Fluidized bed coated pellet agglomeration rate online monitoring method and system

Abstract

The invention relates to the technical field of quality control, and provides a fluidized bed coated pellet agglomeration rate on-line monitoring method and a fluidized bed coated pellet agglomeration rate on-line monitoring system. The method comprises the following steps: respectively acquiring near infrared spectrum and process parameter data of the whole coating process of the fluidized bed according to a preset time interval; splicing the near infrared spectrum and the process parameter data of the whole fluidized bed coating process according to a corresponding preset time interval to obtain fusion data; based on the fusion data, adopting a trained agglomeration rate model to obtain the pellet agglomeration rate; wherein the training process of the agglomeration rate model comprises the following steps: taking the spliced fusion data as input, and taking the pellet agglomeration rate measured offline at each time point as output, and training the agglomeration rate model. According to the method, the change of the pellet agglomeration rate in the fluidized bed coating process can be effectively and accurately monitored under the complex background by fusing the near infrared spectrum and the process parameter data.

Description

Technical Field

[0001] The present invention relates to the technical field of quality control, and in particular to an online monitoring method and system for agglomeration rate of fluidized bed coated micropills. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In the pharmaceutical industry, fluidized bed coating process is widely used in the preparation of micropellets to improve drug stability, release characteristics and taste. However, micropellets are prone to agglomeration during the coating process, which not only affects the appearance and physical properties of the product, but may also lead to inconsistent drug release, thereby affecting the bioavailability and efficacy of the drug. Therefore, real-time monitoring of the agglomeration rate during fluidized bed coating is a key link to ensure product quality.

[0004] Online monitoring of the agglomeration rate of micropellets during the coating process through process analytical technology (PAT) can timely identify and correct agglomeration phenomena and avoid product defects that may occur during the production process. Through real-time monitoring of the agglomeration rate, dynamic adjustment of process parameters can be achieved to ensure the consistency and repeatability of micropellets during the coating process, thereby improving production efficiency and reducing scrap rate. At present, there are mainly the following methods for online monitoring of the agglomeration rate of micropellets in fluidized bed coating processes: (1) Image analysis technology: By using high-resolution cameras and image processing software, the morphology and distribution of micropellets can be monitored in real time. By analyzing the image features of micropellets, the degree of agglomeration can be evaluated. (2) Spatial filtering technology: By using optical phase difference and velocity analysis, the chord length distribution of pellets is directly calculated to measure particle size.

[0005] Although PAT technology has been used to monitor the agglomeration rate of fluidized bed coating processes, the reliability of the micropellet agglomeration rate data obtained under complex backgrounds is reduced due to the diversity of controllable factors such as peristaltic pump flow rate and inlet air temperature, as well as difficult-to-control process parameters such as inlet air humidity. Summary of the invention

[0006] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a method and system for online monitoring of the agglomeration rate of fluidized bed coated micropellets. By integrating near-infrared spectral data and process parameter data, the present invention can effectively and accurately monitor the changes in the agglomeration rate of micropellets in the fluidized bed coating process under complex backgrounds, providing reference and technical means for quality monitoring in the fluidized bed coating process, and therefore has good practical application value.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] The first aspect of the present invention provides an on-line monitoring method for agglomeration rate of fluidized bed coated micropellets.

[0009] An online monitoring method for agglomeration rate of fluidized bed coated pellets, comprising:

[0010] At preset time intervals, the near infrared spectrum and process parameter data of the entire fluidized bed coating process are obtained respectively;

[0011] The near infrared spectrum and process parameter data of the whole process of fluidized bed coating are spliced ​​according to the corresponding preset time intervals to obtain fused data;

[0012] Based on the fusion data, the trained agglomeration rate model was used to obtain the pellet agglomeration rate;

[0013] The training process of the agglomeration rate model includes: taking the spliced ​​fusion data as input and taking the microcapsule agglomeration rate at each time point measured offline as output, and training the agglomeration rate model.

[0014] Furthermore, the process of offline measuring the agglomeration rate of the microcapsules at each time point includes: selecting a sieve larger than the normal particle size of the microcapsules to sieve the samples collected at each time point; and obtaining the agglomeration rate of the microcapsules at each time point based on the ratio of the mass of the microcapsules retained on the sieve to the mass of the sample.

[0015] Furthermore, the samples collected at each time point correspond to the time at which the near-infrared spectrum and process parameter data are acquired.

[0016] Furthermore, before obtaining the near infrared spectrum of the whole fluidized bed coating process, the method includes: inserting a near infrared spectrometer into the fluidized bed cavity, and collecting the near infrared spectrum in diffuse reflection mode.

[0017] Furthermore, the process parameter data include: inlet air temperature, inlet air humidity, exhaust air temperature, material temperature, atomization pressure and peristaltic pump speed.

[0018] Furthermore, the agglomeration rate model adopts a neural network model.

[0019] The second aspect of the present invention provides an online monitoring system for the agglomeration rate of fluidized bed coated micropellets.

[0020] An online monitoring system for agglomeration rate of fluidized bed coated pellets, comprising:

[0021] A data acquisition module is configured to: acquire near infrared spectrum and process parameter data of the whole process of fluidized bed coating at preset time intervals;

[0022] A data splicing module is configured to: splice the near infrared spectrum and process parameter data of the whole process of fluidized bed coating according to corresponding preset time intervals to obtain fused data;

[0023] The output module is configured to: obtain the pellet agglomeration rate based on the fusion data and using the trained agglomeration rate model;

[0024] The training process of the agglomeration rate model includes: taking the spliced ​​fusion data as input and taking the microcapsule agglomeration rate at each time point measured offline as output, and training the agglomeration rate model.

[0025] A third aspect of the present invention provides a computer-readable storage medium.

[0026] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for online monitoring of agglomeration rate of fluidized bed coated micropellets as described in the first aspect above.

[0027] A fourth aspect of the present invention provides a computer device.

[0028] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for online monitoring agglomeration rate of fluidized bed coated micropellets as described in the first aspect above are implemented.

[0029] A fifth aspect of the present invention provides a computer program product or a computer program.

[0030] The present invention provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in the method for online monitoring of agglomeration rate of fluidized bed coated micropellets as described in the first aspect above.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention provides a method and system for online monitoring of the agglomeration rate of micropellets coated in a fluidized bed. By utilizing the concept of multivariate heterogeneous data fusion, near-infrared spectral data and process parameter data are fused for the first time for online monitoring of the agglomeration rate of micropellets in a fluidized bed coating process. The method and system can effectively and accurately monitor the changes in the agglomeration rate of micropellets in a fluidized bed coating process under complex backgrounds.

[0033] Compared with single sensor modeling (predicting pellet agglomeration rate using only near-infrared spectroscopy or process parameters), the present invention greatly improves the prediction accuracy of the model, provides technical support for subsequent fluidized bed coating quality control, and thus improves the safety and effectiveness of drugs. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0035] Figure 1 It is a flow chart of the on-line monitoring method of agglomeration rate of fluidized bed coated micropellets shown in the present invention;

[0036] Figure 2 It is the original spectrum diagram shown in the present invention;

[0037] Figure 3 is an architectural diagram of a fusion network structure shown in the present invention;

[0038] Figure 4 Fusion CNN network feature visualization diagram;

[0039] Figure 5 It is a structural diagram of the on-line monitoring system for agglomeration rate of fluidized bed coated micropellets shown in the present invention. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0041] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0042] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0043] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the methods and systems according to various embodiments of the present disclosure. It should be noted that each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of a code may include one or more executable instructions for implementing the logical functions specified in each embodiment. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the flowchart and / or block diagram, and the combination of boxes in the flowchart and / or block diagram can be implemented using a dedicated hardware-based system that performs a specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0044] As mentioned above, the existing common monitoring of microcapsule agglomeration rate in the fluidized bed coating process uses process analysis technologies, such as visual imaging technology, spatial filtering technology, etc. to monitor the agglomeration rate online, but does not consider the impact of changes in process parameters. Therefore, there may be certain errors in the detection process.

[0045] In view of this, the present application proposes an online monitoring method and system for the agglomeration rate of fluidized bed coated micropellets, which is based on the fusion of near infrared spectroscopy technology and process parameters, combined with CNN modeling method, to predict the agglomeration rate of micropellets online. The present invention is described in detail below through specific embodiments:

[0046] Embodiment 1

[0047] like Figure 1 As shown, this embodiment provides an online monitoring method for agglomeration rate of fluidized bed coated pellets, comprising:

[0048] At preset time intervals, the near infrared spectrum and process parameter data of the entire fluidized bed coating process are obtained respectively;

[0049] The near infrared spectrum and process parameter data of the whole process of fluidized bed coating are spliced ​​according to the corresponding preset time intervals to obtain fused data;

[0050] Based on the fusion data, the trained agglomeration rate model was used to obtain the pellet agglomeration rate;

[0051] The training process of the agglomeration rate model includes: taking the spliced ​​fusion data as input and taking the microcapsule agglomeration rate at each time point measured offline as output, and training the agglomeration rate model.

[0052] By integrating near-infrared spectroscopy with process parameter data, the present invention can effectively and accurately monitor the changes in the agglomeration rate of micropellets in the fluidized bed coating process under complex backgrounds, providing reference and technical means for quality monitoring in the fluidized bed coating process, and therefore has good practical application value.

[0053] The present embodiment is described in detail below. The method for online monitoring agglomeration rate of fluidized bed coated pellets comprises the following steps:

[0054] Step (1): collecting near infrared spectra of the entire fluidized bed coating process;

[0055] Step (2): collecting process parameter data (inlet air temperature, inlet air humidity, exhaust air temperature, material temperature, atomization pressure, peristaltic pump speed) of the entire fluidized bed coating process;

[0056] Step (3): Collect samples at different time points during the fluidized bed coating process (corresponding to the spectral sequence) and measure the pellet agglomeration rate offline;

[0057] Step (4): splicing the fluidized bed near infrared spectroscopy data with the process parameter data, using convolutional neural networks (CNNs) and partial least squares (PLS) methods to establish and verify the model; using convolutional neural networks to extract features, and using partial least squares methods to further adjust and optimize the extracted features to improve the accuracy of data prediction; training the model; using the trained model to predict the agglomeration rate of fluidized bed coated micropellets;

[0058] Step (5): Compare with the near-infrared spectrum modeling alone and the process parameter modeling alone to verify the superiority of the fusion method.

[0059] In another specific embodiment of the present invention, the step (1) collects near infrared spectra of the whole process of fluidized bed coating, and the specific method includes:

[0060] During the fluidized bed coating process, a near infrared spectrometer (Micro NIR PATU) was inserted into the fluidized bed cavity, and the spectrum was collected in diffuse reflectance mode.

[0061] Specifically, before inserting the near-infrared probe, a polytetrafluoroethylene white plate was first collected as a background spectrum, the integration time was adjusted to 8.1ms, the wavelength range was 908.1-1676.0nm, and the number of scans for each spectrum was 100. The spectrum was collected throughout the entire fluidized bed coating process, and the spectrum was collected approximately every 30s. In order to prevent the near-infrared probe from being contaminated, compressed air was blown every 60s.

[0062] In another specific embodiment of the present invention, step (2) collects near-process parameter data (inlet air temperature, inlet air humidity, exhaust air temperature, material temperature, atomization pressure, peristaltic pump speed) of the whole process of fluidized bed coating: the time interval is the same as the near-infrared spectrum recording time interval, ensuring that it can be aligned with the near-infrared spectrum in time sequence, facilitating the establishment of the subsequent agglomeration rate model.

[0063] In another specific embodiment of the present invention, in step (3), samples at different time points (corresponding to the spectral sequence) are collected during the fluidized bed coating process to measure the agglomeration rate offline:

[0064] Among them, the samples collected at each time point were sieved using a sieve larger than the normal particle size of the micropellets (the maximum size of the micropellets in this article does not exceed 300μm, so a 355μm sieve was used). The agglomeration rate of the particles can be obtained by calculating the ratio of the mass of the particles retained on the sieve to the total mass of the sample. This indicator reflects the degree of particle agglomeration during fluidized bed filtration, which is a key factor in evaluating the quality and consistency of the final product.

[0065] In another specific embodiment of the present invention, step (4) combines the fluidized bed data with the process parameter data, and uses CNN and PLS methods to establish and verify the model;

[0066] Specifically, during the sampling process, the time of obtaining the sample, the spectrum and the process parameter time are recorded, and the agglomeration rate of the sample is correlated with the offline measured agglomeration rate. The near infrared spectrum (1×125) and the process parameters (1×6) are spliced ​​to obtain 1×131 dimensional data, and the agglomeration rate model is established using CNN. The model is used to evaluate the parameter R 2 C , R 2 V , R 2 T , RMSEC, RMSEV, RMSET, etc., to determine the optimal prediction model.

[0067] In another specific embodiment of the present invention, the features extracted by the CNN network are plotted in a graph to characterize key quality parameters.

[0068] In another specific embodiment of the present invention, step (5) is compared with the near-infrared spectrum modeling alone and the process parameter modeling alone to verify the superiority of the fusion method.

[0069] Specifically, by comparing different modeling methods (near-infrared spectrum CNN model alone, near-infrared spectrum PLS model alone, process parameter CNN model alone, process parameter PLS model alone), the superiority of integrating near-infrared spectrum and process parameter modeling is further demonstrated.

[0070] Embodiment 2

[0071] This example carried out a modeling experiment design, with a total of 14 batches of experiments. The specific process parameters are shown in Table 1. In order to ensure the robustness of the model, the peristaltic pump speed (X1, rpm / min), atomization pressure (X2, bar), inlet air temperature (X3, °C), inlet air volume (X4, m 3 / s) and spray volume (X5, g) to avoid the randomness of the established model as much as possible.

[0072] Table 1 Modeling model experimental design

[0073]

[0074] Step (1): Collect near infrared spectra of the entire fluidized bed coating process.

[0075] Before each coating batch, preheat the fluidized bed for half an hour to make the fluidized bed granulate in a relatively stable state. After preheating, put the materials in, mix for 10 minutes, and then start spraying the coating liquid.

[0076] During the coating process, the Micro NIR PATU near-infrared spectrometer was inserted into the fluidized bed cavity, and the spectrum was collected in diffuse reflection mode. Before inserting the near-infrared probe, a polytetrafluoroethylene white plate was collected as the background spectrum, the integration time was adjusted to 8.1ms, the wavelength range was 908.1-1676.0nm, and the number of scans for each spectrum was 100. The spectrum was collected throughout the entire fluidized bed coating process, and a spectrum was collected every 30s. The spectrum is shown in the figure below. Figure 2 To prevent the near-infrared probe from being contaminated, blow compressed air every 40 seconds.

[0077] (2) Collect the near-process parameter data of the entire fluidized bed coating process (inlet air temperature, inlet air humidity, exhaust air temperature, material temperature, atomization pressure, and peristaltic pump speed).

[0078] During the fluidized bed coating process, ensure that the collection interval of process parameters (inlet air temperature, inlet air humidity, exhaust air temperature, material temperature, atomization pressure, peristaltic pump speed) is consistent with the spectrum collection interval to ensure that data from different sources are aligned in the time dimension.

[0079] (3) Collect samples at different time points during the fluidized bed coating process (corresponding to the spectral sequence) and measure the pellet agglomeration rate offline.

[0080] During the fluidized bed coating process, approximately 13-14 samples were taken per batch, each sample weighing approximately 10 g. A total of 14 batches of experiments were conducted, with a total of 184 samples taken. The corresponding time was recorded when sampling, corresponding to the online near-infrared spectrum and process parameter time, for subsequent modeling analysis. The data set was randomly divided into a calibration set, a validation set, and a test set in a ratio of 3:1:1.

[0081] Agglomeration rate determination: Use a sieve larger than the normal particle size of the micropellets (the maximum size of the micropellets in this article does not exceed 300μm, so a 355μm sieve is used) to sieve the samples collected at each time point. The agglomeration rate of the particles can be obtained by calculating the ratio of the mass of the particles retained on the sieve to the total mass of the sample.

[0082] (4) The fluidized bed data were combined with the process parameter data, and the CNN and PLS methods were used to establish and verify the model.

[0083] Establishment of agglomeration rate fusion model: record the time of sample acquisition and the time of spectrum and process parameters, and associate it with the agglomeration rate measured offline. After splicing the near infrared spectrum (1×125) and process parameters (1×6), 1×131 dimensional data is obtained, and the CNN method is used for modeling. The CNN structure is a 1D CNN customized based on a ResNet-like architecture, which is designed to process and fuse one-dimensional data signals. The network structure is as follows Figure 3 As shown. The architecture starts with an image input layer, followed by a convolutional layer with a filter size of 1×3, combined with batch normalization and ReLU activation to ensure stability and accelerate training. The maximum pooling layer is used to reduce the dimensionality while retaining the key features of the input data. The network applies three convolutional blocks, each consisting of two 1×3 convolutional layers, batch normalization, and ReLU activation. The use of residual connections (additional layers) ensures that information from earlier layers is retained, which helps alleviate the vanishing gradient problem common in deep networks. This enables deeper layers to learn effectively without degrading performance.

[0084] All models were trained on an NVIDIA GeForce RTX 4070 with 32GB RAM. Model evaluation parameters included: coefficient of determination (R 2 ), root mean square error (RMSE) and performance to deviation ratio (RPD) to evaluate the model performance.

[0085]

[0086]

[0087]

[0088] Among them, y i , actual represents the reference value of the i-th sample, y i , predicted represents the predicted value of the i-th sample, represents the mean of the actual value, and m is the number of samples. Generally speaking, a good model will have a higher R 2 and RPD and lower RMSE.

[0089] The fusion CNN model hyperparameters are as follows: batch size, learning rate and maximum number of iterations are set to 32, 0.001 and 1000 respectively. The fusion model results are shown in Table 2.

[0090] Table 2 Comparison between the clustering rate fusion CNN model and the PLS model

[0091]

[0092] The underline represents the optimal model result.

[0093] In Table 2, R 2 C represents the calibration set determination coefficient; R 2 V represents the coefficient of determination of the validation set; R 2 T It represents the determination coefficient of the test set; RMSEC represents the root mean square error of the calibration set; RMSEP represents the root mean square error of the validation set; RMSET represents the root mean square error of the test set. As can be seen from Table 2, the results of the fusion model based on CNN are far better than those of the PLS model, which proves the superiority of the CNN model.

[0094] (5) Using convolutional neural network to extract features, and using partial least squares method to further adjust and optimize the extracted features;

[0095] The spliced ​​features are input into the fused CNN network, and the features are extracted through the convolution layer. The extracted features are adjusted and optimized using the partial least squares method. The adjusted and optimized features are passed through the activation function layer, pooling layer and fully connected layer to obtain the output result, i.e., the agglomeration rate of fluidized bed coated micropellets. Figure 4 (a) is the original input spectrum plus process parameter diagram, Figure 4 (b) is the feature map after convolution of the fused CNN network. It can be seen from the figure that after convolution processing, the features are mainly concentrated at the end of the spectrum (1450nm) and the temperature of the corresponding process parameters. The end of the spectrum is mainly related to the combination band of OH stretching vibration and HOH bending vibration, which is also consistent with the continuous spraying of coating liquid during the theoretical coating process. The material temperature also reflects a sign of whether the current material is too wet. Overwetting also means that excessive liquid is sprayed, which means that the agglomeration rate may increase, which is consistent with the theoretical trend.

[0096] (6) Compared with the modeling of near-infrared spectroscopy alone and the modeling of process parameters alone, the superiority of this fusion method was verified.

[0097] The hyperparameters of the single sensor CNN model are as follows: batch size, learning rate, and maximum number of iterations are set to 32, 0.001, and 1000, respectively. The results of the single sensor model are shown in Table 3.

[0098] Table 3 Comparison of agglomeration rate single sensor CNN model and PLS model

[0099]

[0100] The underline represents the optimal model result.

[0101] The optimal model for fusing near infrared and process parameters is the fusion CNN model, where the RPD value is 4.95, while the optimal model for a single sensor is the process parameter CNN model, where the RPD value is 2.68. It can be seen that the performance of the agglomeration rate model fusing near infrared and process parameters is much better than the model seen by a single near infrared or process parameter, proving the superiority of the fusion model.

[0102] The present invention combines near-infrared spectral data with process parameter data, and combines CNN and PLS models to monitor the changes in the agglomeration rate of micropills during fluidized bed coating online. The results show that the performance of the CNN model integrating multiple data sources is much better than the model of a single sensor. The establishment of the technical solution of the present invention helps to timely monitor the dynamic behavior of the coating process during fluidized bed coating, ensure the consistency of the quality of the coating process, thereby improving the safety and effectiveness of the drug, and therefore has good practical application value.

[0103] Embodiment 3

[0104] like Figure 5 As shown, this embodiment provides an online monitoring system for the agglomeration rate of fluidized bed coated micropellets.

[0105] An online monitoring system for agglomeration rate of fluidized bed coated pellets, comprising:

[0106] A data acquisition module is configured to: acquire near infrared spectrum and process parameter data of the whole process of fluidized bed coating at preset time intervals;

[0107] A data splicing module is configured to: splice the near infrared spectrum and process parameter data of the whole process of fluidized bed coating according to corresponding preset time intervals to obtain fused data;

[0108] The output module is configured to: obtain the pellet agglomeration rate based on the fusion data and using the trained agglomeration rate model;

[0109] The training process of the agglomeration rate model includes: taking the spliced ​​fusion data as input and taking the microcapsule agglomeration rate at each time point measured offline as output, and training the agglomeration rate model.

[0110] In some embodiments, the system further comprises a training module, which is configured to: select a sieve larger than the normal particle size of the microcapsules to sieve the samples collected at each time point; and obtain the microcapsules agglomeration rate at each time point based on the ratio of the mass of the microcapsules retained on the sieve to the mass of the sample.

[0111] In some embodiments, the samples collected at each time point correspond to the time at which the near infrared spectrum and process parameter data were acquired.

[0112] In some embodiments, the data acquisition module is further configured to: before acquiring the near-infrared spectrum of the entire fluidized bed coating process, include: inserting a near-infrared spectrometer into the fluidized bed cavity and collecting the near-infrared spectrum in diffuse reflection mode.

[0113] In some embodiments, the process parameter data include: inlet air temperature, inlet air humidity, exhaust air temperature, material temperature, atomization pressure and peristaltic pump speed.

[0114] In some embodiments, the agglomeration rate model adopts a neural network model.

[0115] Embodiment 4

[0116] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the method for online monitoring of agglomeration rate of fluidized bed coated micropellets as described in the above-mentioned embodiment 1 or embodiment 2 are implemented.

[0117] Embodiment 5

[0118] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for online monitoring the agglomeration rate of fluidized bed coated microcapsules as described in the above-mentioned embodiment 1 or embodiment 2 are implemented.

[0119] Embodiment 6

[0120] This embodiment provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in the method for online monitoring of agglomeration rate of fluidized bed coated micropellets described in the first or second embodiment above.

[0121] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0122] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0123] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0124] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0125] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0126] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for online monitoring of agglomeration rate of fluidized bed coated pellets, characterized in that: include: At preset time intervals, the near infrared spectrum and process parameter data of the entire fluidized bed coating process are obtained respectively; The near infrared spectrum and process parameter data of the whole process of fluidized bed coating are spliced ​​according to the corresponding preset time intervals to obtain fused data; Based on the fusion data, the trained agglomeration rate model was used to obtain the pellet agglomeration rate; The training process of the agglomeration rate model includes: taking the spliced ​​fusion data as input and taking the microcapsule agglomeration rate at each time point measured offline as output, and training the agglomeration rate model.

2. The method for online monitoring agglomeration rate of fluidized bed coated pellets according to claim 1, characterized in that: The process of offline measuring the agglomeration rate of the microcapsules at each time point includes: selecting a sieve larger than the normal particle size of the microcapsules to sieve the samples collected at each time point; and obtaining the agglomeration rate of the microcapsules at each time point according to the ratio of the mass of the microcapsules retained on the sieve to the mass of the sample.

3. The method for online monitoring agglomeration rate of fluidized bed coated pellets according to claim 2, characterized in that: The samples collected at each time point correspond to the time of the acquired near infrared spectrum and process parameter data.

4. The method for online monitoring agglomeration rate of fluidized bed coated pellets according to claim 1, characterized in that: Before obtaining the near infrared spectrum of the whole process of fluidized bed coating, the method includes: inserting a near infrared spectrometer into the fluidized bed cavity, and collecting the near infrared spectrum in diffuse reflection mode.

5. The method for online monitoring agglomeration rate of fluidized bed coated pellets according to claim 1, characterized in that: The process parameter data include: inlet air temperature, inlet air humidity, exhaust air temperature, material temperature, atomization pressure and peristaltic pump speed.

6. The method for online monitoring agglomeration rate of fluidized bed coated pellets according to claim 1, characterized in that: The agglomeration rate model adopts a neural network model.

7. An online monitoring system for agglomeration rate of fluidized bed coated pellets, characterized in that: include: A data acquisition module is configured to: acquire near infrared spectrum and process parameter data of the whole process of fluidized bed coating at preset time intervals; A data splicing module is configured to: splice the near infrared spectrum and process parameter data of the whole process of fluidized bed coating according to corresponding preset time intervals to obtain fused data; The output module is configured to: obtain the pellet agglomeration rate based on the fusion data and using the trained agglomeration rate model; The training process of the agglomeration rate model includes: taking the spliced ​​fusion data as input and taking the microcapsule agglomeration rate at each time point measured offline as output, and training the agglomeration rate model.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the on-line monitoring method for agglomeration rate of fluidized bed coated micropellets as described in any one of claims 1 to 6 are implemented.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for online monitoring the agglomeration rate of fluidized bed coated micropellets according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps in the method for online monitoring of agglomeration rate of fluidized bed coated micropellets according to any one of claims 1 to 6 are implemented.

Images