A continuous twin-screw wet granulation control system and control method thereof

Abstract

The present invention discloses a continuous twin-screw high-shear wet granulation control system and a control method thereof, including a granulation liquid flow rate and an adhesive concentration control module, a powder weight loss feeding control module, a granulation particle size and particle size yield model prediction control module, and a wetting raw material powder granulation liquid volume fraction content control module. The present invention introduces the surface energy of the wetted raw material powder measured under equivalent equivalent working conditions, and uses the viscoelastic-plastic contact force EEPA constitutive model to truly and quantitatively characterize the essential characteristics of the synergistic coupling effect of the twin-screw pre-compression and the granulation KE block periodic grinding and pressing during the granulation process, thereby constructing a multi-parameter synergistic coupling model prediction control system based on the granulation liquid volume fraction content and surface energy, adhesive concentration, granulation liquid flow rate and screw speed, and a granulation liquid dynamic real-time preparation device with online adjustable adhesive concentration, realizing online fine regulation of the terminal particle granulation particle size and granulation yield, and the granulation yield of the present invention can reach more than 85%.

Description

Technical Field

[0001] The invention relates to the technical field of continuous wet granulation of solid oral preparations, and in particular to a continuous twin-screw wet granulation control system and a control method thereof. Background Art

[0002] In the continuous production of solid oral preparations, the pelleting of solid oral preparations requires a continuous twin-screw high-shear wet granulation process. The consistency of the biological efficacy of solid oral preparations is controlled by the consistency of the quality of the continuous twin-screw wet granulation and the reproducibility of the process. This requires the continuous twin-screw wet granulator to meet the ideal granulation particle size and narrow particle size distribution range required by the user, so that the key quality attribute parameters of the granulation are highly consistent, and at the same time have a high granulation yield, to achieve high-quality and efficient continuous granulation production. Therefore, the continuous twin-screw wet granulation needs to be equipped with an online real-time fine-tuning and controllable online monitoring and control system for the granulation particle size, particle size distribution, granulation morphology and granulation yield.

[0003] The particle size and particle size distribution of the traditional continuous twin-screw wet granulator are both controlled by a single parameter method based on the granulation liquid spray flow rate. The granulation liquid spray has two main effects. Effect 1 is the formation of liquid bridge contact force between adjacent wetted raw material powder particles after spraying, which is mainly controlled by the granulation liquid volume fraction content of the wetted raw material powder particles in the granulation zone, the viscosity of the granulation liquid and the surface tension. Effect 2 is that the spraying wetting can make the raw material powder wet and soften and increase the viscosity of the surface. The wetted, softened and viscous raw material powder particles will form viscoelastic-plastic contact force, which is mainly controlled by the surface energy of the wetted and softened viscous powder, the pre-compression density of the raw material powder of the granulation screw, and the synergistic coupling effect of the periodic high-intensity grinding and high shear stress of the granulation KE block. Obviously, the granulation quality is controlled by the screw structure and granulation process of the continuous twin-screw wet granulation system. How to online regulate the liquid bridge contact force and viscoelastic-plastic contact force is the technical key to achieving consistency of the key quality attributes of the terminal granulation.

[0004] At present, the control system of traditional continuous twin-screw wet granulation mainly configures a granulating liquid with a fixed viscosity, puts it into a granulating liquid tank, and controls the injection flow rate of the granulating liquid under constant viscosity conditions to adjust the granulating liquid volume fraction content of the moistening and softening raw material powder in the granulation zone, which can only achieve online regulation of the liquid bridge contact force. Therefore, the existing traditional control system has two main deficiencies: (1) The key quality attributes of the end point granulation are controlled by the combined effect of the liquid bridge contact force and the viscoelastic-plastic contact force, and does not truly reflect the effect of the surface energy of the moistening and softening viscous powder, the synergistic coupling of the pre-compression density and the periodic grinding and high shear stress on the cohesive viscous contact force. It cannot truly characterize and reflect the effect of the granulation essential characteristics of the synergistic coupling of the pre-compression and periodic grinding and high shear stress in the granulation zone, and this essential characteristic is the driving force source for the twin-screw wet granulation technology to achieve continuous granulation, which will inevitably lead to the current The traditional control system is not accurate enough in predicting and controlling the granulation process, so it is impossible to achieve the online regulation of the granulation particle size, narrow particle size, particle size morphology and granulation yield in a refined, scientific and quantitative manner. It is a technical control method based on trial and error. (2) When the granulation particle size is small, since the viscosity of the granulation liquid in the tank cannot be regulated online, the particle size can only be increased by increasing the injection flow rate of the granulation liquid, that is, increasing the moisture content of the powder and improving the liquid bridge contact force. However, this regulation method will cause the residual humidity at the end of the subsequent continuous drying granulation to fluctuate, affecting its consistency and increasing the energy consumption of continuous drying. In summary, the existing continuous twin-screw wet granulation still lacks an online real-time precision control system that uses science to seek quality and technology to ensure success in a refined, scientific and quantitative manner to achieve consistency in the key quality attributes of granulation.

[0005] The key factors affecting the granulation performance and key granulation properties of the twin-screw high-shear wet granulator are the liquid bridge contact force of the granulation liquid, the viscoelastic-plastic contact force of the wetting and softening viscous raw material powder, and the synergistic coupling effect of the raw material powder pre-compression formed by the rotation of the twin screws during the granulation process and the periodic high-intensity grinding and high shear stress formed by the rotation of the granulation KE block. In order to build a refined, scientific and quantitative granulation key quality attribute and its consistency online and real-time controllable scientific granulation that seeks quality with science and ensures success with technology, it is necessary to achieve online refined regulation of the granulation liquid-liquid bridge contact force and the viscoelastic-plastic contact force of the wetting and softening viscous raw material powder, and to achieve scientific quantitative characterization and refined regulation of the synergistic coupling effect of the raw material powder pre-compression formed by the rotation of the twin screws, the periodic high-intensity grinding and high shear stress formed by the rotation of the granulation KE block.

[0006] The experimental study of the present invention found that the surface energy of the wet and softened viscous raw material powder is extremely sensitive to the high-intensity grinding and pressing, pre-compression and adhesive concentration it has experienced. The surface energy of the wet and softened viscous raw material powder has been significantly improved after the pre-compression of the twin screw co-rotating conjugate and the periodic high-intensity grinding and pressing of the KE block. Taking fructose powder as an example, the dispersion surface energy of fructose powder can be greatly increased by undergoing the pre-compression and periodic high-intensity grinding and pressing of the raw material powder, from 35mJ / m 2 Increased to 95mJ / m 2 , and its increase is as high as 171%, and as the concentration of the binder in the granulation liquid increases, its surface energy also increases. Taking MCC powder as an example, if PVP is used as a granulation binder, when the PVP content of its granulation liquid increases from 0% to 5%, its surface energy increases by as much as 121%. It can be seen that the surface energy of the wetted raw material powder in the continuous twin-screw wet granulation process is much greater than the surface energy of the raw material powder in its natural state. It can be seen that the surface energy of the wetted raw material powder under the equivalent working conditions of twin-screw pre-compression and periodic high-intensity grinding and pressing can be used to scientifically and quantitatively characterize the essential characteristics of the synergistic coupling effect of twin-screw pre-compression and periodic high-intensity grinding and pressing. The present invention introduces this measured surface energy into the EEPA viscoelastic-plastic contact force model, and can scientifically and quantitatively characterize the essential influence of the synergistic coupling effect of twin-screw pre-compression and periodic grinding and pressing on the viscoelastic-plastic contact force of the wetted raw material powder in the granulation zone, thereby improving the prediction and control accuracy of the control system for the granulation process.

[0007] From the above analysis, it can be seen that the dynamic response of the twin screw to the pre-compression of the raw material powder, periodic high-intensity grinding and high shear stress can change the surface energy of the wetted raw material powder after the liquid spraying, which has a crucial impact on the twin screw granulation process and granulation quality. Therefore, in order to achieve the refined, scientific and quantitative online real-time control of the continuous twin screw wet granulation process, it is necessary to construct a new type of granulation particle size, particle size distribution and granulation yield refined online real-time model prediction control model based on the multi-parameter synergistic coupling of the wet raw material powder granulation liquid volume fraction content and surface energy, binder concentration, and screw speed, in order to achieve the above goals. Summary of the invention

[0008] In order to solve the problem that the above-mentioned traditional continuous twin-screw wet granulator based on the control method of granulation liquid spray flow rate is difficult to truly reflect the key influences such as the viscous contact force of the wetted and softened viscous powder, and the dynamic response of the twin-screw rotation to the pre-compression and periodic grinding of the raw material powder, resulting in the inability to achieve online refined scientific and quantitative control of the granulation process, the present invention provides a continuous twin-screw wet granulation control system based on the multi-parameter synergistic coupling of the volume fraction content and surface energy of the granulation liquid for wetting the raw material powder, the concentration of the adhesive, and the screw speed, which realizes the online real-time controllable continuous twin-screw high-shear wet granulation with refined scientific and quantitative key quality attribute consistency, with science for quality and technology for success.

[0009] To achieve the above-mentioned purpose, the present invention provides a continuous twin-screw wet granulation control system, which is applied to a continuous twin-screw wet granulator for solid oral preparations. It is an online real-time control system for granulation particle size, particle size distribution, and particle size yield based on the synergistic coupling of the volume fraction content of the wetting raw material powder granulation liquid and the surface energy, adhesive concentration, and granulation liquid flow rate. It is characterized in that it includes a granulation liquid flow rate and adhesive concentration control module, a raw material powder weight loss feeding control module, a granulation particle size and granulation yield model prediction control module, and a wetting raw material powder granulation liquid volume fraction content control module. The granulation liquid flow rate and adhesive concentration control module, the powder weight loss feeding control module, the granulation particle size and particle size yield model prediction control module, and the wetting raw material powder granulation liquid volume fraction content control module are all connected to the control system host;

[0010] The granulation liquid flow rate and adhesive concentration control module is used for dynamic real-time preparation of granulation liquid with online adjustable adhesive concentration, and online real-time monitoring and control of granulation liquid delivery flow rate and adhesive concentration of granulation liquid;

[0011] The powder weight loss feeding control module is used for online real-time monitoring and control of the weight loss feeding rate of the raw material powder of the granulator;

[0012] The granulation particle size and granulation yield model prediction control module is used to achieve the critical quality attribute d of the granulator endpoint granulation 10 d 50 d 90 , online real-time monitoring and refined scientific quantitative control of particle size span q and granulation yield Y;

[0013] The wetting raw material powder granulation liquid volume fraction content control module is used for online real-time monitoring and refined scientific quantitative control of the granulation liquid volume fraction content in the wetting raw material powder in the granulation area of ​​the granulator, as well as indirect control of the raw material powder conveying flow rate and screw speed of the granulator screw.

[0014] As a further improvement of the present technology, the granulation liquid flow and adhesive concentration control module includes a water solvent storage tank, a water solvent constant flow pump for conveying water solvent, a first frequency converter for regulating the speed of the water solvent constant flow pump, a water solvent flow transmitter, an adhesive storage tank, an adhesive constant flow pump for conveying adhesive, a second frequency converter for regulating the speed of the adhesive constant flow pump, an adhesive flow transmitter, a double closed-loop ratio controller for regulating the granulation liquid flow and the adhesive concentration, and a small single-screw mixing device for quickly and evenly mixing the water solvent and the adhesive to prepare the granulation liquid, wherein the first frequency converter, the water solvent flow transmitter, the second frequency converter and the adhesive flow transmitter are all connected to the double closed-loop ratio controller, and the double closed-loop ratio controller is connected to the control system host;

[0015] The water solvent flow transmitter is arranged at the outlet of the water solvent constant flow pump, and is used to detect the actual flow of the water solvent in real time, and transmit its measurement signal for feedback to the double closed-loop ratio controller;

[0016] The adhesive flow transmitter is arranged at the outlet of the adhesive constant flow pump, and is used to detect the actual flow of the adhesive in real time, and transmit its measurement signal for feedback to the double closed-loop ratio controller;

[0017] The double closed-loop ratio controller includes a flow ratio device, a main flow controller for regulating the flow of a water solvent, and a follow-up flow controller for regulating the flow of an adhesive, wherein both the main flow controller and the follow-up flow controller are connected to the flow ratio device;

[0018] The dual closed-loop ratio controller adjusts the speed of the water solvent constant flow pump with the help of the main flow controller and the first frequency converter according to the main flow setting value of the water solvent and the real-time detection data deviation of the water solvent flow transmitter, so as to realize online regulation of the main flow of the water solvent. The dual closed-loop ratio controller then dynamically updates the adhesive flow setting value of the follow-up flow controller according to the measured main flow of the water solvent and the flow ratio setting value of the flow ratio converter. The dual closed-loop ratio controller adjusts the speed of the adhesive constant flow pump with the help of the follow-up flow controller and the second frequency converter according to the adhesive flow setting value and the real-time detection data deviation of the adhesive transmitter, and finally realizes online real-time dynamic regulation of the water solvent flow, adhesive flow, granulation liquid flow and adhesive concentration, and thereby realizes dynamic real-time preparation of granulation liquid with online adjustable adhesive concentration through the small single-screw mixing device.

[0019] The present invention provides a dynamic online real-time preparation and precise quantitative delivery control technology for granulating liquid with online adjustable adhesive concentration by setting a granulating liquid flow rate and adhesive concentration control module, which can realize dynamic online precise control of granulating liquid viscosity, surface tension and contact angle, so that the prediction and control of liquid bridge contact force in the granulating area are more accurate and stable, and solves the defects that the existing control technology is not accurate enough in predicting and controlling the liquid bridge contact force, and the viscosity, surface tension and contact angle of the granulating liquid cannot be dynamically adjusted during the granulation process.

[0020] The present invention uses a dynamic online real-time preparation and precise quantitative delivery control technology for a granulating liquid with an online adjustable binder concentration. Under the condition of keeping the granulating liquid spray flow rate constant, the binder concentration can be dynamically adjusted online to precisely control the viscosity, surface tension and contact angle of the granulating liquid to increase the liquid bridge contact force and the viscoelastic-plastic contact force, thereby avoiding the inherent defects of the existing control technology that the consistency of the residual moisture at the end point of the dry granulation is difficult to control and the energy efficiency of the system is increased.

[0021] As a further improvement of the present technology, the granulation particle size and granulation yield model prediction control module includes an online detection system for granulation particle size, a synergistic coupling association prediction model of granulation particle size, granulation yield, granulation liquid volume fraction content and surface energy of the wetting raw material powder, and granulation liquid adhesive concentration, and a granulation particle size and granulation yield model prediction controller. The online detection system for granulation particle size and the granulation particle size and granulation yield model prediction controller are all connected to the control system host. The synergistic coupling association prediction model of granulation particle size, granulation yield, granulation liquid volume fraction content of the wetting raw material powder, wetting raw material powder surface energy, and adhesive concentration is constructed through intelligent digital simulation of the granulation process of the granulator.

[0022] As a further improvement of the present technology, the raw powder weight loss feeding control module includes a feeding controller, a weighing sensor and a fourth frequency converter, the fourth frequency converter is used to adjust the speed of the screw drive motor of the screw weight loss feeder, the weighing sensor is connected to the feeding controller, and is used to detect the weight loss of the raw powder of the screw weight loss feeder per unit time online in real time, obtain the online feeding rate of the raw powder of the granulator, and send the online real-time detection data signal to the feeding controller for feedback, the feeding controller sends a control instruction to the fourth frequency converter according to the deviation between the online feeding rate of the raw powder and the feeding rate set by the user, and the fourth frequency converter controls the speed of the screw drive motor of the screw weight loss feeder by adjusting the output frequency, thereby realizing real-time control of the raw powder feeding rate; the feeding controller is connected to the granulation particle size and granulation yield model prediction controller, and the feeding controller feeds back the raw powder online feeding rate signal to the granulation particle size and granulation yield model prediction controller.

[0023] As a further improvement of the present technology, the online particle size detection system for granulation includes an inclined conveying flow channel and an online particle analyzer. The inclined conveying flow channel is arranged below the discharge port of the granulator, and is used to output the terminal particles of the discharge port at a uniform speed and slowly, thereby improving the particle size detection accuracy. The online particle analyzer is arranged directly above the inclined conveying flow channel, and is used to detect the particle size, particle size distribution and particle size morphology of the terminal particles in real time, and obtain the particle size distribution of the terminal particles. 10 d 50 d 90 , particle size span and granulation yield, the online particle analyzer is connected to the control system host, and the endpoint particles d detected in real time are 10 d 50 d 90 , particle size span, and granulation yield data signals are sent back to the control system host.

[0024] As a further improvement of the present technology, the granulation liquid volume fraction content control module for wetting the raw material powder includes an inclined conveying flow channel, a multi-fiber probe near-infrared spectrometer for online real-time detection of the wet granule humidity at the end point, a flow ratio controller for real-time adjustment of the ratio of the granulating liquid spray flow rate to the granulating screw raw material powder conveying flow rate, and a third frequency converter for regulating the speed of the driving motor of the twin screw of the granulator. The inclined conveying flow channel is arranged below the discharge port of the granulator, and is used to output the end granules at the discharge port at a uniform speed and slowly, thereby improving the particle size detection accuracy. The multi-fiber probe near-infrared spectrometer and the flow ratio controller are both connected to the control system host;

[0025] The multiple fiber optic probes of the online multi-fiber probe near-infrared spectrometer are arranged directly above the inclined conveying flow channel, and are used for multi-point real-time detection of the humidity of the wet particles at the outlet of the granulator, obtaining the average humidity of the wet particles at the outlet of the granulator, and feeding back the measured data signal to the control system host. The control system host calculates the real-time data of the granulation liquid volume fraction content C and the fluid-to-solid ratio L / S of the wetting raw material powder according to the measured average humidity of the terminal particles, and feeds back the real-time data signals of the granulation liquid volume fraction content C and the fluid-to-solid ratio L / S to the flow ratio controller. The flow ratio controller sends a granulation screw raw material powder conveying flow rate control instruction to the third frequency converter according to the deviation between the real-time data of the fluid-to-solid ratio L / S and the set value of the fluid-to-solid ratio L / S. The third frequency converter adjusts the rotation speed of the granulation screw by adjusting the output frequency, and finally realizes the online real-time monitoring and refined scientific quantitative control of the granulation liquid volume fraction content of the wetting raw material powder, as well as the indirect control of the raw material powder conveying flow rate and the screw speed of the granulator granulator.

[0026] As a further improvement of the present technology, the method for constructing a synergistic coupling correlation prediction model of the granulation particle size, granulation yield, granulation liquid volume fraction content of the wetted raw material powder, surface energy, and granulation liquid binder concentration comprises the following steps:

[0027] S1. On the control system host, based on the EEPA constitutive model of liquid bridge contact force and viscoelastic-plastic contact force, design and write an embedded C language program for calculating the liquid bridge contact force and the viscoelastic-plastic contact force, and realize a two-way data transmission interface with the discrete element simulation software, wherein the embedded C language program automatically recognizes that the constitutive model is only applied to the granulation zone downstream of the granulation liquid nozzle in the granulator;

[0028] S2. Based on the integration of the embedded C language program, the bidirectional data transmission interface and the discrete element simulation software, a multi-scale intelligent digital simulation platform for the granulation process of the continuous twin-screw wet granulator is secondary developed. The simulation platform can truly and quantitatively characterize the essential characteristics of the synergistic coupling effect of the twin-screw pre-compression and the granulation KE block periodic grinding and pressing in the granulation process through the surface energy of the wetted raw material powder measured under equivalent experimental conditions, and can describe and simulate the macroscopic dynamic granulation process of the continuous twin-screw wet granulator from the dynamic response of the particle scale;

[0029] S3. constructing a three-dimensional digital functional virtual prototype of a continuous twin-screw wet granulator on the simulation platform;

[0030] S4, determining the rheological parameters of the granulation liquid by offline experiments, and constructing a synergistic correlation prediction model between the rheological parameters of the granulation liquid and the binder concentration of the granulation liquid;

[0031] S5. Under the equivalent working conditions of twin-screw pre-compression and periodic grinding and pressing of granulation KE block, the surface energy E of the wetted raw material powder after spraying granulation liquid with different binder concentrations is measured offline by inverse gas chromatography surface energy analyzer, and a synergistic correlation prediction model of the surface energy E and the binder concentration in the granulation liquid is constructed;

[0032] S6. Design and write an embedded computing program on the control system host computer for the synergistic correlation prediction model of the rheological performance parameters of the granulation liquid, the surface energy of the wetted raw material powder and the concentration of the adhesive;

[0033] S7, inputting granulation process parameters and raw material powder physical property parameters on the simulation platform;

[0034] S8, starting the intelligent digital simulation platform for the granulation process developed in step S2;

[0035] S9, automatically and dynamically update the data required for calculating the liquid bridge contact force and viscoelastic-plastic contact force C language embedded program;

[0036] S10, start the embedded calculation program of liquid bridge viscous contact force and EEPA viscoelastic-plastic contact force, and calculate and obtain the liquid bridge viscous contact force and EEPA viscoelastic-plastic contact force;

[0037] S11, dynamically transferring the calculation result data of the liquid bridge viscous contact force and the EEPA viscoelastic-plastic contact force in step S10 to the discrete element software through the bidirectional data transmission interface between the embedded C language program and the discrete element software;

[0038] S12, automatic dynamic update of the liquid bridge contact force and viscoelastic-plastic contact force of the discrete element software;

[0039] S13, under the experimental operating parameters, the granulation process intelligent digital simulation platform developed in step S2 is used to simulate and predict the key quality attribute characterization parameters d of the granulator end granule product. 10 d 50 d 90 , particle size span q, granulation yield Y;

[0040] S14. Conduct a continuous twin-screw wet granulation experiment. The online particle size detection system detects the key quality attribute characterization parameter d of the end-point granule product in real time. 10 d 50 d 90 and particle size span q, and then verify the test results with the simulation prediction results of step S13, if the error between the simulation prediction results and the test results is ≥±5%, modify and improve the intelligent digital simulation platform of the granulation process until the simulation prediction results are consistent with the experimental test results, and the error is controlled within the range of ±5%;

[0041] S15. Characterize the key quality attribute parameters of the end product 10 d 50 d 90 The particle size span q, granulation yield Y and the granulation liquid volume fraction content of the wetted raw material powder and the surface energy, granulation liquid binder concentration, and screw speed are simulated by the synergistic coupling evolution law. The data regression theory modeling of the simulation data is simulated according to the synergistic coupling evolution law, and the granulation particle size d is finally obtained. 90 , the synergistic coupling correlation control prediction model of granulation yield Y and the granulation liquid volume fraction content of the raw material powder, the surface energy of the raw material powder, and the concentration of the granulation liquid binder, and the synergistic coupling correlation control prediction model of the granulation yield Y and the screw speed.

[0042] As a further improvement of the present technology, the granulation particle size and granulation yield model prediction controller is to program the granulation particle size d obtained in step S15 on the control system host. 90, a synergistic coupling correlation control prediction model of granulation yield Y and granulation liquid volume fraction content of wetting raw material powder, surface energy of wetting raw material powder, and concentration of granulation liquid binder, a synergistic coupling correlation control prediction model of granulation yield Y and screw speed, and an embedded computing program of the control algorithm of the corresponding correlation control prediction model, so that the granulation particle size and granulation yield model prediction controller can achieve the key quality attribute d of the endpoint granulation of the continuous twin-screw wet granulator by online real-time regulation of the granulation liquid volume fraction content and surface energy of the wetting raw material powder, the concentration of granulation liquid binder, and the screw speed. 10 d 50 d 90 , online monitoring and refined scientific quantitative control of particle size span q and granulation yield Y;

[0043] The granulation particle size and granulation yield model prediction controller is connected with the feeding controller, the double closed-loop ratio controller and the flow ratio controller.

[0044] As a further improvement of the present technology, the synergistic correlation prediction model formula of the surface energy E of the wetted raw material powder and the concentration of the adhesive in step S5 is:

[0045] E=α0+α1·C V +α2·C V 2 (1)

[0046] Where: E is the surface energy of the wetted raw material powder, C V is the binder concentration of the granulation solution, α0, α1, and α2 are constants for data regression modeling;

[0047] The granulation particle size d in step S15 90 The formula of the synergistic coupling correlation control prediction model of the volume fraction content of the granulation liquid wetting the raw material powder, the surface energy of the wetting raw material powder, and the concentration of the granulation liquid binder is:

[0048]

[0049] In formula (2): d 90 The particle size d at the end of granulation 90 , μm; C is the volume fraction of the granulation liquid that wets the raw material powder, %; E is the surface energy of the wetted raw material powder; d0, A1 to A 15 , B 11 To B 15 They are all model regression constants of the association control prediction model;

[0050] The synergistic coupling correlation control prediction model formula of the granulation yield Y in step S15, the granulation liquid volume fraction content of the wetted raw material powder, the surface energy of the wetted raw material powder, and the concentration of the granulation liquid binder is:

[0051] Y=Y0+A 21 ·C+A 22 ·C 2 +A 23 ·C 3 +A 24 ·C 4 +A 25 ·C 5 +B 21 ·E+B 22 ·E 2 +B 23 ·E 3 +B 24 ·E 4 +B 25 ·E 5 (3)

[0052] In formula (3), Y is the granulation yield, %; C is the granulation liquid volume fraction content of the wetted raw material powder, %; E is the surface energy of the wetted raw material powder; Y0, A 21 To A 25 , B 21 To B 25 They are all model regression constants of the association control prediction model;

[0053] The prediction model formula of the granulation yield Y and the screw speed synergistic coupling correlation control in step S15 is:

[0054] Y=B0+B1·n+B2·n 2 +B3·n 3 (4)

[0055] In formula (4), Y is the granulation yield, %; n is the screw speed, rpm; B0 to B3 are model regression constants of the correlation control prediction model.

[0056] The present invention also provides a control method for the above-mentioned continuous twin-screw wet granulation control system, which is an online real-time monitoring and refined control method for granulation particle size and granulation yield based on the multi-parameter synergistic coupling of the granulation liquid volume fraction content and surface energy of the wetting raw material powder, the granulation liquid binder concentration, and the screw speed, characterized in that it comprises the following steps:

[0057] T1. First, set the raw material powder weight loss feeding rate and the final granulation particle size d on the control system host. 90 , qualified particle size distribution range, expected value of granulation yield Y;

[0058] T2, according to the granulation particle size d 90The expected value of the granulation yield Y, the granulation particle size and the granulation yield model prediction controller are used to predict the optimal parameter combination of the granulation liquid volume fraction content C, surface energy E, and screw speed of the wetted raw material powder to obtain the optimal granulation yield through the global collaborative optimization prediction through the associated control prediction model formula (2), (3), (4) and the optimization algorithm. Then, based on the optimal granulation liquid volume fraction content C and the associated prediction model formula (1), the binder concentration C of the granulation liquid is predicted and calculated. V ;

[0059] T3. Start the continuous twin-screw wet granulation system and start granulation production;

[0060] T4, the granulation particle size and granulation yield model prediction controller automatically calculates and dynamically updates the set value of the granulation liquid spray flow process parameter based on the real-time data of the raw material powder weight loss feeding rate fed back by the feeding controller and the granulation liquid volume fraction content C of the optimal raw material powder wetting predicted in step T2;

[0061] T5, online particle analyzer to detect the endpoint particle size d in real time 10 d 50 d 90 , particle size span q information, and transmit the real-time measurement signal to the control system for feedback;

[0062] T6, the granulation particle size and granulation yield model prediction controller calculates and analyzes the consistency of the granulation particle size and particle size distribution online in real time, and predicts the granulation process of the next time step by associating the control prediction model formulas (1), (2), (3), (4) and the prediction model control strategy based on the deviation between the granulation endpoint particle size detected in real time and the user's expected particle size, and calculates the granulation liquid volume fraction content C and surface energy E of the externally controlled wetted raw material powder, the granulation liquid spray flow rate, and the granulation liquid binder concentration parameter setting values ​​for obtaining the optimal granulation yield through the global collaborative optimization algorithm;

[0063] T7, the double closed-loop ratio controller automatically and dynamically updates the process parameter setting values ​​of the granulation liquid spray flow rate and the granulation liquid adhesive concentration according to the optimal parameter setting value obtained in step T6. The granulation liquid flow rate and adhesive concentration control module realizes online real-time monitoring and fine regulation of the granulation liquid flow rate and the granulation liquid adhesive concentration through the water solvent flow transmitter, the adhesive flow transmitter and the double closed-loop ratio controller, and the dynamic real-time preparation of the granulation liquid with the required optimal adhesive concentration;

[0064] T8, the granulation liquid volume fraction content control module for wetting the raw material powder automatically and dynamically updates the set value of the fluid-solid ratio L / S of the wetting raw material powder according to the set value of the optimal granulation liquid volume fraction content C of the wetting raw material powder and the set value of the optimal granulation liquid spray flow rate obtained in step T6, and the multi-fiber probe near-infrared spectrometer detects the humidity of the wet granules at the outlet of the granulator at multiple points online in real time, and sends the multi-point real-time humidity detection data to the control system host, and the control system host calculates and obtains the average humidity of the granulated wet granules, the granulation liquid volume fraction content C of the wetting raw material powder and the fluid-solid ratio. The real-time data of the ratio L / S is obtained, and the granulation liquid volume fraction content C and the real-time monitoring data signal of the fluid-solid ratio L / S are fed back to the flow ratio controller. The flow ratio controller sends a raw material powder conveying flow control instruction to the third frequency converter according to the deviation between the real-time monitoring data of the fluid-solid ratio L / S and the set value of the fluid-solid ratio L / S. The frequency converter adjusts the output frequency and the rotation speed of the granulating screw to achieve online real-time monitoring and refined control of the granulation liquid volume fraction content C of the wetted raw material powder, as well as indirect control of the raw material powder conveying flow rate and the screw rotation speed of the granulator granulator screw;

[0065] T9, through the control module of the volume fraction content of the granulating liquid wetting the raw material powder, the dynamic online control of the volume fraction content C of the granulating liquid wetting the raw material powder, the raw material powder conveying flow rate, the screw speed, the raw material powder pre-compression working state and the periodic grinding and pressing working state of the granulation KE block can be realized, and then the double closed-loop ratio controller of step T7 is coordinated to realize the dynamic online precision control of the granulating liquid spray flow rate and the solubility of the binder in the granulating liquid, and the control of the surface energy E of step T6. By combining the above-mentioned multi-parameter collaborative coupling control means, the dynamic online real-time monitoring and precise control of the terminal granulation particle size, particle size distribution and granulation yield can be finally realized.

[0066] Compared with the prior art, the beneficial effects of the present invention are:

[0067] 1. The present invention uses the surface energy measured under equivalent working conditions of the raw material powder pre-compression formed by the rotation of the screw and the periodic grinding and pressing of the granulation KE block to truly characterize and reflect the essential characteristics of the synergistic coupling effect of the raw material powder pre-compression formed by the rotation of the twin screw and the periodic grinding and pressing formed by the rotation of the granulation KE block during the granulation process, and introduces this surface energy into the viscoelastic-plastic contact force EEPA model, and constructs a multi-scale continuous twin-screw wet granulator granulation process intelligent digital simulation platform based on the liquid bridge contact force model and the viscoelastic-plastic contact force EEPA constitutive model. The platform can describe and simulate the macroscopic dynamic granulation process of the twin-screw high shear wet granulator from the dynamic response of the particle scale, and can truly and quantitatively characterize the essential characteristics of the synergistic coupling effect of the raw material powder pre-compression formed by the rotation of the twin screws and the periodic grinding and pressing formed by the rotation of the granulation KE block during the granulation process, as well as the liquid bridge contact force and viscoelastic-plastic contact force in the granulation zone, so as to more scientifically and accurately simulate, predict and control the dynamic changes of the granulation process, thereby effectively improving the accuracy and reliability of the simulation prediction and control of the key quality attribute parameters of the twin-screw wet granulator end point;

[0068] 2. The present invention realizes the dynamic real-time preparation of granulating liquid with online adjustable adhesive concentration through a double closed-loop ratio controller of granulating liquid flow rate and adhesive concentration and a granulating liquid uniform mixing device, realizes online real-time monitoring and control of granulating liquid delivery flow rate and adhesive concentration in granulating liquid, and avoids the defects of the existing control method that easily causes fluctuation and consistency of residual humidity at the end point of subsequent continuous drying granulation, as well as increased energy efficiency of the continuous drying system;

[0069] 3. The present invention constructs a synergistic coupling correlation prediction model of granulation particle size, granulation liquid volume fraction content and surface energy of the wetting raw material powder, and granulation liquid binder concentration, and constructs a granulation particle size and granulation yield model prediction controller based on multi-parameter synergistic coupling of granulation liquid volume fraction content and surface energy of the wetting raw material powder, granulation liquid spray flow rate, granulation liquid binder concentration, and screw speed, thereby realizing online real-time dynamic and precise control of the wetting raw material powder liquid bridge contact force and viscoelastic-plastic contact force, ensuring the granulation particle size and particle size distribution. The accuracy and stability of online control of the distribution and granulation yield have greatly improved the control ability of the endpoint key quality attribute consistency and process reproducibility of the continuous twin-screw wet granulator, and provided technical support for the scientific manufacturing of online real-time controllable granules of continuous twin-screw wet granulators that pursue quality with science and ensure success with technology, and solved the common technical problem of low granulation yield of traditional continuous twin-screw wet granulators. The present invention can make the granulation yield of the continuous twin-screw wet granulator reach more than 85%, which is worthy of popularization and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a schematic diagram of the control system of the patent embodiment of the present invention;

[0071] Figure 2 A flow chart for constructing a correlation control prediction model between the intelligent digital simulation platform and the granulation particle size, particle size distribution and granulation yield of the embodiment of the present invention;

[0072] Figure 3 It is the relationship curve between the water solvent delivery flow rate, the adhesive delivery flow rate and the corresponding constant flow pump driving motor speed;

[0073] Figure 4 It is the synergistic correlation curve between the concentration of the PVP adhesive in the granulation liquid and the speed ratio of the driving motor of the water solvent constant flow pump and the driving motor of the adhesive constant flow pump;

[0074] Figure 5 is the correlation curve between the viscosity of the granulation liquid and the concentration of the PVP adhesive;

[0075] Figure 6 This is the correlation curve between the surface energy of MCC raw material powder and the content of adhesive PVP in the embodiment of the present invention;

[0076] Figure 7 The curves of the coordinated evolution of the maximum granulation particle size and the surface energy of the wetted raw material powder under the conditions of the spray liquid volume fraction content C of 5%, 10% and 20% respectively in the embodiment of the present invention;

[0077] Figure 8 The final granulation quality characteristic parameter d is shown in the embodiment of the present invention under the condition of 10% spray liquid volume fraction content C. 90 d 50 and d 10 Co-evolution law curve of surface energy of wetted raw material powder;

[0078] Fig. 9 The final granulation quality characteristic parameter d is shown in the embodiment of the present invention under the condition of 20% spray liquid volume fraction content C. 90 d 50 and d 10 Co-evolution law curve of surface energy of wetted raw material powder;

[0079] Fig.10 For the embodiment of the present invention, the spray liquid volume fraction content C is 10% and the spray liquid volume fraction content C is 20J / m 2 Under the condition of wetted raw material powder surface energy, the correlation evolution curve between the final granulation quality characteristic particle size and the screw speed;

[0080] Fig.11 It is the correlation control curve of the final granulation yield Y and the surface energy of the wetted raw material powder under the condition of 10% spray liquid volume fraction content C in the embodiment of the present invention;

[0081] Fig.12This is the correlation control curve between the final granulation yield Y and the screw speed under the condition of 10% spray liquid volume fraction content C in the embodiment of the present invention;

[0082] Fig.13 This is the correlation control curve of the final granulation yield Y and the surface energy E of the wetted raw material powder under the condition of 20% spray liquid volume fraction content C in the embodiment of the present invention;

[0083] Fig.14 This is the correlation control curve of the final granulation yield Y and the screw speed n under the condition of 20% spray liquid volume fraction content C in the embodiment of the present invention;

[0084] Fig.15 The final granulation particle size d of the embodiment of the present invention 90 The control curve is synergistically coupled with the spray liquid volume fraction content C and the surface energy E of the wetted raw material powder;

[0085] Fig.16 It is the synergistic coupling correlation control curve of the final granulation yield Y, the spray liquid volume fraction content C, and the wetting raw material powder surface energy E of the embodiment of the patent of the present invention.

[0086] In the figure: 1. Continuous twin-screw wet granulator, 2. Granulator drive motor, 3. Third frequency converter, 4. Screw loss-in-weight feeder, 5. Weighing sensor, 6. Feeding controller, 7. Fourth frequency converter, 8. Small single-screw mixing device, 9. Water solvent constant flow pump, 10. First frequency converter, 11. Water solvent flow transmitter, 12. Adhesive constant flow pump, 13. Second frequency converter, 14. Adhesive flow transmitter, 15. Double closed-loop ratio controller, 16. Flow ratio controller, 17. Main flow controller, 18. Follow-up flow controller, 19. Inclined conveying channel, 20. Eyecon2 TM Online particle analyzer, 21. Multi-fiber probe near-infrared spectrometer, 22. Control system host, 23. Granulation particle size and particle size yield model prediction controller, 24. Flow ratio controller. DETAILED DESCRIPTION

[0087] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0088] like Figure 1As shown, the embodiment of the present invention is applied to a continuous twin-screw wet granulator for solid oral preparations, which is an online real-time control system for granulation particle size, particle size distribution, and particle size yield based on the synergistic coupling of multiple parameters of wetting raw material powder granulation liquid volume fraction content and surface energy, adhesive concentration, and granulation liquid flow rate, including a granulation liquid flow rate and adhesive concentration control module, a raw material powder weight loss feeding control module, a granulation particle size and granulation yield model prediction control module, a wetting raw material powder granulation liquid volume fraction content control module, a granulation liquid flow rate and adhesive concentration control module, a powder weight loss feeding control module, a granulation particle size and particle size yield model prediction control module, and a wetting raw material powder granulation liquid volume fraction content control module are all connected to the control system host 22;

[0089] The granulation liquid flow rate and adhesive concentration control module is used for the dynamic real-time preparation of granulation liquid with online adjustable adhesive concentration, and the online real-time monitoring and control of the granulation liquid delivery flow rate and the adhesive concentration of the granulation liquid;

[0090] The powder loss-in-weight feeding control module is used for online real-time monitoring and control of the weight loss feeding rate of the raw material powder of the granulator;

[0091] Granulation particle size and granulation yield model predictive control module is used to achieve the critical quality attribute d of endpoint granulation in a continuous twin-screw wet granulator 10 d 50 d 90 , online real-time monitoring and refined scientific quantitative control of particle size span q and granulation yield Y;

[0092] The wetting raw material powder granulation liquid volume fraction content control module is used for online real-time monitoring and refined scientific quantitative control of the granulation liquid volume fraction content in the wetting raw material powder in the granulation area of ​​the granulator, as well as indirect control of the raw material powder conveying flow rate and screw speed of the granulator granulator screw.

[0093] The granulation liquid flow and adhesive concentration control module includes a water solvent storage tank, a water solvent constant flow pump 9 for conveying water solvent to a small single screw mixing device 8, a first frequency converter 10 for regulating the speed of the water solvent constant flow pump 9, a water solvent flow transmitter 11, an adhesive storage tank, an adhesive constant flow pump 12 for conveying adhesive to the small single screw mixing device 8, a second frequency converter 13 for regulating the speed of the adhesive constant flow pump 12, an adhesive flow transmitter 14, a double closed-loop ratio controller 15 for regulating the granulation liquid flow and adhesive concentration, and a small single screw mixing device 8 for quickly and evenly mixing water solvent and adhesive to prepare granulation liquid, the first frequency converter 10, the water solvent flow transmitter 11, the second frequency converter 13 and the adhesive flow transmitter 14 are all connected to the double closed-loop ratio controller 15, and the double closed-loop ratio controller 15 is connected to the control system host 22;

[0094] The water solvent flow transmitter 11 is arranged at the outlet of the water solvent constant flow pump 9, and is used to detect the actual flow of the water solvent in real time, and transmit its measurement signal to the double closed-loop ratio controller 15 for feedback;

[0095] The adhesive flow transmitter 14 is arranged at the liquid outlet of the adhesive constant flow pump 12, and is used to detect the actual flow of the adhesive in real time, and transmit its measurement signal for feedback to the double closed-loop ratio controller 15;

[0096] The double closed-loop ratio controller 15 includes a flow ratio device 16, a main flow controller 17 for regulating the flow of a water solvent, and a follow-up flow controller 18 for regulating the flow of an adhesive. Both the main flow controller 17 and the follow-up flow controller 18 are connected to the flow ratio device 16.

[0097] The double closed-loop ratio controller 15 sets the set value of the main flow rate of the water solvent and the set value of the flow rate ratio of the water solvent and the adhesive, and sends the set values ​​to the main flow controller 17 and the flow ratio device 16 respectively. The main flow controller 17 sends the main flow control instruction to the first frequency converter 10 according to the deviation between the main flow set value and the main flow data detected in real time by the water solvent flow transmitter 11. The first frequency converter 10 adjusts the output frequency and controls the speed of the driving motor of the water solvent constant flow pump 9 to realize real-time control of the main flow rate of the water solvent; the double closed-loop ratio controller 15 then dynamically updates the adhesive flow set value of the follow-up flow controller 18 online according to the measured main flow rate of the water solvent and the flow ratio set value of the flow ratio device 16, and The adhesive flow set value is sent to the follow-up flow controller 18; the follow-up flow controller 18 sends the adhesive flow control instruction to the second frequency converter 13 according to the deviation between the adhesive flow set value and the adhesive flow data detected in real time by the adhesive flow transmitter 14, and the second frequency converter 13 adjusts the output frequency, regulates the speed of the driving motor of the adhesive variable frequency constant flow pump 12, and realizes the real-time control of the main flow rate of the adhesive, and finally realizes the dynamic online real-time preparation of the granulating liquid with online adjustable adhesive concentration, as well as the online real-time monitoring and precise control of the water solvent flow, adhesive flow, granulating liquid flow and adhesive concentration, and thereby realizes the dynamic real-time preparation of the granulating liquid with online adjustable adhesive concentration through the small single-screw mixing device 8.

[0098] like Figure 3 As shown, the embodiment of the present invention constructs a relationship curve between the water solvent delivery flow rate, the adhesive delivery flow rate and the corresponding constant flow pump driving motor speed by real-time detecting the delivery flow rate of the water solvent constant flow pump 9, the delivery flow rate of the adhesive constant flow pump 12 and the corresponding constant flow pump driving motor speed through the correlation data of the granulation liquid flow rate and the adhesive concentration control module.

[0099] like Figure 4As shown, the embodiment of the present invention simultaneously constructs a synergistic correlation curve between the concentration of the PVP adhesive in the granulation liquid and the speed ratio of the driving motor of the water solvent constant flow pump 9 and the driving motor of the adhesive constant flow pump 12. A synergistic correlation prediction model of the concentration of the PVP adhesive in the granulation liquid and the speed ratio is constructed, and the formula of the synergistic correlation prediction model is obtained as follows:

[0100]

[0101] Where: C V is the concentration of PVP adhesive in the granulation liquid, %; R is the speed ratio of the water solvent constant flow pump 9 to the speed of the PVP adhesive constant flow pump 12, and e is a natural constant.

[0102] Based on the synergistic correlation prediction model formula (7) of the granulation liquid PVP adhesive concentration and the rotation speed ratio, the granulation liquid PVP adhesive concentration C V The set value of the speed ratio R between the speed of the water solvent constant flow pump 9 and the speed of the PVP adhesive constant flow pump 12 can be predicted and calculated, thereby the speed ratio R between the speed of the water solvent constant flow pump 9 and the speed of the PVP adhesive constant flow pump 12 can be controlled by the first frequency converter 10 and the second frequency converter 13. When the above-mentioned speed ratio R set value is met, the online real-time monitoring and fine regulation of the granulation liquid PVP adhesive concentration can be achieved.

[0103] The present invention provides a dynamic preparation and precise quantitative delivery control technology for granulating liquid with online controllable adhesive concentration by setting a granulating liquid flow rate and an adhesive concentration control module, which can realize dynamic online precise regulation of the viscosity, surface tension and contact angle of the granulating liquid, thereby making the prediction and control of the liquid bridge contact force in the granulating area more accurate and stable, solving the defects that the existing control technology is not accurate enough in predicting and controlling the liquid bridge contact force, and the viscosity, surface tension and contact angle of the granulating liquid cannot be dynamically regulated during the granulation process.

[0104] The present invention uses a dynamic preparation and precise quantitative conveying control technology for granulating liquid with online controllable binder concentration. Under the condition of keeping the granulating liquid spray flow rate constant, the binder concentration can be dynamically adjusted online to precisely control the viscosity, surface tension and contact angle of the granulating liquid to increase the liquid bridge contact force, and avoids the inherent defects of the existing control technology that the consistency of the residual moisture at the end point of the dry granulation is difficult to control and the energy efficiency of the system is increased.

[0105] like Figure 1As shown, the granulation particle size and granulation yield model prediction control module includes a granulation particle size online detection system, a granulation particle size, a granulation yield and a granulation liquid volume fraction content and surface energy of the raw material powder, and a granulation liquid adhesive concentration synergistic coupling correlation prediction model, and a granulation particle size and granulation yield model prediction controller 23. The granulation particle size online detection system and the granulation particle size and granulation yield model prediction controller 23 are all connected to the control system host 22. The granulation particle size, the granulation yield and the granulation liquid volume fraction content of the raw material powder, the surface energy of the wet raw material powder, and the adhesive concentration synergistic coupling correlation prediction model is constructed through intelligent digital simulation of the granulation process of the continuous twin-screw wet granulator.

[0106] The raw material powder weight loss feeding control module includes a feeding controller 6, a weighing sensor 5 and a fourth frequency converter 7. The fourth frequency converter 7 is used to adjust the speed of the screw drive motor of the screw weight loss feeder 4. The weighing sensor 5 is connected to the feeding controller 6, and is used to detect the weight loss of the raw material powder of the screw weight loss feeder 4 per unit time online in real time, obtain the online feeding rate of the raw material powder of the granulator, and send the online real-time detection data signal as feedback to the feeding controller 6. The feeding controller 6 sends a control instruction to the fourth frequency converter 7 according to the deviation between the online feeding rate of the raw material powder and the feeding rate set by the user. The fourth frequency converter 7 controls the speed of the screw drive motor of the screw weight loss feeder 4 by adjusting the output frequency, thereby realizing real-time control of the raw material powder feeding rate; and the feeding controller 6 is connected to the granulation particle size and granulation yield model prediction controller 23, and the feeding controller 6 feeds back the raw material powder online feeding rate signal to the granulation particle size and granulation yield model prediction controller 23.

[0107] The online particle size detection system includes an inclined conveying channel 19 and Eyecon2 TM Online particle analyzer 20, inclined conveying channel 19 is set below the discharge port of the granulator, which is used to output the terminal particles of the discharge port at a uniform speed and slowly, so as to improve the particle size detection accuracy, Eyecon2 TM The online particle analyzer 20 is located directly above the inclined conveying channel 19 and is used to detect the particle size, particle size distribution and particle size morphology of the end point particles in real time to obtain the particle size distribution of the end point particles. 10 d 50 d 90 , particle size span and granulation yield values, Eyecon2 TM The online particle analyzer 20 is connected to the control system host 22 and transmits the endpoint particles d detected in real time. 10 d 50 d 90 , particle size span, and granulation yield data signals are sent as feedback to the control system host 22.

[0108] The granulation liquid volume fraction content control module for wetting the raw material powder includes a multi-fiber probe near-infrared spectrum detector 21 for online real-time detection of the wet granule humidity at the end point, a flow ratio controller 24 for real-time adjustment of the ratio of the granulation liquid spray flow rate to the granulation screw raw material powder delivery flow rate, and a third frequency converter 3 for adjusting the speed of the driving motor 2 of the twin screw of the granulator. The multi-fiber probe near-infrared spectrum detector 21 and the flow ratio controller 24 are both connected to the control system host 22;

[0109] The multiple fiber optic probes of the online multi-fiber probe near-infrared spectrometer 21 are arranged directly above the inclined conveying channel 19, and are used for multi-point real-time detection of the humidity of the wet particles at the outlet of the granulator, obtaining the average humidity of the wet particles at the outlet of the granulator, and feeding back the measured data signal to the control system host 22. The control system host 22 calculates the real-time data of the granulation liquid volume fraction content C and the fluid-to-solid ratio L / S of the wetting raw material powder according to the measured average humidity of the terminal particles, and feeds back the real-time data signals of the granulation liquid volume fraction content C and the fluid-to-solid ratio L / S to the flow ratio controller 24. The flow ratio controller 24 sends a granulation screw raw material powder conveying flow rate control instruction to the third frequency converter 3 based on the deviation between the real-time data of the fluid-to-solid ratio L / S and the set value of the fluid-to-solid ratio L / S. The third frequency converter 3 adjusts the rotation speed of the granulation screw by adjusting the output frequency, and finally realizes the online real-time monitoring and refined scientific quantitative control of the granulation liquid volume fraction content of the wetting raw material powder, as well as the indirect control of the raw material powder conveying flow rate and the screw speed of the granulator granulator.

[0110] In order to construct an online real-time controllable control system for the key quality attributes of a continuous twin-screw wet granulator that uses technology to ensure success, and to achieve online real-time controllable granulation of the continuous twin-screw wet granulator, the control system needs to be able to achieve both the refined scientific quantitative online real-time comprehensive regulation of the liquid bridge contact force of the granulating liquid and the cohesive contact force of the wetted raw material powder, and the scientific quantitative characterization and online refined regulation of the essential characteristics of the synergistic coupling effect of the pre-compression of the raw material powder formed by the rotation of the twin screws and the periodic grinding and pressing formed by the rotation of the granulating KE blocks during the granulation process. To this end, the embodiment of the present invention wets the raw material powder in the twin screw The surface energy E measured under the equivalent working conditions of periodic pre-compression and grinding of the rod is used to scientifically and quantitatively characterize the synergistic coupling effect of the raw material powder pre-compression formed by the rotation of the twin screw and the periodic grinding formed by the rotation of the granulating KE block during the granulation process. By developing a multi-parameter synergistic coupling model predictive controller based on the granulation liquid volume fraction content C of the wet raw material powder and the surface energy E, the granulation liquid binder concentration, and the granulation screw speed, the key quality attributes of the continuous twin-screw wet granulator can be controlled in a refined, scientific, quantitative, and online real-time manner. Only in this way can the online real-time control of the continuous twin-screw wet granulator 1 be achieved with science for quality and technology for success.

[0111] Therefore, the embodiment of the present invention constructs a synergistic coupling correlation prediction model of granulation particle size, granulation liquid volume fraction content C of wetting raw material powder, surface energy E, and granulation liquid binder concentration, and the construction method thereof comprises the following steps:

[0112] S1. On the control system host 22, based on the EEPA constitutive model of liquid bridge contact force and viscoelastic-plastic contact force, an embedded C language program for calculating the liquid bridge contact force and the viscoelastic-plastic contact force is designed and written, and a two-way data transmission interface with the discrete element simulation software is realized, wherein the embedded C language program automatically recognizes that the constitutive model is only applied to the granulation zone downstream of the granulation liquid nozzle in the granulator;

[0113] S2. Based on the integration of the embedded C language program, the bidirectional data transmission interface and the discrete element simulation software, a multi-scale intelligent digital simulation platform for the granulation process of the continuous twin-screw wet granulator is secondary developed. The simulation platform can truly and quantitatively characterize the essential characteristics of the synergistic coupling effect of the twin-screw pre-compression and the granulation KE block periodic grinding and pressing in the granulation process through the surface energy of the wetted raw material powder measured under equivalent experimental working conditions, and can describe and simulate the macroscopic dynamic granulation process of the continuous twin-screw wet granulator 1 from the dynamic response of the particle scale;

[0114] S3, constructing a three-dimensional digital functional virtual prototype of the continuous twin-screw wet granulator 1 on the simulation platform;

[0115] S4, determining the rheological parameters of the granulation liquid by offline experiments, and constructing a synergistic correlation prediction model between the rheological parameters of the granulation liquid and the binder concentration of the granulation liquid;

[0116] S5. Under the equivalent working conditions of twin-screw pre-compression and periodic grinding and pressing of granulation KE block, the surface energy E of the wetted raw material powder after spraying granulation liquid with different binder concentrations is measured offline by inverse gas chromatography surface energy analyzer, and a synergistic correlation prediction model of the surface energy E and the binder concentration in the granulation liquid is constructed;

[0117] S6. Design and write an embedded computing program on the control system host 22 for a synergistic correlation prediction model of rheological parameters of the granulating liquid, surface energy of the wetting raw material powder and adhesive concentration;

[0118] S7, inputting granulation process parameters and raw material powder physical property parameters on the simulation platform;

[0119] S8, starting the intelligent digital simulation platform for the granulation process developed in step S2;

[0120] S9, automatically and dynamically update the data required for calculating the liquid bridge contact force and viscoelastic-plastic contact force C language embedded program;

[0121] S10, start the embedded calculation program of liquid bridge viscous contact force and EEPA viscoelastic-plastic contact force, and calculate and obtain the liquid bridge viscous contact force and EEPA viscoelastic-plastic contact force;

[0122] S11, dynamically transferring the calculation result data of the liquid bridge viscous contact force and the EEPA viscoelastic-plastic contact force in step S10 to the discrete element software through the bidirectional data transmission interface between the embedded C language program and the discrete element software;

[0123] S12, automatic dynamic update of the liquid bridge contact force and viscoelastic-plastic contact force of the discrete element software;

[0124] S13, under the experimental operating parameters, the granulation process intelligent digital simulation platform developed in step S2 is used to simulate and predict the key quality attribute characterization parameters d of the granulator end granule product. 10 d 50 d 90 , particle size span q, granulation yield Y;

[0125] S14, continuous twin-screw wet granulation experiment, Eyecon2 TM Online particle analyzer 20 Online real-time detection of endpoint particle product key quality attribute characterization parameters d 10 d 50 d 90 and particle size span q, and then verify the test results with the simulation prediction results of step S13, if the error between the simulation prediction results and the test results is ≥±5%, modify and improve the intelligent digital simulation platform of the granulation process until the simulation prediction results are consistent with the experimental test results, and the error is controlled within the range of ±5%;

[0126] S15. Characterize the key quality attribute parameters of the end product 10 d 50 d 90 The particle size span q, granulation yield Y and the granulation liquid volume fraction content of the wetted raw material powder and the surface energy, granulation liquid binder concentration, and screw speed are simulated by the synergistic coupling evolution law. The data regression theory modeling of the simulation data is simulated according to the synergistic coupling evolution law, and the granulation particle size d is finally obtained. 90 , the synergistic coupling correlation control prediction model of granulation yield Y and the granulation liquid volume fraction content of the raw material powder, the surface energy of the raw material powder, and the concentration of the granulation liquid binder, and the synergistic coupling correlation control prediction model of the granulation yield Y and the screw speed.

[0127] like Figures 5 to 16 Shown are multiple control prediction model curves constructed by an intelligent digital simulation platform according to an embodiment of the present invention.

[0128] like Figure 5As shown, the embodiment of the present invention measures the viscosity of the granulation liquid with different PVP adhesive concentrations by a rheometer, constructs a correlation prediction model between the viscosity of the granulation liquid in step S4 and the PVP adhesive concentration, and obtains the correlation prediction model formula as follows:

[0129]

[0130] In formula (8): μ is the viscosity of the granulation liquid, mPa.s; C V is the concentration of PVP adhesive, %.

[0131] like Figure 6 As shown, in the embodiment of the present invention, the surface energy E of the wetted MCC raw material powder under the condition of cooperative coupling is measured offline by an inverse gas chromatography surface energy analyzer IGC SEA, and a correlation prediction model curve of the wetted MCC raw material powder surface energy E and the binder PVP concentration is constructed, which presents a quadratic parabola. Based on this, the calculation formula of the correlation prediction model in step S5 is E=50.16+26.12﹒ C V -2.59﹒ C V 2 .

[0132] In the embodiment of the present invention, different raw material powders and / or different adhesives are used to construct the above correlation prediction model curve, and the correlation prediction model curve is obtained to be a quadratic parabola. Based on this, the correlation prediction model formula of the surface energy E of the wetted raw material powder particles and the binder concentration is obtained as follows:

[0133] E=α0+α1·C V +α2·C V 2 (1)

[0134] Where E is the surface energy of the wetted raw material powder, C V is the binder concentration of the granulation solution, α0, α1, and α2 are constants for data regression modeling;

[0135] However, the data regression constants α0, α1, and α2 of different raw material powders and / or different adhesives are all different. For example, the data regression constants of raw material powder MCC and adhesive PVP are α0=50.16, α1=26.12, and α2=-2.59;

[0136] The synergistic correlation prediction model of the surface energy E of the wetted raw material powder and the binder concentration of the granulation liquid can truly characterize the key influence of the binder concentration, the pre-compression and grinding coupling effect of the raw material powder on the granulation process and surface energy.

[0137] like Figure 7 , 8 As shown in 9, the maximum particle size d of the granulation in the embodiment of the present invention 90It shows a positive correlation with the wetting powder surface energy E and a nonlinear evolution law. When the wetting powder surface energy E is less than 10 J / m 2 When the maximum particle size and d90 of the granulation are increased, the surface energy E of the wetted powder increases slowly. 2 When the maximum particle size and d 90 It increases rapidly with the increase of the surface energy E of the wet powder. It can be seen that the granulation of the embodiment of the present invention tends to prepare particles with larger particle sizes. The surface energy E of the wet powder has a small influence on the particle size d50 and d10, showing a trend of slight increase or decrease. The solid oral preparation pellets require a specific particle size distribution range. It is now given that the solid oral preparation pellets require a specific particle size distribution range of 250μm to 1050μm. The final product particle size exceeding this range is considered to be unqualified granulation. The granulation of the embodiment of the present invention is prone to a large number of coarse particles, which makes its one-time granulation yield low. How to control and suppress its coarse particles and study the construction of its granulation particle size, particle size yield and process parameters, granulation liquid spray volume, wet powder surface energy E synergistic coupling correlation control prediction model is particularly important.

[0138] like Fig.10 The following is the evolution curve of characteristic particle size and screw speed n (the spray liquid volume fraction C is 10%, the wetted powder surface energy E is 20 J / m 2 ), the study shows that the granulation quality characteristic parameter d of the embodiment of the present invention 90 d 50 It shows a negative correlation evolution law with the screw speed n. As the screw speed n increases, the pelleting quality characteristic parameter d 90 d 50 It can be seen that in order to control and suppress the generation of coarse and unqualified particles and improve the granulation yield Y, it is necessary to construct a synergistic coupling correlation control prediction model of the granulation yield Y and the screw speed n under different wetting raw material powder volume fraction content and surface energy E conditions.

[0139] Fig.11 It is the synergistic coupling correlation control curve of granulation yield Y and wetted powder surface energy E when the volume fraction content C of granulation liquid is 10%; Fig.12 It is the synergistic coupling correlation control curve of granulation yield Y and screw speed n when the volume fraction content C of granulation liquid is 10%; Fig.13 It is the synergistic coupling correlation control curve of granulation yield Y and wetted powder surface energy E when the volume fraction content C of granulation liquid is 20%; Fig.14 It is the synergistic coupling correlation control curve of granulation yield Y and screw speed n when the granulation liquid volume fraction content C is 20%.

[0140] Through the data regression concept modeling method, the calculation formula of the synergistic coupling correlation control prediction model of the granulation yield Y and the screw speed n in the embodiment of the present invention is obtained as follows:

[0141] Y=B0+B1.n+B2.n 2 +B3.n 3 (4)

[0142] Wherein: Y is the granulation yield, %; n is the screw speed, rpm; B0 to B3 are model regression constants, and their values ​​are shown in Table 1.

[0143] Table 1

[0144]

[0145]

[0146] According to the synergistic coupling correlation control prediction model of granulation yield Y and screw speed n in formula (1), when the granulation liquid volume fraction content is 10% and the surface energy is 20 J / m 2 The optimal granulation yield was 78.3% when the screw speed n was 800 rpm, and the optimal granulation yield was 78.3% when the granulation liquid volume fraction content was 20% and the surface energy was 20 J / m 2 When the screw speed is 1200 rpm, the optimal granulation yield is 68.54%. The increases of the primary granulation yield in the embodiment of the present invention are 13% and 7.8% respectively. If the unqualified coarse particles are subjected to secondary granulation, the final granulation yield will exceed 85%.

[0147] Fig.15 and Fig.16 The particle size d 90 , granulation yield Y, granulation liquid volume fraction content C, and wetting powder surface energy E synergistic coupling correlation control curve.

[0148] in accordance with Fig.15 The simulation prediction correlation control curve data is used to obtain the granulation particle size d under different working conditions through the multi-parameter data regression theory modeling method. 90 The calculation formula of the synergistic coupling correlation control prediction model with the granulation liquid volume fraction content C and the surface energy E of the wetted raw material powder is as follows:

[0149]

[0150] In formula (2): d 90 The particle size d is the particle size at the end of granulation. 90 , μm; C is the volume fraction of granulation liquid in wet granules, %; E is the surface energy of wetted raw material powder; d0, A1 to A 15 , B 11 To B15 are all model regression constants of the control prediction model (see Table 2);

[0151] Table 2

[0152] name <![CDATA[d0]]> <![CDATA[A 11 ]]> <![CDATA[A 12 ]]> <![CDATA[A 13 ]]> <![CDATA[A 14 ]]> <![CDATA[A 15 ]]> <![CDATA[B 11 ]]> <![CDATA[B 12 ]]> <![CDATA[B 13 ]]> <![CDATA[B 14 ]]> <![CDATA[B 15 ]]> <![CDATA[d 90 ]]> 263.43 382.75 -65.02 5.065 -0.181 0.00241 -7.64 6.51 -1.052 0.063 -0.0013

[0153] Substituting the regression constants in Table 2 into formula (2), the calculation formula of the synergistic coupling correlation control prediction model of the terminal granulation particle size d90, the granulation liquid volume fraction content C, and the wetting raw material powder surface energy E is obtained as follows:

[0154]

[0155] in accordance with Fig.16 The simulation prediction correlation control curve data is used to obtain the granulation yield Y and the granulation liquid volume fraction content C, and the surface energy E of the wetted raw material powder under different working conditions through the multi-parameter data regression theory modeling method. The calculation formula of the synergistic coupling correlation control prediction model is as follows:

[0156] Y=Y0+A 21 ·C+A 22 ·C 2 +A 23 ·C 3 +A 24 ·C 4 +A 25 ·C 5 +B 21 ·E+B 22 ·E 2 +B 23 ·E 3 +B 24 ·E 4 +B 25 ·E 5 (3)

[0157] In formula (3), Y is the granulation yield, %; C is the volume fraction of the granulation liquid, %; E is the surface energy of the wetted raw material powder; Y0, A 21 To A 25 , B 21 To B 25 are all model regression constants of the control prediction model (see Table 3);

[0158] Table 3

[0159] name <![CDATA[Y0]]> <![CDATA[A 21 ]]> <![CDATA[A 22 ]]> <![CDATA[A 23 ]]> <![CDATA[A 24 ]]> <![CDATA[A 25 ]]> <![CDATA[B 21 ]]> <![CDATA[B 22 ]]> <![CDATA[B 23 ]]> <![CDATA[B 24 ]]> <![CDATA[B 25 ]]> Y 1.1036 31.104 -5.283 0.4353 -0.0177 2.8093 2.552 -0.7075 0.0646 -0.0021 <![CDATA[1.54×10 -5 ]]>

[0160] Substituting the regression constants in Table 3 into formula (3), the calculation formula of the synergistic coupling correlation control prediction model of the endpoint granulation yield Y, the granulation liquid volume fraction content C, and the wetting raw material powder surface energy E is obtained as follows:

[0161] Y=1.1036+31.104·C-5.283·C 2 +0.4353·C 3 -0.0177 C 4 +2.8093·C 5 +2.552·E-0.7075·E 2 +0.0646·E 3 -0.0021·E 4 +1.54×10 -5 ·E 5 (6)

[0162] Note: The volume fraction content C of the granulation liquid is the volume of the granulation liquid sprayed into 100 parts of the wet raw material powder in the granulation area of ​​the granulation device.

[0163] In the embodiment of the present invention, the granulation particle size d obtained above is programmed on the control system host 22. 90 , the granulation yield Y and the granulation liquid volume fraction content C and surface energy E of the raw material powder, the granulation liquid binder concentration, the granulation yield Y and the screw speed n synergistic coupling correlation control prediction model, and the control algorithm (including formulas 1 to 6) of the corresponding correlation control prediction model are finally obtained. The granulation particle size and particle size yield model prediction controller 23 can achieve the key quality attribute d of the terminal granulation of the continuous twin-screw wet granulator 1 by online real-time regulation of the granulation liquid volume fraction content C and surface energy E of the raw material powder, the granulation liquid binder concentration, and the screw speed n. 10 d 50 d 90 , fine scientific quantitative control of particle size span q, granulation yield Y;

[0164] The granulation particle size and particle size yield model prediction controller 23 is connected to the feeding controller 6, the double closed-loop ratio controller 15, and the flow ratio controller 24.

[0165] The control method of the continuous twin-screw wet granulation control system of the embodiment of the present invention is an online fine control method of granulation particle size and particle size yield based on the multi-parameter synergistic coupling of the granulation liquid volume fraction content C and surface energy E of the wetting raw material powder, the granulation liquid binder concentration, and the screw speed n, comprising the following steps:

[0166] T1. First, set the raw material powder weight loss feeding rate and the final granulation particle size d on the control system host 22. 90 , qualified particle size distribution range, expected value of granulation yield Y;

[0167] T2, according to the granulation particle size d90 The controller 23 predicts the particle size and the granulation yield by the model of the associated control prediction model formula (4), (5), (6) and the optimization algorithm, and the global collaborative optimization prediction gives the optimal parameter combination of the granulation liquid volume fraction content C, the surface energy E, and the screw speed of the wetted raw material powder to obtain the optimal granulation yield. Then, according to the optimal granulation liquid volume fraction content C and the associated prediction model formula (1), the binder concentration C of the granulation liquid is predicted and calculated. V ;

[0168] T3. Start the continuous twin-screw wet granulation system and start granulation production;

[0169] T4, the granulation particle size and granulation yield model prediction controller 23 automatically calculates and dynamically updates the set value of the granulation liquid spray flow process parameter according to the real-time data of the raw material powder weight loss feeding rate fed back by the feeding controller 6 and the granulation liquid volume fraction content C of the optimal raw material powder wetting predicted in step T2;

[0170] T5, Eyecon2 TM The online particle analyzer 20 detects the endpoint particle diameter d10, d50, d90, and particle diameter span q information online and in real time, and transmits the real-time measurement signal to the control system for feedback;

[0171] T6, the granulation particle size and granulation yield model prediction controller 23 calculates and analyzes the consistency of the granulation particle size and particle size distribution online in real time, and predicts the granulation process of the next time step by associating the control prediction model formulas (1), (4), (5), (6) with the prediction model control strategy based on the deviation between the granulation endpoint particle size detected in real time and the user's expected particle size, and calculates the granulation liquid volume fraction content C and surface energy E of the externally controlled wetted raw material powder, the granulation liquid spray flow rate, and the granulation liquid adhesive concentration parameter setting values ​​for obtaining the optimal granulation yield through the global collaborative optimization algorithm;

[0172] T7, the double closed-loop ratio controller 15 automatically and dynamically updates the process parameter setting values ​​of the granulation liquid spray flow rate and the granulation liquid adhesive concentration according to the optimal parameter setting value obtained in step T6. The granulation liquid flow rate and adhesive concentration control module realizes the online real-time monitoring and fine regulation of the granulation liquid flow rate and the granulation liquid adhesive concentration through the water solvent flow transmitter 11, the adhesive flow transmitter 14 and the double closed-loop ratio controller 15, and the dynamic real-time preparation of the granulation liquid with the required optimal adhesive concentration;

[0173] T8, the granulation liquid volume fraction content control module for wetting the raw material powder automatically and dynamically updates the set value of the fluid-solid ratio L / S of the wetting raw material powder according to the set value of the optimal granulation liquid volume fraction content C of the wetting raw material powder and the set value of the optimal granulation liquid spray flow rate obtained in step T6, the multi-fiber probe near-infrared spectrometer 21 detects the humidity of the wet granules at the outlet of the granulator at multiple points online in real time, and sends the multi-point real-time humidity detection data to the control system host 22, and the control system host 22 calculates and obtains the average humidity of the granulated wet granules, the granulation liquid volume fraction content C of the wetting raw material powder and the The real-time data of the fluid-solid ratio L / S is collected, and the granulation liquid volume fraction content C and the real-time data signal of the fluid-solid ratio L / S are sent and fed back to the flow ratio controller 24. The flow ratio controller 24 sends a raw material powder conveying flow control instruction to the third frequency converter 3 according to the deviation between the real-time monitoring data of the fluid-solid ratio L / S and the set value of the fluid-solid ratio L / S. The frequency converter adjusts the output frequency and the rotation speed of the granulating screw to achieve online real-time monitoring and refined control of the granulation liquid volume fraction content C of the wetted raw material powder, as well as indirect control of the raw material powder conveying flow rate and the screw rotation speed of the granulator granulator screw;

[0174] T9, through the control module of the volume fraction content of the granulating liquid wetting the raw material powder, the dynamic online control of the volume fraction content C of the granulating liquid wetting the raw material powder, the raw material powder conveying flow rate, the screw speed, and the raw material powder pre-compression working state and the periodic grinding and pressing working state of the granulation KE block can be realized, and then the double closed-loop ratio controller 15 of step T7 is coordinated to realize the dynamic online precision control of the granulating liquid spray flow rate and the granulating liquid binder solubility, as well as the surface energy E control of step T6. Combining the above-mentioned multi-parameter collaborative coupling control means, the dynamic online real-time monitoring and precision control of the terminal granulation particle size, particle size distribution and granulation yield can be finally realized.

[0175] The granulation control system of the embodiment of the present invention indirectly and finely coordinates and controls the granulation liquid volume fraction content C and the surface energy E online through real-time online coordinated coupling control of the granulation liquid volume fraction content C-binder concentration-granulation twin-screw speed n of the wetting raw material powder, and finally realizes the online real-time fine adjustment and control of the granulation endpoint granulation particle size, particle size distribution and granulation yield based on the coordinated coupling of the granulation liquid volume fraction content C and the surface energy E, thereby solving the common technical problem of low granulation yield of traditional continuous twin-screw wet granulation machine. The granulation yield of the granulation control system of the embodiment of the present invention can reach more than 85%.

[0176] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention. These improvements and substitutions should also be regarded as the scope of protection of the present invention.

Claims

1. A continuous twin-screw wet granulation control system, applied to a continuous twin-screw wet granulator for solid oral preparations, is an online real-time control system for granulation particle size, particle size distribution, and particle size yield based on the synergistic coupling of the volume fraction content of the granulation liquid wetting the raw material powder and the surface energy, the concentration of the adhesive, and the flow rate of the granulation liquid, characterized in that: It includes a granulation liquid flow and adhesive concentration control module, a raw material powder weight loss feeding control module, a granulation particle size and granulation yield model prediction control module, and a raw material powder granulation liquid volume fraction content control module, wherein the granulation liquid flow and adhesive concentration control module, the raw material powder weight loss feeding control module, the granulation particle size and particle size yield model prediction control module, and the raw material powder granulation liquid volume fraction content control module are all connected to the control system host; The granulation liquid flow rate and adhesive concentration control module is used for dynamic real-time preparation of granulation liquid with online adjustable adhesive concentration, and online real-time monitoring and control of granulation liquid delivery flow rate and adhesive concentration of granulation liquid; The raw material powder weight loss feeding control module is used for online real-time monitoring and control of the raw material powder weight loss feeding rate of the granulator; The granulation particle size and granulation yield model prediction control module is used to achieve the critical quality attribute d of the granulator endpoint granulation 10 d 50 d 90 , online real-time monitoring and refined scientific quantitative control of particle size span q and granulation yield Y; The wetting raw material powder granulation liquid volume fraction content control module is used for online real-time monitoring and refined scientific quantitative control of the volume fraction content of the wetting raw material powder in the granulation zone of the granulator, and indirect control of the raw material powder conveying flow rate and screw speed of the granulator granulator screw; The granulation particle size and granulation yield model prediction control module includes an online detection system for granulation particle size, a granulation particle size, a granulation yield, a granulation liquid volume fraction content and surface energy of a raw material powder, and a granulation liquid binder concentration synergistic coupling association prediction model, and a granulation particle size and granulation yield model prediction controller. The online detection system for granulation particle size and the granulation particle size and granulation yield model prediction controller are both connected to a control system host. The granulation particle size, granulation yield, a granulation liquid volume fraction content of a raw material powder, a wetting raw material powder surface energy, and a binder concentration synergistic coupling association prediction model is constructed through intelligent digital simulation of the granulation process of the granulator; The simulation construction method comprises the following steps: S1. On the control system host, based on the EEPA constitutive model of liquid bridge contact force and viscoelastic-plastic contact force, design and write an embedded C language program for calculating the liquid bridge contact force and the viscoelastic-plastic contact force, and realize a two-way data transmission interface with the discrete element simulation software, wherein the embedded C language program automatically recognizes that the constitutive model is only applied to the granulation zone downstream of the granulation liquid nozzle in the granulator; S2. Based on the integration of the embedded C language program, the bidirectional data transmission interface and the discrete element simulation software, a multi-scale intelligent digital simulation platform for the granulation process of the continuous twin-screw wet granulator is secondary developed. The simulation platform can truly and quantitatively characterize the essential characteristics of the synergistic coupling effect of the twin-screw pre-compression and the granulation KE block periodic grinding and pressing in the granulation process through the surface energy of the wetted raw material powder measured under equivalent experimental conditions, and can describe and simulate the macroscopic dynamic granulation process of the continuous twin-screw wet granulator from the dynamic response of the particle scale; S3. constructing a three-dimensional digital functional virtual prototype of a continuous twin-screw wet granulator on the simulation platform; S4, determining the rheological parameters of the granulation liquid by offline experiments, and constructing a synergistic correlation prediction model between the rheological parameters of the granulation liquid and the binder concentration of the granulation liquid; S5. Under the equivalent working conditions of twin-screw pre-compression and periodic grinding and pressing of granulation KE block, the surface energy E of the wetted raw material powder after spraying granulation liquid with different binder concentrations is measured offline by inverse gas chromatography surface energy analyzer, and a synergistic correlation prediction model of the surface energy E and the binder concentration in the granulation liquid is constructed; S6. Design and write an embedded computing program on the control system host computer for the synergistic correlation prediction model of the rheological performance parameters of the granulation liquid, the surface energy of the wetted raw material powder and the concentration of the adhesive; S7, inputting granulation process parameters and raw material powder physical property parameters on the simulation platform; S8, starting the intelligent digital simulation platform for the granulation process developed in step S2; S9, automatically and dynamically update the data required for calculating the liquid bridge contact force and viscoelastic-plastic contact force C language embedded program; S10, start the embedded calculation program of liquid bridge viscous contact force and EEPA viscoelastic-plastic contact force, and calculate and obtain the liquid bridge viscous contact force and EEPA viscoelastic-plastic contact force; S11, dynamically transferring the calculation result data of the liquid bridge viscous contact force and the EEPA viscoelastic-plastic contact force in step S10 to the discrete element software through the bidirectional data transmission interface between the embedded C language program and the discrete element software; S12, automatic dynamic update of the liquid bridge contact force and viscoelastic-plastic contact force of the discrete element software; S13, under the experimental operating parameters, the granulation process intelligent digital simulation platform developed in step S2 is used to simulate and predict the key quality attribute characterization parameters d of the granulator end granule product. 10 ,d 50 ,d 90 , particle size span q, granulation yield Y; S14. Conduct a continuous twin-screw wet granulation experiment. The online particle size detection system detects the key quality attribute characterization parameter d of the end-point granule product in real time. 10 d 50 d 90 and particle size span q, and then verify the test results with the simulation prediction results of step S13. If the error between the simulation prediction results and the test results is ≥±5%, modify and improve the intelligent digital simulation platform of the granulation process until the simulation prediction results are consistent with the experimental test results, and the error is controlled within the range of ±5%; S15. Characterize the key quality attribute parameters of the end product 10 ,d 50 ,d 90 The particle size span q, granulation yield Y and the granulation liquid volume fraction content of the wetted raw material powder and the surface energy, granulation liquid binder concentration, and screw speed are simulated by the synergistic coupling evolution law. The data regression theory modeling of the simulation data is simulated according to the synergistic coupling evolution law, and the granulation particle size d is finally obtained. 90 , the synergistic coupling correlation control prediction model of granulation yield Y and the granulation liquid volume fraction content of the raw material powder, the surface energy of the raw material powder, and the concentration of the granulation liquid binder, and the synergistic coupling correlation control prediction model of the granulation yield Y and the screw speed.

2. A continuous twin-screw wet granulation control system according to claim 1, characterized in that: The granulation liquid flow and adhesive concentration control module comprises a water solvent storage tank, a water solvent constant flow pump for conveying water solvent, a first frequency converter for regulating the speed of the water solvent constant flow pump, a water solvent flow transmitter, an adhesive storage tank, an adhesive constant flow pump for conveying adhesive, a second frequency converter for regulating the speed of the adhesive constant flow pump, an adhesive flow transmitter, a double closed-loop ratio controller for regulating the granulation liquid flow and adhesive concentration, and a small single-screw mixing device for quickly and evenly mixing the water solvent and the adhesive to prepare the granulation liquid, wherein the first frequency converter, the water solvent flow transmitter, the second frequency converter and the adhesive flow transmitter are all connected to the double closed-loop ratio controller, and the double closed-loop ratio controller is connected to the control system host; The water solvent flow transmitter is arranged at the outlet of the water solvent constant flow pump, and is used to detect the actual flow of the water solvent in real time, and transmit its measurement signal for feedback to the double closed-loop ratio controller; The adhesive flow transmitter is arranged at the outlet of the adhesive constant flow pump, and is used to detect the actual flow of the adhesive in real time, and transmit its measurement signal for feedback to the double closed-loop ratio controller; The double closed-loop ratio controller includes a flow ratio device, a main flow controller for regulating the flow of a water solvent, and a follow-up flow controller for regulating the flow of an adhesive, wherein both the main flow controller and the follow-up flow controller are connected to the flow ratio device; The dual closed-loop ratio controller adjusts the speed of the water solvent constant flow pump with the help of the main flow controller and the first frequency converter according to the main flow setting value of the water solvent and the real-time detection data deviation of the water solvent flow transmitter, so as to realize online regulation of the main flow of the water solvent. The dual closed-loop ratio controller then dynamically updates the adhesive flow setting value of the follow-up flow controller according to the measured main flow of the water solvent and the flow ratio setting value of the flow ratio converter. The dual closed-loop ratio controller adjusts the speed of the adhesive constant flow pump with the help of the follow-up flow controller and the second frequency converter according to the adhesive flow setting value and the real-time detection data deviation of the adhesive transmitter, and finally realizes online real-time dynamic regulation of the water solvent flow, adhesive flow, granulation liquid flow and adhesive concentration, and thereby realizes dynamic real-time preparation of granulation liquid with online adjustable adhesive concentration through the small single-screw mixing device.

3. A continuous twin-screw wet granulation control system according to claim 1, characterized in that: The raw powder weight loss feeding control module includes a feeding controller, a weighing sensor and a fourth frequency converter, wherein the fourth frequency converter is used to adjust the speed of the screw driving motor of the raw powder weight loss feeder of the granulator, and the weighing sensor is connected to the feeding controller, and is used to detect the weight loss of the raw powder in the raw powder weight loss feeder per unit time online in real time, obtain the online feeding rate of the raw powder of the granulator, and send the online real-time detection data signal to the feeding controller for feedback, and the feeding controller sends a control instruction to the fourth frequency converter according to the deviation between the online feeding rate of the raw powder and the feeding rate set by the user, and the fourth frequency converter controls the speed of the screw driving motor of the raw powder weight loss feeder by adjusting the output frequency, thereby realizing real-time control of the raw powder feeding rate; the feeding controller is connected to the granulation particle size and granulation yield model prediction controller, and the feeding controller feeds back the raw powder online feeding rate signal to the granulation particle size and granulation yield model prediction controller.

4. A continuous twin-screw wet granulation control system according to claim 1, characterized in that: The granulation particle size online detection system includes an inclined conveying flow channel and an online particle analyzer. The inclined conveying flow channel is arranged below the discharge port of the granulator, and is used to output the terminal particles of the discharge port at a uniform speed and slowly, thereby improving the particle size detection accuracy. The online particle analyzer is arranged just above the inclined conveying flow channel, and is used to detect the particle size, particle size distribution and particle size morphology of the terminal particles in real time, and obtain the particle size distribution of the terminal particles. 10 ,d 50 ,d 90 , particle size span and granulation yield, the online particle analyzer is connected to the control system host, and the endpoint particles d detected in real time are 10 ,d 50 ,d 90 , particle size span, and granulation yield data signals are sent back to the control system host.

5. A continuous twin-screw wet granulation control system according to claim 1, characterized in that: The granulation liquid volume fraction content control module for wetting the raw material powder comprises an inclined conveying flow channel, a multi-fiber probe near-infrared spectrometer for online real-time detection of the wet granule humidity at the end point, a flow ratio controller for real-time adjustment of the ratio of the granulation liquid spray flow rate to the granulation screw raw material powder conveying flow rate, and a third frequency converter for adjusting the speed of the twin-screw drive reducer of the granulator. The inclined conveying flow channel is arranged below the discharge port of the granulator, and is used to output the end granules at the discharge port at a uniform speed and slowly, so as to improve the particle size detection accuracy. The multi-fiber probe near-infrared spectrometer and the flow ratio controller are both connected to the control system host. The multiple fiber optic probes of the online multi-fiber probe near-infrared spectrometer are arranged directly above the inclined conveying flow channel, and are used for multi-point real-time detection of the humidity of the wet particles at the outlet of the granulator, obtaining the average humidity of the wet particles at the outlet of the granulator, and feeding back the measured data signal to the control system host. The control system host calculates the real-time data of the granulation liquid volume fraction content C and the fluid-to-solid ratio L / S of the wetting raw material powder according to the measured average humidity of the terminal particles, and feeds back the real-time data signals of the granulation liquid volume fraction content C and the fluid-to-solid ratio L / S to the flow ratio controller. The flow ratio controller sends a granulation screw raw material powder conveying flow rate control instruction to the third frequency converter according to the deviation between the real-time data of the fluid-to-solid ratio L / S and the set value of the fluid-to-solid ratio L / S. The third frequency converter adjusts the rotation speed of the granulation screw by adjusting the output frequency, and finally realizes the online real-time monitoring and refined scientific quantitative control of the granulation liquid volume fraction content of the wetting raw material powder, as well as the indirect control of the raw material powder conveying flow rate and the screw speed of the granulator granulator.

6. A continuous twin-screw wet granulation control system according to claim 1, characterized in that: The granulation particle size and granulation yield model prediction controller is programmed on the control system host to obtain the granulation particle size d obtained in step S15. 90 , a synergistic coupling correlation control prediction model of granulation yield Y and granulation liquid volume fraction content of wetting raw material powder, surface energy of wetting raw material powder, and concentration of granulation liquid binder, a synergistic coupling correlation control prediction model of granulation yield Y and screw speed, and an embedded computing program of the control algorithm of the corresponding correlation control prediction model, so that the granulation particle size and granulation yield model prediction controller can achieve the key quality attribute d of the endpoint granulation of the continuous twin-screw wet granulator by online real-time regulation of the granulation liquid volume fraction content and surface energy of the wetting raw material powder, the concentration of granulation liquid binder, and the screw speed. 10 d 50 d 90 , online monitoring and refined scientific quantitative control of particle size span q and granulation yield Y; The granulation particle size and granulation yield model prediction controller is connected with the feeding controller, the double closed-loop ratio controller and the flow ratio controller.

7. A continuous twin-screw wet granulation control system according to claim 6, characterized in that: The synergistic correlation prediction model formula of the surface energy E of the wetted raw material powder and the concentration of the adhesive in step S5 is: （1） Where: E is the surface energy of the wetted raw material powder, C V is the binder concentration of the granulation solution, α0, α1, and α2 are constants for data regression modeling; The granulation particle size d in step S15 90 The formula of the synergistic coupling correlation control prediction model of the volume fraction content of the granulation liquid wetting the raw material powder, the surface energy of the wetting raw material powder, and the concentration of the granulation liquid binder is: （2） In formula (2): d 90 is the particle size of the final granulation point, μm; C is the volume fraction of the granulation liquid that wets the raw material powder, %; E is the surface energy of the wetted raw material powder; d0, A 11 To A 15 , B 11 To B 15 They are all model regression constants of the association control prediction model; The synergistic coupling correlation control prediction model formula of the granulation yield Y in step S15, the granulation liquid volume fraction content of the wetted raw material powder, the surface energy of the wetted raw material powder, and the concentration of the granulation liquid binder is: （3） In formula (3), Y is the granulation yield, %; C is the volume fraction of the granulation liquid that wets the raw material powder, %; E is the surface energy of the wetted raw material powder; Y0, A 21 To A 25 , B 21 To B 25 They are all model regression constants of the association control prediction model; The prediction model formula of the granulation yield Y and the screw speed synergistic coupling correlation control in step S15 is: （4） In formula (4), Y is the granulation yield, %; n is the screw speed, rpm; B0 to B3 are the model regression constants of the correlation control prediction model.

8. A control method for the continuous twin-screw wet granulation control system according to claim 7, which is an online real-time monitoring and refined control method for granulation particle size and granulation yield based on the multi-parameter synergistic coupling of the granulation liquid volume fraction content and surface energy of the wet raw material powder, the granulation liquid binder concentration, and the screw speed, characterized in that: The steps include: T1. First, set the raw material powder weight loss feeding rate and the final granulation particle size d on the control system host. 90 , qualified particle size distribution range, expected value of granulation yield Y; T2, according to the granulation particle size d 90 The expected value of the granulation yield Y, the granulation particle size and granulation yield model prediction controller uses the associated control prediction model formulas (2), (3), (4) and the optimization algorithm to globally coordinate the optimization prediction to give the optimal parameter combination of the granulation liquid volume fraction content C, surface energy E, and screw speed of the wetted raw material powder to obtain the optimal granulation yield. Then, based on the optimal granulation liquid volume fraction content C and the associated prediction model formula (1), the binder concentration C of the granulation liquid is predicted and calculated. V ; T3. Start the continuous twin-screw wet granulation system and start granulation production; T4, the granulation particle size and granulation yield model prediction controller automatically calculates and dynamically updates the set value of the granulation liquid spray flow process parameter based on the real-time data of the raw material powder weight loss feeding rate fed back by the feeding controller and the granulation liquid volume fraction content C of the optimal raw material powder wetting predicted in step T2; T5, online particle analyzer to detect the endpoint particle size d in real time 10 ,d 50 ,d 90 , particle size span q information, and transmit the real-time measurement signal to the control system for feedback; T6, Granulation particle size and granulation yield model prediction controller online real-time calculation and analysis of granulation particle size and particle size distribution consistency, and according to the deviation between the real-time detected granulation endpoint particle size and the user's expected particle size, through the correlation control prediction model formula (1), (2), (3), (4) and the prediction model control strategy, predict the granulation process of the next time step, and calculate the granulation liquid volume fraction content C and surface energy E of the externally controlled raw material powder, granulation liquid spray flow rate, granulation liquid binder concentration parameter setting values ​​for obtaining the optimal granulation yield through the global collaborative optimization algorithm; T7, the double closed-loop ratio controller automatically and dynamically updates the process parameter setting values ​​of the granulation liquid spray flow rate and the granulation liquid adhesive concentration according to the optimal parameter setting value obtained in step T6. The granulation liquid flow rate and adhesive concentration control module realizes online real-time monitoring and fine regulation of the granulation liquid flow rate and the granulation liquid adhesive concentration through the water solvent flow transmitter, the adhesive flow transmitter and the double closed-loop ratio controller, and the dynamic real-time preparation of the granulation liquid with the required optimal adhesive concentration; T8, the granulation liquid volume fraction content control module for wetting the raw material powder automatically and dynamically updates the set value of the fluid-solid ratio L / S of the wetting raw material powder according to the set value of the optimal granulation liquid volume fraction content C of the wetting raw material powder and the set value of the optimal granulation liquid spray flow rate obtained in step T6, and the multi-fiber probe near-infrared spectrometer detects the humidity of the wet granules at the outlet of the granulator at multiple points online in real time, and sends the multi-point real-time humidity detection data to the control system host, and the control system host calculates and obtains the average humidity of the granulated wet granules, the granulation liquid volume fraction content C of the wetting raw material powder and the fluid-solid ratio. The real-time data of the ratio L / S is obtained, and the granulation liquid volume fraction content C and the real-time monitoring data signal of the fluid-solid ratio L / S are fed back to the flow ratio controller. The flow ratio controller sends a raw material powder conveying flow control instruction to the third frequency converter according to the deviation between the real-time monitoring data of the fluid-solid ratio L / S and the set value of the fluid-solid ratio L / S. The frequency converter adjusts the output frequency and the rotation speed of the granulating screw to achieve online real-time monitoring and refined control of the granulation liquid volume fraction content C of the wetted raw material powder, as well as indirect control of the raw material powder conveying flow rate and the screw rotation speed of the granulator granulator screw; T9, through the control module of the volume fraction content of the granulation liquid wetting the raw material powder, the dynamic online control of the volume fraction content C of the granulation liquid wetting the raw material powder, the raw material powder conveying flow rate, the screw speed, the raw material powder pre-compression working condition and the periodic grinding and pressing working condition of the granulation KE block can be realized, and then the double closed-loop ratio controller of step T7 is coordinated to realize the dynamic online precision control of the granulation liquid spray flow rate and the solubility of the granulation liquid binder, as well as the surface energy E control of step T6. Combining the above-mentioned multi-parameter collaborative coupling control means, the dynamic online real-time monitoring and precise control of the terminal granulation particle size, particle size distribution and granulation yield can be finally realized.

Images