Spray-drying micro-capsule granulation system with built-in fluidized bed and spray-drying micro-capsule granulation method

Abstract

The invention discloses a spray-drying micro-capsule granulation system with a built-in fluidized bed and a spray-drying micro-capsule granulation method, and relates to the technical field of micro-capsule spray granulation. The system comprises a granulation tower, a built-in fluidized bed, a first detection unit, a state sensing unit, an external fluidized bed, a second detection unit, a numerical simulation unit and a control unit. Primary particles generated by the granulation tower are dried by the built-in fluidized bed to form a semi-finished product, and then the semi-finished product is dried by the external fluidized bed to obtain a finished product. The numerical model unit creates and updates a first-order numerical model of the built-in fluidized bed and a second-order numerical model of the external fluidized bed, the control unit evaluates the reliability of the humidity of the semi-finished product through the first-order numerical model, and the second-order numerical model adjusts the operation parameters of the external fluidized bed. The second-order numerical model is updated through the finished product humidity, and then the first-order numerical model is updated based on the updated second-order numerical model. Cooperative and adaptive control of two-stage drying is realized, and the product quality stability and the system robustness are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the microcapsule spray granulation technology, and particularly to a spray drying microcapsule granulation system and method with an internal fluidized bed. BACKGROUND

[0002] The spray drying granulation technology is one of the important methods for preparing microencapsulated products. In order to overcome the problems of uneven particles and excessive fine powder in the traditional single fluidized bed, the existing technology introduces a scheme combining an internal fluidized bed with a granulation tower. For example, a double-fluidized microcapsule granulation system is disclosed in Chinese patent application No. CN101822963A. The system effectively prolongs the residence time of the material by setting an internal fluidized bed in the granulation tower, and improves the fluidization quality by using a stirring device. At the same time, the system recovers fine powder at the top of the granulation tower by using a dust removal and feeding system, effectively improving the uniformity of the particles and the utilization rate of the raw materials. However, the moisture content and other parameters of the semi-finished product produced by the internal fluidized bed directly determine the optimal drying conditions required by the external fluidized bed. The operating parameters of the external fluidized bed are usually set according to fixed empirical values, and cannot be self-adaptively adjusted according to the actual output of the previous process. When the material state fluctuates, the external fluidized bed is difficult to dynamically adjust according to the semi-finished product produced by the previous process due to the fixed operating parameters, which can easily cause the drying effect to deviate from the ideal interval. If the operating parameters of the external fluidized bed are set too high (such as the inlet air temperature is set too high or the inlet air volume is too large), the particles in the external fluidized bed will be overheated when they are close to the drying endpoint, which can cause the inactivation of heat-sensitive core materials, the brittleness or even the fusion and adhesion of the microcapsule shell. On the contrary, if the operating parameters of the external fluidized bed are set too low, the moisture content of the finished product may exceed the standard due to insufficient dehydration, which can affect the flowability and storage stability of the finished product. Therefore, it is necessary to further improve the existing technology. SUMMARY

[0003] In view of the above problems, the present application provides a spray drying microcapsule granulation system and method with an internal fluidized bed. The first-order numerical model of the internal fluidized bed is used to evaluate the reliability of the moisture content of the semi-finished product. Based on the reliability, the second-order numerical model and the moisture content of the semi-finished product are used to accurately adjust the operating parameters of the external fluidized bed in advance, to realize the active cooperation of the two-stage fluidized bed. The two-stage model is updated in reverse according to the accurate moisture content of the finished product, to realize the closed-loop optimization control of the model and the drying process.

[0004] The application objectives of the present application can be achieved by the following technical solutions: A spray drying microcapsule granulation system with an internal fluidized bed, comprising: a granulation tower configured to generate primary particles after atomizing the liquid material; an internal fluidized bed arranged in the granulation tower and configured to form semi-finished products after drying the primary particles; a first detection unit configured to detect the humidity of the semi-finished product; a state sensing unit configured to acquire the state parameters of the internal fluidized bed; an external fluidized bed disposed outside the prilling tower, the external fluidized bed being configured to dry the semi-finished product to obtain a finished product; a second detection unit configured to detect the humidity of the finished product; a numerical simulation unit configured to create and update a first-order numerical model and a second-order numerical model; a control unit configured to adjust at least one set of operating parameters of the external fluidized bed based on the first-order numerical model and the second-order numerical model, wherein, the numerical simulation unit creates the first-order numerical model based on the structural parameters of the internal fluidized bed and creates the second-order numerical model based on the structural parameters of the external fluidized bed; the numerical simulation unit updates the second-order numerical model based on the operating parameters of the external fluidized bed and the humidity of the finished product, and updates the first-order numerical model based on the second-order numerical model and the state parameters of the internal fluidized bed; the control unit calculates the reliability of the humidity of the semi-finished product based on the first-order numerical model and adjusts the operating parameters of the external fluidized bed according to the second-order numerical model.

[0005] In the present application, the semi-finished product outlet of the internal fluidized bed is connected to the semi-finished product inlet of the external fluidized bed through a conveying pipe, the first detection unit is arranged on the conveying pipe, the prilling tower has a wrapping material passage, and the state sensing unit includes temperature sensors arranged at the bed layers of the internal fluidized bed, a humidity sensor arranged at the air inlet of the internal fluidized bed, and temperature sensors and humidity sensors arranged at the air outlets of the prilling tower.

[0006] In the present application, the numerical simulation unit updates the drying kinetics parameters in the second-order numerical model with the humidity of the finished product obtained by the second detection unit, inversely calculates the calibrated value of the humidity of the semi-finished product based on the updated second-order numerical model, and updates the drying kinetics parameters in the first-order numerical model with the calibrated value.

[0007] In the present application, the control unit first predicts the reliability of the humidity of the semi-finished product through the first-order numerical model, and if the reliability is higher than a preset threshold, calculates the operating parameters of the external fluidized bed based on the second-order numerical model, otherwise the external fluidized bed uses fixed operating parameters.

[0008] A prilling method applied to the spray drying microcapsule prilling system, comprising the following steps: Step 1: creating a first-order numerical model based on the structural parameters of the internal fluidized bed and a second-order numerical model based on the structural parameters of the external fluidized bed; Step 2: the material solution enters the prilling tower from the atomizing nozzle, the coating material is sprayed into the prilling tower at the bottom of the prilling tower, and the material solution is coated by the coating material to form primary particles which fall into the built-in fluidized bed at the bottom of the prilling tower; Step 3: the built-in fluidized bed performs primary drying on the primary particles to form semi-finished products, and the humidity of the semi-finished products is detected; Step 4: the state parameters of the built-in fluidized bed are obtained, the reliability of the humidity of the semi-finished products is calculated based on the first-order numerical model, and the operation parameters of the external fluidized bed are adjusted according to the second-order numerical model; Step 5: the semi-finished products enter the external fluidized bed through the conveying pipe, the external fluidized bed performs secondary drying on the semi-finished products to obtain finished products, and the humidity of the finished products is detected; Step 6: the second-order numerical model is updated based on the operation parameters of the external fluidized bed and the humidity of the finished products, the first-order numerical model is updated based on the second-order numerical model and the state parameters of the built-in fluidized bed, and the process returns to Step 2.

[0009] In the present application, in Step 1, the first-order numerical model is used to predict the theoretical humidity of the semi-finished products based on the state parameters of the built-in fluidized bed, and the second-order numerical model is used to predict the theoretical humidity of the finished products based on the humidity of the semi-finished products and the operation parameters of the external fluidized bed, and the first-order numerical model and the second-order numerical model are both composed of a mass balance equation, a drying kinetics equation and an energy balance equation which are coupled with each other.

[0010] In the present application, in Step 4, the theoretical humidity of the semi-finished products is predicted based on the first-order numerical model and the state parameters of the built-in fluidized bed, and the reliability of the humidity of the semi-finished products is the deviation value between the theoretical humidity of the semi-finished products and the humidity of the semi-finished products.

[0011] In the present application, in Step 4, if the reliability of the humidity of the semi-finished products is higher than a preset threshold value, the humidity of the semi-finished products is taken as an input parameter, the operation parameters of the external fluidized bed which make the theoretical humidity of the finished products predicted by the second-order numerical model closest to the target humidity of the finished products are solved by taking the second-order numerical model as a function, otherwise the external fluidized bed adopts fixed operation parameters.

[0012] In the present application, in Step 6, the theoretical humidity of the finished products is predicted based on the second-order numerical model, the humidity of the semi-finished products and the operation parameters of the external fluidized bed, the drying kinetics parameters in the second-order numerical model are updated by taking the minimization of the error between the humidity of the finished products and the theoretical humidity of the finished products as a target, the calibration value of the humidity of the semi-finished products is inversely calculated based on the updated second-order numerical model and the humidity of the finished products, and the drying kinetics parameters in the first-order numerical model are updated based on the state parameters of the built-in fluidized bed and the calibration value.

[0013] The spray drying micro-capsule granulation system and method with the built-in fluidized bed of the present application have the following beneficial effects: the first-order numerical model is created based on the structural parameters of the built-in fluidized bed, the second-order numerical model is created based on the structural parameters of the external fluidized bed, the built-in fluidized bed dries the primary particles to generate the semi-finished product, the external fluidized bed dries the semi-finished product to generate the finished product. The humidity of the semi-finished product is obtained through the first detection unit, the theoretical humidity of the semi-finished product is predicted based on the first-order numerical model, and the reliability of the humidity of the semi-finished product is calculated based on the theoretical humidity of the semi-finished product. When the reliability exceeds the preset threshold, the operation parameters of the external fluidized bed are adjusted according to the humidity of the semi-finished product; otherwise, the preset fixed operation parameters are automatically started to run. To ensure the prediction accuracy of the first-order numerical model, the second-order numerical model is updated by the high-precision finished product humidity, and the first-order numerical model is updated based on the updated second-order numerical model. Through this front and rear cascading control mode, the drying process is accurately controlled, the consistency of the final product humidity is effectively improved, and the stable operation ability of the system under complex working conditions is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The block diagram of the spray drying micro-capsule granulation system with the built-in fluidized bed of the present application is provided; Figure 2 The schematic diagram of the spray drying micro-capsule granulation system of the present application is provided; Figure 3 The schematic diagram of the external fluidized bed of the present application is provided; Figure 4 The curve diagram of the particle temperature in the external fluidized bed of the present application is provided; Figure 5 The production line schematic diagram of the spray drying micro-capsule granulation system of the present application is provided; Figure 6 The equipment model parameter of a part of the spray drying micro-capsule granulation system of the present application is provided; Figure 7 The equipment model parameter of another part of the spray drying micro-capsule granulation system of the present application is provided; Figure 8 The process parameter of a part of the spray drying micro-capsule granulation system of the present application is provided; Figure 9 The flow chart of the granulation method applied to the spray drying micro-capsule granulation system of the present application is provided; Figure 10 The original near-infrared spectrum diagram of the semi-finished product of the present application is provided; Figure 11 The near-infrared spectrum diagram of the semi-finished product after pretreatment of the present application is provided; Figure 12 The schematic diagram of the production cycle of the present application is provided.

[0015] The reference signs in the drawings: prilling tower 1, atomizing nozzle 101, wrapping material channel 102, exhaust port 103, cylinder 104, cone 105, built-in fluidized bed 2, upper chamber 201, lower chamber 202, air distribution plate 203, external fluidized bed 3, spring support 301, rack 302, drying chamber 303, cooling chamber 304, hot air inlet 305, cold air inlet 306, semi-finished product feeding port 307, finished product discharging port 308, hot air device 4, air blower 401, air filter 402, heater 403, first detection unit 5, rotary screen 501, material conveying pipe 502, gas-solid separation equipment 601, air locking device 602, air induction fan 603, liquid material bin 7, powder material bin 8, finished product bin 9. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.

[0017] The traditional spray drying microcapsule prilling generally adopts a separate layout of “prilling tower + external fluidized bed”. The operation of the external fluidized bed in this traditional design is isolated, and the control strategy is based on fixed parameters or only relies on local feedback (such as a single humidity sensor) at the inlet or outlet of the external fluidized bed. This control mode cannot perceive the actual output state of the previous prilling process at all, and it is also impossible to adjust the drying intensity of the current stage in advance according to the real-time characteristics (such as humidity trend) of the output material of the previous stage. Therefore, the traditional system lacks the ability to cooperatively regulate the material state of the entire production line, resulting in poor adaptability to raw material fluctuations and working condition changes, and finally showing insufficient product humidity stability and high energy consumption.

[0018] The present application integrates a built-in fluidized bed inside and below the prilling tower, and sets an external fluidized bed outside, forming a compact and efficient two-stage series drying process. The system obtains the humidity of the semi-finished product reflecting the output of the built-in fluidized bed based on the first detection unit in real time, and adjusts at least one operating parameter (such as inlet air temperature, air volume) of the external fluidized bed based on the humidity of the semi-finished product. Further, the section environment where the first detection unit is located is harsh, and the sensor probe is easy to be contaminated or adhered with dust by the humid and sticky particles, resulting in slow drift or sudden distortion of the measurement signal. In order to verify the accuracy of the humidity of the semi-finished product, the present application calculates the theoretical humidity of the semi-finished product through a first-order numerical model, and compares the theoretical humidity of the semi-finished product with the humidity of the semi-finished product obtained by the first detection unit, so as to evaluate the reliability of the humidity of the semi-finished product. And update the first-order numerical model based on the second-order numerical model, so as to fully exert the efficiency advantage of the two-stage drying process while ensuring the accuracy and robustness of the final control. Embodiment one

[0019] Reference Figures 1 to 8The spray drying micro-capsule granulation system with the built-in fluidized bed of the present application comprises a granulation tower 1, a built-in fluidized bed 2 arranged in the granulation tower 1, a first detection unit 5 for detecting the humidity of the semi-finished product, a state sensing unit for collecting the state parameters of the built-in fluidized bed 2, an external fluidized bed 3 arranged outside the granulation tower 1, a second detection unit for detecting the humidity of the finished product, a numerical simulation unit, and a control unit.

[0020] The granulation tower 1 is configured to generate primary particles after atomizing the liquid material. The top center of the granulation tower 1 is provided with an atomizing nozzle 101 for atomizing the liquid material. The atomizing nozzle 101 is connected with a liquid material bin 7 containing the liquid material. A liquid pump is used to convey the liquid material from the liquid material bin 7 to the atomizing nozzle. The frequency of the liquid pump and the frequency of the atomizing nozzle are adjusted according to the particle size of the target finished product. The frequency of the liquid pump is usually 20 Hz, and the frequency of the atomizing nozzle is usually 25 Hz. The middle of the side of the granulation tower is provided with a wrapping material channel 102, and the top of the side is provided with an air outlet 103 for discharging dry exhaust gas and guiding to subsequent recycling devices. The upper part of the granulation tower 1 is a cylindrical barrel 104, and the lower part of the granulation tower is a cone 105 for guiding the primary particles to converge. The discharge port below the cone 105 is directly connected with the primary particle inlet of the built-in fluidized bed.

[0021] Referring to Figure 5 In this embodiment, before spray granulation, the crude oil and auxiliary materials are put into the dissolving kettle, the auxiliary materials and crude oil in the dissolving kettle are heated to 55℃, and the emulsifying tank is heated to 65-70℃. Then the mixed solution of crude oil and auxiliary materials is slowly added into the emulsifying tank after the stirrer is started. The emulsification time is usually 35-45 minutes, and the temperature is maintained at 65-70℃. The viscosity of the obtained liquid material after emulsification is usually 200-230 centipoise.

[0022] The built-in fluidized bed 2 arranged in the granulation tower 1 is configured to dry the primary particles to form semi-finished products. The cavity of the built-in fluidized bed 2 is divided into an upper chamber 201 at the top and a lower chamber 202 at the bottom. A cloth air distribution plate 203 is arranged between the upper chamber 201 and the lower chamber 202. The upper chamber 201 is an open structure and is directly connected with the lower part of the cone of the granulation tower. A semi-finished product discharge port is arranged at the side of the upper chamber 201, and the semi-finished product discharge port is connected with the semi-finished product inlet 307 of the external fluidized bed 3 through a conveying pipe 502. The side of the lower chamber 202 is provided with an air inlet. Hot air enters the cavity of the built-in fluidized bed 2 through the air inlet, penetrates the primary particles, and then flows upward and is discharged from the air outlet at the top of the granulation tower 1. The inlet air temperature of the built-in fluidized bed 2 is usually controlled between 80-120℃. In this embodiment, a pressure sensor is further arranged in the granulation tower for detecting the negative pressure in the granulation tower. The negative pressure in the granulation tower needs to be maintained at 100-400 Pa at all times, and the negative pressure in the tower is adjusted by the granulation tower exhaust fan.

[0023] A first detection unit 5 is arranged to detect the humidity of the semi-finished product. To prevent abnormal particles with excessively large particle size or agglomerates from impacting, wearing or even clogging the probe of the first detection unit 5, and to ensure that the particles flowing through the probe have uniform physical properties, thereby ensuring the accuracy and stability of the humidity measurement, a rotary screen 501 is further arranged between the semi-finished product discharge port of the built-in fluidized bed 2 and the first detection unit 5. The rotary screen 501 is configured to screen the semi-finished product, and the undersize of the rotary screen 501 is guided to the powder bin, and the oversize is the semi-finished product with qualified particle size, which is transported to the external fluidized bed. The probe of the first detection unit 5 is partially embedded in the wall of the conveying pipe 502, and the detection surface is in direct contact with the semi-finished product.

[0024] A state sensing unit is arranged to collect the state parameters of the built-in fluidized bed 2. The state sensing unit includes temperature sensors arranged in the bed layer of the built-in fluidized bed 2, humidity sensors arranged at the air inlet of the built-in fluidized bed 2, and temperature sensors and humidity sensors arranged at the air outlet of the granulation tower 1. The temperature at the air outlet of the granulation tower 1 cannot be higher than 75°C.

[0025] An external fluidized bed 3 is arranged outside the granulation tower 1, which is configured to dry the semi-finished product to obtain a finished product. Referring to Figure 3 , the external fluidized bed 3 is, for example, a vibrating fluidized bed, and the bed body of the vibrating fluidized bed is installed on the rack 302 through a spring support 301 and is provided with vibration force by a vibration motor installed at the bottom of the bed body. During the secondary drying process, the particle temperature of the semi-finished product is usually lower than the inlet temperature of the hot air, but when the drying is completed, the temperature will rise rapidly until it approaches the inlet temperature of the hot air. At this time, the semi-finished product is in an unstable state, which causes the internal heat-sensitive active ingredients to be inactivated or degraded due to overheating, and also easily causes moisture regain and caking due to temperature difference during subsequent packaging. Referring to Figure 4 , to prevent the semi-finished product from continuously rising in temperature, in this embodiment, the inside of the bed body is divided into a drying chamber 303 and a cooling chamber 304 along the direction of advancement of the semi-finished product, the bottom of the drying chamber is provided with a hot air inlet 305, and the bottom of the cooling chamber is provided with a cold air inlet 306. The semi-finished product inlet 307 at the first end of the top of the bed body is connected to the conveying pipe 502, and the finished product discharge port 308 at the end of the top of the bed body is connected to the finished product bin 9. In this embodiment, a vibrating screen is arranged between the finished product bin 9 and the built-in fluidized bed 2, and the aperture of the vibrating screen is usually 30-100 mesh. The hot air in the drying chamber 303 is provided by the hot air inlet 305, and the cold air in the cooling chamber 304 is provided by the cold air inlet 306, so as to rapidly cool the finished product to a stable state close to the ambient temperature that can be safely packaged and stored. The cold air is dry cold air treated by dehumidification, and the dew point temperature is lower than the temperature of the cooled finished product, so as to prevent moisture from condensing on the surface of the particles.

[0026] a second detection unit configured to detect the moisture of the finished product. The second detection unit comprises an offline automatic sampler and a rapid moisture analyzer. The automatic sampler periodically collects samples from the finished product bin 9, and the moisture analyzer uses a benchmark method such as the thermogravimetric method to analyze the samples with high precision, obtaining the moisture of the finished product with high precision. Offline measurement can avoid the drift of online sensors, providing the only reliable standard data for the update of the first-order numerical model and the second-order numerical model.

[0027] a numerical simulation unit. The numerical simulation unit is configured to create and update the first-order numerical model and the second-order numerical model. The numerical simulation unit creates the first-order numerical model based on the structural parameters of the internal fluidized bed, and creates the second-order numerical model based on the structural parameters of the external fluidized bed. The numerical simulation unit updates the second-order numerical model based on the operating parameters of the external fluidized bed and the moisture of the finished product, and updates the first-order numerical model based on the second-order numerical model and the state parameters of the internal fluidized bed. Specifically, the numerical simulation unit updates the drying kinetics parameters in the second-order numerical model with the moisture of the finished product obtained by the second detection unit, inversely calculates the calibration value of the moisture of the semi-finished product based on the updated second-order numerical model, and updates the drying kinetics parameters in the first-order numerical model with the calibration value.

[0028] a control unit. The control unit is configured to adjust at least one set of operating parameters of the external fluidized bed based on the first-order numerical model and the second-order numerical model. The control unit first calculates the reliability of the moisture of the semi-finished product through the first-order numerical model, and if the reliability is higher than a preset threshold, calculates the operating parameters of the external fluidized bed based on the second-order numerical model, otherwise the external fluidized bed uses fixed operating parameters.

[0029] In the present embodiment, a hot air device 4 is configured to deliver hot air to the built-in fluidized bed 2 and the external fluidized bed 3. The hot air device 4 comprises an air blower 401, an air filter 402 and a heater 403. The air inlet of the air blower 401 is in communication with ambient air, and the air outlet of the air blower 401 is connected to the air filter 402 and the heater 403 in sequence. The air outlet of the heater 403 is connected to the air inlet of the built-in fluidized bed 2 or the hot air inlet 305 of the drying chamber 303 of the external fluidized bed 3. A dryer can be arranged between the air filter 402 and the heater 403, which can be a refrigeration dryer or an adsorption dryer. The recovery device is configured to recover the exhaust gas discharged from the exhaust outlets of the external fluidized bed 3 and the granulation tower 1. The recovery device comprises a gas-solid separation device 601. In the present embodiment, the gas-solid separation device 601 comprises, for example, a cyclone separator and a bag filter. Before spray granulation, the starch (coating material) in the bag filter is completely discharged and conveyed back to the powder bin, which is discharged about once every 20-30 minutes. The air inlets of the gas-solid separation device 601 are connected to the exhaust outlets of the granulation tower 1 and the external fluidized bed 3 through air pipes, respectively. The bottom powder outlet of the gas-solid separation device 601 is provided with an air blower 602 connected to the powder bin 8. The powder bin 8 is filled with coating material such as starch, and the water content of the starch is usually maintained at 10-11%. If the water content of the starch is too low, the risk of dust explosion increases, and if the water content of the starch is too high, the starch will be caked and the flowability will be poor. The top of the gas-solid separation device 601 is connected to the air inlet of an induced draft fan 603, so that the treated exhaust gas is discharged. Embodiment Two

[0030] With reference to Figures 5 to 11 A granulation method applied to the spray drying microcapsule granulation system, comprising the following steps: Step 1: creating a first-order numerical model based on the structural parameters of the built-in fluidized bed and a second-order numerical model based on the structural parameters of the external fluidized bed. The first-order numerical model is used to predict the theoretical semi-finished product humidity based on the state parameters of the built-in fluidized bed, and the second-order numerical model is used to predict the theoretical finished product humidity based on the semi-finished product humidity and the operating parameters of the external fluidized bed. Both the first-order numerical model and the second-order numerical model are composed of a mass balance equation, a drying kinetics equation and an energy balance equation which are coupled with each other. The model structure of the first-order numerical model and the second-order numerical model is specifically described in Embodiment Three.

[0031] Step 2: The feed liquid enters the granulation tower from the atomizing nozzle, and the coating material is sprayed at the bottom of the granulation tower. After the coating material coats the feed liquid to form primary particles, the primary particles fall into the built-in fluidized bed at the bottom of the granulation tower. Figure 5, the oil (e.g. vitamin oil) and the excipient (e.g. ethoxyquin) are stirred in a dissolving kettle and then enter an emulsifying tank to be emulsified, obtaining a liquid with a certain viscosity (e.g. vitamin emulsion). The screw pump is started, the feeding amount of the liquid is adjusted (200-250 liters / hour, gradually increasing the feeding amount from 200 liters / hour), and the feeding amount of the packaging material is adjusted according to the ratio of 1:10 of the liquid to the packaging material. The rotating speed of the atomizing nozzle is 1000-1400 rpm. The atomizing nozzle at the top of the prilling tower atomizes the liquid to form droplets. The droplets are coated with the packaging material and come into contact with hot air in the prilling tower, evaporating about 60% of the water, and rapidly forming primary particles. The primary particles fall into the built-in fluidized bed at the bottom of the prilling tower under the action of gravity.

[0032] Step 3: The built-in fluidized bed performs primary drying on the primary particles to form a semi-finished product, and the humidity of the semi-finished product is detected. The primary particles falling into the built-in fluidized bed form a fluidized state under the action of uniform hot air in the upper chamber, and are subjected to primary drying, so that the internal moisture of the primary particles is further evaporated by about 30%, forming a semi-finished product. The semi-finished product passes through a rotary screen, and the rotary screen guides the semi-finished product with qualified particle size through a conveying pipe to the external fluidized bed. The pore size of the rotary screen is, for example, 120 mesh. The undersize is transported back to the powder bin, and the oversize is the semi-finished product with qualified particle size. Referring to Figure 10 and Figure 11 To meet the requirements of rapid detection, the first detection unit is, for example, a near-infrared sensor. The probe part of the near-infrared sensor is embedded in the wall of the conveying pipe. The near-infrared sensor emits near-infrared light and receives the original near-infrared spectrum of the semi-finished product. The original near-infrared spectrum is corrected to obtain the humidity of the semi-finished product in real time.

[0033] Step 4: Obtain the state parameters of the built-in fluidized bed, calculate the reliability of the humidity of the semi-finished product based on the first-order numerical model, and adjust the operating parameters of the external fluidized bed according to the second-order numerical model. The theoretical humidity of the semi-finished product is predicted based on the first-order numerical model and the state parameters of the built-in fluidized bed. The reliability of the humidity of the semi-finished product is the deviation value between the theoretical humidity of the semi-finished product and the humidity of the semi-finished product. If the reliability of the humidity of the semi-finished product is higher than the preset threshold, the humidity of the semi-finished product is taken as an input parameter (fixed condition), and the operating parameter optimization problem is constructed with the second-order numerical model as the core function. The operating parameters of the external fluidized bed are solved, which make the theoretical product humidity predicted by the second-order numerical model closest to the target product humidity. Otherwise, the external fluidized bed uses fixed operating parameters. Further, referring to Figure 8 the preset range of the operating parameters (e.g. inlet air temperature) of the external fluidized bed in the background art, if the operating parameters predicted based on the second-order numerical model exceed the preset range, the fixed operating parameters are also used.

[0034] In the present embodiment, the operating parameters of the external fluidized bed are, for example, the inlet air temperature of the hot air. In order to prevent the impact on the mechanical structure of the equipment caused by excessively large single adjustment amount of the inlet air temperature of the hot air, after the target inlet air temperature of the external fluidized bed is calculated by the second-order numerical model, the hot air device is not directly instructed to jump to the target inlet air temperature, but a smooth transition setting curve from the current inlet air temperature to the target inlet air temperature is generated based on the preset maximum safe temperature change rate and the thermal inertia of the equipment. The smooth transition setting curve is usually planned by using an S-shaped function or a linear slope, so as to ensure that the instantaneous slope is always not greater than the maximum safe temperature change rate. Then, the smooth transition setting curve is taken as the dynamic setting value for closed-loop tracking, so that the actual inlet air temperature of the external fluidized bed approaches the target inlet air temperature smoothly and without overshoot. This method can avoid the thermal stress impact of temperature step change on the mechanical structure such as refractory lining, thereby ensuring the reliability of long-term operation of the equipment.

[0035] Step 5: The semi-finished product enters the external fluidized bed through the conveying pipe, the external fluidized bed performs secondary drying on the semi-finished product to obtain a finished product, and the humidity of the finished product is detected. In order to ensure the reliability of the humidity of the finished product relied on for updating the second-order numerical model, the second detection unit is an automatic sampler and a rapid moisture meter. The automatic sampler periodically collects samples from the finished product bin in the current production cycle, and inputs the samples into the rapid moisture meter for determination. The rapid moisture meter determines by using the oven drying weighing method, and outputs the humidity of the finished product with high precision.

[0036] Step 6: The second-order numerical model is updated based on the operating parameters of the external fluidized bed and the humidity of the finished product, and the first-order numerical model is updated based on the second-order numerical model and the state parameters of the internal fluidized bed, and the process returns to step 2. In the present application, in step 6, the theoretical humidity of the finished product is predicted based on the second-order numerical model, the humidity of the semi-finished product and the operating parameters of the external fluidized bed, the drying kinetic parameters in the second-order numerical model are updated with the minimum error between the humidity of the finished product and the theoretical humidity of the finished product as the target, the calibrated value of the humidity of the semi-finished product is inversely calculated based on the updated second-order numerical model and the humidity of the finished product, and the drying kinetic parameters in the first-order numerical model are updated based on the state parameters of the internal fluidized bed and the calibrated value. Embodiment Three

[0037] The present embodiment further discloses the model structure of the first-order numerical model and the model structure of the second-order numerical model. The first-order numerical model is composed of a mass balance equation based on the internal fluidized bed and the prilling tower , a drying kinetic equation , and an energy balance equation The mutual coupling constitutes, wherein t1 is a drying time of the semi-finished product formed by the feed liquid, R1 is a feeding rate of the feed liquid, R2 is a total drying rate of the built-in fluidized bed, F1 is a feeding rate of the coating material, F2 is a discharging rate of the semi-finished product, k1 is a drying kinetics parameter of the first-order numerical model, in the physical sense, k1 is a lumped time-varying process parameter, which comprehensively reflects the overall mass transfer coefficient and effective mass transfer area in the built-in fluidized bed and the connected granulation tower, and the specific value needs to be determined through experiments or system identification, X1 is the dry basis moisture content, the initial value of X1 can be measured through experiments, that is, X1=M2 / M1, M1 is the total mass of dry solid materials in the built-in fluidized bed, M2 is the total mass of wet materials in the built-in fluidized bed, the initial value of M1 is the total mass of solid solutes in the coating material and the feed liquid, and the initial value of M2 is the total mass of solvents in the feed liquid, X2 is the material balance moisture content balanced by the inlet gas conditions (inlet gas temperature and inlet gas humidity) of the built-in fluidized bed, C1 is the total heat capacity of the built-in fluidized bed, that is, C1=M1c3+ M2c4, c3 is the specific heat capacity of dry solid materials (the comprehensive average specific heat capacity of solid solutes in the coating material and the feed liquid), c4 is the specific heat capacity of wet materials (the specific heat capacity of solvents in the feed liquid), ρ1 is the inlet gas density of the built-in fluidized bed, V1 is the volume flow rate of the inlet gas of the built-in fluidized bed, c1 is the constant-pressure specific heat capacity of the inlet gas of the built-in fluidized bed, U1 is the total heat dissipation coefficient (device thermal parameter) of the built-in fluidized bed and the granulation tower, A1 is the effective heat dissipation area of the built-in fluidized bed and the granulation tower, T1 is the bed temperature in the built-in fluidized bed (equivalent to the temperature of the primary particles in the built-in fluidized bed), which can be measured in real time by a thermocouple or a thermal resistor deeply embedded in the built-in fluidized bed, the thermocouple or the thermal resistor mainly contacts the primary particles, T2 is the inlet gas temperature of the built-in fluidized bed, T3 is the ambient temperature (room temperature), ΔH is the latent heat of evaporation of the liquid in the primary particles, Q1 represents the sensible heat brought in by the feed liquid and the coating material, and Q2 represents the sensible heat taken away by the semi-finished product. The theoretical semi-finished product humidity (wet basis moisture content) Y1=X1 / (X1+1). In the embodiment, the state parameters of the built-in fluidized bed include the bed temperature T1 in the built-in fluidized bed, the material balance moisture content X2 balanced by the inlet gas humidity and the inlet gas temperature T2 of the built-in fluidized bed, and the total drying rate R2.

[0038] The second-order numerical model is composed of a mass balance equation based on the external fluidized bed , a drying kinetics equation , and an energy balance equation The two coupled models are as follows: in the second-order numerical model, the initial dry basis moisture content X3 of the semi-product is equal to the output X1 of the first-order numerical model, the bed temperature T4 of the external fluidized bed is equal to the temperature of the semi-product in the external fluidized bed, and the initial value of T4 is set based on the temperature of the semi-product at the outlet of the internal fluidized bed, wherein t2 is the drying time of the semi-product to form the product, M3 is the total mass of the dry solid material in the external fluidized bed, M4 is the total mass of the wet material in the external fluidized bed, the initial values of M3 and M4 can be calculated based on the moisture of the semi-product, that is, the initial value of M3 is the total mass of the dry solid material in the semi-product, and the initial value of M4 is the total mass of the wet material in the semi-product, R3 is the total drying rate of the external fluidized bed, F3 is the feeding speed of the semi-product (F3=F2 in the continuous production process of a single batch), k2 is the drying kinetics parameter of the second-order numerical model, in the physical sense, k2 is also a lumped time-varying process parameter, which comprehensively reflects the overall mass transfer coefficient and effective mass transfer area of the external fluidized bed, and the specific value needs to be determined through experiments or system identification, X4 is the material balance moisture content balanced by the inlet conditions (inlet temperature and inlet humidity) of the internal fluidized bed, C2 is the total heat capacity of the external fluidized bed, p2 is the inlet density of the external fluidized bed, V2 is the volume flow rate of the inlet of the external fluidized bed, c2 is the constant-pressure specific heat capacity of the inlet of the external fluidized bed, T5 is the inlet temperature of the external fluidized bed, Q3 represents the sensible heat carried by the semi-product, Q4 represents the sensible heat carried away by the product, U2 is the total heat dissipation coefficient (device thermal parameter) of the external fluidized bed, and A2 is the effective heat dissipation area of the external fluidized bed. In this embodiment, the calculation of Q1, Q2, Q3 and Q4 is based on the simplification with the reference temperature being zero. The theoretical product moisture (wet basis moisture content) Y2=X3 / (X3+1). In this embodiment, the operating parameters of the external fluidized bed are, for example, the inlet temperature of the hot air. Embodiment four

[0039] The semi-product moisture obtained by the first detection unit is instantaneous and local measurement point information, which is easily affected by errors such as drift of the first detection unit, while the external fluidized bed is the forming unit of the final product, and the product moisture obtained by the second detection unit is the only information that can be measured with high precision offline in the production process, so the most reliable product moisture can be used to correct the drying kinetics parameter k2 in the second-order numerical model in reverse, and then the calibrated value of the semi-product moisture is calculated based on the updated second-order numerical model and the product moisture, and the drying kinetics parameter k1 in the first-order numerical model is updated based on the state parameters of the internal fluidized bed and the calibrated value. Refer to Figure 12 , this embodiment further discloses a preferred method for updating the first-order numerical model and the second-order numerical model in step 6.

[0040] The operation parameters of the external fluidized bed in the n-1th production cycle, the semi-product humidity Y3 obtained by the first detection unit, and the product humidity Y4 obtained by the second detection unit are collected in the updating stage 1. The initial value of the dry basis moisture content X3 corresponding to the semi-product humidity Y3 is calculated, and the operation parameters of the external fluidized bed and the initial value of the dry basis moisture content X3 are taken as inputs to run the current second-order numerical model to obtain the theoretical product humidity Y2. The drying kinetics parameter k2 in the second-order numerical model is adjusted by an optimization algorithm to make the theoretical product humidity Y2 infinitely close to the measured product humidity Y4, that is, min(Y2-Y4) is taken as the target to update the second-order numerical model. 2 The first-order numerical model is updated.

[0041] The operation parameters of the external fluidized bed in the n-1th production cycle, the semi-product humidity Y3 obtained by the first detection unit, and the product humidity Y4 obtained by the second detection unit are collected in the updating stage 1. The initial value of the dry basis moisture content X3 corresponding to the semi-product humidity Y3 is calculated, and the operation parameters of the external fluidized bed and the initial value of the dry basis moisture content X3 are taken as inputs to run the current second-order numerical model to obtain the theoretical product humidity Y2. The drying kinetics parameter k2 in the second-order numerical model is adjusted by an optimization algorithm to make the theoretical product humidity Y2 infinitely close to the measured product humidity Y4, that is, min(Y2-Y4) is taken as the target to update the second-order numerical model. 2 The first-order numerical model is updated.

[0042] The nth production cycle is performed based on the updated first-order numerical model and second-order numerical model.

[0043] The above only describes the preferred embodiments of the present application and is not used to limit the present application. Any modification, equivalent replacement and improvement within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A spray-drying microcapsule granulation system with a built-in fluidized bed, characterized in that, include: A granulation tower, configured to atomize liquid to generate primary granules; An internal fluidized bed is provided inside the granulation tower, which is configured to dry the primary particles to form a semi-finished product; The first detection unit for detecting the humidity of semi-finished products; A state sensing unit that collects built-in fluidized bed state parameters; An external fluidized bed is set outside the granulation tower, and this external fluidized bed is configured to dry the semi-finished product to obtain the finished product; The second detection unit for detecting the humidity of the finished product; A numerical simulation unit configured to create and update first-order and second-order numerical models; A control unit configured to adjust at least one set of operating parameters of an external fluidized bed based on the first-order numerical model and the second-order numerical model, wherein, The numerical simulation unit creates a first-order numerical model based on the structural parameters of the built-in fluidized bed and a second-order numerical model based on the structural parameters of the external fluidized bed. The numerical simulation unit updates the second-order numerical model based on the operating parameters of the external fluidized bed and the finished product humidity, and updates the first-order numerical model based on the second-order numerical model and the state parameters of the internal fluidized bed. The control unit calculates the reliability of the semi-finished product humidity based on a first-order numerical model and adjusts the operating parameters of the external fluidized bed according to a second-order numerical model.

2. The spray-drying microcapsule granulation system with a built-in fluidized bed according to claim 1, characterized in that, The semi-finished product outlet of the built-in fluidized bed and the semi-finished product inlet of the external fluidized bed are connected by a conveying pipe. The first detection unit is set on the conveying pipe. The granulation tower has a packaging material channel. The status sensing unit includes a temperature sensor set on the built-in fluidized bed, a humidity sensor set on the air inlet of the built-in fluidized bed, and a temperature sensor and a humidity sensor set on the air outlet of the granulation tower.

3. The spray-drying microcapsule granulation system with a built-in fluidized bed according to claim 1, characterized in that, The numerical simulation unit updates the drying kinetic parameters in the second-order numerical model with the finished product humidity obtained by the second detection unit, calculates the calibration value of the semi-finished product humidity based on the updated second-order numerical model, and updates the drying kinetic parameters in the first-order numerical model with the calibration value.

4. The spray-drying microcapsule granulation system with a built-in fluidized bed according to claim 1, characterized in that, The control unit first predicts the reliability of the semi-finished product humidity using a first-order numerical model. If the reliability is higher than a preset threshold, it calculates the operating parameters of the external fluidized bed based on a second-order numerical model; otherwise, it uses fixed operating parameters for the external fluidized bed.

5. A granulation method applied to the spray-drying microcapsule granulation system of claim 1, characterized in that, Includes the following steps: Step 1: Create a first-order numerical model based on the structural parameters of the built-in fluidized bed, and create a second-order numerical model based on the structural parameters of the external fluidized bed; Step 2: The liquid feed enters the granulation tower from the atomizing nozzle. Packaging material is sprayed into the bottom of the granulation tower. The packaging material coats the liquid feed to form primary particles, which then fall into the built-in fluidized bed at the bottom of the granulation tower. Step 3: The primary particles are dried in a built-in fluidized bed to form a semi-finished product, and the moisture content of the semi-finished product is measured. Step 4: Obtain the state parameters of the built-in fluidized bed, calculate the reliability of the semi-finished product humidity based on the first-order numerical model, and adjust the operating parameters of the external fluidized bed according to the second-order numerical model. Step 5: The semi-finished product enters the external fluidized bed through the conveying pipe. The external fluidized bed performs secondary drying on the semi-finished product to obtain the finished product. The moisture content of the finished product is then measured. Step 6: Update the second-order numerical model based on the operating parameters of the external fluidized bed and the finished product humidity, update the first-order numerical model based on the second-order numerical model and the state parameters of the internal fluidized bed, and return to step 2.

6. The granulation method according to claim 5, characterized in that, In step 1, the first-order numerical model predicts the theoretical humidity of the semi-finished product based on the state parameters of the built-in fluidized bed, and the second-order numerical model predicts the theoretical humidity of the finished product based on the humidity of the semi-finished product and the operating parameters of the external fluidized bed. Both the first-order and second-order numerical models are composed of mutually coupled mass balance equations, drying kinetic equations, and energy balance equations.

7. The granulation method according to claim 5, characterized in that, In step 4, the theoretical semi-finished product humidity is predicted based on the first-order numerical model and the state parameters of the built-in fluidized bed. The reliability of the semi-finished product humidity is the deviation value between the theoretical semi-finished product humidity and the semi-finished product humidity.

8. The granulation method according to claim 5, characterized in that, In step 4, if the confidence level of the semi-finished product humidity is higher than the preset threshold, an operation parameter optimization problem is constructed with the second-order numerical model as the core. The operation parameters of the external fluidized bed that make the theoretical finished product humidity predicted by the second-order numerical model closest to the target finished product humidity are solved. Otherwise, the external fluidized bed adopts fixed operation parameters.

9. The granulation method according to claim 5, characterized in that, In step 6, the theoretical finished product humidity is predicted based on the second-order numerical model, the semi-finished product humidity, and the operating parameters of the external fluidized bed. With the goal of minimizing the error between the finished product humidity and the theoretical finished product humidity, the drying kinetic parameters in the second-order numerical model are updated. Then, the calibration value of the semi-finished product humidity is calculated based on the updated second-order numerical model and the finished product humidity. The drying kinetic parameters in the first-order numerical model are updated based on the state parameters of the built-in fluidized bed and the calibration value.

Images