Supercritical fluid separation and extraction control method

Abstract

The invention discloses a supercritical fluid separation and extraction control method, and belongs to the technical field of computer-aided separation processes. The method is applied to separation and extraction equipment with the following structure: the separation and extraction equipment comprises an extraction tank, a separation tank assembly, a sound wave flow guide device and a heat exchange circulating system; the specific separation and extraction control method comprises the following steps: collecting material granularity, total amount, extraction tank segmented temperature values and interstage pressure difference parameters, and constructing feature vectors; predicting a temperature / pressure set value and extraction time of the separation tank component by using a neural network model, wherein the model input comprises feature vectors and supercritical fluid density; the temperature and the pressure of each stage of separation tank are regulated and controlled according to a prediction result, and a material movement path is optimized through a sound wave flow guide device. Through fusion of neural network prediction and MPC dynamic optimization, the extraction efficiency and the product purity are remarkably improved, the energy consumption is reduced, and the problem of insufficient control precision caused by parameter lag of traditional equipment is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer-aided separation process, in particular to a supercritical fluid separation and extraction control method. BACKGROUND

[0002] Supercritical extraction is an extraction process carried out under supercritical conditions, which is usually applied to extract active ingredients, spices, pigments, etc. from natural plants, animals or synthetic substances. Supercritical extraction takes advantage of the special properties of supercritical fluid. When a substance is in a supercritical state, it has properties similar to both gas and liquid, with low surface tension and high diffusivity of gas, and high density and solubility of liquid. This mixed gas-liquid state enables supercritical fluid to have high solubility and mass transfer rate, effectively dissolving target components and achieving efficient extraction and separation.

[0003] The patent for announcement No. CN118161876B (hereinafter referred to as prior art 1) discloses a supercritical fluid extraction and separation equipment for beta-glucan and a control method thereof, which comprises: an extraction tank for providing a containing space for the material to be extracted; a preheating tank arranged on one side of the extraction tank and communicating with the interior of the extraction tank for preheating the supercritical fluid; and a separation tank arranged above the preheating tank and communicating with the interiors of the extraction tank and the preheating tank respectively for separating the mixed supercritical fluid medium in the extraction tank.

[0004] However, in actual application, the single-stage separation tank of the prior art 1 cannot realize the molecular weight classification collection of beta-glucan and other components, and the temperature control relies on the static preheating tank, which cannot respond to the dynamic changes in the extraction process. SUMMARY

[0005] The present application aims to provide a supercritical fluid separation and extraction control method, which has the advantages of fine separation capability and dynamic temperature control in actual use, and can effectively solve the technical problems of insufficient separation precision and temperature control lag in the prior art.

[0006] To solve the above technical problems, the technical solution adopted by the present application is as follows:

[0007] A supercritical fluid separation and extraction control method applied to a separation and extraction equipment with the following structure, which comprises an extraction tank, a separation tank assembly, a sound wave guide flow device and a heat exchange circulation system; the interior of the extraction tank is divided into a bottom high-temperature zone, a middle constant-temperature zone and a top low-temperature zone, and heating modules and temperature sensors are independently arranged in the bottom high-temperature zone, the middle constant-temperature zone and the top low-temperature zone, respectively.

[0008] The specific separation and extraction control method is as follows:

[0009] Step 1: Collect material particle size, total amount, extraction tank segmented temperature value, interstage pressure difference parameters, and construct a feature vector;

[0010] Step 2: Use a neural network model to predict the temperature / pressure set value and extraction time of the separation tank assembly, the model input includes the feature vector and supercritical fluid density;

[0011] Step 3: Adjust the temperature and pressure of each stage of the separation tank according to the prediction results, and optimize the material movement path through the acoustic wave guide flow device;

[0012] Step 4: Real-time monitoring of system parameters, dynamic adjustment of control parameters through model predictive control algorithm;

[0013] The feature vector construction formula is:

[0014] Wherein, is the material particle size, is the total amount of material, ΔP1 and ΔP2 are the interstage pressure difference, , , is the extraction tank segmented temperature, is the supercritical fluid density.

[0015] Further, in the model predictive control algorithm, COMSOL multi-physics simulation is used to generate training data, and a neural network proxy model is used to replace traditional PDE calculation, achieving temperature field rotation accuracy control within ±0.5℃.

[0016] Further, the separation tank assembly includes a first-stage separation tank, a second-stage separation tank, and a third-stage separation tank, which are connected in series and independently provided with pressure control modules and temperature control modules;

[0017] An intelligent valve group is provided between the first-stage separation tank, the second-stage separation tank, and the third-stage separation tank to control the pressure gradient between the stages; an acoustic wave guide flow device is provided at the top of the extraction tank to emit directional acoustic pressure waves to control the material movement path; a heat exchange circulation system is used to transfer the waste heat from the first-stage separation tank to the high-temperature area at the bottom of the extraction tank.

[0018] Further, the heating module is an annular nano heating film array arranged on the outer wall of the extraction tank, which generates a rotatable temperature field through an independent temperature control unit, and the rotation direction is reverse and synchronous with the material movement direction.

[0019] Further, the sound wave guide flow device comprises a high-frequency sound wave transducer and a low-frequency sound wave transducer, the high-frequency sound wave transducer is used for emitting high-frequency sound waves and pushing fine particles to gather in the high-temperature area, and the low-frequency sound wave transducer is used for emitting low-frequency sound waves and driving coarse particles to suspend in the top low-temperature area.

[0020] Further, the heat exchange circulation system comprises a heat pump assembly, the heat pump assembly is used for transmitting the waste heat of the first separation tank to the high-temperature area at the bottom of the extraction tank, and the low-temperature output of the third separation tank is input into the low-temperature area at the top of the extraction tank after being pressurized by the heat pump assembly.

[0021] Further, the pressure of the first separation tank is 20-30 MPa, the pressure of the second separation tank is 10-20 MPa, and the pressure of the third separation tank is 5-10 MPa; the temperature setting range is that the temperature of the first separation tank is 50-70 DEG C, the temperature of the second separation tank is 40-55 DEG C, and the temperature of the third separation tank is 30-45 DEG C.

[0022] Further, the first separation tank, the second separation tank and the third separation tank are connected in series by using high-pressure corrugated pipes.

[0023] Further, the high-temperature area at the bottom of the extraction tank is provided with a spiral waste heat inlet, and the low-temperature area at the top of the extraction tank is embedded with a serpentine cooling coil.

[0024] Further, the sound wave guide flow device has 6 groups, wherein 4 groups are high-frequency sound wave transducers, 2 groups are low-frequency sound wave transducers, and the high-frequency sound wave transducers and the low-frequency sound wave transducers are arranged at intervals in a ratio of 2:1.

[0025] Further, the outlet of the first separation tank is connected with the heating area at the bottom of the extraction tank through a finned heat exchanger, and the low-temperature end of the third separation tank forms a closed loop with the low-temperature area at the top of the extraction tank through the heat pump assembly.

[0026] Further, the annular nano heating film array is an 8-part independent power supply structure, the resistance temperature coefficient (TCR) of a single heating unit is greater than or equal to 2000 ppm / DEG C, and a clockwise rotating temperature gradient field is generated through PWM pulse width modulation.

[0027] Compared with the prior art, the present application has the following beneficial effects:

[0028] This invention divides the extraction tank into a high-temperature zone at the bottom, a constant-temperature zone in the middle, and a low-temperature zone at the top, with independent thermal modules and temperature sensors. This allows for precise temperature control of different zones, meeting the temperature requirements of materials at different stages of the extraction process and improving extraction efficiency. The multi-stage series design of the separation tank components and the independent pressure and temperature control modules enable the gradual separation of different substances under different pressure and temperature conditions, improving separation purity. The acoustic wave guide device regulates the material movement path and optimizes the extraction process. The heat exchange circulation system recovers waste heat from the first-stage separation tank, achieving efficient energy utilization and reducing equipment operating costs. More importantly, the first stage separates large molecular impurities, the second stage collects the target product, and the third stage recovers small molecular components, achieving segmented collection of β-glucan molecular weight, effectively improving the precision compared to single-stage separation. This invention significantly improves extraction efficiency and product purity, reduces energy consumption, and solves the problem of insufficient control precision caused by parameter lag in traditional equipment by integrating neural network prediction and MPC dynamic optimization. Attached Figure Description

[0029] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0030] Figure 1 This is a schematic diagram of the overall principle of the present invention.

[0031] Figure 2 This is a block diagram illustrating the principle of the supercritical fluid extraction control method of the present invention.

[0032] Figure label:

[0033] 101 Extraction tank, 102 Separation tank assembly, 103 Acoustic wave guide device, 104 High-frequency acoustic transducer, 105 Low-frequency acoustic transducer, 106 Bottom high-temperature zone, 107 Middle constant-temperature zone, 108 Top low-temperature zone, 109 Heating module, 110 Temperature sensor, 111 Primary separation tank, 112 Secondary separation tank, 113 Tertiary separation tank, 114 Intelligent valve assembly, 115 Heat pump assembly, 116 Corrugated pipe, 117 Cooling coil. Detailed Implementation

[0034] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the embodiments of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive. The following is in conjunction with... Figure 1 and Figure 2Embodiments of the present application are described in detail.

[0035] Embodiment 1

[0036] The embodiment discloses a supercritical fluid separation and extraction control method, which is applied to a separation and extraction device with the following structure, the separation and extraction device comprising an extraction tank 101, a separation tank assembly 102, a sound wave guide flow device 103 and a heat exchange circulation system; wherein the extraction tank 101 is internally divided into a bottom high-temperature zone 106, a middle constant-temperature zone 107 and a top low-temperature zone 108, and heating modules 109 and temperature sensors 110 are independently arranged in the bottom high-temperature zone 106, the middle constant-temperature zone 107 and the top low-temperature zone 108 respectively.

[0037] The separation tank assembly 102 comprises a first-stage separation tank 111, a second-stage separation tank 112 and a third-stage separation tank 113, the first-stage separation tank 111, the second-stage separation tank 112 and the third-stage separation tank 113 are connected in series, and pressure control modules and temperature control modules are independently arranged in the first-stage separation tank 111, the second-stage separation tank 112 and the third-stage separation tank 113 respectively.

[0038] Intelligent valve groups 114 are arranged between the first-stage separation tank 111, the second-stage separation tank 112 and the third-stage separation tank 113, for controlling the pressure gradient between the separation tanks; the sound wave guide flow device 103 is arranged at the top of the extraction tank 101, for emitting directional sound pressure waves to control the material movement path; the heat exchange circulation system is used for transmitting the waste heat of the first-stage separation tank 111 to the bottom high-temperature zone 106 of the extraction tank 101.

[0039] The heating module 109 is an annular nano heating film array arranged on the outer wall of the extraction tank 101, the annular nano heating film array generates a rotatable temperature field through an independent temperature control unit, and the rotation direction is reversely synchronized with the material movement direction.

[0040] Further, the sound wave guide flow device 103 comprises a high-frequency sound wave transducer 104 and a low-frequency sound wave transducer 105, the high-frequency sound wave transducer 104 is used for emitting high-frequency sound waves and pushing fine particles to gather in the high-temperature zone, and the low-frequency sound wave transducer 105 is used for emitting low-frequency sound waves to drive coarse particles to float in the top low-temperature zone, the frequency of the high-frequency transducer is 100-150 kHz, and the frequency of the low-frequency transducer is 20-40 kHz.

[0041] The heat exchange circulation system comprises a heat pump assembly 115, the heat pump assembly 115 is used for transmitting the waste heat of the first-stage separation tank 111 to the bottom high-temperature zone 106 of the extraction tank 101, and the low-temperature output of the third-stage separation tank 113 is input into the top low-temperature zone 108 of the extraction tank 101 after being pressurized to 1.5-2 MPa by the heat pump assembly 115.

[0042] Wherein, the pressure of the first separation tank 111 is 20-30 MPa, the pressure of the second separation tank 112 is 10-20 MPa, and the pressure of the third separation tank 113 is 5-10 MPa; the temperature setting range is: the temperature of the first separation tank 111 is 50-70℃, the temperature of the second separation tank 112 is 40-55℃, and the temperature of the third separation tank 113 is 30-45℃.

[0043] Further, the high-pressure corrugated pipe 116 is used in series between the first separation tank 111, the second separation tank 112 and the third separation tank 113.

[0044] Further, the high-temperature area 106 at the bottom of the extraction tank 101 is provided with a spiral waste heat inlet, and the top low-temperature area 108 is embedded with a serpentine cooling coil 117.

[0045] Among them, the acoustic wave flow guide device 103 has 6 groups, among which 4 groups are high-frequency acoustic wave transducers and 2 groups are low-frequency acoustic wave transducers, and the high-frequency acoustic wave transducer 104 and the low-frequency acoustic wave transducer 105 are arranged at an interval of 2:1.

[0046] Further, the outlet of the first separation tank 111 is connected with the heating area at the bottom of the extraction tank 101 through a finned heat exchanger, and the low-temperature end of the third separation tank 113 is connected with the top low-temperature area 108 of the extraction tank 101 through a heat pump assembly 115 to form a closed loop.

[0047] It should be noted that in the embodiment, it is specifically divided into a high-temperature circuit and a low-temperature circuit.

[0048] High-temperature circuit: the waste heat fluid (about 60℃) of the outlet of the first separation tank 111 flows through the finned heat exchanger, and exchanges heat with the CO2 fluid entering the bottom of the extraction tank 101, so that the CO2 is raised from room temperature to 50℃.

[0049] Low-temperature circuit: the low-temperature fluid (about 30℃) of the outlet of the third separation tank 113 is pressurized to 1.8 MPa through the heat pump assembly 115 and then divided into two paths: the main path enters the refrigeration system to release cold energy, and the branch path inputs the serpentine cooling coil 117 of the top low-temperature area 108 of the extraction tank 101 through the closed loop pipeline to maintain the top low temperature.

[0050] Further, the annular nano heating film array is an 8-part independent power supply structure, the resistance temperature coefficient (TCR) of a single heating unit is ≥2000ppm / ℃, and a clockwise rotating temperature gradient field is generated through PWM pulse width modulation.

[0051] Among them, the acoustic wave parameter coupling rule of the acoustic wave flow guide device in the application is as follows:

[0052] 1. Basic frequency range:

[0053] High-frequency acoustic transducer 104: 100-150 kHz (preferably 130±5 kHz);

[0054] Low-frequency acoustic transducer 105: 20-40 kHz (preferably 30±5 kHz).

[0055] 2. The power matching formula is as follows:

[0056] Wherein is the average particle size of the material (μm), = 80 μm, = 800 W.

[0057] The specific separation and extraction control method is as follows:

[0058] Step 1: Collect the particle size, total amount, temperature value of the extraction tank 101, and interstage pressure difference parameters, and construct a feature vector;

[0059] Step 2: Use a neural network model to predict the temperature / pressure set value of the separation tank assembly and the extraction time, and the model input includes the feature vector and the supercritical fluid density;

[0060] Step 3: Adjust the temperature and pressure of each stage of the separation tank according to the prediction result, and optimize the material movement path through the acoustic flow guide device 103;

[0061] Step 4: Real-time monitoring of system parameters, dynamic adjustment of control parameters through model predictive control (MPC) algorithm.

[0062] Further, the feature vector construction formula is as follows:

[0063] Wherein, is the particle size of the material, is the total amount of the material, ΔP1 and ΔP2 are the interstage pressure difference, , , is the temperature of the extraction tank 101, is the density of the supercritical fluid.

[0064] Further, the neural network loss function is as follows:

[0065] Wherein, is the Gaussian weighted time loss term, is the purity loss term based on the pressure gradient, is the temperature and pressure coupling energy consumption loss term, and the weight / / Adaptively adjusted by a softmax function.

[0066] In the model predictive control (MPC) algorithm, training data is generated by COMSOL multi-physics simulation, a neural network proxy model is used to replace traditional PDE calculation, and temperature field rotation precision control is realized within ±0.5℃.

[0067] Further, in actual application, step 5 is further included, that is, the purity of β-glucan is monitored in real time by HPLC detection, when the purity fluctuation exceeds 5%, the model is retrained, and the neural network is incrementally trained by using newly collected data (including particle size, pressure, temperature, and purity).

[0068] After each 50 extraction cycles are completed, the boundary conditions of the COMSOL simulation model are automatically updated, the acoustic frequency combination strategy is optimized, and the switching logic of high frequency 125 / 135 kHz and low frequency 25 / 35 kHz is realized.

[0069] The neural network model is as follows:

[0070] Input layer: 7 nodes (feature vector dimension);

[0071] Hidden layer: 2 layers, 32 nodes in each layer (ReLU activation function);

[0072] Output layer: 3 nodes (three-stage separation tank temperature / pressure set value).

[0073] The model training data is generated based on COMSOL multi-physics simulation, 10,000 groups of data are generated, covering material particle size (10-200 μm), pressure (5-40 MPa), and temperature (30-80℃) full parameter space.

[0074] The boundary conditions are set as follows: the rotation temperature field speed of the extraction tank is 0.5-2 rpm, and the acoustic frequency is 25-150 kHz.

[0075] The PWM pulse width is adjusted in real time by a PID controller, and the average value of the 8-zone temperature sensor is used as the feedback signal; when the temperature difference between adjacent zones is >0.3℃, the power of the annular nano heating film is automatically compensated, and the compensation formula is ΔW=Kp×|ΔT|, Kp=15W / ℃.

[0076] In order to facilitate further understanding of the present application by those skilled in the art, the present embodiment is further described in detail below.

[0077] In this example, the extraction tank 101 body is made of 316L stainless steel material, the inner diameter is 600 mm, the height is 1800 mm, and the effective volume is 500 L; the extraction tank 101 adopts a three-section jacket structure: a bottom high-temperature zone 106 with a height of 500 mm, a middle constant-temperature zone 107 with a height of 800 mm, and a top low-temperature zone 108 with a height of 500 mm.

[0078] The outer wall of the extraction tank 101 is provided with 8 partitioned annular nano heating films with a thickness of 80 nm and a single-zone power of 15 kW, which are powered by independent constant current sources, model DR-60-300, with a current accuracy of ±0.2%.

[0079] The temperature sensor 110 is configured as follows: three groups of PT100 sensors are arranged in each temperature zone, with an accuracy of ±0.1°C, distributed in a 120° circle, and a sampling frequency of 10 Hz. The pressure sensor is a Rosemount 3051 type with a range of 0-40 MPa and an accuracy of ±0.075%, installed at the top, middle and bottom of the extraction tank 101.

[0080] Among them, the first-stage separation tank 111 has an inner diameter of 400 mm, a height of 1200 mm, a volume of 150 L, a working pressure of 25 MPa, and a temperature of 60°C.

[0081] The second-stage separation tank 112 has an inner diameter of 350 mm, a height of 1000 mm, a volume of 100 L, a working pressure of 15 MPa, and a temperature of 45°C.

[0082] The third-stage separation tank 113 has an inner diameter of 300 mm, a height of 800 mm, a volume of 60 L, a working pressure of 8 MPa, and a temperature of 35°C.

[0083] DN25 high-pressure corrugated pipe 116 is used between the first-stage separation tank 111, the second-stage separation tank 112 and the third-stage separation tank 113, equipped with an electric liquid servo proportional regulating valve (intelligent valve group), model FESTOMPPE-3-1 / 8, flow range 0-15 L / min, response time ≤80 ms.

[0084] Six groups of ultrasonic transducers (acoustic wave guide devices 103) are installed in an array at the top, four groups of 130 kHz high-frequency transducers are installed in the center area of the tank top to emit focused sound beams, and two groups of 30 kHz low-frequency transducers are installed in the edge area of the tank top to form a ring-shaped standing wave field; in actual application, the ultrasonic generator is connected to the ultrasonic transducer through an optical fiber transmission system.

[0085] In the heat exchange circulation system, the outlet of the first-stage separation tank 111 is connected to a finned heat exchanger, and the low-temperature outlet of the third-stage separation tank 113 is connected to a heat pump assembly 115. The heat exchange pipeline is made of DN50 stainless steel pipe with a wall thickness of 5 mm and a pressure resistance of 30 MPa, equipped with an electric regulating valve, model SAMSON 3271, with a regulating range of 0-50 m³ / h.

[0086] The material (average particle size 80 μm, water content ≤5%) is transported to the extraction tank 101 by a vacuum feeding machine, the extraction tank 101 is closed, and the pre-evacuation system is started to a tank pressure ≤0.5 mbar.

[0087] The valve of the extraction tank 101 is opened, liquid CO2 is injected into the extraction tank 101 by a high-pressure plunger pump, the temperature is raised to 45°C, and the pressure is raised to 28 MPa to form a supercritical state; during injection, the initial stage is 20 L / min, and it is reduced to 5 L / min when approaching the target pressure to avoid pressure impact.

[0088] When dynamic extraction is performed, the annular nano-heating film array is started to generate a clockwise rotating temperature field with a rotating speed of 1.2 rpm; at the same time, the acoustic flow guide device 103 is started, the high-frequency transducer emits 130 kHz acoustic waves, and the fine particles with d < 50 μm obtain a directional migration speed of 0.2 m / s; the low-frequency transducer emits 30 kHz acoustic waves, and the suspension time of coarse particles with d > 100 μm is prolonged to more than 30 minutes.

[0089] Under the synergistic action of the rotating temperature field and the acoustic flow guide, a high-temperature, high-pressure, and high-frequency acoustic wave reinforced extraction environment is formed, and the actual detection shows that the dissolution rate of β-glucan is increased to 95%.

[0090] When multi-stage separation is performed, the extraction liquid enters the first separation tank 111 through the top pipeline, the pressure is reduced to 25 MPa, and the temperature is maintained at 60°C, at which time the macromolecular protein (molecular weight > 200 kDa) is first precipitated.

[0091] The first separation liquid enters the second separation tank 112 through the inter-stage valve (depressurization rate 0.8 MPa / min), the pressure is reduced to 15 MPa, and the temperature is reduced to 45°C, at which time the target product β-glucan (molecular weight 50-100 kDa) is precipitated in large quantities.

[0092] The second separation liquid continues to pass through the inter-stage valve (depressurization rate 0.3 MPa / min) into the third separation tank 113, the pressure is reduced to 8 MPa, and the temperature is reduced to 35°C, at which time the remaining small molecular oligosaccharides (molecular weight < 50 kDa) are precipitated.

[0093] Among them, the first separation tank 111, the second separation tank 112, and the third separation tank 113 are all equipped with automatic discharge valves for discharging.

[0094] In the waste heat recovery stage, the outlet fluid temperature of the first separation tank 111 is about 60°C, and the heat is transferred to the CO2 fluid entering the extraction tank 101 through a heat exchanger, so that the temperature of the CO2 fluid is raised from room temperature to about 50°C;

[0095] The fluid temperature at the outlet of the third separation tank 113 is about 30℃, and after being pressurized to 1.8 MPa by the heat pump, heat is transferred to the low-temperature area at the top of the extraction tank, while the top area of the extraction tank 101 is cooled to 35℃;

[0096] The entire heat exchange circulation system is controlled in a closed loop by a PLC, and the temperature control accuracy is ±0.5℃. The overall control accuracy of the system is ±0.5℃, and the independent accuracy of the temperature sensor is ±0.1℃.

[0097] After testing, the indicators in the prior art 1 are as follows:

[0098] This embodiment realizes accurate segmented collection of the molecular weight of β-glucan by precisely controlling the pressure and temperature of each stage of the separation tank, and the molecular weight concentration is improved by 45% compared with the prior art. The synergistic effect of the rotating temperature field and the acoustic wave guide flow improves the mass transfer efficiency by 42%, shortens the extraction time by 30%, and the application of the heat exchange circulation system reduces the unit energy consumption by 36%.

[0099] In example 2, when processing heat-sensitive materials (such as ginsenosides), the following protective parameter strategy is adopted:

[0100] Separation tank pressure / temperature down-regulation: the pressure of the first separation tank 111 is ≤22 MPa, and the temperature is ≤55℃;

[0101] Acoustic wave frequency adaptation: the low-frequency transducer 105 can be reduced to 25 kHz;

[0102] The extraction tank temperature field rotation speed is reduced to 0.8 rpm to avoid thermal degradation.

[0103] In this embodiment, the device parameters are adjusted as follows:

[0104] Extraction tank 101 temperature configuration: bottom high-temperature area 55℃, middle constant-temperature area 40℃, top low-temperature area 32℃;

[0105] Acoustic wave parameter optimization: high-frequency transducer power is reduced to 700W (sound pressure level 120dB), and low-frequency transducer frequency is adjusted to 25kHz;

[0106] Third separation tank 113 parameters: first stage (22 MPa, 55℃), second stage (12 MPa, 40℃), and third stage (6 MPa, 30℃);

[0107] An adaptive weight neural network model is used, and the purity loss term weight β is improved to 0.6.

[0108] 50 kg of ginseng extract with an average particle size of 60 μm was mixed with 20 kg of entrainer (ethanol / water = 9:1), mixed well by a static mixer, and then injected into the extraction tank 101. A stepwise pressure increase strategy was adopted, with a 0.5 MPa / min pressure increase rate at 0-15 MPa and a 0.2 MPa / min pressure decrease rate at 15-28 MPa to avoid thermal sensitive component degradation caused by pressure shock.

[0109] The temperature field rotation speed was reduced to 0.8 rpm, and the acoustic wave guide flow adopted an intermittent working mode (working for 30 s and stopping for 10 s) to avoid damage to active ingredients caused by long-time ultrasound; the extraction liquid temperature was monitored in real time, and the temperature was controlled below 40°C by the top cooling system.

[0110] The separation process was optimized as follows:

[0111] The pressure decrease rate of the first separation tank 111 was reduced to 0.5 MPa / min; the second separation tank 112 adopted a gradient pressure decrease strategy, first quickly decreased to 15 MPa (rate 0.3 MPa / min), and then slowly decreased to 12 MPa (rate 0.1 MPa / min) to improve the selective precipitation of saponin components; the third separation tank 113 increased the stirring device (rotation speed 30 rpm) to promote the complete separation of small molecule components.

[0112] In this embodiment and the conventional constant temperature extraction, the specific indicators are as follows:

[0113] Through the synergistic effect of segmented temperature control and acoustic wave guide flow, efficient extraction of thermal sensitive components was realized under low temperature conditions, and the activity retention rate was improved. At the same time, by adopting a segmented pressure decrease strategy, the selective precipitation of target components was significantly improved, and the purity was improved.

[0114] Example 3, this embodiment is mainly to verify the self-learning ability of the intelligent control system, and the experimental design is as follows:

[0115] 50 batches of β-glucan extraction experiments were continuously carried out, and the particle size of each batch of material randomly fluctuated (60-120 μm);

[0116] Initial model parameters: α = 0.3, β = 0.5, γ = 0.2;

[0117] Triggering condition: when the β-glucan purity fluctuation of 3 consecutive batches exceeds 5%, the online learning mechanism is started.

[0118] The overall response process of the system is as follows:

[0119] Batch 1-20: The material particle size is relatively stable (average 80 μm), the system runs stably, the β-glucan purity is maintained at 98.5±0.5%, the model predicts the temperature field rotation speed as 1.2 rpm, and the acoustic wave power combination is high frequency 130 kHz x 700 W + low frequency 30 kHz x 500 W;

[0120] Batch 21-30: The material particle size gradually increases (average 105 μm), the β-glucan purity starts to decrease, and the lowest decreases to 93.2%; the system triggers the abnormal detection mechanism, and automatically collects and analyzes the data of nearly 10 batches.

[0121] Model diagnosis result: The current parameter configuration is insufficient for the extraction efficiency of large particle materials, and the acoustic wave parameters need to be adjusted.

[0122] Batch 31-40: The system automatically updates the control parameters: the temperature field rotation speed is increased to 1.5 rpm, the acoustic wave power combination is adjusted to high frequency 135 kHz x 800 W + low frequency 25 kHz x 600 W; at the same time, the HPLC online detection system is started, and the detection period is shortened to every 10 minutes; the β-glucan purity gradually rises to 97.8%, proving the effectiveness of the parameter adjustment.

[0123] Batch 41-50: The system automatically triggers the model incremental training, optimizes the neural network using the newly collected 200 groups of data, and the data collection covers material particle size, pressure gradient, temperature field rotation speed and real-time purity feedback; the updated model weight is adjusted to α=0.25, β=0.6, γ=0.15, and the purity loss term weight is further improved; finally, the β-glucan purity is stable at 98.2±0.3%, proving the self-learning ability of the system.

[0124] The specific experimental data is as follows:

[0125] By establishing a purity fluctuation warning model, real-time monitoring of the system running state is realized, the abnormal response time is shortened by 50%, the system can automatically adjust the control parameters according to the material characteristics, so that the extraction effect of different particle size materials reaches the best state, through the online learning mechanism, the system can continuously optimize the control strategy, and the long-term running stability is improved by 60%.

[0126] Example 4, this example is mainly used to realize the suitability verification of special materials, and the experimental material is ganoderma spore powder with an average particle size of 10 μm.

[0127] Adjust the device parameters as follows: the extraction tank extraction pressure is 30 MPa, the temperature is 50℃, and the acoustic wave guide flow high frequency power is increased to 900 W.

[0128] The specific test results are as follows:

[0129] For ganoderma spore powder and other microparticle materials, the extraction efficiency and active ingredient retention rate are significantly improved by optimizing the sound wave parameters and temperature field control. Through precise control of the extraction conditions, the synergistic extraction of ganoderma triterpenes and polysaccharides is realized, proving the universality of the present invention technology.

[0130] It should be noted that in Examples 2-4, improvements are made on the basis of the examples, such as adjustment of equipment parameters.

[0131] Although the preferred embodiments of the present application have been described, those skilled in the art can make further changes and modifications to these embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0132] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. It should be noted that any modifications, equivalent replacements and improvements made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A supercritical fluid separation and extraction control method applied to a separation and extraction apparatus of the following structure, the separation and extraction apparatus comprising an extraction tank, a separation tank assembly, an acoustic waveguide flow device, and a heat exchange circulation system; characterized in that: The extraction tank is divided into a bottom high-temperature zone, a middle constant-temperature zone and a top low-temperature zone, and heating modules and temperature sensors are independently arranged in the bottom high-temperature zone, the middle constant-temperature zone and the top low-temperature zone, respectively. The specific separation and extraction control method is as follows: Step 1: Collect the material particle size, total amount, extraction tank segmented temperature value and interstage pressure difference parameters to construct a feature vector; Step 2: Use a neural network model to predict the temperature / pressure set value of the separation tank assembly and the extraction time, wherein the model input includes the feature vector and the supercritical fluid density; Step 3: Adjust the temperature and pressure of each stage of the separation tank according to the prediction result, and optimize the material movement path through the acoustic wave flow guide device; Step 4: Real-time monitor the system parameters and dynamically adjust the control parameters through the model predictive control algorithm; The feature vector construction formula is: wherein, is the material particle size, is the total amount of material, ΔP1, ΔP2 is the inter-stage pressure difference, , , is the extraction tank segment temperature, is the supercritical fluid density.

2. A supercritical fluid separation extraction control method according to claim 1, wherein: In the model predictive control algorithm, training data is generated by COMSOL multi-physics simulation, and a neural network proxy model is used to replace traditional PDE calculation to realize temperature field rotation accuracy control within ±0.5℃.

3. The method of claim 1 wherein: The separation tank assembly includes a first-stage separation tank, a second-stage separation tank and a third-stage separation tank, which are connected in series and independently provided with pressure regulation modules and temperature regulation modules, respectively. An intelligent valve group is arranged between the first-stage separation tank, the second-stage separation tank and the third-stage separation tank to control the pressure gradient between the stages; an acoustic wave flow guide device is arranged at the top of the extraction tank to emit directional acoustic pressure waves to control the material movement path; and a heat exchange circulation system is used to transfer the waste heat of the first-stage separation tank to the bottom high-temperature zone of the extraction tank.

4. The method of claim 3 wherein: The heating module is a ring-shaped nano heating film array arranged on the outer wall of the extraction tank, which generates a rotatable temperature field through an independent temperature control unit, and the rotation direction is opposite to the material movement direction.

5. The method of claim 3 wherein: the supercritical fluid separation extraction control method further comprises: determining a plurality of extraction parameters; and determining a plurality of extraction conditions. The acoustic wave flow guide device includes a high-frequency acoustic wave transducer and a low-frequency acoustic wave transducer, the high-frequency acoustic wave transducer is used to emit high-frequency acoustic waves to push fine particles to the high-temperature zone, and the low-frequency acoustic wave transducer is used to emit low-frequency acoustic waves to drive coarse particles to float in the top low-temperature zone.

6. The method of claim 3 wherein: The heat exchange circulation system includes a heat pump assembly, which is used to transfer the waste heat of the first-stage separation tank to the bottom high-temperature zone of the extraction tank, and to input the low-temperature output of the third-stage separation tank into the top low-temperature zone of the extraction tank after being pressurized by the heat pump assembly.

7. The method of claim 3 wherein: the supercritical fluid separation extraction control method further comprises: determining a plurality of extraction parameters; and determining a plurality of extraction conditions. The pressure of the first-stage separation tank is 20-30MPa, the pressure of the second-stage separation tank is 10-20MPa, and the pressure of the third-stage separation tank is 5-10MPa; the temperature setting range is: the temperature of the first-stage separation tank is 50-70℃, the temperature of the second-stage separation tank is 40-55℃, and the temperature of the third-stage separation tank is 30-45℃.

8. The method of claim 3 wherein: the supercritical fluid separation extraction control method further comprises: determining a plurality of extraction parameters; and determining a plurality of extraction conditions. High-pressure corrugated pipes are used to connect the first-stage separation tank, the second-stage separation tank and the third-stage separation tank.

9. The method of claim 3 wherein: the supercritical fluid separation extraction control method further comprises: determining a plurality of extraction parameters; and determining a plurality of extraction conditions. A spiral waste heat inlet is arranged at the bottom of the extraction tank, and a serpentine cooling coil is embedded in the top low-temperature zone.

Images