Efficient spray drying process for plant extract

Abstract

The invention relates to the technical field of non-electrical variable control systems, in particular to an efficient spray drying process for plant extracts, which comprises the following steps: correcting wet bulb temperature according to real-time atomization pressure of a tower top, determining a real-time wet bulb temperature correction value, determining a real-time tower top temperature difference value in combination with an initial Tg value, and performing spray drying on the plant extracts according to the real-time tower top temperature difference value. The system is used for implementing closed-loop control on the inlet air temperature of a tower top constant-speed drying area, determining the real-time water content in a tower according to the difference between a real-time material actually-measured water content detection value and a water content speculative value, obtaining a real-time Tg value corresponding to the water content in the tower, determining a real-time temperature difference value of a speed-reduction curing area by combining the real-time material temperature of the speed-reduction curing area, and controlling the temperature of the tower top constant-speed drying area according to the real-time temperature difference value. The method is used for carrying out closed-loop control on the inlet air temperature of a tower-bottom speed reduction curing area in the middle of the tower. Through self-adaptive control of the efficient spray drying process, the production stability is high, the energy consumption is lower, and the product quality is improved.

Description

Technical Field

[0001] This invention relates to the field of control systems technology for non-electrical variables, and more specifically to a high-efficiency spray drying process for plant extracts. Background Technology

[0002] Plant extracts, as core raw materials in food, pharmaceuticals, and cosmetics, are experiencing continuous global market expansion. Spray drying, due to its high efficiency, continuous processing, and suitability for large-scale production, has become the mainstream process for their powdering. However, plant extract complexes contain low-molecular-weight components such as polysaccharides and polyphenols, resulting in low Tg (Glass Transition Temperature) that dynamically changes with composition and moisture content. During spray drying, these complexes are prone to agglomeration due to material softening and adhesion to the drying walls, leading to low product yield and uneven quality. This has become a key bottleneck restricting efficient production in the industry.

[0003] Existing problems: Traditional spray drying processes for plant extracts often employ a fixed combination of parameters, including inlet air temperature, atomization pressure, and feed rate, relying on the addition of carriers such as maltodextrin and starch to increase the material's heat of growth (Tg) and improve drying characteristics. This process does not address the dynamic changes in the Tg of plant extracts. During drying, the material is prone to entering a viscous flow state due to temperature and moisture fluctuations, leading to adhesion to the tower walls and particle agglomeration. This not only reduces product yield but also results in poor powder flowability and decreased solubility. Furthermore, excessive carrier addition dilutes the active ingredients, making it difficult to meet the drying requirements of high-purity, high-quality plant extracts. Summary of the Invention

[0004] This invention provides a highly efficient spray drying process for plant extracts to solve existing problems.

[0005] The high-efficiency spray drying process for plant extracts of the present invention adopts the following technical solution: One embodiment of the present invention provides a highly efficient spray drying process for plant extracts, the process comprising the following steps: Obtain the initial Tg value of the plant extract; The system acquires the real-time inlet air temperature, inlet air humidity, and atomization pressure at the top of the tower; acquires the wet-bulb temperature corresponding to the inlet air temperature and inlet air humidity; corrects the wet-bulb temperature based on the atomization pressure to determine a real-time wet-bulb temperature correction value; determines the real-time temperature difference at the top of the tower based on the wet-bulb temperature correction value and the initial Tg value; and implements closed-loop control of the inlet air temperature in the constant-rate drying zone at the top of the tower based on the temperature difference at the top of the tower. The system acquires real-time measured moisture content and estimated moisture content of the material; determines the real-time moisture content in the tower based on the difference between the measured and estimated moisture content; acquires the real-time Tg value corresponding to the moisture content in the tower; acquires the real-time temperature in the tower and the outlet air temperature at the bottom of the tower; adjusts the difference between the temperature in the tower and the outlet air temperature at the bottom of the tower using the moisture content in the tower, and determines the real-time material temperature in the slow-deceleration curing zone based on the outlet air temperature at the bottom of the tower; determines the real-time temperature difference in the slow-deceleration curing zone based on the material temperature in the slow-deceleration curing zone and the real-time Tg value; and implements closed-loop control of the inlet air temperature in the slow-deceleration curing zone from the middle to the bottom of the tower based on the temperature difference in the slow-deceleration curing zone.

[0006] Furthermore, the specific steps for determining the real-time wet-bulb temperature correction value are as follows: If the atomization pressure is greater than a preset pressure threshold, the wet-bulb temperature is compensated according to the amount by which the atomization pressure exceeds the preset pressure threshold to obtain a real-time wet-bulb temperature correction value. If the atomization pressure is less than or equal to a preset pressure threshold, then the real-time wet-bulb temperature correction value is set to equal the wet-bulb temperature.

[0007] Furthermore, the specific steps for compensating the wet-bulb temperature based on the amount by which the atomization pressure exceeds a preset pressure threshold to obtain a real-time wet-bulb temperature correction value are as follows: Calculate the difference between the atomization pressure and the preset pressure threshold, multiply the difference by the preset correction coefficient to obtain a compensation term, subtract the compensation term from the wet-bulb temperature, and record the result as the real-time wet-bulb temperature correction value.

[0008] Furthermore, the specific steps for determining the real-time temperature difference at the top of the tower are as follows: The difference between the initial Tg value and the preset tower top safety threshold is recorded as the real-time tower top target temperature. The difference between the real-time wet-bulb temperature correction value and the target temperature at the top of the column is recorded as the real-time temperature difference at the top of the column.

[0009] Furthermore, the specific steps for determining the real-time water content in the tower are as follows: The real-time moisture content deviation is determined based on the difference between the measured moisture content of the material and the estimated moisture content. If the real-time moisture content deviation is less than or equal to the preset deviation threshold, then the real-time moisture content in the tower is set to equal the estimated moisture content. If the real-time moisture content deviation is greater than the preset deviation threshold, the real-time moisture content in the tower is determined based on the difference between the measured moisture content of the material and the estimated moisture content.

[0010] Furthermore, the specific steps for determining the real-time moisture content deviation are as follows: Calculate the absolute value of the difference between the measured moisture content of the material and the estimated moisture content, calculate the ratio of the absolute value of the difference to the measured moisture content of the material, convert the ratio into a percentage, and record it as the real-time moisture content deviation.

[0011] Furthermore, the specific steps for determining the real-time moisture content in the tower based on the difference between the measured moisture content and the estimated moisture content of the material are as follows: Calculate the ratio of the measured moisture content of the material at the previous sampling time to the estimated moisture content, and multiply the ratio by the real-time estimated moisture content to record the real-time moisture content in the tower.

[0012] Furthermore, the specific steps for determining the real-time material temperature in the slow-curing zone are as follows: Calculate the difference between the temperature in the tower and the temperature of the air outlet at the bottom of the tower. Then calculate the product of the normalized value of the water content in the tower and the difference. The sum of the product and the temperature of the air outlet at the bottom of the tower is recorded as the real-time temperature of the material in the slow-down curing zone.

[0013] Furthermore, the specific steps for determining the real-time temperature difference in the slow-curing zone are as follows: Based on the real-time Tg value, the real-time target temperature of the slow-curing zone is determined; The difference between the material temperature in the slow-curing zone and the target temperature in the slow-curing zone is recorded as the real-time temperature difference in the slow-curing zone.

[0014] Furthermore, the specific steps for determining the real-time target temperature of the slow-curing zone are as follows: The difference between the real-time Tg value and the preset safety threshold in the tower is recorded as the real-time target temperature of the slow-down curing zone.

[0015] The beneficial effects of the technical solution of the present invention are: In this embodiment of the invention, the real-time inlet air temperature and wet-bulb temperature corresponding to the inlet air humidity at the top of the tower are corrected according to the real-time atomization pressure at the top of the tower to determine the real-time wet-bulb temperature correction value. Combined with the initial Tg value, the real-time temperature difference at the top of the tower is determined to implement closed-loop control of the inlet air temperature in the constant-rate drying zone at the top of the tower. According to the real-time Tg value corresponding to the moisture content in the tower, combined with the real-time material temperature in the falling-rate curing zone, the real-time temperature difference in the falling-rate curing zone is determined to implement closed-loop control of the inlet air temperature in the falling-rate curing zone from the middle to the bottom of the tower. This invention achieves precise matching between drying temperature and material state through zoned temperature control of the spray drying tower, real-time prediction of material glass transition temperature, online calculation of moisture content, and closed-loop adaptive regulation. This effectively prevents materials from entering a viscous flow state, significantly improves the problems of wall adhesion and particle agglomeration within the tower, and increases product yield. It can greatly reduce or eliminate the need for carriers such as maltodextrin and starch, avoids dilution of active ingredients, improves the purity of plant extracts, and results in powders with better flowability and solubility, stronger batch stability. At the same time, it reduces the thermal degradation of heat-sensitive active ingredients, improves product quality, and achieves intelligent adaptive control of the entire process. This results in high production stability, lower energy consumption, and is more suitable for the industrial drying production of high-purity, high-quality plant extracts. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the steps of the efficient spray drying process for plant extracts according to the present invention. Figure 2 This is a flow chart of the spray drying process in a spray drying tower. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the high-efficiency spray drying process for plant extracts proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific solution of the high-efficiency spray drying process for plant extracts provided by this invention.

[0021] Please see Figure 1 The diagram illustrates a flowchart of a highly efficient spray drying process for plant extracts according to an embodiment of the present invention, the process comprising the following steps: Step S001: Obtain the initial Tg value of the plant extract.

[0022] In this embodiment, the spray drying tower is first divided into two independent temperature control zones: a constant-rate drying zone at the top of the tower and a decreasing-rate curing zone in the middle to bottom of the tower. Then, a pre-trained CNN (Convolutional Neural Network) model is used to predict the Tg value of the material at different stages in real time. Combined with the humidity difference between the incoming air and the tower, the moisture content of the material is estimated online through material balance. Based on the difference between the actual temperature of the material and the Tg threshold, the input temperature of each temperature zone is dynamically adjusted by a PID controller. At the same time, the model and parameters are optimized through closed-loop feedback of product agglomeration rate to achieve process self-adaptation.

[0023] It should be noted that Tg is the glass transition temperature, the critical temperature at which an amorphous material transitions from a hard, brittle, non-sticky, and stable glassy state to a soft, sticky, and easily agglomerated rubbery state. For plant extract composite systems, Tg is not simply an average of the components; it is significantly lowered by components with low Tg and moisture. Whether the material temperature exceeds Tg directly determines whether the powder sticks to the wall, absorbs moisture, or agglomerates, and is the only thermodynamic critical basis for temperature control in spray drying. In plant extract composite systems, the component ratio and moisture content directly affect Tg; therefore, the initial Tg is first predicted based on the component ratio and moisture content. The specific steps for raw material pretreatment and initial Tg prediction are as follows: (1) Determination of component percentage: In plant extracts, the proportion of polysaccharides, flavonoids, alkaloids, and other components directly affects the Tg value (the higher the polysaccharide content, the lower the Tg). The component proportion is one of the core input parameters for CNN to predict Tg.

[0024] In this embodiment, high-performance liquid chromatography (HPLC) was used to determine the component proportions of the plant extract complex system: 10 g of dried plant extract sample was taken and ultrasonically extracted with 50% ethanol solution for 30 min, filtered, and then diluted to 100 mL. HPLC parameters were set as follows: C18 column (250 mm × 4.6 mm), column temperature 30 °C, mobile phase methanol-water gradient elution (methanol content 20% to 50% from 0 to 10 min), flow rate 1 mL / min, polysaccharide was detected by differential refractive index detector, and flavonoids were detected by 254 nm ultraviolet detector. The mass proportions of each component were calculated by standard comparison method, which is a well-known technique and the specific method is not described here.

[0025] The composition of the plant extract complex system can be determined from the HPLC analysis results, such as the proportion of polysaccharides. flavonoid ratio alkaloid ratio Etc., keep two decimal places.

[0026] (2) Determine the initial moisture content: The water content (IE, or initial water content) in plant extracts affects the Tg value, and the Tg value changes with the water content.

[0027] In this embodiment, a halogen moisture analyzer is used to determine the initial moisture content: 2 to 3 g of uniform sample is spread evenly on the sample tray of the instrument. The instrument temperature is set to 105°C, and the measurement is started until the weight stabilizes (weight change ≤ 0.001 g for 30 consecutive seconds). The instrument automatically calculates the initial moisture content; this is a known technique, and the specific method will not be described here.

[0028] Obtain the initial moisture content Round to one decimal place.

[0029] (3) Predict the initial Tg value using a pre-trained CNN model: CNN models are trained with a large amount of offline Tg data, which can accurately capture the non-linear relationship between component ratio and moisture content and Tg value, and can quickly and accurately predict the Tg value of materials with different components and moisture content.

[0030] The training set contains more than 1,000 sets of offline Tg data corresponding to different parameters, which are measured by differential scanning calorimetry.

[0031] Single data point in the training set , , , Input the pre-trained CNN model for training, and the model outputs the initial Tg value. (Unit: °C), the model input-output relationship is as follows:

[0032] in The input to the trained convolutional neural network model is the proportion of each component and the initial water content, and the output is the initial Tg value. This is a well-known technique, and the specific method will not be introduced here.

[0033] The pre-collected component ratio data of the plant extracts and the initial water content are input into the trained convolutional neural network model to obtain the initial Tg value. .

[0034] Step S002: Obtain the real-time inlet air temperature, inlet air humidity, and atomization pressure at the top of the tower; obtain the wet-bulb temperature corresponding to the inlet air temperature and the inlet air humidity; correct the wet-bulb temperature according to the magnitude of the atomization pressure to determine the real-time wet-bulb temperature correction value; determine the real-time temperature difference at the top of the tower according to the magnitude of the wet-bulb temperature correction value and the initial Tg value; implement closed-loop control of the inlet air temperature of the constant-speed drying zone at the top of the tower according to the temperature difference at the top of the tower.

[0035] It should be noted that: the initial Tg was determined based on the plant extract data before spray drying. Further, closed-loop temperature control was implemented in the constant-rate drying zone at the top of the column. The constant-rate drying zone at the top of the column is the initial drying stage where the plant extract droplets come into contact with hot air after atomization. The process principle is that the free water on the droplet surface evaporates rapidly and steadily, locking the particle temperature at the wet-bulb temperature due to evaporation and heat dissipation. The heat and mass transfer rates remain constant. During this process, due to the extremely high water content of the extract, the real-time Tg is extremely low and it is in a high-viscosity rubber state under the effects of water plasticization and the shift of low-Tg components. As the surface free water continues to be removed, the particle water content slowly decreases, and the system Tg increases slightly and steadily without forming a dense hard shell. The core purpose of temperature control in this zone is to control the heating intensity of the hot air, preventing the droplets from instantly forming a shell and burying water due to excessively high temperature, or from sticking to the wall due to insufficient heating and prolonged low-Tg, high-viscosity conditions, thus ensuring stable drying.

[0036] It should be further noted that in the constant-rate drying zone at the top of the tower, the inlet air temperature and humidity directly affect the material surface temperature, while the atomization pressure affects the droplet size; the higher the pressure, the smaller the droplets and the faster the evaporation. Since the material surface temperature is close to the wet-bulb temperature during the constant-rate drying stage, the inlet air temperature and humidity, as well as the atomization pressure, directly determine the material temperature. Therefore, by measuring the inlet air temperature and humidity and the atomization pressure in this zone, the material temperature in this temperature range can be obtained. Adjusting the inlet air temperature in this zone ensures that the material temperature is within a reasonable range from the initial Tg, thereby improving the evaporation stability of the plant extract droplets.

[0037] Preferably, in one embodiment of the present invention, the method for implementing closed-loop control of the inlet air temperature of the constant-speed drying zone at the top of the tower includes: (1) Multiple sensors collect inlet air temperature and humidity and atomization pressure: In this embodiment, 3 to 5 temperature and humidity sensors are evenly arranged in the air inlet duct at the top of the tower, and 1 pressure sensor is arranged at the atomizer outlet. Data is collected every 1 second. The temperature and humidity data are smoothed using a moving average method (taking the average of 10 consecutive sets of data, which is a known technique, and the specific method will not be described here). The real-time temperature, humidity, and atomization pressure data collected by each sensor are obtained. The data are aligned using timestamps, and the temperature and humidity data from multiple sensors are fused using a weighted average method to obtain the real-time air inlet temperature. The stable airflow region at the top of the tower was determined using CFD (Computational Fluid Dynamics, a well-known technique; specific methods will not be described here) simulation. The weights of the temperature and humidity sensors in the stable airflow region were set to 0.3, while those in other regions were set to 0.175. Using this as an example, the weighted average formula is as follows:

[0038] In the formula, For the first Real-time temperature data from a temperature and humidity sensor. The number of temperature and humidity sensors. For the first The weight of each temperature and humidity sensor.

[0039] According to the real-time intake air temperature The method of obtaining the real-time intake air humidity. .

[0040] This allows us to obtain the real-time air intake temperature at the top of the tower. (Unit: °C), Inlet air humidity (Unit: g / kg dry air, mass of water vapor (g) contained in 1 kg of absolutely dry air), atomization pressure) (Unit: MPa (megapascal)).

[0041] It should be noted that in this embodiment, the sensor is calibrated monthly using a standard temperature and humidity calibrator to eliminate system errors.

[0042] (2) Calculate the real-time temperature of the material: It should be noted that during the constant-rate drying stage, the surface temperature of the material is close to the wet-bulb temperature; therefore, the material temperature can be directly represented by the wet-bulb temperature. Under normal atmospheric pressure, the wet-bulb temperature can be directly obtained from a wet-air enthalpy-humidity chart, specifically: Based on the real-time air intake temperature at the top of the tower With air intake humidity The corresponding real-time wet-bulb temperature can be directly found in the enthalpy-humidity chart of moist air. The enthalpy-humidity diagram of moist air is a well-known technique, and its specific method will not be described here.

[0043] It should be noted that during the constant-rate drying stage, the atomization pressure will affect the droplet evaporation rate, so the wet-bulb temperature should be corrected according to the atomization pressure.

[0044] Preset pressure threshold The pressure is 0.2 MPa, and the preset correction factor is [value missing]. The value is 10℃ / MPa (degrees Celsius per megapascal), and this will be used as an example for explanation.

[0045] If the real-time atomization pressure at the top of the tower greater than the preset pressure threshold The real-time wet-bulb temperature correction value .in, This is a compensation item.

[0046] It should be noted that: when Above 0.2 MPa, the droplets are smaller, evaporation is faster, and the material temperature is higher. Lower by 1 to 2 degrees Celsius. Ready to use. minus The difference is used as a real-time wet-bulb temperature correction value.

[0047] If the real-time atomization pressure at the top of the tower Less than or equal to the preset pressure threshold The real-time wet-bulb temperature correction value .

[0048] It should be noted that the real-time wet-bulb temperature correction value is... This refers to the real-time material temperature during the constant-rate drying stage.

[0049] (3) Control of inlet air temperature in constant speed drying zone: It should be noted that Tg is the glass transition temperature of the material. Whether the material temperature exceeds Tg directly determines whether the powder will stick to the wall, absorb moisture, or clump. Therefore, during the constant-rate drying stage, the inlet air temperature needs to be controlled to maintain a high evaporation rate as much as possible while keeping the material temperature below Tg. If Tg is directly used as the target temperature, fluctuations in the inlet air temperature due to data errors and PID algorithm control will cause short-term fluctuations in the wet-bulb temperature, thus affecting process stability. Therefore, a temperature threshold needs to be set as a safety threshold, and the target temperature adjustment should be based on this threshold.

[0050] Preset tower top safety threshold The temperature is 3℃, and we will use this as an example for explanation.

[0051] Real-time target temperature at the top of the tower .in, This is the initial Tg value.

[0052] Real-time temperature difference at the top of the tower .

[0053] It should be noted that: It directly reflects the difference between the temperature of the material at the top of the tower and Tg.

[0054] Based on real-time temperature difference at the top of the tower Construct the PID temperature control computational components. Among them, the real-time tower top temperature difference is used. Construct the proportional term for PID control, and base it on the real-time temperature difference at the top of the tower. Real-time calculation of the corresponding integral terms and differential terms The obtained proportional, integral, and derivative terms are imported into the preset PID temperature control algorithm, and combined with the tuned PID control parameters, the inlet air temperature of the constant-speed drying zone at the top of the tower is subjected to closed-loop steady-state regulation and automatic control.

[0055] It should be noted that the PID controller is a very common and well-known type of controller used to control industrial processes, mechanical systems, and various other systems. PID stands for Proportional, Integral, and Derivative, representing the three main components of the controller. The linear PID calculation rule, which derives the integral and derivative terms from the base temperature difference, is a generally known model in automation control.

[0056] Step S003: Obtain the real-time measured moisture content and the estimated moisture content of the material; determine the real-time moisture content in the tower based on the difference between the measured moisture content and the estimated moisture content; obtain the real-time Tg value corresponding to the moisture content in the tower; obtain the real-time temperature in the tower and the outlet air temperature at the bottom of the tower; adjust the difference between the temperature in the tower and the outlet air temperature at the bottom of the tower using the moisture content in the tower, and determine the real-time material temperature in the slow-deceleration curing zone based on the outlet air temperature at the bottom of the tower; determine the real-time temperature difference in the slow-deceleration curing zone based on the material temperature in the slow-deceleration curing zone and the real-time Tg value; implement closed-loop control of the inlet air temperature of the slow-deceleration curing zone from the middle to the bottom of the tower based on the temperature difference in the slow-deceleration curing zone.

[0057] It should be noted that: furthermore, dynamic temperature control is performed in the mid-to-bottom decreasing curing zone of the tower. When the material is in the constant-rate drying zone at the top of the tower, the initial moisture content is extremely high. The strong plasticizing effect of moisture, combined with the characteristic of the composite system shifting towards the low Tg component, results in a very low real-time Tg. As the surface free water in the constant-rate zone continues to evaporate steadily, the moisture content of the material gradually decreases, and the plasticizing effect of moisture on the system continuously weakens. The Tg of the material then shows a continuous and steady upward trend as the moisture content decreases, transforming from a highly viscous state that easily sticks to the wall to a state that can be stably cured. In the slow-deceleration solidification zone from the middle to the bottom of the tower, the free water in the material has been largely removed, the moisture content has dropped to a low level, and the temperature gradient (Tg) rises and gradually stabilizes. Temperature control in this zone must be based on the material's Tg, with the temperature setting adjusted synchronously to the increase in the input Tg of the slow-deceleration solidification zone. This requires both ensuring sufficient removal of bound water through appropriate heating to prevent moisture entrapment and strictly controlling the material temperature below the input Tg to prevent particles from reverting to a rubbery state due to excessive temperature, causing adhesion and agglomeration. Ultimately, the material is completely solidified into a stable glassy powder. In the aforementioned drying tower, the core reason for the change in the material's Tg is the change in its moisture content; therefore, the Tg is determined by inferring the moisture content of the material in the solidification zone.

[0058] Preferably, in one embodiment of the present invention, the method for implementing closed-loop control of the inlet air temperature in the mid-to-bottom slow-deceleration curing zone of the tower includes: (1) Temperature, humidity and make-up air temperature in the multi-sensor acquisition tower: Three temperature and humidity sensors were placed in the tower (at half its height) and at the bottom of the tower. Data was collected every 1 second and smoothed using a moving average method (10 consecutive sets of data).

[0059] CFD simulations were used to determine the stable airflow regions in and at the bottom of the tower. The weights of the temperature and humidity sensors in the stable airflow regions were set to 0.4, while those in other regions were set to 0.15. Using this as an example, a weighted average was calculated on the real-time temperature and humidity data from all sensors in and at the bottom of the tower to obtain the real-time temperature in the tower. (Unit: °C) Humidity in the tower (Unit: g / kg dry air), the calculation method is the same as the real-time intake air temperature and humidity data at the top of the tower.

[0060] (2) Calculate the real-time moisture content of the material: It should be noted that the humidity difference between the air intake and the tower reflects the amount of water evaporated from the material, and the moisture content can be estimated in real time through material balance.

[0061] Obtain the real-time air volume of the drying tower (m) 3 / min (cubic meters per minute), which is determined by the fan parameters. Then, the real-time material feed flow rate is obtained. (kg / h, kilograms per hour).

[0062] Real-time humidity difference ,in, This refers to the real-time humidity of the incoming air at the top of the tower.

[0063] Based on existing, well-known online near-infrared moisture detection devices at the bottom of the tower, real-time measured moisture content values ​​of the material are obtained. .

[0064] The real-time estimated moisture content is output based on the material and heat balance model. The calculation formula is:

[0065] In the formula, This represents the initial water content of the plant extract. Given the density of water, in this embodiment, kg / m 3 . In the molecule: The unit is g / min, so 60 is used to convert min time to h, and the numerator unit is g / h; in the denominator... The unit is kg / h, so 1000 is used to convert kg to g, and the denominator is set to g / h to eliminate dimensions. The material and heat balance model is a well-known technique, and the specific method will not be described here.

[0066] Constructing real-time water content deviation .in, It is an absolute value function.

[0067] The preset deviation threshold is 5%, which will be used as an example for explanation.

[0068] If the real-time moisture content deviation If the water content in the tower is less than or equal to a preset deviation threshold, then the real-time water content in the tower will be set to... .

[0069] It should be noted that when the real-time moisture content deviation exceeds 5%, the balance factor should be adjusted.

[0070] If the real-time moisture content deviation If the deviation exceeds the preset threshold, the real-time water content in the tower will be adjusted. .in, and These are the measured moisture content and the estimated moisture content of the material at the previous sampling time, respectively. That is, if... and Corresponding to the Each sampling time, then and Corresponding to the Each sampling time.

[0071] It should be noted that: because the online near-infrared moisture detection at the bottom of the tower is easily affected by dust, water vapor, material layer distribution, and ambient temperature, the measured value at a single point exhibits instantaneous fluctuations and measurement drift, and is not suitable as a closed-loop main control variable. Therefore, in this embodiment, when the deviation is small, the real-time measured moisture content of the material is directly used as the final real-time moisture content of the material. When the deviation is large, the ratio of the historical measured moisture content to the estimated moisture content is used to correct the current estimated moisture content, thereby achieving steady-state smooth control.

[0072] (3) Predicting real-time Tg values ​​using a pre-trained CNN model : By obtaining the real-time water content in the tower By combining the above-mentioned CNN prediction model, the Tg change of the plant extract composite system can be predicted, thereby obtaining the Tg of the plant extract composite system in the tower after evaporation, that is, the Tg when entering the slow-down solidification zone.

[0073] The initial component proportion (polysaccharide proportion) flavonoid ratio alkaloid ratio ) and real-time water content in the tower Input the CNN model and output the real-time Tg value. .

[0074] (4) Calculation of material temperature: It should be noted that during the deceleration phase, the evaporation and heat dissipation of the particles are extremely weak. The heat is mainly used to raise the material temperature, which is between the temperature of the hot air in the second zone and the outlet air temperature. The residual effect of evaporation and heat dissipation is reflected by moisture correction, and finally the true material temperature that can be used for temperature control is accurately obtained.

[0075] The real-time air temperature at the bottom of the tower is obtained by installing a temperature sensor at the air outlet. .

[0076] It should be noted that: since the material temperature is between the in-tower temperature and the outlet air temperature, the real-time in-tower temperature is used. The upper limit is the outlet air temperature. As the lower limit, and combined with the real-time water content in the tower obtained above, the real-time material temperature in the falling-rate solidification zone is calculated using a linear equation, specifically: Real-time material temperature in the deceleration curing zone ,in, Real-time water content in the tower The normalized value is used as a correction factor.

[0077] In this embodiment, the initial water content of the plant extract is used. The upper limit is set at the standard moisture content at the air outlet, and the lower limit is set at the standard moisture content at the outlet (which can be obtained from equipment technical manuals, industry standards, and material process specifications). The minimum-maximum standard method is used to determine the real-time moisture content in the tower. Normalize to between 0 and 1.

[0078] (5) Temperature control in the slow-curing zone: Preset safety threshold in tower The temperature is 3℃, and we will use this as an example for explanation.

[0079] Real-time target temperature of the deceleration curing zone .in, This is the real-time Tg value.

[0080] Real-time temperature difference in the deceleration curing zone .in, This is for the real-time temperature of the material in the deceleration curing zone.

[0081] Based on real-time temperature difference in the slow-curing zone Construct the PID temperature control computational components. Among them, the real-time temperature difference in the deceleration curing zone is used. Construct the proportional term for PID control, and base it on the real-time temperature difference in the deceleration curing zone. Real-time calculation of the corresponding integral terms and differential terms The obtained proportional, integral, and derivative terms are imported into the preset PID temperature control algorithm, and combined with the tuned PID control parameters, the inlet air temperature of the slow-deceleration curing zone in the middle to bottom of the tower is subjected to closed-loop steady-state regulation and automatic control.

[0082] It should be noted that in this embodiment, by monitoring the product agglomeration rate online, the CNN model and temperature control parameters are optimized in reverse to achieve process self-adaptation and improve long-term stability. Product monitoring and closed-loop feedback optimization are specifically as follows: First, the product's agglomeration rate and final moisture content are detected online, specifically as follows: An online screening device (1mm screen diameter) is installed at the outlet of the drying tower. The mass of material on the screen is collected every 10 minutes, and the agglomeration rate is calculated.

[0083] This clumping rate The calculation formula is:

[0084] In the formula, This refers to the mass of material that passed through the sieve in this test (unit: g), specifically the mass of agglomerated material with a particle size greater than 1 mm that was screened out in this test. This refers to the total mass of all materials collected in this instance (unit: g).

[0085] Final moisture content was determined using an online near-infrared moisture meter. .

[0086] Then, the closed-loop feedback optimization model and parameters are determined.

[0087] When the agglomeration rate exceeds the standard, it indicates that there is a deviation in the current Tg prediction or temperature control parameters. Fine-tuning the model through feedback data can improve the accuracy of subsequent predictions.

[0088] The default clumping rate is 2%, which will be used as an example for explanation.

[0089] like The predicted agglomeration rate, temperature, humidity, and Tg were input into the CNN online learning module. The model weights were fine-tuned using mini-batch gradient descent with a learning rate of 0.001 and a batch size of 32. This will be used as an example for explanation. Simultaneously, the sensor data and material balance parameters were verified to eliminate interference. The model weight update formula is:

[0090] In the formula, For the new weights of the model, For the old weights of the model, For learning rate, , The model loss function is the mean square error between the predicted Tg and the actual Tg. This is the partial derivative of the loss function with respect to the weights. This is a well-known technique, and the specific method will not be described here.

[0091] like We collect process parameters and product data, and store them in a database for subsequent model training.

[0092] In this embodiment, the spray drying process flow diagram of the spray drying tower is as follows: Figure 2 As shown, Figure 2 In the process, the raw materials are first pretreated and the initial Tg is predicted. Then, the temperature of the constant-rate drying zone at the top of the tower is controlled in a closed loop. Next, the temperature of the cooling-rate curing zone in the middle and bottom of the tower is controlled dynamically. Finally, the product is monitored and the closed-loop feedback is optimized.

[0093] This invention is now complete.

[0094] In summary, in this embodiment of the invention, the wet-bulb temperature corresponding to the real-time inlet air temperature and humidity at the top of the tower is obtained. Based on the real-time atomization pressure at the top of the tower, the wet-bulb temperature is corrected to determine the real-time wet-bulb temperature correction value. Combined with the initial Tg value, the real-time temperature difference at the top of the tower is determined, which is used to implement closed-loop control of the inlet air temperature in the constant-rate drying zone at the top of the tower. Based on the difference between the real-time measured moisture content and the estimated moisture content, the real-time moisture content in the tower is determined, and the real-time Tg value corresponding to the moisture content in the tower is obtained. Using the moisture content in the tower, the difference between the temperature in the tower and the outlet air temperature at the bottom of the tower is adjusted. Combined with the outlet air temperature at the bottom of the tower, the real-time material temperature in the slow-rate curing zone is determined. Then, combined with the real-time Tg value, the real-time temperature difference in the slow-rate curing zone is determined, which is used to implement closed-loop control of the inlet air temperature in the slow-rate curing zone from the middle to the bottom of the tower. This invention, through adaptive control of the efficient spray drying process, achieves high production stability, lower energy consumption, and improved product quality.

[0095] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-efficiency spray drying process for plant extracts, characterized in that, The process includes the following steps: Obtain the initial Tg value of the plant extract; The system acquires the real-time inlet air temperature, inlet air humidity, and atomization pressure at the top of the tower; acquires the wet-bulb temperature corresponding to the inlet air temperature and inlet air humidity; corrects the wet-bulb temperature based on the atomization pressure to determine a real-time wet-bulb temperature correction value; determines the real-time temperature difference at the top of the tower based on the wet-bulb temperature correction value and the initial Tg value; and implements closed-loop control of the inlet air temperature in the constant-rate drying zone at the top of the tower based on the temperature difference at the top of the tower. The system acquires real-time measured moisture content and estimated moisture content of the material; determines the real-time moisture content in the tower based on the difference between the measured and estimated moisture content; acquires the real-time Tg value corresponding to the moisture content in the tower; acquires the real-time temperature in the tower and the outlet air temperature at the bottom of the tower; adjusts the difference between the temperature in the tower and the outlet air temperature at the bottom of the tower using the moisture content in the tower, and determines the real-time material temperature in the slow-deceleration curing zone based on the outlet air temperature at the bottom of the tower; determines the real-time temperature difference in the slow-deceleration curing zone based on the material temperature in the slow-deceleration curing zone and the real-time Tg value; and implements closed-loop control of the inlet air temperature in the slow-deceleration curing zone from the middle to the bottom of the tower based on the temperature difference in the slow-deceleration curing zone.

2. The high-efficiency spray drying process for plant extracts according to claim 1, characterized in that, The specific steps for determining the real-time wet-bulb temperature correction value are as follows: If the atomization pressure is greater than a preset pressure threshold, the wet-bulb temperature is compensated according to the amount by which the atomization pressure exceeds the preset pressure threshold to obtain a real-time wet-bulb temperature correction value. If the atomization pressure is less than or equal to a preset pressure threshold, then the real-time wet-bulb temperature correction value is set to equal the wet-bulb temperature.

3. The high-efficiency spray drying process for plant extracts according to claim 2, characterized in that, The specific steps for compensating the wet-bulb temperature based on the amount by which the atomization pressure exceeds a preset pressure threshold to obtain a real-time wet-bulb temperature correction value are as follows: Calculate the difference between the atomization pressure and the preset pressure threshold, multiply the difference by the preset correction coefficient to obtain a compensation term, subtract the compensation term from the wet-bulb temperature, and record the result as the real-time wet-bulb temperature correction value.

4. The high-efficiency spray drying process for plant extracts according to claim 1, characterized in that, The specific steps for determining the real-time temperature difference at the top of the tower are as follows: The difference between the initial Tg value and the preset tower top safety threshold is recorded as the real-time tower top target temperature. The difference between the real-time wet-bulb temperature correction value and the target temperature at the top of the column is recorded as the real-time temperature difference at the top of the column.

5. The high-efficiency spray drying process for plant extracts according to claim 1, characterized in that, The specific steps for determining the real-time water content in the tower are as follows: The real-time moisture content deviation is determined based on the difference between the measured moisture content of the material and the estimated moisture content. If the real-time moisture content deviation is less than or equal to the preset deviation threshold, then the real-time moisture content in the tower is set to equal the estimated moisture content. If the real-time moisture content deviation is greater than the preset deviation threshold, the real-time moisture content in the tower is determined based on the magnitude of the measured moisture content of the material and the estimated moisture content.

6. The high-efficiency spray drying process for plant extracts according to claim 5, characterized in that, The specific steps involved in determining the real-time moisture content deviation are as follows: Calculate the absolute value of the difference between the measured moisture content of the material and the estimated moisture content, calculate the ratio of the absolute value of the difference to the measured moisture content of the material, convert the ratio into a percentage, and record it as the real-time moisture content deviation.

7. The high-efficiency spray drying process for plant extracts according to claim 5, characterized in that, The specific steps for determining the real-time moisture content in the tower based on the difference between the measured moisture content and the estimated moisture content of the material are as follows: Calculate the ratio of the measured moisture content of the material at the previous sampling time to the estimated moisture content, and multiply the ratio by the real-time estimated moisture content to record the real-time moisture content in the tower.

8. The high-efficiency spray drying process for plant extracts according to claim 1, characterized in that, The specific steps for determining the real-time material temperature in the slow-curing zone are as follows: Calculate the difference between the temperature in the tower and the temperature of the air outlet at the bottom of the tower. Then calculate the product of the normalized value of the water content in the tower and the difference. The sum of the product and the temperature of the air outlet at the bottom of the tower is recorded as the real-time temperature of the material in the slow-down curing zone.

9. The high-efficiency spray drying process for plant extracts according to claim 1, characterized in that, The specific steps for determining the real-time temperature difference in the slow-curing zone are as follows: Based on the real-time Tg value, the real-time target temperature of the slow-curing zone is determined; The difference between the material temperature in the slow-curing zone and the target temperature in the slow-curing zone is recorded as the real-time temperature difference in the slow-curing zone.

10. The high-efficiency spray drying process for plant extracts according to claim 9, characterized in that, The specific steps for determining the real-time target temperature of the slow-curing zone are as follows: The difference between the real-time Tg value and the preset safety threshold in the tower is recorded as the real-time target temperature of the slow-down curing zone.

Images