A pharmaceutical coating condition detection system and method

Abstract

This invention relates to the field of tablet preparation technology, specifically to a drug coating state detection system and method. The system includes a surface pretreatment module, a coating execution module, an online monitoring module, a data processing module, and a control module. The pretreatment module removes contaminants and enhances activity through plasma treatment, thereby improving coating adhesion. The coating execution module uses a fluidized bed environment, combined with a multi-channel spray gun and attitude control, to achieve uniform coating. The online monitoring module simultaneously acquires three-dimensional morphology, spectral, and image data, suppressing interference to ensure reliability. The data processing module calculates thickness and uniformity indicators, using machine learning to correlate process and quality. The control module dynamically adjusts parameters based on quality parameters using a PID algorithm, initiating emergency handling in case of anomalies. The method corresponds to the above steps. This invention improves coating adhesion and uniformity, achieves real-time monitoring and precise control, reduces the difficulty and cost of quality control, and is applicable to the production of various functional tablet coatings.

Description

Technical Field

[0001] This invention belongs to the field of tablet preparation technology, and specifically relates to a drug coating state detection system and method. Background Technology

[0002] Tablet coating is a key process in modern pharmaceutical manufacturing. By coating the tablet surface with specific materials to form a functional coating layer, multiple functions can be achieved, including masking the bitterness of the drug, controlling the release rate, improving stability, and enhancing appearance. Its quality directly affects the drug's efficacy, safety, and patient compliance. With the development of formulation technology, the requirements for the precision and stability of the coating process are increasingly stringent. Existing technologies have the following problems:

[0003] (1) Trace amounts of contaminants are easily left on the surface of tablets during preparation and transport. At the same time, most tablet substrates have low surface energy, resulting in insufficient interfacial bonding between the coating material and the tablet core. This directly causes defects such as peeling and cracking of the coating layer during subsequent processing, storage and transportation, affecting the stability and consistency of drug release behavior;

[0004] (2) Coating equipment often uses fixed process parameters for batch processing, which makes it difficult to adapt to changes in the posture of tablets during the coating process and differences in the physical properties of tablets within a batch. This limitation of the process leads to significant differences in the coating layer thickness of tablets in the same batch, with some products having thickness deviations exceeding the quality standard requirements, affecting the uniformity of drug release;

[0005] (3) Traditional coating quality inspection mainly relies on offline sampling inspection methods, such as weighing and slicing measurement. These methods have detection lag and cannot reflect quality changes in the coating process in real time; at the same time, they are limited by the sampling ratio and cannot fully represent the quality status of the entire batch of products, which may lead to potential quality risks not being detected and corrected in time, increasing the difficulty and cost of quality control in the production process.

[0006] In view of this, the present invention is hereby proposed. Summary of the Invention

[0007] To address the aforementioned technical problems in existing technologies, this invention provides a drug coating state detection system and method. This system solves the problems of tablets having easily residual contaminants and low surface energy, leading to insufficient adhesion between the coating material and the tablet core, resulting in peeling and cracking, and affecting drug stability and release consistency. Furthermore, coating equipment often uses fixed process parameters, making it difficult to adapt to changes in tablet posture and individual differences, causing significant variations in coating layer thickness within the same batch, affecting drug release uniformity. Traditional offline sampling detection methods are also problematic due to their lag, making it difficult to comprehensively reflect the quality of the entire batch, increasing quality risks and control costs.

[0008] To achieve the above objectives, the technical solution of the present invention is as follows:

[0009] In a first aspect, a drug coating status detection system includes:

[0010] Surface pretreatment module: used to transport the pretreated tablets to the coating execution module;

[0011] Coating execution module: Used to receive the control parameters from the control module and execute the coating operation, while simultaneously transporting the tablets during the coating process to the online monitoring module;

[0012] Online monitoring module: used to transmit the collected coating layer data to the data processing module;

[0013] Data processing module: used to transmit the analyzed and processed quality parameters to the control module;

[0014] Control module: Used to adjust the working parameters of the coating execution module according to the quality parameters, and transmit them to the coating execution module.

[0015] Furthermore, the surface pretreatment module includes:

[0016] Plasma processing unit: The output end of the plasma processing unit is connected to the vacuum adsorption conveying unit, which is used to convey the processed tablets by vacuum adsorption.

[0017] Vacuum adsorption conveying unit: The output end of the vacuum adsorption conveying unit is connected to the pretreatment quality detection unit, which is used to accurately convey the tablets to the detection station;

[0018] Pre-processing quality inspection unit: The output of the pre-processing quality inspection unit is connected to the coating execution module and the data processing module respectively, and is used to transport qualified tablets to the coating execution module and upload the test data to the data processing module.

[0019] Furthermore, the plasma processing unit employs dielectric barrier discharge technology, including a gas mixing device and discharge electrodes;

[0020] The output end of the gas mixing device is connected to the discharge electrode and is used to provide a mixture of inert gas and active gas. The operating power of the discharge electrode is 50-200W.

[0021] Furthermore, the coating execution module includes:

[0022] Fluidized bed coating chamber unit: The fluidized bed coating chamber is connected to the multi-channel spray gun array and the tablet attitude control unit respectively, and is used to provide coating operation space;

[0023] Multi-channel spray gun array unit: The multi-channel spray gun array is connected to a closed-loop control module to receive control parameters and perform spraying operations;

[0024] Tablet posture control unit: The tablet posture control unit is connected to the data processing module and is used to adjust the vibration frequency and airflow intensity according to the tablet position data.

[0025] Furthermore, the online monitoring module includes a synchronization triggering unit, and a laser scanning unit, a spectral imaging unit, and a multi-view camera connected to the synchronization triggering unit;

[0026] The synchronous triggering unit is used to control the acquisition timing of the laser scanning unit, the spectral imaging unit, and the multi-view camera;

[0027] The outputs of the laser scanning unit, the spectral imaging unit, and the multi-view camera are all connected to the data processing module, which is used to synchronously transmit the acquired three-dimensional topographic data, spectral data, and image data to the data processing module.

[0028] Furthermore, the laser scanning unit uses a helium-neon laser light source with a wavelength of 633nm, the scanning range covers the entire surface of the tablet, the scanning rate is ≥1000 points / second, and the longitudinal measurement accuracy is ≤±0.1μm.

[0029] Furthermore, the data processing module includes:

[0030] Thickness calculation unit: used to receive raw data and calculate the coating layer thickness;

[0031] Uniformity analysis unit: used to calculate uniformity indices based on thickness data;

[0032] Machine learning unit: Used to optimize preprocessing parameters based on uniformity metrics.

[0033] Furthermore, the uniformity analysis unit calculates the uniformity index using the following formula:

[0034]

[0035] in, The uniformity index, For thickness standard deviation, This represents the average thickness.

[0036] Furthermore, the control module includes:

[0037] Parameter comparison unit: The input end of the parameter comparison unit is connected to the data processing module, and the output end is connected to the multivariable adjustment unit. It is used to compare the real-time quality parameters with the preset threshold and generate a deviation signal.

[0038] Multivariable adjustment unit: The multivariable adjustment unit adopts the PID algorithm to adjust the spray gun flow rate, angle and moving speed of the coating execution module according to the deviation signal.

[0039] Secondly, a method for detecting the state of drug coating includes:

[0040] S1. Activate and purify the tablet surface to enhance the adhesion of the coating material. This includes plasma treatment of the tablets, transporting the tablets by vacuum adsorption, detecting the surface quality of the pretreated tablets, and screening qualified tablets for subsequent processes.

[0041] S2. Coating the qualified pre-treated tablets is carried out by providing a stable coating environment through a fluidized bed, using a multi-channel spray gun array to spray the coating material, and adjusting the tablet posture to ensure uniform coating. At the same time, the tablets in the coating process are transported to the online monitoring stage.

[0042] S3. Simultaneously collect three-dimensional morphology data, spectral data and multi-view images of the tablet surface during the coating process, suppress environmental interference to ensure data reliability, and transmit the collected data to the data processing stage.

[0043] S4. Calculate the coating layer thickness based on the monitoring data, analyze the thickness uniformity index, and generate quality assessment results by linking process parameters with coating quality through a machine learning model.

[0044] S5. Compare the real-time quality assessment results with the preset thresholds, dynamically adjust the coating process parameters based on the comparison results, and initiate emergency handling when persistent quality abnormalities occur.

[0045] Compared with existing technologies, the present invention provides a drug coating state detection system and method, comprising a surface pretreatment module, a coating execution module, an online monitoring module, a data processing module, and a control module. The surface pretreatment module removes contaminants from the tablet surface and enhances surface activity through plasma treatment, thereby improving coating adhesion. The coating execution module utilizes a fluidized bed to provide a stable environment, combined with a multi-channel spray gun array and tablet posture control to achieve uniform coating. The online monitoring module simultaneously acquires three-dimensional morphology, spectral, and multi-view image data, suppressing environmental interference to ensure data reliability. The data processing module calculates the coating layer thickness and uniformity indicators, and correlates process parameters with coating quality through machine learning. The control module dynamically adjusts coating parameters based on quality parameters using a PID algorithm, initiating emergency handling when persistent abnormalities occur. The method includes surface pretreatment, coating execution, online monitoring, data processing, and dynamic adjustment steps. This invention, through a collaborative design across the entire process, effectively improves coating adhesion and uniformity, achieves real-time monitoring and precise control of coating quality, reduces the difficulty of quality control and production costs, and is suitable for the functional coating production of various tablets. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the structure of the drug coating status detection system provided in an embodiment of the present invention;

[0047] Figure 2 This is a flowchart of a drug coating state detection method provided in an embodiment of the present invention. Detailed Implementation

[0048] The technical solution of the present invention will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are not all embodiments of the present invention. All other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0049] It should be noted that, unless otherwise specifically stated, the relative arrangement and numerical expressions of the components and steps described in these embodiments should not be construed as limiting the scope of the invention.

[0050] The following description of exemplary embodiments is merely illustrative and is not intended to limit the invention or its application or use in any way. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail herein, but where applicable, such techniques, methods, and apparatus should be considered part of this specification.

[0051] Example 1

[0052] See Figure 1 , Figure 1 This invention proposes a drug coating state detection system that addresses issues such as the ease with which contaminants remain on tablet surfaces and the low surface energy leading to insufficient coating adhesion; the difficulty in adapting fixed parameters of coating equipment to tablet posture and individual differences causing uneven thickness; and the lag in traditional offline detection methods, which fail to comprehensively reflect the overall batch quality. Specifically, this drug coating state detection system may include:

[0053] M1, Surface Pretreatment Module: Used for tablet surface activation and purification, enhancing the adhesion of coating materials, specifically including:

[0054] M11, Plasma Processing Unit: The output of the plasma processing unit is connected to the vacuum adsorption conveying unit, which is used to convey the processed tablets by vacuum adsorption.

[0055] This method removes surface contaminants and introduces active groups through high-energy particle action, thereby enhancing the adhesion of the coating material. It employs dielectric barrier discharge (DBD) technology, using an Ar / O2 mixed gas (volume ratio 9:1) as the working gas. The discharge power is continuously adjustable from 50-200W, and the treatment distance is controlled within 5-15mm. By bombarding the tablet surface with high-energy plasma particles, microscopic contaminants such as release agent residue and environmental dust can be removed (removal rate >99%). Simultaneously, active groups such as hydroxyl and carboxyl groups are introduced onto the surface, increasing the surface energy of the tablet from 35-45 mN / m before treatment to 65-75 mN / m.

[0056] This unit interfaces with the vacuum adsorption conveying unit to receive tablets that have been stably fixed and then transfers them to the pretreatment quality detection unit after processing.

[0057] M12, Vacuum Adsorption Conveying Unit: The output of the vacuum adsorption conveying unit is connected to the pretreatment quality inspection unit to accurately convey tablets to the inspection station, ensuring the accuracy of plasma treatment and quality inspection. It uses a porous ceramic suction cup (pore size 50-100μm) and provides an adjustable adsorption pressure of -0.03 to -0.05MPa through a negative pressure system. This allows for firm adsorption of tablets without damaging their surface, ensuring that the tablet deflection angle during conveying and processing is <0.5°. One end of this unit connects to the tablet feeding device to receive tablets to be pretreated; the other end connects sequentially to the plasma treatment unit and the pretreatment quality inspection unit, accurately conveying tablets to each processing station according to a preset rhythm, achieving continuous pretreatment operations.

[0058] M13, Pre-treatment Quality Inspection Unit: The output of the pre-treatment quality inspection unit is connected to the coating execution module and the data processing module respectively. It is used to transport qualified tablets to the coating execution module and upload the test data to the data processing module.

[0059] It integrates a high-precision contact angle measuring instrument (measurement accuracy ±0.1°), which detects the water contact angle on the surface of the treated tablets by adding deionized water (droplet volume 5μL), with a contact angle ≤30° as the acceptable threshold. For tablets that do not meet the standard, the unit automatically pushes them to the waste collection box via a pneumatic pusher (rejection accuracy 100%); qualified tablets are conveyed to the coating execution module via a conveyor track. At the same time, the unit uploads real-time detection data (such as contact angle value, pass rate) to the data processing module, providing a basis for optimizing plasma processing parameters.

[0060] M2, Coating Execution Module: This module receives control parameters from the control module and, through coordinated control of the coating environment, spraying precision, and tablet posture, achieves uniform coating. It also precisely delivers the tablets during the coating process to the online monitoring module for quality inspection. Specifically, this module may include:

[0061] M21, Fluidized Bed Coating Chamber Unit: Connected to the multi-channel spray gun array and tablet posture control unit, this unit provides the coating operation space. The fluidized bed coating chamber offers a stable, sealed environment for coating operations. Constructed of stainless steel, the chamber has a diameter of 300-500mm and an internal volume suitable for batch coating needs of 500-2000 tablets. A perforated distribution plate (0.5-1mm pore size) is installed at the bottom, and a variable frequency fan adjusts the airflow speed to 0.8-2.0m / s, suspending the tablets under the airflow to prevent stacking and contact. An internal infrared heating device (500-1000W power) and a precision humidifier are included. A PID control algorithm stabilizes the chamber temperature at 30-40℃ (temperature control accuracy ±1℃) and the relative humidity at 30-45% (humidity control accuracy ±2%), ensuring rapid film formation without cracking.

[0062] The top of the unit is connected to the nozzle group of the multi-channel spray gun array unit, the side is connected to the vibration device and airflow disturbance device of the tablet posture control unit, and the bottom is provided with a tablet conveying channel, which can directionally convey tablets during the coating process to the online monitoring module.

[0063] M22, Multi-channel spray gun array unit: The multi-channel spray gun array is connected to the closed-loop control module to receive control parameters and execute spraying operations. Directional and quantitative spraying is achieved by independently adjusting the parameters of each spray gun. It contains 4-6 air atomizing spray guns, evenly distributed in a ring at the top of the fluidized bed coating chamber, covering the entire angle range of the tablet suspension area; the nozzle diameter is infinitely adjustable from 0.5-1.0 mm, adaptable to coating materials of different viscosities (such as aqueous or organic coating agents); the atomization pressure is adjustable from 0.15-0.3 MPa, using compressed air to atomize the coating liquid into 50-100 μm droplets; each spray gun is equipped with a high-precision metering pump, with a flow rate adjustment range of 0.5-5 mL / min (accuracy ±0.05 mL / min), and an adjustable spray angle of 30°-60°. Directional spraying of different areas of the tablet is achieved by driving the spray gun rotation via a servo motor.

[0064] This unit is connected to the multivariable adjustment unit of the closed-loop control module via a signal cable. It can receive control parameters such as flow rate, angle, and pressure in real time, and can also feed back the current working status (such as actual flow rate and nozzle temperature) to the data processing module.

[0065] M23, Tablet Posture Control Unit: The tablet posture control unit is connected to the data processing module and is used to adjust the vibration frequency and airflow intensity according to the tablet position data to ensure that the tablets continuously rotate during the coating process to avoid uneven local spraying. An electromagnetic vibration device is installed at the bottom of the fluidized bed coating chamber, with an adjustable vibration frequency of 20-50Hz (amplitude 0.5-2mm), which changes the tablet suspension trajectory through periodic vibration; simultaneously, 3-4 auxiliary airflow nozzles are set on the side wall of the chamber, with the spray direction forming a 30° angle with the main airflow to create a rotating airflow field to enhance the tablet rotation effect.

[0066] This unit works in conjunction with the uniformity analysis module of the data processing module. Based on the tablet position distribution data (such as whether there is local aggregation) fed back by the online monitoring module, it dynamically adjusts the vibration frequency and auxiliary airflow intensity (0.2-0.5 m / s) to ensure that the tablet flipping frequency is ≥5 times / second and the spraying time difference between different areas on the surface of each tablet is ≤0.5 seconds. It receives attitude optimization commands output by the data processing module and simultaneously uploads parameters such as real-time vibration frequency and airflow intensity to the data processing module.

[0067] M3, Online Monitoring Module: Used to transmit the collected coating layer data to the data processing module. This module may specifically include:

[0068] M31, Synchronous Trigger Unit: This unit assists in determining tablet posture and surface integrity through full-angle image acquisition, ensuring comprehensive thickness detection. It is equipped with 3-4 high-speed cameras (50fps), covering 0°, 90°, 180°, and 270° viewing angles respectively, capable of completely capturing the circumferential surface state of the tablet. Image stitching technology fuses multi-view images into a complete tablet surface map, identifying surface defects (such as bubbles and scratches) and posture anomalies (such as local occlusion), providing a basis for judging the effectiveness of laser scanning and spectral imaging. It ensures the temporal consistency between image acquisition and laser / spectral detection, and the acquired image data is uploaded to the data processing module to assist in correcting the thickness calculation results.

[0069] M32, Laser Scanning Unit: This unit is the core device for three-dimensional thickness detection of the coating layer. It reconstructs the contour of the coating layer on the tablet surface through high-precision laser scanning. It uses a 633nm helium-neon laser as the light source, achieving a lateral resolution of 0.5μm and a longitudinal measurement accuracy of ±0.1μm. The scanning rate is 1000 points / second, enabling rapid acquisition of microscopic morphological data of the tablet surface. Through focused tracking technology, it adapts to the curvature of the tablet surface in real time, accurately reconstructing the three-dimensional contour before and after coating, providing raw point cloud data for the thickness calculation unit.

[0070] This unit is connected to the thickness calculation unit of the data processing module, and transmits the three-dimensional coordinate information obtained from the scan in real time to calculate the coating layer thickness value at each point.

[0071] M33, Spectral Imaging Unit: This unit assists in verifying the coating layer thickness distribution through spectral feature analysis, improving detection reliability. Its operating wavelength covers 900-1700nm, with a spectral resolution of 5nm and a spatial resolution of 0.1mm. It can simultaneously acquire spectral images and spatial location information of the tablet surface. Utilizing characteristic absorption peaks of the coating material in specific wavelength bands (such as the absorption peak at 1390nm), the local coating thickness is inverted through a correlation model between peak intensity and thickness, forming cross-validation with laser scanning data.

[0072] The spectral imaging data output by this unit is transmitted to the uniformity analysis module of the data processing module to optimize the analysis accuracy of the thickness distribution.

[0073] M34, Environmental Interference Suppression Unit: Used to reduce the impact of external factors on monitoring accuracy and ensure data reliability. It is equipped with a laser power stabilizer (power fluctuation <1%) to ensure stable laser source intensity; and uses a temperature-controlled optical lens (operating temperature 25±0.5℃) to avoid lens focal length shift caused by temperature changes. Through the above design, measurement errors caused by ambient light, temperature fluctuations, and other factors can be controlled within <2%.

[0074] This unit is integrated with the laser scanning unit and the spectral imaging unit, providing a stable working environment for each detection device and indirectly improving the analytical accuracy of the data processing module.

[0075] M4, Data Processing Module: Used to transmit the analyzed and processed quality parameters to the control module. This module may specifically include:

[0076] M41, Thickness Calculation Unit: Connected to the online monitoring module, it receives raw data and calculates the coating layer thickness; it converts the 3D point cloud data acquired by the laser scanning unit into coating layer thickness values. Its core function is to calculate the thickness by comparing the surface height difference before and after coating. The specific calculation formula is as follows:

[0077]

[0078] in, Coordinates of the tablet surface after coating Height data at the location; This provides the baseline height data after tablet pretreatment. It receives real-time point cloud data from the laser scanning unit, performs coordinate matching and difference calculations, generates a thickness distribution matrix covering the entire tablet surface (with resolution consistent with laser scanning accuracy), and synchronously transmits the thickness data to the uniformity analysis module.

[0079] M42, Uniformity Analysis Unit: Connected to the thickness calculation unit, it calculates uniformity indices based on thickness data, providing a quantitative basis for quality assessment and parameter adjustment. It performs analysis using average thickness, thickness standard deviation, and uniformity index.

[0080] M421. Calculate the average thickness of all measurement points on the tablet surface. The specific formula is as follows:

[0081]

[0082] in, The total number of measurement points. These are the thickness values ​​at each point.

[0083] M422. Calculate the standard deviation of thickness at all measurement points on the tablet surface. The specific formula is as follows:

[0084]

[0085] in, The total number of measurement points. For the first The coating thickness at each measurement point This represents the average thickness of the coating layer. The total number of measurement points is determined by the scanning accuracy of the laser scanning unit and the tablet surface area, typically covering the entire tablet surface, with a typical value of 10,000-50,000 points; the coating layer thickness is obtained through the thickness calculation unit.

[0086] The denominator is adopted Instead It is an unbiased estimation correction based on the sample standard deviation, which is suitable for measurement scenarios with limited sample size (such as surface scanning of a single tablet) and ensures that the standard deviation can more accurately reflect the thickness fluctuation characteristics of the entire batch.

[0087] M423. Calculate the uniformity index of all measurement points on the tablet surface. The specific formula is as follows:

[0088]

[0089] in, The uniformity index, For thickness standard deviation, This represents the average thickness. The preset threshold is... When the value falls below this threshold, a parameter adjustment signal is triggered.

[0090] M43, Machine Learning Unit: Connected to the uniformity analysis unit and surface pretreatment module, it optimizes pretreatment parameters based on uniformity indicators. Employing a BP neural network model, it intelligently predicts and optimizes the adhesion of the coating material by correlating process parameters with coating quality.

[0091] Its input characteristics include 12 key process parameters: plasma processing power, processing time, spray gun flow rate, atomization pressure, spray angle, chamber temperature, humidity, tablet tumbling frequency, coating solution viscosity, initial surface energy of tablets, and average thickness. and uniformity index .

[0092] M5, Control Module: Used to adjust the working parameters of the coating execution module according to the quality parameters and transmit them to the coating execution module.

[0093] M51, Parameter Comparison Unit: The input end of the parameter comparison unit is connected to the data processing module, and the output end is connected to the multivariable adjustment unit. It is used to compare the real-time quality parameters with the preset threshold and generate a deviation signal.

[0094] The system includes a built-in preset threshold database, containing average thickness thresholds, uniformity index thresholds, and local deviation thresholds. The average thickness threshold, for example, is 50±5μm, and this threshold can be adjusted according to different dosage forms. The uniformity index threshold includes normal production control values ​​and emergency warning values. The normal production control value is... Emergency warning value The local deviation threshold is the deviation of the thickness at a single measurement point from the mean. .

[0095] This unit generates three types of signals through point-by-point comparison and statistical analysis: a normal signal, generated when all indicators meet the standards, which does not trigger adjustment; a fine-tuning signal, generated when local deviations exceed the standard; and a normal signal, generated when all indicators meet the standards and does not trigger adjustment. Or the uniformity index is at The signal generated within the specified range is transmitted to the multivariable control unit; the emergency signal refers to the uniformity index of five consecutive tablets. The signal generated in time is simultaneously sent to the emergency response unit.

[0096] M52, Multivariable Adjustment Unit: The multivariable adjustment unit employs a PID algorithm to adjust the spray gun flow rate, angle, and movement speed of the coating execution module based on the deviation signal. It receives the fine-tuning signal output from the parameter comparison unit and, in conjunction with the coordinates of the deviation area provided by the laser scanning unit (e.g., the edge or top of the tablet), adjusts the spray gun parameters for the corresponding area accordingly.

[0097] Flow rate adjustment is achieved by controlling the metering pump speed. The spray gun flow rate can be adjusted from 0.5 to 5 mL / min, with an adjustment accuracy of ±0.1 mL / min. When the deviation area is too thick, the flow rate is reduced; when the deviation area is too thin, the flow rate is increased.

[0098] For angle adjustment, the spray gun is rotated via a servo motor, with an adjustment range of 30°-60° and an adjustment accuracy of ±1°. This adjustment allows for targeted respraying of areas with localized obstruction caused by the tablet's posture.

[0099] Regarding the adjustment of the moving speed, the speed of the spray gun along the circular track is controlled. The adjustment range of the moving speed is 5-20mm / s. By adjusting the speed, the spraying time can be extended or shortened for the deviation area.

[0100] During the adjustment process, the PID algorithm continuously optimizes the adjustment amount by collecting the adjusted thickness feedback in real time, with a feedback lag time of no more than 0.5s, thus avoiding overshoot or undershoot and ensuring that local deviations return to within ±3μm within three adjustment cycles. This unit is directly connected to the multi-channel spray gun array unit of the coating execution module, transmitting adjustment commands via digital signals.

[0101] M53 Emergency Handling Unit: Used to respond to persistent quality anomalies, prevent batch defects, and ensure production safety. Its trigger condition is that the parameter comparison unit continuously detects the uniformity index of 5 tablets. This situation has exceeded the normal control range. Upon triggering, the emergency response unit will perform the following actions:

[0102] In terms of audible and visual alarms, a red warning light and a buzzer are activated, with the buzzer frequency at 1kHz, to alert operators to intervene.

[0103] Regarding process speed reduction, instructions are sent to the coating execution module to automatically reduce the coating speed by 30%. For example, if the original batch processing speed is 100 pieces / minute, it is reduced to 70 pieces / minute to allow time for parameter adjustment.

[0104] In terms of data recording, the system automatically records the time of the anomaly, batch number, real-time process parameters (including temperature, humidity, and spray gun flow rate), and quality indicators (including average thickness and uniformity index U value). This data is stored in the local database and simultaneously uploaded to the production management system for subsequent traceability and analysis.

[0105] For manual interaction, the interface displays possible causes of the anomaly, such as "possible spray gun blockage" or "possible airflow instability," along with suggested troubleshooting steps. After operator confirmation, the operator can choose to continue production, which requires parameter recalibration; or they can choose to pause the batch.

[0106] Example 2

[0107] See Figure 2 , Figure 2 This invention proposes a method for detecting the state of drug coating, which solves problems such as insufficient adhesion of coating materials, poor thickness uniformity, and monitoring lag. It is applicable to the production of functional coatings for various tablets. Specific steps include:

[0108] S1. Surface Pretreatment: This involves activating and purifying the tablet surface to enhance the adhesion of the coating material. Specifically, this includes plasma treatment of the tablets, transporting the tablets via vacuum adsorption, inspecting the surface quality of the pretreated tablets, and selecting qualified tablets for subsequent processes. Specific steps may include:

[0109] S11. Tablet loading: The tablets to be coated are conveyed to the pretreatment station by a vibrating feeder. The vacuum adsorption device is started, and a porous ceramic suction cup with an 80μm pore size generates a negative pressure of -0.04MPa to stably adsorb and fix the tablets, ensuring that the tablet deflection angle is <0.5°, so as to provide a stable posture for subsequent plasma treatment.

[0110] S12. Plasma activation treatment: Turn on the dielectric barrier discharge device and introduce a mixed gas of Ar / O2 with a volume ratio of 9:1 (flow rate 200 sccm). Set the discharge power to 120W, the treatment distance to 10mm, and the treatment time to 10s. High-energy plasma particles bombard the tablet surface to remove mold release agent residue and environmental dust (pollutant removal rate > 99%). At the same time, active groups such as hydroxyl and carboxyl groups are introduced to increase the surface energy and enhance the adhesion of the coating material.

[0111] S13. Pre-treatment quality screening: After treatment, the tablets are conveyed to the detection station via vacuum. The contact angle measuring instrument is activated and 5μL of deionized water is automatically added to detect the surface water contact angle. If the contact angle is ≤30° (corresponding to a surface energy ≥65mN / m), it is judged as qualified and is conveyed to the coating station via the conveyor track. If the contact angle is >30°, the pneumatic pusher is activated to push the unqualified tablets to the waste box. The contact angle value and the pass rate are recorded simultaneously and uploaded to the production system to provide data support for process optimization.

[0112] S2. Coating Execution: The pretreated, qualified tablets are coated using a fluidized bed to provide a stable coating environment. A multi-channel spray gun array is used to apply the coating material, and the tablet orientation is adjusted to ensure uniform coating. Simultaneously, the tablets during the coating process are conveyed to an online monitoring station. Specific steps may include:

[0113] S21. Fluidized Bed Environment Preparation: Start the fluidized bed coating chamber, adjust the variable frequency fan corresponding to the 0.8mm perforated distribution plate at the bottom to stabilize the airflow velocity at 1.2m / s; turn on the 800W infrared heating device and precision humidifier, and maintain the chamber temperature at [temperature value missing] using PID control. The relative humidity is maintained at This provides a stable temperature and humidity environment for coating.

[0114] S22. Spray gun parameter initialization: Arrange 6 air atomizing spray guns (nozzle diameter 0.8mm) in a ring at the top of the fluidized bed. Set the initial parameters as follows: atomization pressure 0.2MPa, spray flow rate 2mL / min, spray angle 45°, and vertical distance between the nozzle and the tablet suspension area 150mm to ensure that the initial spraying range of the coating liquid covers the tablet suspension area.

[0115] S23. Tablet posture control: Activate the electromagnetic vibration device at the bottom of the fluidized bed (frequency 30Hz, amplitude 1mm), and simultaneously open the four auxiliary airflow nozzles on the side wall of the chamber at an angle of 30° to the main airflow to introduce airflow with an intensity of 0.3m / s to form a rotating airflow field; through the synergistic effect of vibration and airflow, the tablets are suspended and turned over, ensuring that the turning frequency is ≥5 times / second and the spray time difference between different areas of the tablet surface is ≤0.5s, thereby reducing uneven local coating.

[0116] S24. Coating solution spraying: A water-based enteric coating solution with a viscosity of 200 cP is delivered to the spray gun through a metering pump, and the spraying operation is started; the tablets are continuously suspended and turned in the fluidized bed to receive all-round coating and achieve uniform coverage of the coating material.

[0117] S3. Online Monitoring: Simultaneously collect three-dimensional morphology data, spectral data, and multi-view images of the tablet surface during the coating process, suppressing environmental interference to ensure data reliability, and transmitting the collected data to the data processing stage. Specific steps may include:

[0118] S31. Multi-source data synchronous acquisition: Activate the timing controller (synchronization error ≤1ms) to coordinate the collaborative work of multiple devices: Laser scanning uses a 633nm helium-neon laser device to scan the tablet surface at a rate of 1000 points / second to obtain lateral resolution. Longitudinal accuracy The 3D point cloud data generates a surface height matrix after coating every 2 seconds. Spectral imaging uses a 900-1700nm band camera to acquire spectral images every 5 seconds, focusing on capturing the characteristic absorption peak of the coating material at 1390nm; image acquisition uses four high-speed cameras with a frame rate of 50fps to photograph tablets from 0°, 90°, 180° and 270° angles to identify surface defects (such as bubbles, scratches) and abnormal posture.

[0119] S32. Environmental Interference Suppression: Activate the laser power stabilizer (ensure power fluctuation <1%) and control the optical lens operating temperature. The measurement error caused by changes in ambient light and temperature is controlled to less than 2%, ensuring the accuracy of the collected data.

[0120] S33. Real-time data transmission: Laser point cloud, spectral image and camera data are transmitted to the data processing terminal via Ethernet with a transmission delay of <0.5s to ensure data timeliness and support real-time analysis.

[0121] S4. Data Processing: Calculate the coating layer thickness based on monitoring data, analyze thickness uniformity indicators, and use a machine learning model to correlate process parameters with coating quality to generate quality assessment results. Specific steps may include:

[0122] S41. Coating layer thickness calculation: based on the tablet reference height recorded during the pretreatment stage. By calculating the thickness at each point, a thickness distribution matrix containing 10,000 measurement points is generated to quantify the local thickness of the coating layer. The specific formula is as follows:

[0123]

[0124] in, Coordinates of the tablet surface after coating Height data at the location; This is the baseline height data after tablet pretreatment.

[0125] S42. Uniformity Index Analysis: Calculate the average thickness. The specific formula is as follows:

[0126]

[0127] in, The total number of measurement points. These are the thickness values ​​at each point.

[0128] Calculate the standard deviation of thickness The formula reflecting the degree of thickness dispersion is as follows:

[0129]

[0130] in, The total number of measurement points. For the first The coating thickness at each measurement point This represents the average thickness of the coating layer.

[0131] Calculate the uniformity index of all measurement points on the tablet surface. The specific formula is as follows:

[0132]

[0133] in, The uniformity index, For thickness standard deviation, This represents the average thickness. The preset threshold is... When the value falls below this threshold, a parameter adjustment signal is triggered; the coating uniformity is comprehensively evaluated.

[0134] S43. Adhesion grade prediction: Twelve process parameters are input into the BP neural network model, including plasma power, processing time, spray gun flow rate, atomization pressure, chamber temperature, humidity, tablet flipping frequency, coating solution viscosity, initial surface energy of tablets, average thickness, uniformity index, and spray angle. The output adhesion grade prediction result is level 4 (level 5 is the optimal level), which realizes the early prediction of coating quality.

[0135] S5. Compare the real-time quality assessment results with preset thresholds, dynamically adjust the coating process parameters based on the comparison results, and initiate emergency handling when persistent quality abnormalities occur. Specific steps may include:

[0136] S51. Routine Parameter Comparison and Adjustment: Compare the real-time calculated average thickness, uniformity index, and local deviation with preset thresholds: If all indicators meet the standards, maintain the current process parameters; if a thickness deviation of 3.2μm (exceeding the threshold) is detected in the tablet edge area, activate PID control: reduce the spray gun flow rate in the corresponding area to... Adjust the spray angle to 48°, and after 3 adjustment cycles, the deviation returns to 2.0μm, ensuring that the local thickness meets the standard.

[0137] S52. Emergency Handling: If the uniformity index of 5 consecutive tablets is reduced due to spray gun blockage, take emergency measures: activate the red warning light and 1kHz buzzer to prompt the operator to check; automatically reduce the coating speed from 10 tablets / minute to 7 tablets / minute to allow time for parameter adjustment.

[0138] Record the abnormal time, batch number, real-time process parameters and quality indicators, and upload them to the production management system simultaneously; the operation interface displays "spray gun blockage may occur" and cleaning steps. After the operator confirms that the cleaning is completed and the parameters are calibrated, normal production is resumed to avoid batch defects.

[0139] Example 3

[0140] In this embodiment, the drug coating state detection system provided by the present invention is used, and the drug coating state detection method provided by the present invention is used. The object to be processed is A tablets with a diameter of 6 mm and a weight of 100 mg. The target is to achieve enteric coating through a coating device and an online detection method. The required coating layer thickness is 40 ± 4 μm, the uniformity index is ≥ 0.9, and it does not disintegrate in the gastric acid environment within 2 hours after coating, meeting the enteric preparation standard.

[0141] B1. Surface pretreatment

[0142] The vibrating feeder conveys the aspirin tablets to the pretreatment station, and the vacuum adsorption device is started. The pore diameter of the porous ceramic suction cup is 60 μm, generating a negative pressure of -0.035 MPa to stably adsorb the tablets, and the deflection angle of the tablets is controlled within 0.5° to avoid position deviation during processing.

[0143] The dielectric barrier discharge device is turned on, and a mixed gas of Ar and O2 with a volume ratio of 9:1 and a flow rate of 150 sccm is introduced. The discharge power is set to 80 W, the processing distance is 8 mm, and the processing time is 8 s. High-energy plasma particles bombard the tablet surface to remove the mold release agent and dust remaining from production, with a removal rate exceeding 99%; at the same time, hydroxyl groups are introduced on the surface to increase the surface energy.

[0144] The pretreatment quality inspection is started. The contact angle measuring instrument drops 5 μL of deionized water onto the surface of the processed tablets to detect the contact angle. If the contact angle ≤ 30°, at this time the surface energy is increased from 38 mN / m before processing to 68 mN / m, it is judged as qualified and sent into the coating chamber through the conveying track; if the contact angle > 30°, the pneumatic push rod pushes the tablets into the waste box, and simultaneously uploads the "unqualified" mark and the specific contact angle value to the production data recording system.

[0145] B2. Coating execution

[0146] The fluidized bed coating chamber is started, and the bottom porous distribution plate is adjusted. Its pore diameter is 0.5 mm, and the corresponding variable frequency fan makes the air flow velocity stable at 1.0 m / s to ensure that the tablets are suspended without stacking. The infrared heating device is turned on with a power of 600 W, and at the same time the precision humidifier is turned on. The temperature in the chamber is maintained at 32°C ± 1°C and the relative humidity is maintained at 35% ± 2% through PID control to meet the film-forming requirements of the enteric coating solution.

[0147] Six air atomizing spray guns are arranged in a ring at the top of the fluidized bed. The nozzle diameter is 0.6 mm, and the initial parameters are: atomizing pressure 0.18 MPa, spraying flow rate 1.5 mL / min, spraying angle 40°, and the distance between the nozzle and the tablet suspension area is 120 mm. The electromagnetic vibration device is started with a frequency of 25 Hz and an amplitude of 0.8 mm. At the same time, three auxiliary air flow nozzles on the side wall of the chamber are turned on, with an included angle of 30° with the main air flow, and an air flow of 0.25 m / s is introduced to form a rotating field.

[0148] Under the combined action of vibration and airflow, the tablets tumble at a stable frequency of 6 times per second, with the spraying time difference between different areas of the surface controlled within 0.4 seconds. The enteric coating solution, with a viscosity of 150 cP and primarily composed of acrylic resin, is delivered to the spray gun via a metering pump, initiating the coating operation. The tablets remain suspended and tumble under the influence of airflow and vibration, receiving coating from all directions.

[0149] B3. Online Monitoring

[0150] The timing controller is activated with a synchronization error of ≤1ms, coordinating the collaborative work of multiple devices: the laser scanner scans the tablet surface at a rate of 1000 points / second, generating a surface height matrix after coating every 1.5 seconds, with a lateral resolution of 0.5μm and a longitudinal accuracy of ±0.1μm; the spectroscopic camera acquires images in the 900-1700nm band every 4 seconds, focusing on capturing the characteristic absorption peak of the enteric coating material at 1390nm to verify the thickness distribution; four high-speed cameras take pictures from 0°, 90°, 180°, and 270° angles respectively to identify bubbles on the tablet surface. When the bubble diameter is >50μm, an early warning is triggered, or when abnormal posture is identified, an alarm is triggered when the static time is >0.8s.

[0151] The laser power stabilizer is activated to ensure power fluctuations are <1%; the operating temperature of the constant-temperature optical lens is controlled at 25±0.5℃, keeping measurement errors caused by ambient light and temperature fluctuations within 2%. Laser point cloud, spectral images, and camera data are transmitted to the industrial PC via Ethernet with a transmission delay of <0.3s, ensuring data timeliness.

[0152] B4. Data Processing

[0153] An industrial PC calls thickness calculation software to calculate thickness data based on the raw film reference height recorded in the preprocessing stage, generating a thickness distribution matrix with 8000 measurement points. Uniformity analysis software automatically calculates the following indicators: average thickness is 39.5 μm, within the target range of 40 ± 4 μm; thickness standard deviation is 1.8 μm; uniformity index is 0.95, ≥0.9, meeting quality standards.

[0154] The BP neural network model terminal receives 12 process parameters, including plasma power of 80W, processing time of 8s, spray gun flow rate of 1.5mL / min, atomization pressure of 0.18MPa, chamber temperature of 32℃, humidity of 35%, tablet tumbling frequency of 6 times / second, coating solution viscosity of 150cP, initial surface energy of tablet of 68mN / m, average thickness of 39.5μm, uniformity index of 0.95, and spray angle of 40°. The output adhesion level prediction result is level 4, with level 5 being the optimal level, and the prediction meets the requirements.

[0155] B5. Dynamic Adjustment

[0156] The PLC controller compares the real-time indicators, namely the average thickness of 39.5 μm and the uniformity index of 0.95, with the preset thresholds and determines that it is "qualified". It then maintains the current parameters, namely the spray gun flow rate of 1.5 mL / min and the vibration frequency of 25 Hz.

[0157] Laser scanning revealed that the thickness of the tablet edge region reached 43.2 μm, exceeding the threshold of 40 ± 4 μm. The PLC initiated PID control: the flow rate of the two spray guns in the corresponding region was reduced from 1.5 mL / min to 1.3 mL / min, with an adjustment accuracy of ±0.1 mL / min; the spray gun angle was increased from 40° to 42°, with an adjustment accuracy of ±1°; after two adjustment cycles, each lasting 1.0 s, the edge thickness returned to 41.0 μm, meeting the threshold.

[0158] Ultimately, the quality indicators of this batch of A tablets after coating were as follows: average coating thickness of 39.5 μm, standard deviation of thickness of 1.8 μm, uniformity index of 0.95; surface contact angle of 28°, coating peeling rate of <0.1%; no disintegration in gastric acid environment (pH 1.2) for 2 hours, meeting the requirements for enteric coating; batch pass rate of 99.5%, an improvement of 12% compared with traditional process.

[0159] In summary, the present invention has the following advantages:

[0160] 1. Through surface activation and purification processes, contaminants on the tablet surface are effectively removed and surface activity is enhanced, thereby strengthening the interfacial bonding between the coating material and the tablet core, reducing defects such as coating peeling and cracking, and ensuring the consistency of drug stability and release performance.

[0161] 2. By adopting a dynamic adaptation and precise control mechanism, the tablet posture and spraying parameters are adjusted in real time to adapt to the dynamic changes and individual differences of the tablets during the coating process. Combined with closed-loop control, local deviations are corrected in a timely manner, which greatly improves the coating uniformity of the same batch of products and ensures the uniformity of drug release.

[0162] 3. Construct a real-time monitoring and response system for the entire process to replace the traditional offline sampling and testing mode, so as to achieve comprehensive and real-time control over the coating quality. Combined with the anomaly handling mechanism, potential risks can be detected and corrected in a timely manner, reducing the difficulty of quality control and production costs.

[0163] The above specific embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to examples, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A drug coating status detection system, characterized in that, include: Surface pretreatment module: used to transport the pretreated tablets to the coating execution module; Coating execution module: Used to receive the control parameters from the control module and execute the coating operation, while simultaneously transporting the tablets during the coating process to the online monitoring module; Online monitoring module: used to transmit the collected coating layer data to the data processing module. The online monitoring module includes a synchronization triggering unit, and a laser scanning unit, a spectral imaging unit and a multi-view camera connected to the synchronization triggering unit. Data processing module: used to transmit the analyzed and processed quality parameters to the control module, the data processing module includes: Thickness calculation unit: used to receive raw data and calculate the coating layer thickness; Uniformity analysis unit: used to calculate uniformity index based on thickness data; the uniformity index is a uniformity index, and the specific calculation formula is as follows: in, The uniformity index, For thickness standard deviation, Average thickness; Machine learning unit: used to optimize preprocessing parameters based on uniformity metrics; Control module: used to adjust the working parameters of the coating execution module according to the quality parameters, and transmit the parameters to the coating execution module; The surface pretreatment module includes: Plasma processing unit: The output end of the plasma processing unit is connected to the vacuum adsorption conveying unit, which is used to convey the processed tablets by vacuum adsorption. Vacuum adsorption conveying unit: The output end of the vacuum adsorption conveying unit is connected to the pretreatment quality detection unit, which is used to accurately convey the tablets to the detection station; Pre-processing quality inspection unit: The output of the pre-processing quality inspection unit is connected to the coating execution module and the data processing module respectively, and is used to transport qualified tablets to the coating execution module and upload the test data to the data processing module.

2. The drug coating status detection system according to claim 1, characterized in that, The plasma processing unit employs dielectric barrier discharge technology and includes a gas mixing device and discharge electrodes. The output end of the gas mixing device is connected to the discharge electrode and is used to provide a mixture of inert gas and active gas. The operating power of the discharge electrode is 50-200W.

3. The drug coating status detection system according to claim 1, characterized in that, The coating execution module includes: Fluidized bed coating chamber unit: The fluidized bed coating chamber is connected to the multi-channel spray gun array and the tablet attitude control unit respectively, and is used to provide coating operation space; Multi-channel spray gun array unit: The multi-channel spray gun array is connected to a closed-loop control module to receive control parameters and perform spraying operations; Tablet posture control unit: The tablet posture control unit is connected to the data processing module and is used to adjust the vibration frequency and airflow intensity according to the tablet position data.

4. The drug coating status detection system according to claim 1, characterized in that, The synchronous triggering unit is used to control the acquisition timing of the laser scanning unit, the spectral imaging unit, and the multi-view camera; The outputs of the laser scanning unit, the spectral imaging unit, and the multi-view camera are all connected to the data processing module, which is used to synchronously transmit the acquired three-dimensional topographic data, spectral data, and image data to the data processing module.

5. The drug coating status detection system according to claim 4, characterized in that, The laser scanning unit uses a helium-neon laser light source with a wavelength of 633nm. The scanning range covers the entire surface of the tablet, the scanning rate is ≥1000 points / second, and the longitudinal measurement accuracy is ≤±0.1μm.

6. The drug coating status detection system according to claim 1, characterized in that, The control module includes: Parameter comparison unit: The input end of the parameter comparison unit is connected to the data processing module, and the output end is connected to the multivariable adjustment unit. It is used to compare the real-time quality parameters with the preset threshold and generate a deviation signal. Multivariable adjustment unit: The multivariable adjustment unit adopts the PID algorithm to adjust the spray gun flow rate, angle and moving speed of the coating execution module according to the deviation signal.

7. A method for detecting the state of drug coating, characterized in that, include: S1. Activate and purify the tablet surface to enhance the adhesion of the coating material. This includes plasma treatment of the tablets, transporting the tablets by vacuum adsorption, detecting the surface quality of the pretreated tablets, and screening qualified tablets for subsequent processes. S2. Coating the qualified pre-treated tablets is carried out by providing a stable coating environment through a fluidized bed, using a multi-channel spray gun array to spray the coating material, and adjusting the tablet posture to ensure uniform coating. At the same time, the tablets in the coating process are transported to the online monitoring stage. S3. Simultaneously collect three-dimensional morphology data, spectral data and multi-view images of the tablet surface during the coating process, suppress environmental interference to ensure data reliability, and transmit the collected data to the data processing stage. S4. Calculate the coating layer thickness based on the monitoring data, analyze the thickness uniformity index, and generate a quality assessment result by correlating process parameters with coating quality through a machine learning model; the uniformity index is the uniformity index, and the specific calculation formula is as follows: in, The uniformity index, For thickness standard deviation, Average thickness; S5. Compare the real-time quality assessment results with the preset thresholds, dynamically adjust the coating process parameters based on the comparison results, and initiate emergency handling when persistent quality abnormalities occur.

Images