Particle size measurement method of traditional Chinese medicine particles, manufacturing process control method, granulation equipment and electronic equipment

Abstract

The invention provides a particle size measurement method of traditional Chinese medicine particles, a manufacturing process control method, granulation equipment and electronic equipment, and belongs to the field of online measurement control. The real-time scanning spectrum of the traditional Chinese medicine particles in the granulation equipment is input into the preset particle size detection model, and the particle size distribution corresponding to the real-time scanning spectrum is generated by using the particle size detection model, so that real-time online detection of the particle size of the traditional Chinese medicine particles and real-time recording of particle size data in the fluidized bed granulation process are effectively realized; and display information that the particle size distribution changes along with the granulation time is automatically generated and displayed, so that visualization of the particle size change condition in the fluidized bed granulation process is facilitated. Furthermore, according to the real-time particle size distribution, the production parameters of the granulation equipment are autonomously optimized and adjusted, and the online process control of the granulation process is realized, so that the traditional Chinese medicine particles with good compressibility and superior particle size range can be more stably obtained, and the production quality of the traditional Chinese medicine particles is ensured.

Description

Technical Field

[0001] The present invention relates to the field of online measurement and control, and in particular to a particle size measurement method, a manufacturing process control method, a granulation device and an electronic device for traditional Chinese medicine particles. Background Art

[0002] In the traditional Chinese medicine manufacturing process, the particle size distribution of traditional Chinese medicine has a significant impact on the intrinsic quality such as granule packaging, capsule and tablet filling difference and content, tablet surface appearance, etc. Therefore, the measurement of the particle size of traditional Chinese medicine particles is crucial to ensure the quality and efficacy of the medicine.

[0003] At present, the methods for measuring the particle size of traditional Chinese medicine granules include: manual screening method, microscope method, mechanical screening method, air jet screening method, and light scattering method. However, these methods all have the following defects: (1) they cannot detect continuously in real time; (2) they cannot monitor the real-time particle size changes during the granulation process; (3) the samples are easily contaminated after testing and cannot be recycled.

[0004] Therefore, there is an urgent need for a method to detect particle size changes in real time without contaminating the Chinese medicine particles. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a particle size measurement method, a manufacturing process control method, a granulation equipment and an electronic device for Chinese medicine granules, so as to monitor the changes in particle size during the fluidized bed granulation process of Chinese medicine in real time, and automatically adjust the production parameters in real time according to the real-time particle size data, so as to stably obtain Chinese medicine granules with good compressibility and high-quality particle size range.

[0006] In order to achieve the above purpose, the technical solution adopted by the embodiment of the present invention is as follows: In a first aspect, the present invention provides a method for measuring the particle size of traditional Chinese medicine particles, the method comprising: Receive the real-time scanning spectrum of the Chinese medicine granules in the granulation equipment by the near infrared spectrometer; Inputting the real-time scanning spectrum into a preset particle size detection model, and using the particle size detection model to generate a particle size distribution corresponding to the real-time scanning spectrum; According to each of the particle size distributions, information on the change of the particle size distribution with the granulation time is displayed.

[0007] Optionally, the particle size detection model includes a large particle detection model and a small particle detection model; The step of inputting the real-time scanning spectrum into a preset particle size detection model and using the particle size detection model to generate a particle size distribution corresponding to the real-time scanning spectrum comprises: Preprocessing the real-time scanning spectrum to obtain an effective scanning band; Based on the real-time scanning spectrum, determining a representative particle size of the real-time scanning spectrum; Determining whether the representative particle size is greater than a large particle threshold; If yes, the effective scanning band is input into the large particle detection model, and the large particle detection model is used to generate a particle size distribution corresponding to the effective scanning band; If not, the effective scanning band is input into the small particle detection model, and the small particle detection model is used to generate the particle size distribution corresponding to the effective scanning band.

[0008] Optionally, the step of determining the representative particle size of the real-time scanning spectrum based on the real-time scanning spectrum comprises: The real-time scanning spectrum is processed using a preset component analysis algorithm to obtain a representative particle size of the real-time scanning spectrum.

[0009] Optionally, the method for constructing the particle size detection model includes: Controlling the near infrared spectrometer to scan the sample in the granulation process to obtain a plurality of sample spectra, and sampling the sample in the granulation process to obtain a plurality of sampled samples; wherein the sample comprises a large particle sample and a small particle sample, and the main particle size of the large particle sample is larger than the main particle size of the small particle sample; Preprocessing the sample spectrum, and extracting effective bands from the preprocessed sample spectrum to obtain a sample spectrum; The sampled samples are tested by laser particle size method to obtain a distribution reference value of the sample particle size corresponding to each sample spectrum; Based on the sample spectra of the large particle samples and the distribution reference values, model building and optimization are performed to obtain a large particle detection model; Based on the sample spectra of the small particle samples and the distribution reference values, a model is constructed and optimized to obtain a small particle detection model.

[0010] Optionally, the sample spectrum is preprocessed, and an effective band is extracted from the preprocessed sample spectrum to obtain a sample spectrum; Using a denoising algorithm to denoise the sample spectrum to obtain a denoised spectrum; The denoised spectrum that meets the effective conditions is subjected to signal amplification to obtain a sample spectrum.

[0011] In a second aspect, the present invention provides a method for controlling a granulation process of traditional Chinese medicine granules, the method comprising: Using the particle size measurement method of the traditional Chinese medicine granules as described in the first aspect to obtain the current particle size distribution; According to the particle size distribution, judging whether the quality condition is met; If not, the particle size distribution is input into an optimization decision model, and the optimization decision model is used to generate control optimization parameters; According to the control optimization parameters, the production parameters of the granulation equipment are adjusted, and the granulation equipment is controlled to continue granulation.

[0012] Optionally, after the step of determining whether the high-quality condition is met according to the particle size distribution, the method further comprises: If it is determined that the quality condition is met, the granulation is terminated.

[0013] Optionally, the method for constructing the optimization decision model includes: Obtaining multiple groups of control sample parameters and particle size distribution corresponding to each group of control sample parameters; For each particle size distribution, the corresponding control sample parameter is used as a reference control parameter of the particle size distribution, and the particle size distribution is labeled to obtain a training sample; Based on each of the training samples, the preset neural network model is iteratively trained to obtain a mature optimization decision model.

[0014] In a third aspect, the present invention provides a granulation device, comprising a body, a near infrared spectrometer, a functional component and a processing device, wherein the near infrared spectrometer is arranged on a window of the body; The near-infrared spectrometer is used to scan particles in the body in real time to obtain a real-time scanning spectrum, and transmit the real-time scanning spectrum to the processing device; The processing equipment is used to implement the particle size measurement method of the traditional Chinese medicine granules as described in the first aspect, or the granulation process control method of the traditional Chinese medicine granules as described in the second aspect.

[0015] In a fourth aspect, the present invention provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and the processor can execute the computer program to implement the particle size measurement method of the traditional Chinese medicine particles as described in the first aspect, or the granulation process control method of the traditional Chinese medicine particles as described in the second aspect.

[0016] In a fifth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the particle size measurement method of the traditional Chinese medicine particles as described in the first aspect, or the granulation process control method of the traditional Chinese medicine particles as described in the second aspect.

[0017] The present invention has at least the following beneficial effects: The particle size measurement method, manufacturing process control method, granulation equipment and electronic equipment of the Chinese medicine granules provided by the present invention can effectively realize the real-time online detection of the particle size of the Chinese medicine granules in the fluidized bed granulation process and the real-time recording of the particle size data, as well as automatically generate and display the display information of the particle size distribution changing with the granulation time, which is helpful to realize the visualization of the particle size change in the fluidized bed granulation process. Thus, the traditional granulation production process is digitized, visualized and concretized, and the real-time continuous detection of the Chinese medicine granules in the Chinese medicine production process and the detection of the particle size change are realized, while no pollution is generated to the Chinese medicine granules, reducing waste. In addition, the particle size quality index of the whole process can be displayed in front of the operator with actual data to guide the employee's operation, avoiding the employee's operation technique based on experience and feel.

[0018] During the granulation process, online near-infrared real-time detection and visualization of the particle size changes of the Chinese medicine granules are used to determine whether the quality conditions are met based on the real-time particle size distribution. When the quality conditions are not met, the production parameters of the granulation equipment (such as adhesive flow, spray pressure, air inlet temperature, air inlet volume, etc.) are autonomously optimized and adjusted by combining the optimization decision model with the particle size prediction model to control the particle size of the Chinese medicine granules to be stable within the quality standard range, avoiding the drawbacks of relying entirely on the operator's feel and experience. In this way, Chinese medicine granules with a good compressibility and quality range can be stably obtained to ensure the production quality of Chinese medicine granules. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 A schematic structural diagram of a Chinese medicine preparation system provided in an embodiment of the present invention is shown.

[0021] Figure 2 A block diagram of an electronic device provided by an embodiment of the present invention is shown.

[0022] Figure 3 One of the flow charts of the particle size measurement method of traditional Chinese medicine particles provided in an embodiment of the present invention is shown.

[0023] Figure 4 The second schematic flow chart of the particle size measurement method of traditional Chinese medicine particles provided in an embodiment of the present invention is shown.

[0024] Figure 5 A graph showing the relationship between the predicted value of the particle size detection model and the reference value.

[0025] Figure 6 Shows Figure 3 Schematic diagram of the process flow of some sub-steps of step 13.

[0026] Figure 7 One of the flow charts of the granulation process control method of traditional Chinese medicine granules provided in an embodiment of the present invention is shown.

[0027] Figure 8 The second flow chart of the granulation process control method of the traditional Chinese medicine granules provided in the embodiment of the present invention is shown.

[0028] Explanation of the reference numerals: 10 - traditional Chinese medicine preparation system; 110 - granulation equipment; 120 - processing equipment; 130 - near-infrared spectrometer; 140 - cooling sleeve; 141 - air inlet; 142 - air outlet; 150 - model building equipment; 20 - electronic equipment; 210 - memory; 220 - processor; 230 - communication module. DETAILED DESCRIPTION

[0029] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0030] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0031] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0032] The particle size measurement method of traditional Chinese medicine granules provided by the embodiment of the present invention can be applied to Figure 1In the illustrated Chinese medicine preparation system 10, the Chinese medicine preparation system 10 includes a granulation device 110 and a model building device 150, a processing device 120, a near infrared spectrometer 130 and a cooling sleeve 140 of the granulation device 110. A viewing window is provided on the body of the granulation device 110, and the probe of the near infrared spectrometer 130 is sleeved in the cooling sleeve 140, and the near infrared spectrometer 130 is fixed on the viewing window through the cooling sleeve 140, so that the detection head of the near infrared spectrometer 130 extends into the granulation device 110. An air inlet 141 and an air outlet 142 are provided on the cooling sleeve 140, and the air inlet 141 is connected to a compressed air pipe, and compressed air is passed from the compressed air pipe into the cooling sleeve 140 and then discharged through the air outlet 142.

[0033] In addition, the granulation device 110 also includes functional components that provide binder flow, spray pressure, inlet air temperature, inlet air volume, humidity, etc., and a display screen.

[0034] The processing device 120 can be connected to the near-infrared spectrometer 130, the functional components and the model building device 150 through any communication method such as data line, network, WiFi, Bluetooth, etc.

[0035] The model building device 150 is used to construct the particle size detection model and optimization decision model used in the particle size measurement method and granulation process control method of the traditional Chinese medicine particles provided in the embodiment of the present invention, deploy the constructed particle size detection model and optimization decision model to the processing device 120, and update the particle size detection model and optimization decision model deployed on the processing device 120.

[0036] The near infrared spectrometer 130 is used to scan particles in the body in real time, obtain a real-time scanning spectrum, and transmit the real-time scanning spectrum to the processing device 120 .

[0037] The processing device 120 is used to implement the particle size measurement method of the traditional Chinese medicine granules provided in the embodiment of the present invention, or the granulation process control method of the traditional Chinese medicine granules provided in the embodiment of the present invention.

[0038] The display screen is used to display the real-time measured particle size distribution and the trend graph of the particle size distribution versus granulation time. The particle size distribution refers to the particle size distribution of the Chinese medicine granules, specifically including the measured particle size value of each particle.

[0039] In the above-mentioned traditional Chinese medicine preparation system 10, the function of the cooling sleeve 140 is to take away the heat generated by the heat conduction of the granulation equipment 110 and the near-infrared spectrometer 130 itself by continuously passing clean compressed air at room temperature, so that the temperature of the near-infrared spectrometer 130 is kept stable, thereby ensuring the high precision of the online near-infrared spectrometer 130 in measuring the particle size.

[0040] Here, the cooling sleeve 140 may be made of stainless steel or any other conventional material.

[0041] The processing device 120 may be, but is not limited to, a programmable logic controller, a single chip microcomputer, a field programmable gate array, a computer, a server, etc. The granulating device 110 may be a fluidized bed or any other granulating machine.

[0042] Please refer to Figure 2 , is a block diagram of an electronic device 20, which may be Figure 1 The processing device 120 in the Chinese medicine preparation system 10 shown. The electronic device 20 includes a memory 210, a processor 220 and a communication module 230. The memory 210, the processor 220 and the communication module 230 are electrically connected to each other directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0043] The memory 210 is used to store programs or data and can be, but not limited to, a random access memory, a read-only memory, a programmable read-only memory, an erasable read-only memory, an electrically erasable read-only memory, and the like.

[0044] The processor 220 is used to read / write data or programs stored in the memory 210 and execute corresponding functions. For example, Figure 1 In the traditional Chinese medicine preparation system 10 shown, the processor 220 of the processing device 120 reads the computer program stored in the memory 210 to implement the particle size measurement method of the traditional Chinese medicine particles provided in the embodiment of the present invention, or the manufacturing process control method of the traditional Chinese medicine particles provided in the embodiment of the present invention.

[0045] The communication module 230 is used to establish a communication connection between the electronic device 20 and other communication terminals, and to send and receive data. For example, Figure 1 In the traditional Chinese medicine preparation system 10 shown, the processing device 120 can send and receive data with the near-infrared spectrometer 130.

[0046] It should be understood that Figure 2 The structure shown is only a schematic diagram of the structure of the electronic device 20. The electronic device 20 may also include Figure 2 More or fewer components as shown, or with Figure 2 Different configurations shown. Figure 2 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0047] In order to solve the current problem of lack of effective methods for real-time continuous monitoring of particle size changes in the fluidized bed granulation process of traditional Chinese medicine, and the problem that samples are easily contaminated after testing and cannot be recycled, reference Figure 3 The embodiment of the present invention provides a method for measuring the particle size of traditional Chinese medicine particles, comprising steps 11 to 15. Figure 1 In the Chinese medicine preparation system 10 shown, the processing device 120 can Figure 2 In the structure shown, when the processor 220 executes the computer program executed in the memory 210, steps 11 to 15 are implemented.

[0048] Step 11, receiving a real-time scanning spectrum of the Chinese medicine granules in the granulation equipment by a near infrared spectrometer.

[0049] Step 13, input the real-time scanning spectrum into a preset particle size detection model, and use the particle size detection model to generate a particle size distribution corresponding to the real-time scanning spectrum.

[0050] Step 15, based on each particle size distribution, display information of the change of particle size distribution with granulation time.

[0051] The displayed information may be a trend chart, a table, or a report. The above-mentioned display information forms are only examples and are not limited to any form.

[0052] For example, in combination Figure 1 The Chinese medicine preparation system 10 shown in the figure preheats the near infrared spectrometer 130 before granulation. When it is preheated to a preset temperature and stabilized, the parameters of the near infrared spectrometer 130 are adjusted to the optimal scanning parameters. Among them, due to the difference in pharmaceutical particle size, pharmaceutical parameters, etc., the setting of the optimal scanning parameters is not limited. For example, the optimal scanning parameters can be: integration time: 7.6ms; number of scans: 100 times; scanning mode: diffuse reflection; trigger mode: continuous scanning mode.

[0053] After adjusting the parameters of the near infrared spectrometer 130 and the feeding of the granulation device 110 (such as a fluidized bed) is completed, the granulation device 110 enters the granulation mode, and the near infrared spectrometer 130 continuously scans the Chinese medicine particles inside the granulation device 110 according to the preset scanning cycle (such as scanning once every 3 seconds), obtains a real-time scanning spectrum, and transmits the real-time scanning spectrum to the processing device 120. After receiving the real-time scanning spectrum, the processing device 120 inputs the real-time scanning spectrum into the preset particle size detection model, and uses the particle size detection model to generate the particle size distribution corresponding to the real-time scanning spectrum. In this way, the above process is repeatedly performed.

[0054] At the same time, after obtaining the particle size distribution detected at least twice, the processing device 120 generates display information of the particle size distribution changing with the granulation time according to the current particle size distribution, and sends the display information to the display screen for display.

[0055] At present, the existing methods for measuring the particle size of traditional Chinese medicine include manual screening, microscopy, mechanical screening, air jet screening, light scattering, etc. The manual screening method is to use sieves of different mesh sizes to sieve and separate the traditional Chinese medicine particles, weigh the weight of coarse particles that do not pass through the large-aperture sieve, and the weight of fine particles that pass through the small-aperture sieve, and finally calculate the proportion (%) of coarse particles, fine particles and intermediate particles respectively, so as to characterize the particle size of the traditional Chinese medicine particles. These measurement methods have at least the following defects: (1) Real-time continuous detection is not possible; (2) Real-time particle size changes during the granulation process cannot be monitored; (3) Samples are easily contaminated after detection and cannot be recycled.

[0056] In addition, there is a process monitoring method for preparing Chinese medicine granules in a fluidized bed based on near-infrared spectroscopy. The comprehensive use of different types of control charts can quickly identify abnormal conditions in the granulation process. However, it cannot continuously monitor the particle size changes during the granulation process of Chinese medicine granules in real time.

[0057] Through the above steps 11 to 15, the particle size measurement method of the Chinese medicine granules provided by the embodiment of the present invention can effectively realize the real-time online detection of the particle size of the Chinese medicine granules in the fluidized bed granulation process and the real-time recording of the particle size data, as well as automatically generate and display the display information of the particle size distribution changing with the granulation time, which is helpful to realize the visualization of the particle size change in the fluidized bed granulation process. Thus, the traditional granulation production process is digitized, visualized, and concretized, and the real-time continuous detection of the Chinese medicine granules in the Chinese medicine production process and the detection of the particle size change are realized, while no pollution is generated to the Chinese medicine granules, reducing waste. In addition, the particle size quality index of the whole process can be displayed in front of the operator with actual data to guide the employee's operation, avoiding the employee's operation method based on experience and feel.

[0058] In order to obtain a particle size detection model which takes a scanning spectrum of a near infrared spectrometer as input and a particle size distribution as output, a method for constructing a particle size detection model is provided.

[0059] The method of constructing the particle size detection model can be flexibly selected. For example, the particle size detection model can be obtained by iteratively training the neural network model, or by data fitting. The above method is only an example, and its implementation method is not limited.

[0060] In order to detect the detection deviation of the particle size detection model and improve the accuracy of the detection result, the particle size measurement method of the traditional Chinese medicine granules provided in the embodiment of the present invention introduces the idea of ​​using sample data of large particle segments and small particle segments to obtain the particle size detection model by segment modeling. Figure 4 , in the particle size measurement method of traditional Chinese medicine granules, Figure 1 The model building device 150 in the illustrated traditional Chinese medicine preparation system 10 can build a particle size detection model through steps 21 to 29 .

[0061] Step 21, controlling the near infrared spectrometer to scan the samples in the granulation process to obtain a plurality of sample spectra, and sampling the samples in the granulation process to obtain a plurality of sampled samples.

[0062] The samples include large particle samples and small particle samples, and the main particle size of the large particle samples is larger than the main particle size of the small particle samples. The large particle samples are Chinese medicine particles with a particle size larger than a preset threshold, and the small particle samples are Chinese medicine particles with a particle size smaller than the preset threshold.

[0063] Step 23, preprocessing the sample spectrum and extracting effective bands from the preprocessed sample spectrum to obtain a sample spectrum.

[0064] Step 25, using a laser particle size method to detect the sampled samples, and obtain a distribution reference value of the sample particle size corresponding to each sample spectrum.

[0065] Step 27, based on the spectra and distribution reference values ​​of each sample of the large particle sample, a model is constructed and optimized to obtain a large particle detection model.

[0066] Step 29, based on the spectra and distribution reference values ​​of each sample of the small particle sample, a model is constructed and optimized to obtain a small particle detection model.

[0067] In step 21, the process of acquiring the sample spectrum can be as follows: the near-infrared spectrometer is preheated to a stable temperature before granulation by the granulating equipment, the parameters of the near-infrared spectrometer are adjusted to the best spectrometer scanning parameters, and the near-infrared spectrometer is controlled to scan once every 3 seconds during the entire granulation process from the end of feeding of the granulating equipment to the end of drying, so as to continuously scan and obtain the sample spectrum.

[0068] In the process of data collection in step 21, representative samples are collected at the same time. The collection process can be: sampling is performed at each stage during the granulation process to obtain sampled samples. The sampled samples at each stage can be multiple.

[0069] For example, in the spraying stage of the granulation process, a sample is taken every 5 minutes from the start of spraying until the end of spraying, so as to obtain multiple sampling samples of the spraying stage. In addition, in the drying stage of the granulation process, a sample is taken every 3 minutes from the start of drying and a sample is taken at the end of drying, so as to obtain multiple sampling samples of the drying stage. The sampling amount of each sampling sample is greater than or equal to 10g.

[0070] It should be noted that during the granulation process, the raw materials in the granulation equipment will gradually change from powder to small particles of Chinese medicine, and then to large particles of Chinese medicine. Therefore, through a scan from the end of feeding to the end of drying, the sample spectra of large particle samples and small particle samples can be obtained at the same time.

[0071] In step 23, the method of preprocessing the sample spectrum can be flexibly selected, for example, it can be denoising or data cleaning, and the implementation method is not limited.

[0072] In order to avoid the presence of abnormal samples and noise in the sample spectrum, which may affect the detection accuracy of the subsequently constructed model, in step 23, a denoising algorithm is used to denoise the sample spectrum to obtain a denoised spectrum; the denoised spectrum that meets the effective conditions is signal amplified to obtain a sample spectrum.

[0073] Here, the denoising algorithm may be, but is not limited to, principal component analysis (PCA), first-order derivative, standard normal transformation, polynomial smoothing or any other denoising method. The effective condition may be that there is a band with a large amplitude threshold in the denoised spectrum, or that the wavelength of the valid data segment is greater than the wavelength threshold, or that the data in a preset band (such as 914-1670nm) is used as data that meets the effective condition. The above method is only an example, and its implementation method is not limited.

[0074] In order to make the distribution reference value more accurate, in step 25, for each sampled sample, the sampled sample can be divided into multiple parallel samples, and the parallel samples are tested by laser particle size method to obtain the particle size distribution of each sample particle in the parallel sample, and then based on the superior grade standard of the sample and the particle size distribution of each sample particle in the parallel sample, the measurement result of each parallel sample is calculated, such as Dx (50) measurement, Dx (60) measurement, etc. Then, the measurement results of each parallel sample of the sampled sample are averaged to obtain the distribution reference value of the sampled sample.

[0075] For each sample spectrum, the distribution reference value of the sample taken at the same sampling time is used as the distribution reference value of the sample particle size corresponding to the sample spectrum.

[0076] The high-quality standard of the sample can be a pre-set value, a value fitted from a calibration experiment, or a value converted from an existing mature sieving particle size standard. The method of obtaining it is not limited.

[0077] In order to further improve the accuracy of the distribution reference value, the particle size obtained by the screening method is matched one by one with the particle size data of the online near-infrared detection, and a linear regression fitting is performed to obtain the fitting formula. Then, the fitting formula is used to convert the mature screening method particle size standard of the traditional Chinese medicine granules to obtain the superior grade standard (also called the intermediate standard).

[0078] Take the Chinese medicine granules, Jianweixiaoshi tablets granules, as an example to determine Dx(50).

[0079] ① The mature sieving method particle size standard for Jianwei Xiaoshi Tablets Granules is shown in Table 1.

[0080] Table 1 Test items Internal Control Standards Premium product standards Amount of fine powder passing through 100 mesh sieve (unit: g / 100g) (unit: g / 100g) 30～62 37~52 ② The particle size (-100 mesh) obtained by sieving the same sample of multiple batches of traditional Chinese medicine granules was matched one by one with the particle size data of Dx (50), and a linear fit was performed. The fitting formula can be: y=-4.7884x+363.07, where y represents the particle size of Dx (50) in μm, and x represents the amount of particles passing through the 100-mesh sieve measured by the sieving method (-100 mesh) in g / 100g.

[0081] ③ According to the above fitting formula, the upper and lower limits of the internal control standard 30, 62 and the upper and lower limits of the premium product standard 37, 52 in Table 1 are respectively substituted into the x in the formula, and converted into the upper and lower limits of the internal control standard of the near-infrared method Dx (50) particle size 66μm, 219μm, and the upper and lower limits of the premium product standard 114μm, 186μm. The final near-infrared method Dx (50) particle size detection standard is shown in Table 2.

[0082] Table 2

[0083] After obtaining the distribution reference value and sample spectrum of each sample by adopting the above steps 21 to 25, the sample spectrum can be segmented by using the principal component analysis technology and the linear discriminant analysis technology to obtain the sample spectra of each large particle sample and the sample spectra corresponding to each small particle sample. For example, the particle size of the main component particles in the sample spectrum can be obtained by using the principal component analysis technology and the linear discriminant analysis technology, and then segmentation can be performed according to the particle size of the main component particles.

[0084] For steps 27 to 29, the least square method, maximum likelihood estimation, nonlinear least square method and other methods can be used to construct and optimize the large particle detection model and the small particle detection model. It is also possible to iteratively train the neural network model based on the sample spectrum and the distribution reference value to obtain the large particle detection model and the small particle detection model. Here, there is no restriction on the method of constructing and optimizing the large particle detection model and the small particle detection model.

[0085] Taking the least squares algorithm modeling as an example, the 914-1670nm band in the denoised spectrum is screened for signal amplification to obtain a sample spectrum. Based on the sample spectrum, the least squares algorithm is used for segmented modeling. The accuracy of the obtained model is as follows: Figure 5 As shown. Figure 5 It can be seen that the model error is around 2.4, and the detection accuracy is greatly improved.

[0086] Through the above steps 21 to 29, the sample is segmented (small particle segment, large particle segment) and modeled to obtain a large particle detection model suitable for detecting large particle size particles and a small particle detection model suitable for detecting small particle size particles, thereby greatly reducing the model's detection deviation for small particle size and improving the accuracy of the model's detection results.

[0087] After the particle size detection model (i.e., large particle detection model and small particle detection model) is obtained by the above method, the particle size detection model (i.e., large particle detection model and small particle detection model) is deployed in the processing equipment. Thus, in step 13, the particle size detection model is used to detect the real-time scanning spectrum to achieve real-time and continuous monitoring of the Chinese medicine particles.

[0088] On the basis of the above, in order to improve the accuracy of the particle size distribution, the representative particle size of the effective scanning band of the real-time scanning spectrum is introduced in step 13, and the model adapted to the representative particle size is selected to detect the particle size distribution of the real-time scanning spectrum. Figure 6 , step 13 may include steps 131 to 139.

[0089] Step 131, preprocessing the real-time scanning spectrum to obtain an effective scanning band.

[0090] Here, the preprocessing of the real-time scanning spectrum is the same as the preprocessing in the above step 23, that is, a denoising algorithm is used to denoise the real-time scanning spectrum to obtain a denoised spectrum; and the denoised spectrum that meets the effective conditions is signal amplified to obtain an effective scanning band.

[0091] Step 133, based on the real-time scanning spectrum, determine the representative particle size of the real-time scanning spectrum.

[0092] Step 135 , determining whether the representative particle size is greater than the large particle threshold. If yes, proceed to step 137 , if no, proceed to step 139 .

[0093] Step 137, input the effective scanning band into the large particle detection model, and use the large particle detection model to generate the particle size distribution corresponding to the effective scanning band.

[0094] Step 139, input the effective scanning band into the small particle detection model, and use the small particle detection model to generate the particle size distribution corresponding to the effective scanning band.

[0095] In step 133, there are multiple implementations to determine the representative particle size of the real-time scanning spectrum. For example, a stage-particle size table can be pre-constructed, in which the main particle sizes of the Chinese medicine granules at various stages of the granulation process are recorded, so that based on the stage when the real-time scanning spectrum is obtained, the stage-particle size table is queried to obtain the representative particle size.

[0096] Alternatively, the particle size distribution of the last real-time scanning spectrum is retrieved, and when the particle size distribution is greater than the distribution rate threshold, the large particle size is used as the representative particle size, otherwise, the small particle size is used as the representative particle size. The above method is only an example, and its implementation method is not limited.

[0097] In order to make the representative particle size more accurate, in step 133, a preset component analysis algorithm can be used to process the real-time scanning spectrum to obtain the representative particle size of the real-time scanning spectrum.

[0098] The component analysis algorithm may be, but is not limited to, a linear discriminant analysis algorithm and a principal component analysis algorithm.

[0099] Based on the same concept of the above-mentioned particle size measurement method of Chinese medicine particles, refer to Figure 7 The embodiment of the present invention also provides a method for controlling the granulation process of traditional Chinese medicine granules, comprising the above steps 11 to 15, and steps 31 to 35. Figure 1 In the Chinese medicine preparation system 10 shown, the processing device 120 can Figure 2 In the structure shown, the processor 220 executes the computer program executed in the memory 210 to implement the above steps 11 to 15, and the following steps 31 to 35.

[0100] Here, the specific contents and effects of steps 11 to 15 are as described above and will not be repeated here.

[0101] Step 31, judging whether the high-quality condition is met according to the particle size distribution. If not, proceed to step 33.

[0102] Here, when the particle size distribution is greater than the preset high-quality distribution rate threshold, it is determined that the high-quality conditions are met, which means that the granulation end point has been reached and the Chinese medicine granules in the granulation equipment 110 can be taken out of the pot, which also means that the Chinese medicine granules in the granulation equipment 110 have reached high-quality products.

[0103] Step 33, input the particle size distribution into the optimization decision model, and use the optimization decision model to generate control optimization parameters.

[0104] Step 35, adjusting the production parameters of the granulation equipment according to the control optimization parameters, and controlling the granulation equipment to continue granulation.

[0105] For example, in combination Figure 1 In the illustrated Chinese medicine preparation system 10, during the granulation process, the processing device 120 uses the particle size measurement method of the Chinese medicine particles provided above to obtain the current real-time particle size distribution of the Chinese medicine particles in the granulation device 110, and then the processing device 120 determines whether the excellent product conditions are met. If it is determined that the excellent product conditions are not met, the particle size distribution is input into the optimization decision model, and the control optimization parameters are generated using the optimization decision model. Then, the processing device 120 adjusts the production parameters of the granulation device 110 according to the control optimization parameters, and controls the granulation to continue, and at the same time, the processing device 120 continues to measure the particle size of the Chinese medicine particles in the granulation device 110 until the excellent product conditions are met.

[0106] The above-mentioned granulation process control method detects and visualizes the particle size changes of Chinese medicine granules in real time during the granulation process, so as to determine whether the quality conditions are met based on the real-time particle size distribution. When the quality conditions are not met, the production parameters of the granulation equipment are autonomously optimized and adjusted by combining the optimization decision model with the particle size prediction model to control the particle size of the Chinese medicine granules to be stable within the quality standard range, avoiding the drawbacks of relying entirely on the operator's feel and experience. In this way, Chinese medicine granules with a quality particle size range with good compressibility can be stably obtained to ensure the production quality of Chinese medicine granules.

[0107] Among them, the high-quality product distribution rate threshold used to determine whether the high-quality product conditions are met in step 31 can be a pre-set value, a value fitted from a calibration experiment, or a mature high-quality product distribution rate standard value, and the method of obtaining it is not restricted.

[0108] Please continue to refer to Figure 7 When it is determined in step 31 that the superior product condition is met, step 37 is executed.

[0109] Step 37, ending granulation.

[0110] The optimization decision model used in the above step 33 can be a generative large language model or a pre-trained mature neural network model, and the implementation method is not limited.

[0111] In order to make the results of the optimization decision model more accurate, a deep reinforcement learning algorithm and historical data iterative training are used to construct an optimized decision model. Figure 8 , Figure 1 The process of constructing the optimized decision model by the model building device 150 in the traditional Chinese medicine preparation system 10 shown includes steps 41 to 45.

[0112] Step 41, obtaining multiple groups of control sample parameters and the particle size distribution corresponding to each group of control sample parameters.

[0113] The control sample parameters and particle size distribution are data extracted from historical production data. The control sample parameters include but are not limited to: adhesive flow, spray pressure, air inlet temperature, air inlet volume, humidity, etc.

[0114] Step 43 , for each particle size distribution, the corresponding control sample parameter is used as the reference control parameter of the particle size distribution, and the particle size distribution is labeled to obtain a training sample.

[0115] For example, the corresponding relationship between the control sample parameters and the particle size distribution can be shown in Table 3.

[0116] Table 3 Test items Premium product standards Real-time particle size (μm) Control sample parameters (granulation parameter control method) Particle size Dx (50) 114~186μm 114≤Dx(50)≤186 Adjust the adhesive flow rate and gradually reduce it to zero Particle size Dx (50) 114~186μm Dx(50)<114 Increase adhesive flow and reduce air inlet temperature Particle size Dx (50) 114~186μm Dx (50) > 186 Adjust the adhesive flow to zero Step 45, based on each training sample, iteratively train the preset neural network model to obtain a mature optimization decision model.

[0117] In order to allow for flexible selection of the setting of the neural network model, for example, it can be a deep reinforcement learning algorithm or a deep feedforward neural network, and its implementation method is not restricted.

[0118] In one possible implementation, the neural network model is a deep Q-network algorithm, the Q value is learned by estimating the sum r of rewards for each state St, a deep neural network is applied to approximate the optimal action-value function, and the optimization strategy is trained and updated by the difference between the value obtained from previous experience and the discounted reward obtained from the current state.

[0119] It is expressed in mathematical formula as: .

[0120] in, is the loss value of the loss function, The network parameters updated by the artificial neural network algorithm, is the environmental state, is the model action, t is the current moment, t-1 is the previous moment, is the target time, In Strategy The expected value under is the reward discount factor, From time t-1 to time Reward.

[0121] The state of each step in the optimization decision model is the process condition and the corresponding particle size, the action is to increase or decrease a single process condition at each step, and the reward is determined according to the change in product quality. The control variable method is used to select the number of hidden layers in the network, the number of hidden layer nodes, and the number of data sets used in each training to complete the training of the granulation optimization decision model.

[0122] The input layer of the neural network model contains 5 hidden layers, and the linear correction function ReLU is used as the activation function; the output layer contains 6 neurons corresponding to 6 different actions, and the linear function Linear is used as the activation function. The hyperparameters of reinforcement learning are set as follows: the learning rate is set to 0.001, the reward discount factor is set to 0.95, the experience pool size is 500, and the number of samples per batch is set to 64.

[0123] In order to verify the accuracy of the particle size measurement method of traditional Chinese medicine particles provided by the embodiment of the present invention (hereinafter referred to as the near-infrared particle size detection method), a comparative experiment between the manual screening method and the near-infrared particle size detection method was conducted. The precision verification (repeatability verification) results of the near-infrared particle size detection are shown in Table 4, and the precision verification (repeatability verification) results of the manual screening method particle size (-100) are shown in Table 5.

[0124] Table 4

[0125] Table 5

[0126] As shown in Tables 4 and 5, for the same sample under the same scanning conditions, the relative deviation of the 10 particle size determination results by near infrared detection was less than 0.3%, and the average relative deviation was 0.15%. The relative deviation of the 6 particle size determination results by manual screening method was greater than 5%, and the average relative deviation was 6.39%. In comparison, the repeatability of particle size determination based on near infrared is excellent, while the repeatability of particle size determination by manual screening method is poor.

[0127] Based on the same inventive concept of the particle size measurement method of traditional Chinese medicine particles, the embodiment of the present invention also provides a particle size measurement device for traditional Chinese medicine particles, which can be applied to Figure 1 The processing device 120 includes a scanning module, a distribution detection module and a display module.

[0128] The scanning module is used to receive the real-time scanning spectrum of the Chinese medicine particles in the granulation equipment by the near-infrared spectrometer.

[0129] The distribution detection module is used to input the real-time scanning spectrum into a preset particle size detection model, and use the particle size detection model to generate a particle size distribution corresponding to the real-time scanning spectrum.

[0130] The display module is used to display information on the change of particle size distribution with granulation time according to each particle size distribution.

[0131] Furthermore, the particle size measurement device may further include a model building module, and the model building module is used to execute the above steps 21 to 29 to obtain a particle size detection model.

[0132] Based on the same inventive concept of the granulation process control method of traditional Chinese medicine granules, the embodiment of the present invention also provides a granulation process control device of traditional Chinese medicine granules, which can be applied to Figure 1 The processing equipment in the system includes a particle size measurement module, a judgment module, an optimization decision module and a parameter adjustment module.

[0133] The particle size measurement module is used to obtain the current particle size distribution using the particle size measurement method of traditional Chinese medicine particles.

[0134] The judgment module is used to judge whether the high-quality condition is met according to the particle size distribution.

[0135] The optimization decision module is used to input the particle size distribution into the optimization decision model if the quality conditions are not met, and use the optimization decision model to generate control optimization parameters.

[0136] The parameter adjustment module is used to adjust the production parameters of the granulation equipment according to the control optimization parameters, and control the granulation equipment to continue granulation.

[0137] Furthermore, the granulation process control method may further include a model building module, and the model building module is used to execute the above steps 41 to 45 to obtain an optimization decision model.

[0138] In this way, during the granulation process, the online near-infrared real-time detection and visualization of the particle size changes of the Chinese medicine granules are used to determine whether the granulation endpoint has been reached based on the real-time particle size and the online near-infrared detection standard. The adhesive flow, spray pressure, air inlet temperature, air inlet volume and other production parameters are adjusted accordingly to control the particle size of the Chinese medicine granules to be stable within the standard range of premium products, avoiding the drawbacks of relying entirely on the operator's feel and experience.

[0139] The specific implementation and effect of the particle size measuring device and the granulation process control device can be found in the above description of the implementation of the particle size measuring method and the granulation process control method, which will not be repeated here.

[0140] Each module of the particle size measuring device and the granulation process control device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor 220 in the electronic device 20 in the form of hardware, or can be stored in the memory 210 of the electronic device 20 in the form of software, so that the processor 220 can call and execute the operations corresponding to each module to implement the particle size measurement method or granulation process control method provided above.

[0141] The embodiment of the present invention further provides a granulation device 110, including a body, a near-infrared spectrometer 130, functional components and a processing device 120, wherein the near-infrared spectrometer 130 is disposed on a window of the body.

[0142] The near infrared spectrometer 130 is used to scan particles in the body in real time, obtain a real-time scanning spectrum, and transmit the real-time scanning spectrum to the processing device 120 .

[0143] The processing device 120 is used to implement the particle size measurement method of the traditional Chinese medicine granules as provided above, or the granulation process control method of the traditional Chinese medicine granules as provided above.

[0144] Based on the same inventive concept, an embodiment of the present invention also provides an electronic device 20, including a processor 220 and a memory 210, wherein the memory 210 stores a computer program that can be executed by the processor 220, and the processor 220 can execute the computer program to implement the particle size measurement method of the traditional Chinese medicine particles provided above, or the granulation process control method of the traditional Chinese medicine particles provided above.

[0145] In summary, the particle size measurement method, manufacturing process control method, granulation equipment and electronic equipment of traditional Chinese medicine granules provided by the embodiments of the present invention have at least the following beneficial effects: (I) It solves the problem of lack of effective methods to monitor the change of particle size in the fluidized bed granulation process of traditional Chinese medicine in real time, and realizes the online real-time detection of traditional Chinese medicine particles without polluting the traditional Chinese medicine particles and reducing waste; (ii) Technical issues related to autonomously adjusting the production parameters of the fluidized bed based on the detected real-time particle size distribution, thereby stably obtaining Chinese medicine granules with good compressibility and high-quality particle size range.

[0146] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0147] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0148] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk.

[0149] 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 measuring the particle size of traditional Chinese medicine particles, characterized in that: The method comprises: Receive the real-time scanning spectrum of the Chinese medicine granules in the granulation equipment by the near infrared spectrometer; Inputting the real-time scanning spectrum into a preset particle size detection model, and using the particle size detection model to generate a particle size distribution corresponding to the real-time scanning spectrum; According to each of the particle size distributions, information on the change of the particle size distribution with the granulation time is displayed.

2. The method for measuring the particle size of Chinese medicine particles according to claim 1, characterized in that: The particle size detection model includes a large particle detection model and a small particle detection model; The step of inputting the real-time scanning spectrum into a preset particle size detection model and using the particle size detection model to generate a particle size distribution corresponding to the real-time scanning spectrum comprises: Preprocessing the real-time scanning spectrum to obtain an effective scanning band; Based on the real-time scanning spectrum, determining a representative particle size of the real-time scanning spectrum; Determining whether the representative particle size is greater than a large particle threshold; If yes, the effective scanning band is input into the large particle detection model, and the large particle detection model is used to generate a particle size distribution corresponding to the effective scanning band; If not, the effective scanning band is input into the small particle detection model, and the small particle detection model is used to generate the particle size distribution corresponding to the effective scanning band.

3. The particle size measurement method of Chinese medicine granules according to claim 2, characterized in that: The step of determining the representative particle size of the real-time scanning spectrum based on the real-time scanning spectrum comprises: The real-time scanning spectrum is processed using a preset component analysis algorithm to obtain a representative particle size of the real-time scanning spectrum.

4. The method for measuring the particle size of Chinese medicine particles according to claim 2, characterized in that: The method for constructing the particle size detection model includes: Controlling a near infrared spectrometer to scan samples in the granulation process to obtain a plurality of sample spectra, and sampling the samples in the granulation process to obtain a plurality of sampled samples; wherein the samples include large particle samples and small particle samples, and the main particle size of the large particle sample is larger than the main particle size of the small particle sample; Preprocessing the sample spectrum, and extracting effective bands from the preprocessed sample spectrum to obtain a sample spectrum; The sampled samples are tested by laser particle size method to obtain a distribution reference value of the sample particle size corresponding to each sample spectrum; Based on the sample spectra of the large particle samples and the distribution reference values, model building and optimization are performed to obtain a large particle detection model; Based on the sample spectra of the small particle samples and the distribution reference values, a model is constructed and optimized to obtain a small particle detection model.

5. The method for measuring the particle size of Chinese medicine particles according to claim 4, characterized in that: The sample spectrum is preprocessed, and effective bands are extracted from the preprocessed sample spectrum to obtain a sample spectrum; Using a denoising algorithm to denoise the sample spectrum to obtain a denoised spectrum; The denoised spectrum that meets the effective conditions is subjected to signal amplification to obtain a sample spectrum.

6. A method for controlling the granulation process of traditional Chinese medicine granules, characterized in that: The method comprises: Using the particle size measurement method of the traditional Chinese medicine particles according to any one of claims 1 to 5 to obtain the current particle size distribution; According to the particle size distribution, judging whether the quality condition is met; If not, the particle size distribution is input into an optimization decision model, and the optimization decision model is used to generate control optimization parameters; According to the control optimization parameters, the production parameters of the granulation equipment are adjusted, and the granulation equipment is controlled to continue granulation.

7. The method for controlling the granulation process of Chinese medicine granules according to claim 6, characterized in that: After the step of determining whether the quality condition is met according to the particle size distribution, the method further comprises: If it is determined that the quality condition is met, the granulation is terminated.

8. The method for controlling the granulation process of Chinese medicine granules according to claim 6 or 7, characterized in that: The method for constructing the optimization decision model includes: Obtaining multiple groups of control sample parameters and particle size distribution corresponding to each group of control sample parameters; For each particle size distribution, the corresponding control sample parameter is used as a reference control parameter of the particle size distribution, and the particle size distribution is labeled to obtain a training sample; Based on each of the training samples, the preset neural network model is iteratively trained to obtain a mature optimization decision model.

9. A granulation device, characterized in that: It comprises a body, a near-infrared spectrometer, functional components and a processing device, wherein the near-infrared spectrometer is arranged on a window of the body; The near-infrared spectrometer is used to scan particles in the body in real time to obtain a real-time scanning spectrum, and transmit the real-time scanning spectrum to the processing device; The processing equipment is used to implement the particle size measurement method of the traditional Chinese medicine particles as described in any one of claims 1 to 5, or the granulation process control method of the traditional Chinese medicine particles as described in any one of claims 6 to 8.

10. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and the processor can execute the computer program to implement the particle size measurement method of the traditional Chinese medicine particles as described in any one of claims 1 to 5, or the granulation process control method of the traditional Chinese medicine particles as described in any one of claims 6 to 8.

Images