CN116414095A - Data-driven optimization method for technological parameters in traditional Chinese medicine manufacturing process - Google Patents
Data-driven optimization method for technological parameters in traditional Chinese medicine manufacturing process Download PDFInfo
- Publication number
- CN116414095A CN116414095A CN202310413626.4A CN202310413626A CN116414095A CN 116414095 A CN116414095 A CN 116414095A CN 202310413626 A CN202310413626 A CN 202310413626A CN 116414095 A CN116414095 A CN 116414095A
- Authority
- CN
- China
- Prior art keywords
- chinese medicine
- traditional chinese
- parameters
- data
- quality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 106
- 239000003814 drug Substances 0.000 title claims abstract description 68
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 64
- 238000005457 optimization Methods 0.000 title claims abstract description 37
- 230000008569 process Effects 0.000 claims abstract description 74
- 239000002245 particle Substances 0.000 claims abstract description 73
- 230000006870 function Effects 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 17
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 239000000825 pharmaceutical preparation Substances 0.000 claims 1
- 229940127557 pharmaceutical product Drugs 0.000 claims 1
- 238000012216 screening Methods 0.000 abstract description 2
- 239000000047 product Substances 0.000 description 31
- 201000004569 Blindness Diseases 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000009776 industrial production Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004321 preservation Methods 0.000 description 2
- 238000003756 stirring Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000036760 body temperature Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000003292 glue Substances 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by quality surveillance of production
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32368—Quality control
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A data-driven optimization method for technological parameters in a traditional Chinese medicine manufacturing process is characterized in that a maximum Mutual Information Coefficient (MIC) between historical technological parameters in the traditional Chinese medicine production process and the quality of the product is calculated, a quality prediction model (PM-AdaBoost) is constructed at the same time, an fitness function is calculated according to the maximum mutual information coefficient and the mean square error of the quality prediction model, the key technological parameters and an optimized quality prediction model in the traditional Chinese medicine production process are obtained through initializing a particle swarm, calculating the fitness function, and updating the speed and the position of particles in a technological parameter search space for multiple iterations, and the mean square error of the optimized quality prediction model is further used as the fitness function, and the particle swarm optimization algorithm is utilized to obtain the optimized key technological parameters. The invention measures the linear and nonlinear relations among variables through the maximum information coefficient, selects the maximum information coefficient and the quality prediction mean square error as the standard for screening key process parameters and constructing a quality prediction model, and has absolute advantages in accuracy; the optimization key is realized through a particle swarm algorithm based on the quality prediction model, and the particle swarm is used as a process parameter optimization algorithm and has absolute advantages in convergence rate.
Description
Technical Field
The invention relates to a technology in the field of information processing in a traditional Chinese medicine manufacturing process, in particular to a data-driven traditional Chinese medicine manufacturing process technological parameter optimization method.
Background
The existing method for optimizing the technological parameters in the traditional Chinese medicine manufacturing process is mainly divided into an artificial experience method and a data driving method. The parameter values obtained by means of manual experience are only suitable for specific traditional Chinese medicine products, experience needs to be accumulated again when production conditions change, and a large amount of manpower and time resources are wasted. The process parameter optimization driven by data depends on the acquired production data and the technologies of machine learning, deep learning and the like, key process parameters influencing the product quality need to be screened in advance before the process parameters are optimized, a quality prediction model is established by means of the screened parameters, and finally the process parameters are optimized by adopting an optimization algorithm based on the prediction model.
The difficulties of optimizing the technological parameters of the traditional Chinese medicine product manufacturing process driven by the current data mainly comprise: 1) It is difficult to quantify the relationship between the process parameters and the quality index during the manufacturing process; 2) The production of traditional Chinese medicine products goes through a plurality of stages, and it is difficult to determine key process parameters affecting the quality of the products in each stage, and if all the process parameters are adopted, the accuracy of a quality prediction model can be affected, and the difficulty of optimizing subsequent process parameters is increased. The prior art parameter selection method for the traditional Chinese medicine manufacturing process still has a plurality of defects: 1) The method based on artificial experience is only suitable for specific products, and the artificial experience has a certain subjectivity and blindness, when new important products appear, the method faces new technology, has no experience and manual reference, only continuously accumulated data, and is difficult to quickly optimize the technological parameters. 2) The method for optimizing the process parameters based on data driving is difficult to screen out key parameters influencing the product quality from a plurality of parameters in advance, an accurate quality prediction model is established, and the process parameters are optimized. 3) It is difficult to quantify the correlation between process parameters and quality indicators during the manufacturing process. 4) When optimizing the technological parameters, the actual production current situation is not considered, and the verification of the real data is lacking.
Disclosure of Invention
Aiming at the defects that data acquired in the existing traditional Chinese medicine product manufacturing process often have data deficiency, abnormal values, multiple collinearity and the like, so that the production data are difficult to directly analyze, and the difficulty of influencing factors influencing the quality of the traditional Chinese medicine in the traditional Chinese medicine product manufacturing process is numerous, the invention provides a data-driven traditional Chinese medicine manufacturing process parameter optimization method, wherein the linear and nonlinear relations among variables are measured through the maximum information coefficient, the maximum information coefficient and the quality prediction mean square error are selected as the standards for screening key process parameters and constructing a quality prediction model, and absolute advantages are achieved in accuracy; the optimization key is realized through a particle swarm algorithm based on the quality prediction model, and the particle swarm is used as a process parameter optimization algorithm and has absolute advantages in convergence rate.
The invention is realized by the following technical scheme:
the invention relates to a data-driven traditional Chinese medicine manufacturing process technological parameter optimization method, which comprises the steps of calculating the maximum Mutual Information Coefficient (MIC) between historical technological parameters of a traditional Chinese medicine product production process and product quality, constructing a quality prediction model (PM-AdaBoost), calculating an fitness function according to the maximum mutual information coefficient and the mean square error of the quality prediction model, initializing a particle swarm, calculating the fitness function, updating the speed and the position of particles in a technological parameter search space in a repeated iteration mode, obtaining key technological parameters and an optimized quality prediction model in the traditional Chinese medicine product production process, further taking the mean square error of the optimized quality prediction model as the fitness function, and obtaining the optimized key technological parameters by utilizing a particle swarm optimization algorithm.
The historical technological parameters of the production process of the traditional Chinese medicine product comprise: technological parameter X in the production process of traditional Chinese medicine and quality Y, X of traditional Chinese medicine products 1 ,x 2 ,…,x i, …,x n E X represents technological parameters of the production process of the traditional Chinese medicine product, wherein Represents a vector formed by samples acquired by Cheng Mou parameters produced by traditional Chinese medicine, N represents the number of process parameters, N represents the number of samples, and y= (Y) (1) ,Y (2) ,…Y (N) ) T Representing the vector formed by the quality samples of the traditional Chinese medicine products.
The historical technological parameters of the production process of the traditional Chinese medicine product are preferably pretreated, and specifically comprise: missing value processing, outlier processing, and data normalization processing.
The missing value processing, namely a single filling method, refers to: in order to reduce the influence of incomplete data on the modeling of the production process, the sample data mean value of the same type of parameters is adopted for filling.
The abnormal value processing means: abnormal data is removed by adopting a 3 sigma method, and the data obtained in the manufacturing process of the traditional Chinese medicine product is unreliable due to various interference, response drift of a test instrument, equipment failure, data recording deviation, misoperation of an analyst and other reasons during acquisition, so that the quality of the data is reduced. Outliers are outliers in a data set that deviate more from other data, deviations in such data are typically not caused by normal fluctuations in the parameters. The abnormal value can generate larger interference on data modeling, so that the model accuracy is not high, the generalization capability is reduced, and the abnormal data is removed by adopting a 3 sigma method.
The data standardization process is as follows: the data standardization processing is carried out by adopting a maximum and minimum method, parameters of different dimensions such as various stirring speeds, vacuum degrees, temperatures, viscosity values, heat preservation time and the like are involved in the production process of the traditional Chinese medicine, the level difference among different indexes is large, the dimensions are different, and the data standardization processing is carried out by adopting the maximum and minimum method in order to reduce the deviation of the analysis results caused by the difference among the variables and improve the convergence speed and the training precision in the prediction model training.
The maximum mutual information coefficient is used for measuring the production of traditional Chinese medicineThe degree of correlation between the process parameters and the product quality in the process is obtained by: process parameter x i And the product quality Y is scattered in a two-dimensional space and expressed by using a scatter diagram, dividing the current two-dimensional space into a certain number of intervals in the X and Y directions respectively, and then checking the condition that the current scatter falls into each square, namely MIC=MAX XY<B {I(x i ,Y)/(log 2 (min(x i Y)), wherein: i (x) i Y) represents a process parameter x i And product quality Y, wherein I (x i ,Y)=-∑∑(P(x i ,Y)log(P(x i ,Y)/(P(x i ) P (Y))), P (Y) being mass y=y j Probability density function, P (x) i ) Is a technological parameterProbability density function of (a).
The calculation fitness function according to the maximum mutual information coefficient and the mean square error of the quality prediction model refers to: wherein: f (f) i Is the ith technological parameter, Y is the quality index parameter, n s For the number of co-selected parameters, MSE is the mean square error between the predicted value of the quality prediction model and the true value of the quality.
The initialization of the particle swarm is as follows: when solving the problems of key process parameter selection and quality prediction model construction by using a particle swarm optimization algorithm, firstly, setting the number m of particles in a process parameter search space of the particles, and randomly assigning values for the position and the speed of each particle in the search space.
The speed and the position of the updated particles in the process parameter search space are specifically as follows: wherein: />The velocity v of the particle swarm is taken as the weight i =(v i1 ,v i2 ,…,v im ) T ,i=1,2,…,m,/>Representing the j-dimensional position vector of particle i in the t-th iteration, the position z of the particle in space i =(z i1 ,z i2 ,…,z im ) T I=1, 2, …, m, the optimal position p for particle search i =(p i1 ,p i2 ,…,p im ) T I=1, 2, …, m, the optimum position p searched by the population g =(p g1 ,p g2 ,…,p gm ) T ,i=1,2,…,m,r 1 r 2 Is a random number from 0 to 1, c 1 c 2 The individual learning factors and the group learning factors are respectively particles; and after updating, calculating the fitness function fitness again, stopping iteration when the iteration times are reached, and outputting the current technological parameter set of the traditional Chinese medicine production process and the optimized quality prediction model PM-AdaBoost.
The method for obtaining the optimized key technological parameters by using the particle swarm optimization algorithm comprises the following steps: initializing particle swarm population by taking mean square error of a traditional Chinese medicine quality true value and a predicted value of an optimized quality prediction model as fitness function, namely, setting the number of particles in a critical process parameter search space of the particles, randomly assigning a value for the position and the speed of each particle in the search space, iterating the particles for a plurality of times, namely, updating the speed and the position of the particle in the process parameter search space each time, wherein: />The velocity v 'of the particle swarm is taken as the weight' i =(v′ i1 ,v′ i2 ,…,v′ im ) T ,i=1,2,…,m,/>Representing the position vector of particle i in the j-dimension of the t-th iteration, the position z 'of the particle in space' i =(z′ i1 ,z′ i2 ,…,z′ im ) T I=1, 2, …, m, representing the position of the particle in space, the optimal position p 'for the particle search' i =(p′ i1 ,p′ i2 ,…,p′ im ) T I=1, 2, …, m ', the optimum position p ' searched by the population ' g =(p′ g1 ,p′ g2 ,…,p′ gm ) T ,i=1,2,…,m′,r′ 1 r′ 2 Is a random number from 0 to 1, c' 1 c′ 2 The individual learning factors and the group learning factors are respectively particles; and after updating, calculating the fitness function again, stopping iteration when the iteration times are reached, and outputting the process parameter optimization value of the current traditional Chinese medicine production process, namely optimizing the key process parameters.
The invention relates to a system for realizing the method, which comprises the following steps: the system comprises a data acquisition and preprocessing module, a key variable selection and quality prediction model construction module and a traditional Chinese medicine product key process parameter optimization module, wherein: the data acquisition and preprocessing module acquires historical data obtained by an industrial field instrument, and obtains a historical data set through missing value processing, abnormal value detection processing and data standardization; the key variable selection and quality prediction module selects a variable set closely related to the quality variable according to the historical data set and constructs a quality prediction model, so that redundant information is removed, quality prediction modeling difficulty and model complexity are reduced, and nonlinear function relations between key parameters and quality are learned to realize accurate quality prediction; and the key process parameter optimization module for producing the traditional Chinese medicine products optimizes the key process parameters by taking the mean square error of quality prediction as an fitness function according to the quality prediction model so as to realize the optimal process parameter combination in the production process of the traditional Chinese medicine products and ensure the production of the traditional Chinese medicine products with high quality.
Technical effects
The invention integrates the key parameter selection and the quality prediction model construction of the traditional Chinese medicine product production through the key variable selection and the quality prediction model construction module. Compared with the prior art, the invention realizes the key parameter selection in the data-driven traditional Chinese medicine industrial production, avoids the blindness of subjective selection of workers, pre-screens key parameters influencing the product quality from a plurality of parameters, establishes an accurate quality prediction model, optimizes the process parameters, obviously improves the accuracy of the prediction model, and ensures that the particle swarm optimization of the key process parameters avoids the subjectivity and blindness of manual experience setting.
Drawings
FIG. 1 is a flow chart of PM-AdaBoost model construction in the present invention;
FIG. 2 is a schematic flow chart of an AdaBoost algorithm;
FIG. 3 is a graph showing experimental comparisons of example quality predictions;
FIG. 4 is an example process parameter optimization iteration curve.
Detailed Description
As shown in fig. 1, the method for optimizing process parameters in the manufacturing process of the data-driven traditional Chinese medicine according to the present embodiment includes:
step A: in this embodiment, 13 traditional Chinese medicine manufacturing process parameters including a feeding amount, a wall-included rotating speed, a homogenizing rotating speed, a vacuum degree 1, a vacuum degree 2, a stirring rotating speed, a heating temperature, a viscosity value, a heat preservation temperature, a standing time, a spray body temperature, a left and right glue storage box temperature, a rolling die rotating speed and the like and 1 traditional Chinese medicine product quality are collected, and in order to be suitable for subsequent calculation and analysis, the collected industrial data set is subjected to missing value supplement, abnormal value removal and standardization treatment.
And (B) step (B): when the initializing particle swarm population utilizes a particle swarm optimization algorithm to solve the problem of key process parameter selection and quality prediction model construction, firstly, determining a process parameter search space, setting the number of particles, and initializing the position and speed of the particles in the search space.
Step C: and (3) updating the speed and the position of the particles in the process parameter search space each time, calculating the fitness function fitness again after updating, judging whether the iteration times are reached, stopping the iteration if the iteration times are reached, otherwise continuing the iteration, and outputting the process parameter set and the quality prediction model PM-AdaBoost in the current traditional Chinese medicine production process if the iteration times are ended.
And D, optimizing the process parameters by using a particle swarm optimization algorithm based on the key process parameters and a quality prediction model and taking the mean square error of the quality prediction as an fitness function, determining a process parameter search space, setting the number of particles, and initializing the initial position and the speed of the particle swarm.
Step E: and updating the speed and the position of the particles in the process parameter search space each time, calculating the fitness function MSE again after updating, judging whether the iteration times are reached, stopping the iteration if the iteration times are reached, otherwise continuing the iteration, and outputting the key process parameter optimization result in the traditional Chinese medicine production process if the iteration times are ended.
As shown in fig. 3, the mse= 0.01738 calculated by the model of this embodiment is reduced by 0.14412 compared to mse= 0.1615 without process parameter selection.
Based on the key process parameters and the quality prediction model, the process parameters are optimized by using a particle swarm optimization algorithm by taking the mean square error of the quality prediction as an fitness function, and an optimization result of the key process parameters is output.
The optimization results of the final optimization of each key process parameter are shown in fig. 4 and table 1 after iterative optimization of particle swarm.
TABLE 1
Through specific practical experiments, running in Python 3.7, the computer was configured as Intel (R) Core (TM) i7-8700CPU@3.20GHz 32.00G RAM. Based on 13 process parameters and quality samples of the Chinese medicine products provided by Beijing certain Chinese medicine manufacturer, the data are shown in Table 2.
TABLE 2
Compared with the prior art, the invention integrally realizes the key parameter selection in the data-driven traditional Chinese medicine industrial production, takes the maximum mutual information coefficient between the technological parameters and the quality variables in the traditional Chinese medicine manufacturing process and the mean square error of the quality prediction model as selection criteria by means of information science, machine learning and intelligent optimization algorithm, avoids blindness of subjective selection of workers, selects key parameters which have important influence on the quality prediction from a plurality of parameters, constructs a quality prediction model with higher prediction precision, intelligently optimizes the technological parameters in the traditional Chinese medicine manufacturing process based on the constructed quality prediction model, obtains the optimal set value of the key technological parameters, and improves the rationality and reliability of the decision in the traditional Chinese medicine product production process.
The foregoing embodiments may be partially modified in numerous ways by those skilled in the art without departing from the principles and spirit of the invention, the scope of which is defined in the claims and not by the foregoing embodiments, and all such implementations are within the scope of the invention.
Claims (9)
1. A data-driven traditional Chinese medicine manufacturing process technological parameter optimization method is characterized in that a quality prediction model is constructed through calculating the maximum mutual information coefficient between historical technological parameters of a traditional Chinese medicine product production process and product quality, an fitness function is calculated according to the maximum mutual information coefficient and the mean square error of the quality prediction model, key technological parameters and an optimized quality prediction model in the traditional Chinese medicine product production process are obtained through initializing a particle swarm, calculating the fitness function, and updating the speed and the position of particles in a technological parameter search space for multiple iterations, and the mean square error of the optimized quality prediction model is further used as the fitness function, and the particle swarm optimization algorithm is utilized to obtain optimized key technological parameters;
the traditional Chinese medicine product production process history technological parameters are preprocessed, and specifically comprise: missing value processing, outlier processing, and data normalization processing.
2. The method for optimizing process parameters of a data-driven pharmaceutical manufacturing process according to claim 1, wherein the historical process parameters of the pharmaceutical product production process comprise: technological parameter X in the production process of traditional Chinese medicine and quality Y, X of traditional Chinese medicine products 1 ,x 2, …,x i ,…,x n E X represents technological parameters of the production process of the traditional Chinese medicine product, whereinRepresents a vector formed by samples acquired by Cheng Mou parameters produced by traditional Chinese medicine, N represents the number of process parameters, N represents the number of samples, and y= (Y) (1) ,Y (2) ,…Y (N) ) T Representing the vector formed by the quality samples of the traditional Chinese medicine products.
3. The method for optimizing process parameters in a data-driven pharmaceutical manufacturing process according to claim 1, wherein the missing value processing, i.e., the single filling method, means: in order to reduce the influence of incomplete data on the modeling of the production process, filling by adopting a sample data mean value of the same type of parameters;
the abnormal value processing means: removing abnormal data by adopting a 3 sigma method;
the data standardization process is as follows: and (5) carrying out data standardization processing by adopting a maximum and minimum method.
4. The method for optimizing technological parameters in the manufacturing process of the traditional Chinese medicine driven by data according to claim 1, wherein the maximum mutual information coefficient is used for measuring the degree of correlation between technological parameters and product quality in the production process of the traditional Chinese medicine, and is obtained by the following steps: process parameter x i And the product quality Y is scattered in a two-dimensional space and expressed by using a scatter diagram, dividing the current two-dimensional space into a certain number of intervals in the X and Y directions respectively, and then checking the condition that the current scatter falls into each square, namely MIC=MAX XY<B {I(x i ,Y)/(log 2 (min(x i Y)), wherein: i (x) i Y) represents a process parameter x i And product quality Y, wherein I (x i ,Y)=-∑∑(P(x i ,Y)log(P(x i ,Y)/(P(x i ) P (Y))), P (Y) being mass y=y j Probability density function, P (x) i ) Is a technological parameterProbability density function of (a).
5. The method for optimizing process parameters in a data-driven pharmaceutical manufacturing process according to claim 1, wherein the calculating the fitness function according to the maximum mutual information coefficient and the mean square error of the quality prediction model means: wherein: f (f) i Is the ith technological parameter, Y is the quality index parameter, n s For the number of co-selected parameters, MSE is the mean square error between the predicted value of the quality prediction model and the true value of the quality。
6. The method for optimizing technological parameters in a data-driven traditional Chinese medicine manufacturing process according to claim 1, wherein the initialization of the particle swarm is as follows: when solving the problems of key process parameter selection and quality prediction model construction by using a particle swarm optimization algorithm, firstly, setting the number m of particles in a process parameter search space of the particles, and randomly assigning values for the position and the speed of each particle in the search space.
7. The method for optimizing process parameters in a data-driven pharmaceutical manufacturing process according to claim 1, wherein the speed and position of the updated particles in the process parameter search space are specifically: wherein: w is the weight, the velocity v of the particle swarm i =(v i1 ,v i2 ,…,v im ) T ,i=1,2,…,m,/>Representing the j-dimensional position vector of particle i in the t-th iteration, the position z of the particle in space i =(z i1 ,z i2 ,…,z im ) T I=1, 2, …, m, the optimal position p for particle search i =(p i1 ,p i2 ,…,p im ) T I=1, 2, …, m, the optimum position p searched by the population g =(p g1 ,p g2 ,…,p gm ) T ,i=1,2,…,m,r 1 r 2 Is a random number from 0 to 1, c 1 c 2 The individual learning factors and the group learning factors are respectively particles; after updating, calculating fitness function fitness again, stopping iteration when the iteration number is reached, and outputting the current traditional Chinese medicineA production process parameter set and an optimized quality prediction model PM-AdaBoost.
8. The method for optimizing process parameters in a data-driven Chinese medicine manufacturing process according to claim 1, wherein the obtaining the optimized key process parameters by using a particle swarm optimization algorithm is as follows: initializing particle swarm population by taking mean square error of a traditional Chinese medicine quality true value and a predicted value of an optimized quality prediction model as fitness function, namely, setting the number of particles in a critical process parameter search space of the particles, randomly assigning a value for the position and the speed of each particle in the search space, iterating the particles for a plurality of times, namely, updating the speed and the position of the particle in the process parameter search space each time, wherein: w ' is the weight, the velocity v ' of the particle swarm ' i =(v′ i1 ,v′ i2 ,…,v′ im ) T ,i=1,2,…,m,/>Representing the position vector of particle i in the j-dimension of the t-th iteration, the position z 'of the particle in space' i =(z′ i1 ,z′ i2 ,…,z′ im ) T I=1, 2, …, m, representing the position of the particle in space, the optimal position p 'for the particle search' i =(p′ i1 ,p′ i2 ,…,p′ im ) T I=1, 2, …, m ', the optimum position p ' searched by the population ' g =(p′ g1 ,p′ g2 ,…,p′ gm ) T ,i=1,2,…,m′,r′ 1 r′ 2 Is a random number from 0 to 1, c' 1 c′ 2 The individual learning factors and the group learning factors are respectively particles; after updating, the fitness function is calculated again, when the fitness function reachesAnd stopping iteration when the iteration times are counted, and outputting the process parameter optimization value of the current traditional Chinese medicine production process, namely optimizing the key process parameter.
9. A system for implementing the data-driven method for optimizing process parameters of a manufacturing process of a traditional Chinese medicine according to any one of claims 1 to 8, comprising: the system comprises a data acquisition and preprocessing module, a key variable selection and quality prediction model construction module and a traditional Chinese medicine product key process parameter optimization module, wherein: the data acquisition and preprocessing module acquires historical data obtained by an industrial field instrument, and obtains a historical data set through missing value processing, abnormal value detection processing and data standardization; the key variable selection and quality prediction module selects a variable set closely related to the quality variable according to the historical data set and constructs a quality prediction model, so that redundant information is removed, quality prediction modeling difficulty and model complexity are reduced, and nonlinear function relations between key parameters and quality are learned to realize accurate quality prediction; and the key process parameter optimization module for producing the traditional Chinese medicine products optimizes the key process parameters by taking the mean square error of quality prediction as an fitness function according to the quality prediction model so as to realize the optimal process parameter combination in the production process of the traditional Chinese medicine products and ensure the production of the traditional Chinese medicine products with high quality.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310413626.4A CN116414095A (en) | 2023-04-18 | 2023-04-18 | Data-driven optimization method for technological parameters in traditional Chinese medicine manufacturing process |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310413626.4A CN116414095A (en) | 2023-04-18 | 2023-04-18 | Data-driven optimization method for technological parameters in traditional Chinese medicine manufacturing process |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116414095A true CN116414095A (en) | 2023-07-11 |
Family
ID=87054456
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310413626.4A Pending CN116414095A (en) | 2023-04-18 | 2023-04-18 | Data-driven optimization method for technological parameters in traditional Chinese medicine manufacturing process |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116414095A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117631627A (en) * | 2023-12-01 | 2024-03-01 | 邵东智能制造技术研究院有限公司 | Digital transformation method based on industrial Internet |
-
2023
- 2023-04-18 CN CN202310413626.4A patent/CN116414095A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117631627A (en) * | 2023-12-01 | 2024-03-01 | 邵东智能制造技术研究院有限公司 | Digital transformation method based on industrial Internet |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110245802B (en) | Cigarette empty-head rate prediction method and system based on improved gradient lifting decision tree | |
CN109902861B (en) | Order production progress real-time prediction method based on double-layer transfer learning | |
CN113454552A (en) | Sensor metering data integration | |
JP7102941B2 (en) | Information processing methods, information processing devices, and programs | |
CN110757510B (en) | Method and system for predicting remaining life of robot | |
CN116011894B (en) | Aluminum alloy rod production data management system | |
CN116414095A (en) | Data-driven optimization method for technological parameters in traditional Chinese medicine manufacturing process | |
CN115630839A (en) | Production intelligent feedback regulation and control system based on data mining | |
CN113673679A (en) | Cut tobacco drying process parameter selection method based on particle swarm optimization neural network | |
CN117152119A (en) | Profile flaw visual detection method based on image processing | |
CN115795343A (en) | Method for analyzing state of unbalanced sample ammeter production equipment based on digital twin | |
EP4302248A1 (en) | System and method of monitoring an industrial environment | |
CN108537249B (en) | Industrial process data clustering method for density peak clustering | |
CN111428329A (en) | Model-based machine learning system | |
CN113177578A (en) | Agricultural product quality classification method based on LSTM | |
JP2023518055A (en) | Systems, methods and computer program products for optimizing manufacturing processes | |
CN117113233A (en) | Hierarchical energy structure scene construction method and energy consumption abnormal link tracing method | |
US20230009725A1 (en) | Use of genetic algorithms to determine a model to identity sample properties based on raman spectra | |
CN114206599A (en) | Machine learning for joint improvement | |
CN113377630B (en) | Universal KPI anomaly detection framework implementation method | |
CN115274004A (en) | Knowledge reuse-based fermentation process thallus concentration prediction method and system | |
CN111062118B (en) | Multilayer soft measurement modeling system and method based on neural network prediction layering | |
CN115115213A (en) | Method and device for retrospective analysis of quality characteristics of steel products | |
CN112508320B (en) | Automatic process stage division workflow for batch production | |
Zhang et al. | Research on full-process product defect traceability analysis technology based on workshop big data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |