extraction tank is maintained, and the convection turbulence of the medicine solution caused by sudden temperature changes is suppressed.

[0036] This invention also provides an intelligent control system for temperature gradient in a traditional Chinese medicine extraction tank, comprising:

[0037] The layered heating unit includes multiple layers of electric heating jackets distributed longitudinally along the wall of the extraction tank. Each layer of electric heating jacket is independently controlled and is used to create multiple longitudinally distributed independent temperature zones inside the extraction tank.

[0038] The multi-layer detection unit includes temperature detection elements and active ingredient concentration detection probes arranged for each temperature zone, which are used to collect real-time data on the temperature of the medicine solution and the concentration of the active ingredient in each temperature zone.

[0039] The storage unit is used to store a database of extraction process parameters for Chinese medicinal materials, preset target dissolution curves, and control parameters;

[0040] The main control unit has a built-in dual closed-loop control module and a gradient smoothing constraint module, which are respectively used to perform dual closed-loop control calculations and gradient smoothing constraint calculations of the concentration outer loop and the temperature inner loop. The main control unit is used to execute the temperature gradient intelligent control method as described in any one of claims 1-6.

[0041] The power drive unit receives the power adjustment signal output by the main control unit and drives each layer of electric heating jacket to adjust the output power accordingly.

[0042] Preferably, the active ingredient concentration detection probe is a near-infrared spectroscopy online detection probe, with at least one detection point set in each temperature zone; the optical window of the probe is made of a light-transmitting material that is resistant to high temperature and high pressure, the probe is fixed to the wall of the extraction tank through a flange sealing structure, and the detection end extends into the drug liquid area inside the tank.

[0043] Preferably, a heat insulation layer is provided between two adjacent electric heating jackets. The heat insulation layer is made of fire-resistant and heat-insulating fiber material to block the lateral heat conduction between adjacent temperature zones, reduce thermal coupling interference between temperature zones, and improve the independence of temperature control in each temperature zone.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] 1. This invention significantly improves the accuracy and stability of temperature control. It constructs a dual closed-loop control architecture of concentration outer loop and temperature inner loop, and combines it with a gradient smoothing constraint mechanism. Through the fuzzy PID composite algorithm, it adapts to the large lag and nonlinear process characteristics of traditional Chinese medicine extraction. This not only realizes independent high-precision closed-loop control of each layer of electric heating temperature zone, but also suppresses thermal coupling interference and temperature change in adjacent temperature zones, ensuring a continuous and stable transition of the longitudinal temperature field.

[0046] 2. This invention simultaneously optimizes solid-liquid extraction efficiency and product quality. Relying on a vertically layered temperature gradient field design, it can simultaneously match the high-temperature dissolution requirements of fat-soluble components and the low-temperature protection requirements of water-soluble heat-sensitive components. Combined with real-time feedback from multi-layer, multi-point near-infrared online concentration detection, the temperature parameters can be dynamically corrected according to the actual dissolution progress. This breaks through the temperature adaptation limitations of traditional constant-temperature extraction, significantly improves the simultaneous extraction rate of multiple effective components, and reduces the high-temperature degradation of heat-sensitive components, ensuring the efficacy and activity of the extracted products.

[0047] 3. This invention synergistically improves production energy consumption and batch consistency. Through a power adjustment strategy that links gradient power allocation with dissolution progress, redundant energy consumption caused by constant temperature heating of the entire tank can be avoided, reducing the production energy consumption per unit extraction amount. At the same time, the adaptive control capability of temperature parameters can offset the impact of batch differences such as herbal particle size and moisture content, improving the stability of extraction quality of different batches. Attached Figure Description

[0048] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0049] In the attached diagram:

[0050] Figure 1 This is a schematic diagram of the intelligent temperature gradient control system described in this embodiment;

[0051] Figure 2 This is a flowchart of the control method described in this embodiment;

[0052] Figure 3 This is a schematic diagram of the dual closed-loop control architecture. Detailed Implementation

[0053] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0054] This embodiment is applied to a vertical Chinese medicine extraction tank. The tank wall is equipped with three independent electric heating jackets, which correspond to the formation of three independent temperature zones distributed longitudinally inside the tank: upper, middle, and lower.

[0055] like Figure 1 The diagram shown illustrates the structure of the intelligent temperature gradient control system for the herbal extraction tank described in this embodiment. The system comprises five parts: a layered heating unit, a multi-layer detection unit, a storage unit, a main control unit, and a power drive unit. The structure and parameters of each unit are as follows:

[0056] The layered heating unit consists of three layers of electrically heated jackets evenly arranged longitudinally along the tank wall. Each jacket covers one-third of the effective height of the tank, with a rated power of 15kW per jacket. It employs a nickel-chromium resistance wire heating structure, ensuring a tight fit between the heating surface and the tank wall. A 20mm thick aluminum silicate refractory fiber insulation layer with a thermal conductivity ≤0.04W / m² is installed between adjacent heating jackets. It is used to block lateral heat conduction between adjacent temperature zones, reduce thermal coupling interference between temperature zones, and improve the independence of temperature control in each temperature zone. The three jackets correspond to the lower, middle, and upper temperature zones inside the tank from bottom to top.

[0057] The multi-layer detection unit is configured with one set of detection units for each of the three temperature zones. Each set includes one temperature detection element and one active ingredient concentration detection probe. The detection points are all located at the center of the main liquid area of ​​the corresponding temperature zone.

[0058] The temperature sensing element uses a PT100A platinum resistance thermometer with a measurement accuracy of ±0.15℃ and an insertion depth of 150mm into the container. It is used to collect the actual temperature of the liquid in each temperature zone.

[0059] The active ingredient concentration detection probe is a near-infrared spectroscopy online detection probe with a wavelength detection range of 1000nm~2500nm and a wavelength resolution ≤2nm. The optical window is made of high-temperature and high-pressure resistant sapphire material, withstanding a pressure of 1.6MPa and a temperature of 150℃. The probe is fixed to the extraction tank wall through a flange sealing structure, with the detection end flush with the inner wall of the tank, enabling real-time acquisition of the diffuse reflectance spectrum of the drug solution and calculation of the active ingredient concentration value. The sampling period of all detection units is uniformly set to 1s.

[0060] The storage unit, using industrial-grade solid-state memory, is integrated into the main control unit and is used to store a database of extraction process parameters for Chinese medicinal materials, target dissolution curves for various medicinal materials, near-infrared quantitative analysis models, control algorithm parameters, and historical operating data.

[0061] The main control unit adopts an industrial programmable logic controller (PLC), with a built-in dual closed-loop control module and gradient smoothing constraint module. The operation cycle and sampling cycle are synchronized at 1 second. It has functions such as analog quantity acquisition, logic operation, and control quantity output, and is used to execute all the temperature gradient intelligent control methods described in this embodiment.

[0062] The power drive unit uses three independent three-phase thyristor power regulators. The input signal is a standard analog quantity of 4mA~20mA, which corresponds to an output of 0~100% heating power, and each independently drives the three-layer electric heating jacket to work.

[0063] This embodiment uses a partial least squares (PLS) quantitative analysis model to achieve online detection of baicalin and baicalein concentrations. The model construction and calibration process is as follows:

[0064] 1) Sample set construction: 30 aliquots each of baicalin and baicalein standard solutions with different concentration gradients were prepared, covering a concentration range of 0.1 mg / mL to 1.5 mg / mL, to simulate the drug concentration levels at each stage of the extraction process. Simultaneously, 20 mixed solutions with different proportions were prepared to simulate the coexistence of multiple components in the actual extract. All samples were prepared using purified water solvent consistent with actual extraction and acidic conditions at pH 5.0–5.5 to ensure the model adapts to actual working conditions.

[0065] 2) Spectral preprocessing: Diffuse reflectance spectra of all samples were acquired using a near-infrared spectrometer of the same model as the online probe, with a scanning interval of 1 nm. Each sample was scanned three times, and the average spectrum was taken. The raw spectra were then subjected to standard normal variable transformation (SNV) and first derivative processing to eliminate particle scattering interference and baseline drift. The standard normal variable transformation was calculated using the following formula:

[0066] ,

[0067] Preprocessed spectra can effectively reduce the impact of drug turbidity and container wall reflection on detection results, and improve the predictive stability of the model.

[0068] 3) PLS quantitative model establishment: Using preprocessed spectral data as independent variables and the actual concentration determined by high-performance liquid chromatography (HPLC) as the dependent variable, a partial least squares quantitative model was established. The number of principal factors was determined using leave-one-out cross-validation; the number of principal factors for the baicalin model was set to 7, and the number of principal factors for the baicalein model was set to 6. After the model was established, it was validated using 20 independent validation samples. The validation results showed a coefficient of determination R² ≥ 0.95 and a root mean square error of prediction RMSEP ≤ 0.03 mg / mL, meeting the accuracy requirements for industrial online detection.

[0069] 4) Online calibration and incremental update: The trained PLS model is embedded into the main control unit. Before production, a standard solution with a concentration of 0.6 mg / mL is used for on-site calibration to correct system deviations caused by differences in optical path installation. During continuous production, drug solution samples are extracted from each batch at three time points (early, middle, and late stages) for HPLC verification. When the relative deviation of the prediction exceeds 3%, the model is incrementally updated using new samples to ensure the detection accuracy during long-term operation.

[0070] like Figure 2 The diagram illustrates the specific implementation steps of the intelligent temperature gradient control method. This embodiment uses the Scutellaria baicalensis extraction process as an example, with the target effective components being the water-soluble baicalin and the fat-soluble baicalein. The total extraction time is set to 60 minutes, and the specific control process is executed according to the following steps:

[0071] S1. Generation of the initial temperature gradient curve

[0072] The pre-constructed database of extraction process parameters for Chinese medicinal materials stores the corresponding process parameters for dozens of commonly used Chinese medicinal materials such as Scutellaria baicalensis, Forsythia suspensa, and Angelica sinensis. These parameters include the number of suitable temperature zones, the initial temperature range of each zone, the dissolution characteristic curves of the target components, and the correction coefficients for the particle size and moisture content of the medicinal materials.

[0073] In this embodiment, the operator inputs the characteristic parameters of the Scutellaria baicalensis to be extracted into the human-machine interface: particle size range of 2mm~5mm, moisture content range of 8%~12%, and the target effective components are water-soluble baicalin and fat-soluble baicalein. The main control unit matches and retrieves the corresponding process parameters from the database of traditional Chinese medicine extraction process parameters, generating an initial temperature gradient curve of three temperature zones distributed along the longitudinal height of the extraction tank.

[0074] The initial temperature of the lower temperature zone is set at 90℃, which is suitable for the optimal dissolution temperature range of fat-soluble baicalein.

[0075] The initial temperature of the middle temperature zone is set at 80℃, forming a transitional temperature zone between the upper and lower temperature zones;

[0076] The initial temperature of the upper temperature zone is set at 70℃ to meet the mild extraction requirements of water-soluble baicalin and reduce the high-temperature decomposition of heat-sensitive components.

[0077] This creates a longitudinal initial temperature gradient field inside the tank, with the temperature gradually decreasing from bottom to top.

[0078] S2. Real-time data acquisition and filtering preprocessing

[0079] After the extraction process is started, the temperature sensors in the three temperature zones and the near-infrared spectral probe synchronously collect data at a sampling period of 1 second:

[0080] 1) The temperature signal is converted into a 4mA~20mA standard electrical signal by the transmitter and input to the main control unit;

[0081] 2) The raw spectral data collected by the near-infrared spectral probe is solved in real time by the above PLS quantitative analysis model to output the concentration values ​​of baicalin and baicalein in the corresponding temperature zone of the drug solution;

[0082] 3) The collected temperature and concentration values ​​were both processed using a 5-point moving average filter to remove random noise caused by electromagnetic interference and drug fluctuations. The filter was calculated using the following formula:

[0083] ,

[0084] The filtered data is fed into the dual closed-loop control module for calculation, which can effectively suppress frequent fluctuations in control quantities caused by random noise.

[0085] S3, Concentration Outer Ring Fuzzy Inference Calculation

[0086] like Figure 3 As shown, the system constructs a dual closed-loop control architecture coupling an outer concentration loop and an inner temperature loop. The outer concentration loop dynamically adjusts the temperature setpoints for each temperature zone based on the actual dissolution state of the active ingredient. Fuzzy inference calculations are executed sequentially in four steps, with each calculation corresponding one-to-one with the control logic:

[0087] The first step is to calculate the concentration deviation at the current sampling time, which characterizes the degree of deviation between the actual dissolution state of the active ingredient and the preset dissolution target, and serves as the basic deviation input for the outer loop control.

[0088] ,

[0089] In this embodiment, the total extraction time is 60 minutes, corresponding to a total of 3600 sampling points. The target dissolution curve is set according to the dissolution kinetics. For example, the target concentration of baicalin is 0.3 mg / mL at the 10th minute of extraction and 1.2 mg / mL at the 60th minute of extraction.

[0090] The second step is to calculate the concentration deviation change rate to characterize the dynamic trend of dissolution deviation, providing an advance prediction basis for temperature regulation and suppressing the control overshoot of the large lag process in the extraction of traditional Chinese medicine.

[0091] ,

[0092] The sampling period is 1 second. The trend of change is calculated by the difference in deviation between adjacent time points, so as to realize the advanced perception of the dissolution rate.

[0093] The third step involves using concentration deviation and the rate of change of concentration deviation as two-dimensional input variables, and obtaining the temperature setting correction value through fuzzy inference mapping calculation. The temperature setting value of the temperature zone is then dynamically adjusted according to the actual dissolution progress to achieve adaptive optimization of dissolution efficiency.

[0094] ,

[0095] This embodiment uses a Mamdani-type fuzzy controller to implement the above mapping operation. The specific execution process is as follows:

[0096] 1) Partitioning of the universe of discourse and fuzzy subsets: The basic universe of discourse for concentration deviation is defined as [-0.2, 0.2] mg / mL, and the basic universe of discourse for the rate of change of concentration deviation is defined as [-0.01, 0.01] mg / mL. The basic domain of the output temperature correction is [-5,5]℃, which is divided into 7 fuzzy subsets: {negative large, negative medium, negative small, zero, positive small, positive medium, positive large}. The membership function adopts a triangular function.

[0097] 2) Establishment of fuzzy rules: A fuzzy rule base is established based on the dissolution kinetics of the effective components of Scutellaria baicalensis. The rules follow the principle of "the greater the deviation and the more the trend deviates from the target, the greater the temperature correction magnitude": when the concentration deviation is positive and the deviation change rate is positive, it indicates that the dissolution rate is significantly lagging behind the target curve, and a large positive temperature correction is output to increase the heating intensity and accelerate dissolution; when the concentration deviation is close to zero and the deviation change rate is negative, it indicates that the dissolution is close to the target saturation value, and a small negative temperature correction is output to reduce the heating intensity and avoid overheating and decomposition of heat-sensitive components; for other conditions, the correction magnitude is linearly matched according to the degree of deviation and the trend of change.

[0098] 3) Defuzzification calculation: The centroid method is used for defuzzification calculation, outputting a precise temperature setpoint correction. The centroid method is calculated using the following formula:

[0099] ,

[0100] The fourth step is to add the initial temperature setpoint to the temperature setpoint correction value to obtain the corrected target temperature value, which is then used as the input for the inner loop temperature control loop, completing the transfer of control quantity from the concentration outer loop to the temperature inner loop.

[0101] ,

[0102] S4, Temperature Inner Loop PID Closed-Loop Control

[0103] The temperature deviation between the corrected target temperature value and the actual temperature of the liquid in each temperature zone is input into the inner loop PID controller. The controller calculates and outputs the corresponding power adjustment signal for the layered electric heating jacket, thereby achieving independent closed-loop tracking control of the temperature in each temperature zone. The position-type PID control quantity is calculated using the following formula:

[0104] ,

[0105] In this embodiment, the heating power control value ranges from 0 to 100%; the proportional coefficient is 2.8, the integral coefficient is 0.0025s⁻¹, the derivative coefficient is 60s, and the sampling period is 1s, consistent with the sampling period of the concentration outer loop calculation.

[0106] The calculated heating power control quantity is converted from digital to analog signal into a standard analog signal of 4mA~20mA, and output to the corresponding layer's thyristor power regulator to drive the electric heating jacket to adjust the output power, achieving rapid and zero steady-state error tracking of the temperature zone. The PID control loops of the three temperature zones are independent of each other and do not interfere with each other.

[0107] S5, Gradient Smoothing Constraint Adjustment

[0108] While operating under dual closed-loop control, the gradient smoothing constraint module calculates the actual temperature difference between adjacent temperature zones in real time, and maintains a continuous and smooth transition of the longitudinal temperature gradient inside the extraction tank by dynamically adjusting the heating rate.

[0109] The system's preset temperature difference constraint is executed according to the following formula:

[0110] ,

[0111] In this embodiment, the maximum allowable temperature difference threshold is set to 15℃, and the system judges the temperature difference status of adjacent temperature zones in real time:

[0112] When the actual temperature difference between adjacent temperature zones is ≤15℃, the gradient state is determined to be normal, and the dual closed-loop control operates according to the original logic without applying additional constraints.

[0113] When the actual temperature difference between adjacent temperature zones is greater than 15℃, the gradient is deemed too steep. A limiting adjustment is applied to the heating rate of the high-temperature zone. The limited heating rate is calculated using the following formula:

[0114] ,

[0115] The heating rate limiting coefficient is set to 0.6. For example, if the original heating rate in the high-temperature zone is 2℃ / min, it will be adjusted to 1.2℃ / min after limiting. The original control logic in the low-temperature zone remains unchanged until the temperature difference between adjacent temperature zones drops to within 15℃.

[0116] Through the above constraints and adjustments, the violent convection of the medicinal solution and uneven distribution of components caused by excessive longitudinal temperature gradient can be avoided, the damage of heat-sensitive active ingredients to sudden temperature changes can be suppressed, and the stability of the extraction process can be improved.

[0117] The extraction process of Scutellaria baicalensis was validated using the method described in this embodiment. Each batch used 100 kg of the herb, with a liquid volume of 800 L and an extraction time of 60 min. Compared to the traditional 90℃ constant-temperature extraction process, the combined extraction rate of baicalin and baicalein was increased by 14%–17%, the energy consumption per unit volume of the extraction liquid was reduced by 11%–13%, the temperature control accuracy in each temperature zone reached ±0.5℃, the longitudinal temperature gradient transition was smooth, and the relative error of near-infrared concentration detection was ≤3%, fully meeting the process requirements for large-scale production of traditional Chinese medicine extracts.

[0118] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent temperature gradient control in a traditional Chinese medicine extraction tank, applied to a traditional Chinese medicine extraction tank with a layered electric heating jacket on the tank wall, characterized in that, Includes the following steps: S1. Obtain the characteristic parameters of the Chinese medicinal materials to be extracted, and match them to generate the initial temperature gradient curves of multiple independent temperature zones distributed longitudinally in the extraction tank. Each temperature zone corresponds to each layer of electric heating jacket. S2. During the extraction process, the actual temperature value of the medicine liquid and the measured value of the effective ingredient concentration in each temperature zone are collected in real time according to the preset sampling cycle through multi-layer and multi-point detection units arranged for each temperature zone. S3. Construct a dual closed-loop control architecture that couples the concentration outer loop and the temperature inner loop. Use the measured concentration of the effective component in each temperature zone and the concentration deviation of the preset target dissolution curve as the outer loop input, and output the temperature setting correction amount of the corresponding temperature zone through fuzzy inference calculation. S4. Input the temperature deviation between the target temperature value after the superimposed temperature setting correction and the actual temperature value of the liquid medicine into the inner loop PID controller, and output the power adjustment signal of the corresponding layered electric heating jacket to realize independent closed-loop tracking control of the temperature of each temperature zone. S5. Calculate the actual temperature difference between adjacent temperature zones in real time, and dynamically adjust the heating rate of each temperature zone through the gradient smoothing constraint algorithm to maintain a continuous and smooth transition of the longitudinal temperature gradient in the extraction tank.

2. The intelligent control method for temperature gradient in a traditional Chinese medicine extraction tank according to claim 1, characterized in that, Step S1 specifically includes: A pre-constructed database of extraction process parameters for traditional Chinese medicinal materials is used. The database stores temperature zone configuration parameters, initial temperature parameters for each temperature zone, and dissolution characteristic parameters of the target active ingredients for various traditional Chinese medicinal materials. Based on the input particle size range, moisture content range, and dissolution properties of the target active ingredients of the traditional Chinese medicinal materials to be extracted, the corresponding process parameters are retrieved from the database to generate an initial temperature gradient curve distributed along the longitudinal height of the extraction tank. The dissolution properties of the target active ingredients include water solubility and lipid solubility.

3. The intelligent control method for temperature gradient in a traditional Chinese medicine extraction tank according to claim 1, characterized in that: In step S2, the multi-layer multi-point detection unit includes temperature detection elements and near-infrared spectroscopy detection probes arranged at intervals along the longitudinal height of the extraction tank. Each temperature zone is equipped with at least one detection unit consisting of temperature detection elements and concentration detection probes. The detection points are located in the mainstream area of ​​the drug solution in the corresponding temperature zone. The collected temperature and concentration values ​​are digitally filtered and then input into the control unit. The preset sampling period is within a preset time interval.

4. The intelligent control method for temperature gradient in a traditional Chinese medicine extraction tank according to claim 1, characterized in that: In step S3, the fuzzy inference operation is performed sequentially according to the following levels: The first step is to calculate the concentration deviation at the current sampling time. This deviation characterizes the degree of discrepancy between the actual dissolution state of the active ingredient and the preset dissolution target, serving as the basic deviation input for outer-loop control. The concentration deviation calculation satisfies the following: ， The second step is to calculate the concentration deviation change rate, which characterizes the dynamic trend of dissolution deviation, provides an advance prediction basis for temperature regulation, and suppresses control lag. The concentration deviation change rate calculation satisfies the following: ， The third step involves using concentration deviation and the rate of change of concentration deviation as two-dimensional input variables. A fuzzy inference mapping operation is then performed to obtain the temperature setting correction value. This correction is used to dynamically adjust the temperature setting value of the temperature zone based on the actual dissolution progress, achieving adaptive optimization of dissolution efficiency. The temperature setting correction value calculation satisfies the following: ， The fourth step involves superimposing the initial temperature setpoint with the temperature setpoint correction to obtain the corrected target temperature value. This corrected target temperature value serves as the input for the inner-loop temperature control, completing the control transfer from the concentration outer loop to the temperature inner loop. The target temperature value calculation satisfies the following: ， Where k is the sampling time number; The concentration deviation at the k-th sampling time; The target concentration value on the target dissolution curve is preset for the k-th sampling time; This represents the measured concentration of the effective component at the k-th sampling time. Let be the rate of change of concentration deviation at the k-th sampling time; For the first Concentration deviation at the sampling time; The preset sampling period; Set a correction amount for the temperature output at the k-th sampling time; This is a fuzzy inference mapping function established based on the dissolution kinetics rules of traditional Chinese medicinal materials; This is the target temperature value corrected at the k-th sampling time; This is the initial temperature setting value at the k-th sampling time.

5. The intelligent control method for temperature gradient in a traditional Chinese medicine extraction tank according to claim 1, characterized in that: In step S4, the power regulation signal output of the inner loop PID controller satisfies the following relationship: ， in, This represents the heating power control quantity output at the k-th sampling time. This is the proportionality coefficient; The integral coefficient; These are the differential coefficients; The temperature deviation at the k-th sampling time is equal to the difference between the corrected target temperature value and the actual temperature value of the liquid medicine. For the first Temperature deviation at the sampling time; The preset sampling period is consistent with the sampling period used in the concentration outer loop calculation; The heating power control quantity is converted from digital to analog and output as a power adjustment signal, which drives the corresponding layered electric heating jacket to adjust the output power, thereby realizing closed-loop tracking control of the temperature zone.

6. The intelligent control method for temperature gradient in a traditional Chinese medicine extraction tank according to claim 1, characterized in that: In step S5, the gradient smoothing constraint algorithm satisfies the following temperature difference constraint relationship: ， in, This represents the actual temperature value of the i-th temperature zone at the k-th sampling time. The k-th sampling time is the longitudinally adjacent temperature zone i. The actual temperature value of each temperature zone; This is the preset maximum allowable temperature difference threshold; When the actual temperature difference between adjacent temperature zones exceeds the maximum allowable temperature difference threshold, a limiting adjustment is applied to the heating rate of the high-temperature zone. The limited heating rate satisfies the following: ， in, The heating rate of the i-th temperature zone before limiting; This represents the heating rate of the i-th temperature zone after limiting. This is the heating rate limiting coefficient; through the above constraints and adjustments, the continuous transition of the longitudinal temperature gradient inside the extraction tank is maintained, and the convection turbulence of the drug solution caused by sudden temperature changes is suppressed.

7. A temperature gradient intelligent control system for a traditional Chinese medicine extraction tank, applied to the method according to any one of claims 1-6, characterized in that, include: The layered heating unit includes multiple layers of electric heating jackets distributed longitudinally along the wall of the extraction tank. Each layer of electric heating jacket is independently controlled and is used to create multiple longitudinally distributed independent temperature zones inside the extraction tank. The multi-layer detection unit includes temperature detection elements and active ingredient concentration detection probes arranged for each temperature zone, which are used to collect real-time data on the temperature of the medicine solution and the concentration of the active ingredient in each temperature zone. The storage unit is used to store a database of extraction process parameters for Chinese medicinal materials, preset target dissolution curves, and control parameters; The main control unit has a built-in dual closed-loop control module and a gradient smoothing constraint module, which are used to perform dual closed-loop control calculations and gradient smoothing constraint calculations for the concentration outer loop and temperature inner loop, respectively. The power drive unit receives the power adjustment signal output by the main control unit and drives each layer of electric heating jacket to adjust the output power accordingly.

8. The intelligent control system for temperature gradient in a traditional Chinese medicine extraction tank according to claim 7, characterized in that: The active ingredient concentration detection probe is a near-infrared spectroscopy online detection probe, with at least one detection point set in each temperature zone.

9. The intelligent control system for temperature gradient in a traditional Chinese medicine extraction tank according to claim 7, characterized in that: A heat insulation layer is provided between two adjacent electric heating jackets, and the heat insulation layer is made of fire-resistant and heat-insulating fiber material.

Images