Method and system for coupling optimization of soaking and drying parameters in processing of traditional chinese medicine decoction pieces

Abstract

The application provides a method and system for coupling and optimizing parameters of soaking and drying in processing of traditional Chinese medicine decoction pieces. The method comprises the following steps: obtaining basic information of specific medicinal materials; constructing N tentative processing parameters flow T'; identifying one tentative processing parameters flow T' with the highest confidence degree from the unbiased parameters flow T by a data detection model M1, and obtaining the soaking time, drying time and drying temperature after coupling and optimization. The system comprises a forward subsystem and an auxiliary subsystem. In the application, the worker intelligence auxiliary technology is adopted, the data detection model M1 is introduced, and the production workers only need to input part of information, and then the soaking time, drying time and drying temperature more biased to the standard processing process can be automatically obtained for the production workers to refer to and execute, so that the production fluctuation is reduced. Meanwhile, the application is also suitable for new production workers, and the training period can be shortened.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine processing technology, and in particular to a method and system for coupling and optimizing soaking and drying parameters in the processing of prepared Chinese medicinal materials. Background Technology

[0002] In the processing of traditional Chinese medicine decoction pieces, soaking and drying are both conventional softening and drying methods, and they are usually consecutive operations. The soaking time during the soaking process, and the drying temperature and drying time during the drying process, have a significant impact on the processing of traditional Chinese medicine decoction pieces, and may even affect the final efficacy.

[0003] Although the *Pharmacopoeia of the People's Republic of China* records the standard processing procedures for various Chinese medicinal herbs, including standard soaking times, drying temperatures, and drying times, the actual production process for these herbs varies in origin, size, and moisture content. Furthermore, production workers typically lack access to or are unable to use reference books like the *Pharmacopoeia*. Therefore, in practice, the control of processing parameters, such as soaking time, drying temperature, and drying time, relies entirely on the experience of the production workers. This approach leads to the following inherent technical problems: (1) The processing of Chinese herbal medicine slices may fluctuate greatly due to the experience of the production workers and their own work status (such as fatigue, mood, etc.), which will lead to a large fluctuation in the final efficacy.

[0004] (2) Newly hired production workers may not be able to operate independently due to a lack of experience and need long-term training to accumulate experience. Summary of the Invention

[0005] This invention addresses the aforementioned problems by providing a method and system for coupled optimization of soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces. This invention employs intelligent worker assistance technology and introduces a data detection model M1, providing pharmaceutical factories and production personnel with refined parameter control and data optimization. Production personnel only need to input basic information about the medicinal materials and approximate ranges for soaking, drying, and drying temperatures. The system then automatically generates soaking, drying, and drying times that more closely resemble the standard processing procedure, allowing production personnel to refer to and execute these parameters. This significantly reduces the impact of the production personnel's experience and work status on production, minimizing fluctuations. Furthermore, this invention is also applicable to newly hired production personnel, shortening their training period.

[0006] The technical solution adopted in this invention is: A method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, the method comprising the following steps: Step S1: Confirm the Chinese medicinal herbs to be processed and the corresponding specific Chinese medicinal materials; Step S2: Obtain the basic information of the specific Chinese medicinal material, which shall include at least the type of medicinal material. Step S3: Based on the type of medicinal material, determine the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range for the specific Chinese medicinal material. Step S4: Gather the basic information of the medicinal materials, as well as the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generate N different proposed processing process parameter streams T' according to the rules, where N is an integer greater than or equal to 6; the proposed processing process parameter stream T' includes the basic information of the medicinal materials, the proposed soaking time, the proposed drying time, and the proposed drying temperature; Step S5: Import the N proposed processing process parameter streams T' into the data detection model M1 one by one to obtain the type and confidence level of each proposed processing process parameter stream T'; the type is negative deviation parameter stream -T, no deviation parameter stream T, or positive deviation parameter stream +T; Step S6: According to each type, sort the proposed processing process parameter flow T' from high to low confidence level, select the proposed processing process parameter flow T' with the highest confidence level among the proposed processing process parameter flows T' identified as unbiased parameter flows T', analyze the proposed processing process parameter flow T' to obtain the coupled and optimized soaking time, drying time and drying temperature, and determine them as the soaking and drying parameters in the processing of Chinese herbal medicine pieces.

[0007] Furthermore, in step S2, the specific process of obtaining the basic information of a specific Chinese medicinal herb includes: The basic information of specific Chinese medicinal materials is obtained by manually entering the data. Alternatively, the basic information of a specific Chinese medicinal herb can be obtained by automatically importing relevant testing data from previous testing processes and / or the control parameters to be achieved in subsequent slicing processes.

[0008] Furthermore, in step S2, the basic information of the medicinal material also includes one or more of the following: medicinal material type, medicinal material thickness, medicinal material texture, medicinal material size, medicinal material origin, medicinal material grade, medicinal material quantity, medicinal material drying degree, soaking method, soaking pressure, liquid temperature, ambient temperature, simulated slicing method, and simulated slice thickness.

[0009] Furthermore, in step S4, the specific process of collecting basic information about the medicinal materials, as well as the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generating N different proposed processing process parameter flows T' according to rules includes: Step S41: Set the first step length. Based on the upper limit, lower limit and center point value of the proposed soaking time range, expand according to the first step length to obtain X proposed soaking times, where X is an integer greater than or equal to 2. Step S42: Set the second step length. Based on the upper limit, lower limit and center point value of the proposed drying time interval, expand according to the second step length to obtain Y proposed drying times, where Y is an integer greater than or equal to 2. Step S43: Set the third step length. Based on the upper limit, lower limit and center point value of the proposed drying temperature range, expand according to the third step length to obtain Z proposed drying temperatures, where Z is an integer greater than or equal to 2. Step S44: Keeping the basic information of the medicinal materials unchanged, select the proposed soaking time in step S41, the proposed drying time in step S42, and the proposed drying temperature in step S43 to form a proposed processing parameter flow T'. Step S45: Repeat step S44, keeping the basic information of the medicinal materials unchanged, and in the same combination order, change the proposed soaking time in step S41, the proposed drying time in step S42, and the proposed drying temperature in step S43, and arrange and combine them to form N different proposed processing process parameter flows T', where N=X*Y*Z.

[0010] Furthermore, in step S5, the data detection model M1 is obtained using machine deep learning, and the specific process includes: Step S51: Retrieve the basic information of medicinal materials involved in the historical processing of Chinese herbal medicine slices, as well as the corresponding soaking time, drying time, and drying temperature, to form ordinary processing process parameter streams T1, T2, and Tp. The positions of each parameter in ordinary processing process parameter streams T1, T2, and Tp are the same as the positions of the parameters with the same name in the proposed processing process parameter stream T', where p is an integer greater than or equal to 3. Step S52: Obtain the basic information of medicinal materials involved in the standard Chinese herbal medicine processing process and the corresponding soaking time, drying time and drying temperature, and form a standard processing process parameter flow T0; Step S53: Manually compare the ordinary processing process parameter stream T1, ordinary processing process parameter stream T2, and ordinary processing process parameter stream Tp with the standard processing process parameter stream T0 one by one. If the parameters of the ordinary processing process parameter flow T1 are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow T1 of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow T1 of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. If the parameters of the ordinary processing process parameter flow T2 are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow T2 of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow T2 of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. If the parameters of the ordinary processing process parameter flow Tp are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow Tp of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow Tp of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. Step S54: The ordinary processing process parameter streams T1, T2, and Tp, which are labeled as negative deviation parameter stream -T, no deviation parameter stream T, or positive deviation parameter stream +T, are divided into training sets and test sets, and fed into a convolutional neural network to train multiple models M0. The model M0 with the highest recognition accuracy is retained as the data detection model M1.

[0011] Furthermore, if in steps S51 and S52, the ordinary processing process parameter streams T1, T2, Tp, and T0 are enlarged using the same method of padding with "0", then in step S4, the proposed processing process parameter stream T' is also enlarged using the same method of padding with "0".

[0012] Furthermore, in step S54, the convolutional neural network adopts the PyTorch architecture, consisting of three convolutional layers and 3*3 convolutional kernels, with a kernel operation stride of L, where L is an integer greater than or equal to 1.

[0013] Furthermore, the method also includes: Step S7: After processing the Chinese herbal medicine slices based on the optimized soaking time, drying time and drying temperature in step S6, obtain the basic information of the medicinal materials involved in the actual processing of the herbal medicine slices and the corresponding soaking time, drying time and drying temperature, and form the parameter flow Tp' of the ordinary processing process. Step S8: Manually compare the ordinary processing process parameter flow Tp' with the standard processing process parameter flow T0, and mark it as negative deviation parameter flow -T, no deviation parameter flow T, or positive deviation parameter flow +T to augment the training data and optimize the training of the data detection model M1.

[0014] A system for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, to implement the aforementioned method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, the system comprising: The forward subsystem includes a data acquisition module, a data analysis module, and a result output module. The data acquisition module imports basic information about the specific medicinal materials corresponding to the prepared Chinese herbal medicine slices, as well as the proposed soaking time interval, proposed drying time interval, and proposed drying temperature interval determined based on the type of medicinal material. It also collects the soaking time, drying time, and drying temperature during the actual processing. The data analysis module aggregates the basic information about the medicinal materials, the proposed soaking time interval, proposed drying time interval, and proposed drying temperature interval, generates N proposed processing process parameter streams T' according to rules, and imports each of the N proposed processing process parameter streams T' into a data detection model M1 to obtain the type and confidence level of each proposed processing process parameter stream T'. The result output module displays the type and confidence level results of each proposed processing process parameter stream T' after identification by the data detection model M1, and sorts the proposed processing process parameter streams T' according to each type, from highest to lowest confidence level. An auxiliary subsystem interacts with the forward subsystem and stores training data for training the data detection model M1.

[0015] Furthermore, the system also includes: A feedback subsystem interacts with the forward subsystem, provides feedback on the output of the result output module, and uses the feedback results to augment the training data and optimize the data detection model M1.

[0016] The beneficial effects of this invention are: To address the challenges of fluctuations in the processing of traditional Chinese medicine decoction pieces due to the varying experience and work status (such as fatigue and mood) of production personnel, and the potential inexperience of newly hired production personnel who may be unable to operate independently, this invention provides a method and system for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces. The method includes the following steps: Step S1, confirming the Chinese medicinal herbs to be processed and the corresponding specific medicinal materials; Step S2, obtaining the basic information of the specific medicinal materials, which includes at least the type of medicinal material; Step S3, determining the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range for the specific medicinal material based on the type of medicinal material; Step S4, combining the basic information of the medicinal materials, the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generating N different proposed processing process parameter flows T' according to rules; the proposed processing process parameter flows T' include the basic information of the medicinal materials, the proposed soaking time, the proposed drying time, and the proposed drying temperature. Step S5: Import N proposed processing process parameter flows T' one by one into the data detection model M1 to obtain the type and confidence level of each proposed processing process parameter flow T'; the type is negative deviation parameter flow -T, no deviation parameter flow T, or positive deviation parameter flow +T. Step S6: According to each type, sort the proposed processing process parameter flows T' from high to low confidence level, select the proposed processing process parameter flow T' with the highest confidence level among those identified as no deviation parameter flow T', analyze the proposed processing process parameter flow T' to obtain the coupled and optimized soaking time, drying time, and drying temperature, and determine them as the soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces. This system includes a forward subsystem and an auxiliary subsystem. In this invention, worker intelligent assistance technology is adopted, and the data detection model M1 is introduced to provide pharmaceutical factories and production operators with refined parameter control and data optimization. Production staff only need to input basic information about the medicinal materials and approximate ranges for soaking, drying, and drying temperatures. The system then automatically generates soaking, drying, and drying times that more closely resemble standard processing procedures for reference. This significantly reduces the impact of staff experience and individual work habits on production, minimizing fluctuations. Furthermore, this invention is also suitable for newly hired production staff, shortening their training period. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, as shown in Example 1.

[0019] Figure 2 This is a schematic diagram of the structure of the convolutional neural network in Example 1.

[0020] Figure 3 This is a flowchart of the system for coupling and optimizing the soaking and drying parameters in the processing of Chinese herbal medicine slices in Example 2. Detailed Implementation

[0021] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] The following disclosure provides many different embodiments or examples for implementing different structures of the present invention. To simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Of course, these are merely examples and are not intended to limit the present invention.

[0023] The embodiments of the invention will now be described in detail with reference to the accompanying drawings.

[0024] Example 1 A method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces is presented in the appendix. Figure 1 As shown in the figure. The method includes the following steps: Step S1: Confirm the Chinese medicinal herbs to be processed and the corresponding specific Chinese medicinal materials; Step S2: Obtain the basic information of the specific Chinese medicinal material. The basic information of the medicinal material shall include at least the type of medicinal material (such as block, spherical, strip, etc.). Step S3: Based on the type of medicinal material, determine the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range for the specific Chinese medicinal material. Step S4: Gather the basic information of the medicinal materials, as well as the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generate N different proposed processing process parameter flows T' according to the rules, where N≥6; the proposed processing process parameter flow T' includes the basic information of the medicinal materials, the proposed soaking time, the proposed drying time, and the proposed drying temperature. Step S5: Import the N proposed processing process parameter streams T' into the data detection model M1 one by one to obtain the type and confidence level of each proposed processing process parameter stream T'; the type is negative deviation parameter stream -T, no deviation parameter stream T, or positive deviation parameter stream +T; Step S6: According to each type, sort the proposed processing process parameter flow T' from high to low confidence level, select the proposed processing process parameter flow T' with the highest confidence level among the proposed processing process parameter flows T' identified as unbiased parameter flows T', analyze the proposed processing process parameter flow T' to obtain the coupled and optimized soaking time, drying time and drying temperature, and determine them as the soaking and drying parameters in the processing of Chinese herbal medicine pieces.

[0025] The beneficial effects of adopting the above technical solutions are: To address the challenges of fluctuations in the processing of traditional Chinese medicine decoction pieces due to the varying experience and work status (such as fatigue and mood) of production personnel, and the potential inexperience of newly hired production personnel who may be unable to operate independently, this invention provides a method and system for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces. The method includes the following steps: Step S1, confirming the Chinese medicinal herbs to be processed and the corresponding specific medicinal materials; Step S2, obtaining the basic information of the specific medicinal materials, which includes at least the type of medicinal material; Step S3, determining the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range for the specific medicinal material based on the type of medicinal material; Step S4, combining the basic information of the medicinal materials, the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generating N different proposed processing process parameter flows T' according to rules; the proposed processing process parameter flows T' include the basic information of the medicinal materials, the proposed soaking time, the proposed drying time, and the proposed drying temperature. Step S5: Import N proposed processing process parameter flows T' one by one into the data detection model M1 to obtain the type and confidence level of each proposed processing process parameter flow T'; the type is negative deviation parameter flow -T, no deviation parameter flow T, or positive deviation parameter flow +T. Step S6: According to each type, sort the proposed processing process parameter flows T' from high to low confidence level, select the proposed processing process parameter flow T' with the highest confidence level among the proposed processing process parameter flows T' identified as no deviation parameter flow T, analyze the proposed processing process parameter flow T' to obtain the coupled and optimized soaking time, drying time, and drying temperature, and determine them as the soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces. In this invention, the use of worker intelligent assistance technology and the introduction of the data detection model M1 can provide pharmaceutical factories and production operators with refined parameter control and data optimization. Production staff only need to input basic information about the medicinal materials and approximate ranges for soaking, drying, and drying temperatures. The system then automatically generates soaking, drying, and drying times that more closely resemble standard processing procedures for reference. This significantly reduces the impact of staff experience and individual work habits on production, minimizing fluctuations. Furthermore, this invention is also suitable for newly hired production staff, shortening their training period.

[0026] Furthermore, in step S2, the specific process of obtaining the basic information of a specific Chinese medicinal herb includes: The basic information of specific Chinese medicinal materials is obtained by manually entering the data. Alternatively, the basic information of a specific Chinese medicinal herb can be obtained by automatically importing relevant testing data from previous testing processes and / or the control parameters to be achieved in subsequent slicing processes.

[0027] The choice between manual entry and automatic import can be made and configured according to the actual needs of the pharmaceutical company.

[0028] Furthermore, in step S2, the basic information of the medicinal materials includes, but is not limited to, one or more of the following: medicinal material type (such as Notopterygium incisum, rhubarb, etc.), medicinal material thickness, medicinal material texture, medicinal material size, medicinal material origin, medicinal material grade, medicinal material quantity, medicinal material drying degree, soaking method, soaking pressure, liquid temperature, ambient temperature, simulated slicing method, and simulated slice thickness.

[0029] For the basic information of medicinal materials listed above, as well as the basic information of medicinal materials not listed, relevant technical personnel can add, delete, or adjust them according to the actual situation.

[0030] Furthermore, in step S4, the specific process of collecting basic information about the medicinal materials, as well as the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generating N different proposed processing process parameter flows T' according to rules includes: Step S41: Set the first step length. Based on the upper limit, lower limit and center point value of the proposed soaking time range, expand according to the first step length to obtain X proposed soaking times, where X is an integer greater than or equal to 2. Step S42: Set the second step length. Based on the upper limit, lower limit and center point value of the proposed drying time interval, expand according to the second step length to obtain Y proposed drying times, where Y is an integer greater than or equal to 2. Step S43: Set the third step length. Based on the upper limit, lower limit and center point value of the proposed drying temperature range, expand according to the third step length to obtain Z proposed drying temperatures, where Z is an integer greater than or equal to 2. Step S44: Keeping the basic information of the medicinal materials unchanged, select the proposed soaking time in step S41, the proposed drying time in step S42, and the proposed drying temperature in step S43 to form a proposed processing parameter flow T'. Step S45: Repeat step S44, keeping the basic information of the medicinal materials unchanged, and in the same combination order, change the proposed soaking time in step S41, the proposed drying time in step S42, and the proposed drying temperature in step S43, and arrange and combine them to form N different proposed processing process parameter flows T', where N=X*Y*Z.

[0031] For example, if the approximate soaking time range for a certain medicinal herb is set to be between 2 hours and 30 minutes and 3 hours and 30 minutes, then the lower limit is 2 hours and 30 minutes, the upper limit is 3 hours and 30 minutes, and the center value is 3 hours. Assuming the first step is five minutes long, data for 2 hours and 30 minutes, 2 hours and 35 minutes, ..., 3 hours and 25 minutes, and 3 hours and 30 minutes are generated, resulting in a total of 13 proposed soaking times. The specific data is converted and presented in hours.

[0032] Similarly, for the proposed drying temperature range of 65~75℃, following a second step of 1℃, we can obtain 11 proposed drying temperatures: 65℃, 66℃, 67℃, ..., 74℃, and 75℃. Specific data are in degrees Celsius.

[0033] Similarly, for the proposed drying time range of 30 minutes to 120 minutes, following the third step length of 10 minutes, a total of 10 proposed drying times are generated for 30 minutes, 40 minutes, ..., 110 minutes, and 120 minutes. The specific data is in minutes.

[0034] Assume the basic information of the medicinal materials is: type a, variety b, and size c. The basic information is stored in relevant txt or Excel documents according to certain principles and order. The input basic information remains constant, while the generated variable data is matched to generate multiple proposed processing parameter streams T', as follows: [a,b,c,2.50,65,30]; [a,b,c,2.58,65,30]; [a,b,c,2.67,65,30]; [a,b,c,2.75,65,30]; ······ [a,b,c,3.5,75,120]; A total of 1430 proposed processing parameter streams T' were generated.

[0035] Furthermore, in step S5, the data detection model M1 is obtained using machine deep learning, and the specific process includes: Step S51: Retrieve the basic information of medicinal materials involved in the historical processing of Chinese herbal medicine slices, as well as the corresponding soaking time, drying time, and drying temperature, to form ordinary processing process parameter streams T1, T2, and Tp. The positions of each parameter in ordinary processing process parameter streams T1, T2, and Tp are the same as the positions of the parameters with the same name in the proposed processing process parameter stream T', where p is an integer greater than or equal to 3. Step S52: Obtain the basic information of medicinal materials involved in the standard Chinese herbal medicine processing process and the corresponding soaking time, drying time and drying temperature, and form a standard processing process parameter flow T0; Step S53: Manually compare the ordinary processing process parameter stream T1, ordinary processing process parameter stream T2, and ordinary processing process parameter stream Tp with the standard processing process parameter stream T0 one by one. If the parameters of the ordinary processing process parameter flow T1 are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow T1 of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow T1 of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. If the parameters of the ordinary processing process parameter flow T2 are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow T2 of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow T2 of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. If the parameters of the ordinary processing process parameter flow Tp are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow Tp of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow Tp of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. Step S54: The ordinary processing process parameter streams T1, T2, and Tp, which are labeled as negative deviation parameter stream -T, no deviation parameter stream T, or positive deviation parameter stream +T, are divided into training sets and test sets, and fed into a convolutional neural network to train multiple models M0. The model M0 with the highest recognition accuracy is retained as the data detection model M1.

[0036] Furthermore, if in steps S51 and S52, the ordinary processing process parameter streams T1, T2, Tp, and T0 are enlarged using the same method of padding with "0", then in step S4, the proposed processing process parameter stream T' is also enlarged using the same method of padding with "0".

[0037] For example, before the "0" padding operation, the matrix formed by the normal processing parameter streams T1, T2, and Tp is as follows: .

[0038] After padding with zeros, the matrix becomes: .

[0039] Alternatively, the matrix becomes: .

[0040] Similarly, the same "0" padding operation is used for both the magnitude of the standard processing process parameter flow T0 and the proposed processing process parameter flow T'.

[0041] The specific rules for padding with zeros are set based on factors such as the amount of data processed. Once the method for padding with zeros is determined, all other data will be processed in the same way to avoid data corruption.

[0042] In this embodiment, the matrix size is increased by padding with "0", giving each value an equal chance to be convolved and improving the accuracy of the data detection model M1.

[0043] Further, in step S54, the convolutional neural network adopts a PyTorch architecture, consisting of three convolutional layers and 3*3 convolutional kernels, with a kernel stride of L, where L is an integer greater than or equal to 1. The structure of the convolutional neural network is as follows: Figure 2 As shown in the image.

[0044] Furthermore, the method also includes: Step S7: After processing the Chinese herbal medicine slices based on the optimized soaking time, drying time and drying temperature in step S6, obtain the basic information of the medicinal materials involved in the actual processing of the herbal medicine slices and the corresponding soaking time, drying time and drying temperature, and form the parameter flow Tp' of the ordinary processing process. Step S8: Manually compare the ordinary processing process parameter flow Tp' with the standard processing process parameter flow T0, and mark it as negative deviation parameter flow -T, no deviation parameter flow T, or positive deviation parameter flow +T to augment the training data and optimize the training of the data detection model M1.

[0045] In this embodiment, production operators apply the soaking time, drying time, and drying temperature obtained based on the data detection model M1 to actual production, resulting in practically applicable soaking time, drying time, and drying temperature, forming a general processing process parameter stream Tp'. Using the general processing process parameter stream Tp' as training data for the data detection model M1 can gradually improve the accuracy of the data detection model M1.

[0046] Example 2 A system for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, to implement the aforementioned method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, the system comprising: The forward subsystem includes a data acquisition module, a data analysis module, and a result output module. The data acquisition module imports basic information about the specific medicinal materials corresponding to the prepared Chinese herbal medicine slices, as well as the proposed soaking time interval, proposed drying time interval, and proposed drying temperature interval determined based on the type of medicinal material. It also collects the soaking time, drying time, and drying temperature during the actual processing. The data analysis module aggregates the basic information about the medicinal materials, the proposed soaking time interval, proposed drying time interval, and proposed drying temperature interval, generates N proposed processing process parameter streams T' according to rules, and imports each of the N proposed processing process parameter streams T' into a data detection model M1 to obtain the type and confidence level of each proposed processing process parameter stream T'. The result output module displays the type and confidence level results of each proposed processing process parameter stream T' after identification by the data detection model M1, and sorts the proposed processing process parameter streams T' according to each type, from highest to lowest confidence level. An auxiliary subsystem interacts with the forward subsystem and stores training data for training the data detection model M1.

[0047] Furthermore, the system also includes: A feedback subsystem interacts with the forward subsystem, provides feedback on the output of the result output module, and uses the feedback results to augment the training data and optimize the data detection model M1.

[0048] The system workflow in this embodiment is as follows: Figure 3As shown, firstly, the data analysis module of the forward subsystem processes the training data provided by the auxiliary subsystem to form complete processing procedure parameter data, and then trains it to obtain the data detection model M1. Next, the data acquisition module collects the basic information of the medicinal materials corresponding to the Chinese herbal medicine slices to be processed, as well as information on the proposed soaking time interval, proposed drying time interval, and proposed drying temperature interval, forming a proposed processing procedure parameter flow T'. This flow is then identified by the data detection model M1, which finds the proposed processing procedure parameter flow T' with the highest confidence among the proposed processing procedure parameter flows T' without deviation. After analysis, the coupled and optimized soaking time, drying time, and drying temperature are obtained and output. Simultaneously, the data acquisition module continues to collect and store the actual soaking time, drying time, and drying temperature parameters during the actual processing procedure, controlled based on the coupled and optimized soaking time, drying time, and drying temperature. Furthermore, the system uses the actual parameter information monitored by the data acquisition module as data for augmenting and optimizing the data detection model M1, further improving the accuracy of the data detection model M1.

[0049] In this embodiment, intelligent worker assistance technology is employed, introducing a data detection model M1 to provide pharmaceutical factories and production personnel with refined parameter control and data optimization. Production personnel only need to input basic information about the medicinal materials and approximate ranges for soaking, drying, and drying temperatures. The system then automatically generates soaking, drying, and drying times that more closely resemble standard processing procedures for reference and execution. This significantly reduces the impact of the production personnel's experience and work status on production, minimizing fluctuations. Furthermore, this invention is also suitable for newly hired production personnel, shortening their training period.

Claims

1. A method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, characterized in that, The method includes the following steps: Step S1: Confirm the Chinese medicinal herbs to be processed and the corresponding specific Chinese medicinal materials; Step S2: Obtain the basic information of the specific Chinese medicinal material, which shall include at least the type of medicinal material. Step S3: Based on the type of medicinal material, determine the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range for the specific Chinese medicinal material. Step S4: Gather the basic information of the medicinal materials, as well as the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generate N different proposed processing process parameter streams T' according to the rules, where N is an integer greater than or equal to 6; the proposed processing process parameter stream T' includes the basic information of the medicinal materials, the proposed soaking time, the proposed drying time, and the proposed drying temperature; Step S5: Import the N proposed processing process parameter streams T' into the data detection model M1 one by one to obtain the type and confidence of each proposed processing process parameter stream T'; the type is negative deviation parameter stream -T, no deviation parameter stream T, or positive deviation parameter stream +T; wherein, the data detection model M1 is obtained by machine deep learning. Step S6: According to each type, sort the proposed processing process parameter flow T' from high to low confidence level, select the proposed processing process parameter flow T' with the highest confidence level among the proposed processing process parameter flows T' identified as unbiased parameter flows T', analyze the proposed processing process parameter flow T' to obtain the coupled and optimized soaking time, drying time and drying temperature, and determine them as the soaking and drying parameters in the processing of Chinese herbal medicine pieces.

2. The method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 1, characterized in that, In step S2, the specific process of obtaining the basic information of a specific Chinese medicinal herb includes: The basic information of specific Chinese medicinal materials is obtained by manually entering the data. Alternatively, the basic information of a specific Chinese medicinal herb can be obtained by automatically importing relevant testing data from previous testing processes and / or the control parameters to be achieved in subsequent slicing processes.

3. The method for coupled optimization of soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 1 or 2, characterized in that, In step S2, the basic information of the medicinal material also includes one or more of the following: medicinal material type, medicinal material thickness, medicinal material texture, medicinal material size, medicinal material origin, medicinal material grade, medicinal material quantity, medicinal material drying degree, soaking method, soaking pressure, liquid temperature, ambient temperature, simulated slicing method, and simulated slice thickness.

4. The method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 1, characterized in that, In step S4, the specific process of collecting basic information about the medicinal materials, as well as the proposed soaking time range, the proposed drying time range, and the proposed drying temperature range, and generating N different proposed processing process parameter flows T' according to rules includes: Step S41: Set the first step length. Based on the upper limit, lower limit and center point value of the proposed soaking time range, expand according to the first step length to obtain X proposed soaking times, where X is an integer greater than or equal to 2. Step S42: Set the second step length. Based on the upper limit, lower limit and center point value of the proposed drying time interval, expand according to the second step length to obtain Y proposed drying times, where Y is an integer greater than or equal to 2. Step S43: Set the third step length. Based on the upper limit, lower limit and center point value of the proposed drying temperature range, expand according to the third step length to obtain Z proposed drying temperatures, where Z is an integer greater than or equal to 2. Step S44: Keeping the basic information of the medicinal materials unchanged, select the proposed soaking time in step S41, the proposed drying time in step S42, and the proposed drying temperature in step S43 to form a proposed processing parameter flow T'. Step S45: Repeat step S44, keeping the basic information of the medicinal materials unchanged, and in the same combination order, change the proposed soaking time in step S41, the proposed drying time in step S42, and the proposed drying temperature in step S43, and arrange and combine them to form N different proposed processing process parameter flows T', where N=X×Y×Z.

5. The method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 1, characterized in that, In step S5, the specific process of obtaining the data detection model M1 using machine deep learning includes: Step S51: Retrieve the basic information of medicinal materials involved in the historical processing of Chinese herbal medicine slices, as well as the corresponding soaking time, drying time, and drying temperature, to form ordinary processing process parameter streams T1, T2, and Tp. The positions of each parameter in ordinary processing process parameter streams T1, T2, and Tp are the same as the positions of the parameters with the same name in the proposed processing process parameter stream T', where p is an integer greater than or equal to 3. Step S52: Obtain the basic information of medicinal materials involved in the standard Chinese herbal medicine processing process and the corresponding soaking time, drying time and drying temperature, and form a standard processing process parameter flow T0; Step S53: Manually compare the ordinary processing process parameter stream T1, ordinary processing process parameter stream T2, and ordinary processing process parameter stream Tp with the standard processing process parameter stream T0 one by one. If the parameters of the ordinary processing process parameter flow T1 are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow T1 of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow T1 of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. If the parameters of the ordinary processing process parameter flow T2 are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow T2 of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow T2 of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. If the parameters of the ordinary processing process parameter flow Tp are the same as those of the standard processing process parameter flow T0, then it is marked as a parameter flow T without deviation. If the parameter flow Tp of the ordinary processing procedure has a negative deviation from the parameter flow T0 of the standard processing procedure, it is marked as the negative deviation parameter flow -T. If the parameter flow Tp of the ordinary processing procedure has a positive deviation from the parameter flow T0 of the standard processing procedure, it is marked as the positive deviation parameter flow +T. Step S54: The ordinary processing process parameter streams T1, T2, and Tp, which are labeled as negative deviation parameter stream -T, no deviation parameter stream T, or positive deviation parameter stream +T, are divided into training sets and test sets, and fed into a convolutional neural network to train multiple models M0. The model M0 with the highest recognition accuracy is retained as the data detection model M1.

6. The method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 5, characterized in that, If, in steps S51 and S52, the ordinary processing process parameter streams T1, T2, Tp, and T0 are enlarged using the same method of padding with "0", then in step S4, the proposed processing process parameter stream T' will also be enlarged using the same method of padding with "0".

7. The method for coupled optimization of soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 5 or 6, characterized in that, In step S54, the convolutional neural network adopts the PyTorch architecture, which consists of three convolutional layers and a 3×3 convolutional kernel. The convolutional kernel operation stride is L, where L is an integer greater than or equal to 1.

8. The method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 6, characterized in that, The method further includes: Step S7: After processing the Chinese herbal medicine slices based on the optimized soaking time, drying time and drying temperature in step S6, obtain the basic information of the medicinal materials involved in the actual processing of the herbal medicine slices and the corresponding soaking time, drying time and drying temperature, and form the parameter flow Tp' of the ordinary processing process. Step S8: Manually compare the ordinary processing process parameter flow Tp' with the standard processing process parameter flow T0, and mark it as negative deviation parameter flow -T, no deviation parameter flow T, or positive deviation parameter flow +T to augment the training data and optimize the training of the data detection model M1.

9. A system for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces, to implement the method for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces as described in any one of claims 1 to 8, characterized in that, The system includes: The forward subsystem includes a data acquisition module, a data analysis module, and a result output module. The data acquisition module imports basic information about the specific medicinal materials corresponding to the prepared Chinese herbal medicine slices, as well as the proposed soaking time interval, proposed drying time interval, and proposed drying temperature interval determined based on the type of medicinal material. It also collects the soaking time, drying time, and drying temperature during the actual processing. The data analysis module aggregates the basic information about the medicinal materials, the proposed soaking time interval, proposed drying time interval, and proposed drying temperature interval, generates N proposed processing process parameter streams T' according to rules, and imports each of the N proposed processing process parameter streams T' into a data detection model M1 to obtain the type and confidence level of each proposed processing process parameter stream T'. The result output module displays the type and confidence level results of each proposed processing process parameter stream T' after identification by the data detection model M1, and sorts the proposed processing process parameter streams T' according to each type, from highest to lowest confidence level. An auxiliary subsystem interacts with the forward subsystem and stores training data for training the data detection model M1.

10. The system for coupling and optimizing soaking and drying parameters in the processing of traditional Chinese medicine decoction pieces according to claim 9, characterized in that, The system also includes: A feedback subsystem interacts with the forward subsystem, provides feedback on the output of the result output module, and uses the feedback results to augment the training data and optimize the data detection model M1.

Images