CN116187507A - Traditional chinese medicine production system of adjusting based on artificial intelligence - Google Patents
Traditional chinese medicine production system of adjusting based on artificial intelligence Download PDFInfo
- Publication number
- CN116187507A CN116187507A CN202211566056.4A CN202211566056A CN116187507A CN 116187507 A CN116187507 A CN 116187507A CN 202211566056 A CN202211566056 A CN 202211566056A CN 116187507 A CN116187507 A CN 116187507A
- Authority
- CN
- China
- Prior art keywords
- data
- chinese medicine
- traditional chinese
- production
- 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
- 239000003814 drug Substances 0.000 title claims abstract description 167
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 151
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 38
- 238000005457 optimization Methods 0.000 claims abstract description 26
- 238000005516 engineering process Methods 0.000 claims abstract description 7
- 238000007405 data analysis Methods 0.000 claims abstract description 6
- 238000011958 production data acquisition Methods 0.000 claims abstract description 6
- 230000003993 interaction Effects 0.000 claims abstract description 5
- 238000000034 method Methods 0.000 claims description 54
- 239000000047 product Substances 0.000 claims description 45
- 238000013531 bayesian neural network Methods 0.000 claims description 35
- 230000008569 process Effects 0.000 claims description 29
- 239000000463 material Substances 0.000 claims description 27
- 229940079593 drug Drugs 0.000 claims description 25
- 238000007726 management method Methods 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 11
- 239000013067 intermediate product Substances 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 10
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 230000010354 integration Effects 0.000 claims description 9
- 230000007246 mechanism Effects 0.000 claims description 9
- 230000008439 repair process Effects 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 5
- 230000002068 genetic effect Effects 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 2
- 230000006978 adaptation Effects 0.000 claims 1
- 230000036541 health Effects 0.000 abstract description 3
- 238000013528 artificial neural network Methods 0.000 description 10
- 238000003860 storage Methods 0.000 description 10
- 239000004973 liquid crystal related substance Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000005856 abnormality Effects 0.000 description 6
- 238000007621 cluster analysis Methods 0.000 description 6
- 238000013507 mapping Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 4
- 238000007635 classification algorithm Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 239000012467 final product Substances 0.000 description 3
- 230000035772 mutation Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- 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
Abstract
A traditional Chinese medicine production optimization system based on artificial intelligence comprises a traditional Chinese medicine production data acquisition and database module, a traditional Chinese medicine production data analysis module and an intelligent feedback optimization module. Aiming at the industrial problems of low control level, unstable quality, difficult timely feedback of production process information data and the like of the traditional Chinese medicine production process, the cloud platform database, the heuristic multi-objective optimization algorithm and the user feedback system are utilized to combine, so that full-period dynamic scheduling and optimization of the traditional Chinese medicine production based on production data, artificial intelligence technology and man-machine interaction are realized, the quality of the produced product is ensured, and powerful guarantee is provided for the health of people.
Description
Technical Field
The invention relates to the field of medicine production, in particular to a traditional Chinese medicine production optimization system based on artificial intelligence.
Background
The traditional Chinese medicine industry is taken as an important component of the medical industry in China, and is one of the most important national industries in China. The traditional Chinese medicine production process is complicated, and the quality of the final product can be influenced by each production operation, so that the fine control of each production process is required to ensure the quality of the final product. The technological parameters of the traditional Chinese medicine production process play a vital role in the safety, effectiveness, stability, uniformity and the like of the final product. The traditional Chinese medicine production process is generally based on standard production operation rules, manual experience is still relied on in actual operation, and technological parameters are used as unique control points, so that the quality difference of products among batches is large.
Disclosure of Invention
The invention aims to provide a traditional Chinese medicine production optimization system based on artificial intelligence so as to solve the problems in the background technology.
In order to achieve the above purpose, the traditional Chinese medicine production optimization system based on artificial intelligence is provided, which comprises a traditional Chinese medicine production data acquisition and database module, a traditional Chinese medicine production data analysis module and an intelligent feedback optimization module;
s1, taking a traditional Chinese medicine production as an application target, collecting production data of the whole period of traditional Chinese medicine production based on a traditional Chinese medicine product formula and a production process, uniformly storing the production data into a traditional Chinese medicine production management system database, and establishing a correlation model of various production data and traditional Chinese medicine product quality;
s2, constructing a multi-element heuristic artificial intelligent algorithm based on parameters and calculation modes in a correlation model of various production data and traditional Chinese medicine product quality, taking information in a system database as a training set of an algorithm model, and increasing data quantity to avoid the occurrence of a fitting phenomenon;
s3, deploying a trained algorithm on a traditional Chinese medicine production management platform, taking material information data and real-time production data as input, and outputting quality prediction values of intermediate products and finished products obtained in a unit workshop section;
s4, combining an internal control quality standard of a target traditional Chinese medicine product as a basis, and if the predicted quality is not in a specified range, carrying out reverse dynamic adjustment on production parameters until the quality of intermediate products and finished products in a unit working section meets the requirement;
s5, inputting the traditional Chinese medicine production information data meeting the production requirements into a system database in batches, and taking the traditional Chinese medicine production information data as a data basis provided for the intelligent feedback tuning module;
s6, feeding the expert opinion in the expert optimizing system back to the traditional Chinese medicine product production management platform and the data display platform, updating the data display in real time, providing decision support for a manager, and constructing an intelligent traditional Chinese medicine product production mechanism and system.
Further, the step S1 of collecting production data of the whole period of the production of the traditional Chinese medicine product mainly comprises the following steps: information data of required materials, parameter data of production process and quality parameter data of medicines obtained in unit working section.
Furthermore, the database of the traditional Chinese medicine production management system in S1 is mainly based on a relational database, and particularly refers to a MySQL database constructed by Python, java and C language programming applications under a computer platform based on Linux, windows and IOS operating systems.
Further, the step of establishing a correlation model between various production data and the quality of the traditional Chinese medicine product in the step S1 is to establish functional relations between the parameter data of the production process and the quality parameter data of the medicine obtained in the unit section on the basis of claim 2, and obtain a comprehensive function affecting the quality of the traditional Chinese medicine product on the basis of a multi-data fusion mode, wherein the detailed process is as follows:
(1) And (5) traversing parallel time sequences. After data is input, multi-thread verification is carried out on a plurality of batches of data according to time sequence according to the data category and the relativity among the data, so that the problems of data missing and data abnormality can be found, and the verification efficiency can be improved.
(2) Index range. For the information data of most of required materials, the parameter data of the production process and the quality parameter data of the medicines obtained in the unit working section, a certain reasonable interval exists in the numerical range, and the normal value range is as follows:
X i，min ≤X i，t ≤X i，max
wherein X is i,t The data value of the data index i (information data of required materials, parameter data of production process technology, quality parameter data of medicines obtained in unit working section, and the like) at the time t; x is X i,min And X is i,max The lower limit and the upper limit of the index i are respectively, and abnormal data are obtained when the index i is not in the range.
(3) The Laida criterion. The information data of most materials, the parameter data of the production process and the quality parameter data of the medicines obtained in the unit working section approximately obey the statistical characteristics of normal distribution, the time sequence data can be checked by the Laida criterion, namely the data is unsuitable to be different from the time sequence mean value by more than 3 times of standard deviation, and the normal value range is
Wherein, the liquid crystal display device comprises a liquid crystal display device,
the time sequence data mean value of the index i; sigma (sigma) i The standard deviation of the time series data of the index i. And if the data exceeds the data, the data is regarded as abnormal data.
(4) And (5) logic association. The data of partial different time sequences have certain logic relevance, can be converted by a related formula and can be summarized as
F(X i,t ,X 2,t ,…,X n,t ,δ)＝0
Wherein F is various physical association relations which are required to be met among various data, delta is an error threshold, and the error is an allowable error of logic association calculation in an analysis and judgment flow, and proper numerical values are used according to the data association calculation of different types; and taking 0 in the missing filling and abnormality repairing processes, and predicting missing or abnormal data through known data. The analysis and judgment of the data and the deletion supplement and the abnormality repair of partial data can be carried out through the logic association.
(5) Sequence mean. And the single data values which are missing or deviate from a reasonable interval can be repaired through a sequence mean value so as to increase the integrity and accuracy of the data.
(6) Adjacent repair. For a single data value which is not easy to cause severe change in a short time and is in a reasonable interval, when the data is in a defect or abnormal state, the adjacent compliance data is taken as repair.
(7) And (5) predicting the medicine quality. And carrying out a parameter-optimized medicine quality prediction model on support vector machine regression (SVR) based on an Adaptive Genetic Algorithm (AGA), and filling the missing data set through the acquired information data of materials, the parameter data of a production process and the quality parameters of medicines obtained in a unit section.
The invention uses the fitness function as the fitness index of the medicine quality in the algorithm under the given condition, and can optimize the path from multiple aspects according to the multi-objective fitness function. The information data of the required materials, the parameter data of the production process and the quality parameter data of the medicines obtained in the unit working section are taken as the basis. The fitness function is expressed as:
wherein D (k) is information data of required materials, S (k) is parameter data of production process, and E (k) is quality parameter number of medicines obtained in unit sectionAccording to the above. ρ 1 、ρ 2 、ρ 3 The coefficients corresponding to the three indexes are respectively.
The adaptive cross mutation operation is the main operation of adaptive adjustment in the algorithm, and needs to maintain the diversity of the algorithm population, so that the crossed population is better than the crossed population as much as possible, and simultaneously, the better population is ensured to be obtained and the population convergence speed is increased. Because the full-adaptive cross variation probability has smaller influence on the population with medium fitness and increases the calculation complexity of the algorithm, the method balances the influence of the cross variation probability adaptability and the calculation complexity through the segmentation threshold thresh. The invention adopts a self-adaptive mode which is self-adaptively changed according to the individual fitness, is not influenced by the iteration times, carries out the same self-adaptive operation every iteration, and changes the crossing rate according to the individual fitness value; the first section is used for adaptively improving the cross variation probability of individuals with poor fitness, the middle section is the group with the largest number, and the fixed cross variation probability is adopted. The last section aims at the high-fitness individuals, and the probability of loss of the high-fitness individuals is reduced by reducing the cross variation probability of the high-fitness individuals.
Wherein, fit max Fit, the maximum fitness of the current population min For the minimum fitness of the current population, fit (i) is the fitness of the individual i, and thresh is a probability segmentation threshold;
SVR is based on the concept of statistical regression, and the prediction capability is stronger than BP neural network, RBF neural network and linear regression under the condition of less sample data; the AGA has faster searching capability and the capability of jumping out a local optimal solution by changing a genetic operator and a mutation operator along with the updating of iteration algebra, so that the problem of SVR model parameter selection can be solved.
Further, the multivariate heuristic artificial intelligence algorithm in S2 is mainly a bayesian neural network algorithm model, and can process the target prediction problem of the multi-factor driven relationship, and the detailed process is as follows:
aiming at the problems that the model complexity cannot be controlled and the data fitting is excessive in the actual use process of a standard neural network model, so that the generalization capability of the standard neural network in the actual use process is poor; under the condition of determining the input vector, the output can be determined with different output probabilities by the probability layer unit. A bayesian neural network may be defined as an integrated network comprising a plurality of sub-networks, wherein a certain correlation exists between different sub-networks, all sub-networks may be optimized simultaneously by different training procedures; when the Bayesian neural network is used for carrying out the association model simulation of the medicine quality, a plurality of forward propagation processes are used for ensuring that simulation results are obtained from different sub-networks, so that the Bayesian neural network can obtain a remarkable regularization effect, and the problem of excessive fitting possibly generated in the standard neural network can be prevented. The Bayesian neural network is obtained according to the BP neural network principle and Bayesian rules, and the Bayesian neural network training process needs to pass through 3 reasoning links:
(1) H and X are used for respectively representing Bayesian neural network model and medicine quality historical sample data, alpha and beta are used for respectively representing Bayesian neural network hyper-parameters, and posterior probability of the weight w is calculated, so that the weight w meeting the upper limit of the posterior probability is obtained up ；
(2) Combining with medicine quality historical sample data X, determining the posterior probability of the superparameter, updating alpha and beta, and corresponding expected E D And E is W ；
(3) Determining a model with the maximum posterior probability by comparing the saliency of different models, thereby obtaining an optimal Bayesian neural network;
based on the reasoning, the prior probability z is set M Under the condition that (alpha, beta) and likelihood both meet the exponential distribution, the posterior distribution of the Bayesian neural network weight taking the posterior probability deviation M (w) into consideration is obtained as follows:
can be obtained through the Bayesian reasoning process
And->
The upper limit of (2) can update the super-parameters and determine the optimal Bayesian neural network model;
to obtain the optimal weight w up The upper limit value of the delay probability, namely the lower limit value of the total error, needs to be obtained. The present invention uses the monte carlo numerical integration method of updating weights. In order to update the super-parameters, the Bayesian neural network needs to continuously iterate and train the drug quality historical sample data in the training process, so as to optimize the weight and determine the optimal Bayesian neural network model. In the training process, the complexity of the Bayesian neural network model structure is reduced by updating the super parameters, so that the problem of excessive fitting is avoided.
Further, the information in the system database in S2 is used as a training set of the algorithm model, and mainly refers to sub-factor data related to quality of the traditional Chinese medicine product, and mainly includes: information data of required materials, parameter data of production process and quality parameter data of intermediate products obtained in unit working section.
Further, in the step S3, the trained algorithm is deployed to a traditional Chinese medicine production management platform, where the platform includes a lightweight management platform at the web page end of the internet, an application program set at the computer end, customized software at the mobile phone end, and a front-end operation management interface and a back-end control service related to the platforms.
Further, the specific way of performing reverse dynamic adjustment on the drug production parameters in the step S4 is as follows:
the K-Means cluster analysis method is introduced to preprocess the observed data, cluster the observed data into sets, and use the cluster center point as the representative of the data, the K-Means cluster analysis method is an unsupervised classification algorithm, and the data sets of n samples are assumed:
X＝{x 1 ,x 2 ,…,x n }
the algorithm targets clustering the dataset into k clusters c= { C 1 ,c 2 ,…,c k Firstly, randomly selecting k initial centroids in a sample, and comparing and calculating the distance d between the sample point and each centroid i ＝‖x-μ i ‖ 2 Then, the sample points are marked into the nearest clusters; then, according to the sample points marked in each cluster, the cluster center mu is recalculated i The process is repeated until the cluster center converges.
The sum of squares of total errors E is:
wherein mu i Is cluster C i Is a centroid of (c).
Wherein, the liquid crystal display device comprises a liquid crystal display device,
is cluster C i The number of samples in (a). The function can be used for evaluating the value of k, selecting different k values to respectively calculate E values, and comparing the E changes among the different k values, if the E value is reduced from the beginning to the beginning, the current k value is the optimal k value;
the invention adopts contour coefficients to evaluate clustering effects because the K-Means clustering algorithm belongs to an unsupervised algorithm and cannot be evaluated by adopting a cross-validation method, and the method comprises the following steps of:
(1) Calculating sample x i Average distance a (i) from the rest of the samples in the same cluster. Definition of a (i) as sample x i The smaller a (i) the intra-cluster dissimilarity of (a), the more the sample x is described i The more should be clustered into such clusters;
(2) Calculating sample x i To other cluster Y j Average distance b of all samples of (a) ij Referred to as sample x i And cluster Y j Is not similar to the degree of dissimilarity of (a). Definition b ij For sample x i Inter-cluster dissimilarity of b (i) =min { b } i1 ,b i2 ,…,b in }；
(3) According to sample x i Intra-cluster dissimilarity a (i) and inter-cluster dissimilarity b (i) of the sample, the profile coefficient s (i) of the sample is expressed as:
the average value of all samples s (i) is called the contour coefficient of the clustering result;
the same distance calculation method is used by a K nearest neighbor algorithm (KNN for short, K-Neaeest Neighbors) classification algorithm and K-Means, a point x is selected in the space, the distance between the point x and each known point in the plane is calculated, K points closest to the point x are selected, and the category to which the x belongs is judged according to the labels of the K points closest to the point x. And judging the predicted quality by taking the combination of the internal control quality standard of the target traditional Chinese medicine product as a basis, and if the predicted quality is not in a specified range, carrying out reverse dynamic adjustment on the medicine production parameters until the medicine quality in a unit working section meets the requirement. If the predicted quality is within the specified range, the traditional Chinese medicine production information data meeting the production requirements is input into a system database in batch, and is used as a data basis for the intelligent feedback optimization module.
Further, the intelligent feedback optimization module in S5 is based on a man-machine interaction mode under an expert evaluation algorithm, and the feedback mode includes one or more combinations of a web service platform and a WeChat applet platform.
Further, in the step S6, expert opinions in the expert optimizing system are fed back to the traditional Chinese medicine production management platform and the data display platform, data display is updated in real time, decision support is provided for a manager, and an ecological intelligent traditional Chinese medicine product production mechanism and system are constructed; the detailed process is as follows:
in order to realize the automatic integration and update of the expert opinion in the expert optimizing system fed back to the traditional Chinese medicine production management platform and the data display platform, the data quality problem caused in the data transmission is avoided, and an index automatic update workflow is designed. The technical process can realize the integration and storage of expert tuning data across heterogeneous databases, complex library table structures and various coding rules and meets different index types and customization conditions:
(1) And (5) coding mapping. The method mainly solves the problem that the coding rules of the identification codes of the independent systems are different, and can realize the unification of the coding rules of all the systems functionally by establishing a coding mapping table, thereby improving the access and interoperation efficiency of the databases of the independent systems;
(2) And (5) expanding the index. Aiming at the problems of inconsistent progress of expert tuning systems, newly-built systems in the future, newly-increased index types and changing requirements, a standard interface of an extensible data source and the index types is provided, and the real requirements of continuous change of management indexes are met;
(3) And (5) establishing an index and warehousing. The index establishment and storage are the final purposes of information integration, the required data are obtained from each independent system by means of links such as code mapping, index expansion, heterogeneous database access and the like, the received data are converted into standard format data conforming to an index table structure by means of a data preprocessing link of an index establishment module, and then the accurate and efficient storage of the index is realized by means of a parallel storage algorithm;
(4) And (5) updating the timing. The method aims at reducing manual operation, realizing automatic index updating and warehousing operation when no person value is on the spot, setting different updating frequencies for different index types according to actual production requirements, updating data display in real time, increasing the weight influence of medicine proportion, realizing the maximum conversion of medicine effect of a user on the premise of ensuring safety quality, and constructing an ecological medicine production mechanism and system.
The invention has the beneficial effects that: the invention discloses a traditional Chinese medicine production optimization system based on artificial intelligence, which is characterized in that a MySQL database module, a traditional Chinese medicine production data analysis module and an intelligent feedback optimization module which are constructed by programming applications based on Python, java and C languages under a computer platform of Linux, windows and IOS operation systems are established by taking traditional Chinese medicine production data acquisition and a relational database as the basis, the prediction quality is judged by taking the information data of required materials, the function relation between the parameter data of a production process and the quality parameter data of intermediate products obtained in a unit section, the comprehensive function affecting the quality of products is obtained by taking a multi-data fusion mode, a medicine quality prediction model with parameter optimization is carried out by an AGA-SVR method, the medicine quality prediction is carried out by taking the information data of the acquired materials, the parameter data of the production process and the quality parameters of medicines obtained in the unit section as the basis, and carrying out inverse dynamic adjustment on the medicine production parameters by combining with the target traditional Chinese medicine product, and a K-Means cluster analysis method is introduced with the medicine production quality as the basis, and if the predicted quality is not within a specified range, and the inverse dynamic adjustment is carried out on the medicine quality until the medicine quality is required in the unit section. And feeding back expert opinions in the expert optimizing system to the traditional Chinese medicine production management platform and the data display platform, and updating data display in real time so as to provide decision support and intelligent optimizing of the production process for a manager on the premise of ensuring the safety quality, thereby constructing an ecological intelligent traditional Chinese medicine product production mechanism and system. . Aiming at the industrial problems of low control level, unstable quality, difficult timely feedback of production process information data and the like in the traditional Chinese medicine production process, the invention combines a cloud platform database, a heuristic multi-objective optimization algorithm and a user feedback system to realize full-period dynamic scheduling and optimization of the traditional Chinese medicine production based on production data, artificial intelligence technology and man-machine interaction, ensure the quality of the produced product and provide powerful guarantee for the health of people.
Drawings
The invention will be further described with reference to the accompanying drawings, in which embodiments do not constitute any limitation on the invention, and other drawings can be obtained by one of ordinary skill in the art without undue effort from the following drawings.
Fig. 1 is a schematic diagram of the structure of the present invention.
Detailed Description
The invention is further described in connection with the following examples.
Referring to fig. 1, the present invention aims to provide an artificial intelligence-based traditional Chinese medicine production optimization system, so as to solve the problems set forth in the background art.
In order to achieve the above purpose, the traditional Chinese medicine production optimization system based on artificial intelligence is provided, and comprises a traditional Chinese medicine production data acquisition and database module, a traditional Chinese medicine production data analysis module and an intelligent feedback optimization module.
S1, taking a traditional Chinese medicine production as an application target, collecting production data of the whole period of traditional Chinese medicine production on the basis of a traditional Chinese medicine product formula and a production process, taking a relational database as a basis, and uniformly storing the MySQL database constructed by programming application of a C language class under a computer platform based on a Windows operating system into a traditional Chinese medicine production management system database to establish a correlation model of various production data and traditional Chinese medicine product quality:
constructing information data of required materials, and obtaining a comprehensive function affecting the production quality of the medicine based on a mode of data fusion by functional relation between parameter data of a production process and quality parameter data of medicines obtained in a unit section, wherein the detailed process is as follows:
(1) And (5) traversing parallel time sequences. After data is input, multi-thread verification is carried out on a plurality of batches of data according to time sequence according to the data category and the relativity among the data, so that the problems of data missing and data abnormality can be found, and the verification efficiency can be improved.
(2) Index range. For the information data of most of required materials, the parameter data of the production process and the quality parameter data of the medicines obtained in the unit working section, a certain reasonable interval exists in the numerical range, and the normal value range is as follows:
X i,min ≤X i,≤ ≤X i,max
wherein X is i,t The data value of the data index i (information data of required materials, parameter data of production process technology, quality parameter data of medicines obtained in unit working section, and the like) at the time t; x is X i,min And X is i,max The lower limit and the upper limit of the index i are respectively, and abnormal data are obtained when the index i is not in the range.
(3) The Laida criterion. The information data of most materials, the parameter data of the production process and the quality parameter data of the medicines obtained in the unit working section approximately obey the statistical characteristics of normal distribution, the time sequence data can be checked by the Laida criterion, namely the data is unsuitable to be different from the time sequence mean value by more than 3 times of standard deviation, and the normal value range is
Wherein, the liquid crystal display device comprises a liquid crystal display device,
the time sequence data mean value of the index i; sigma (sigma) i The standard deviation of the time series data of the index i. And if the data exceeds the data, the data is regarded as abnormal data.
(4) And (5) logic association. The data of partial different time sequences have certain logic relevance, can be converted by a related formula and can be summarized as
F(X i,t ,X 2,t ,…,X n,t ,δ)＝0
Wherein F is various physical association relations which are required to be met among various data, delta is an error threshold, and the error is an allowable error of logic association calculation in an analysis and judgment flow, and proper numerical values are used according to the data association calculation of different types; and taking 0 in the missing filling and abnormality repairing processes, and predicting missing or abnormal data through known data. The analysis and judgment of the data and the deletion supplement and the abnormality repair of partial data can be carried out through the logic association.
(5) Sequence mean. And the single data values which are missing or deviate from a reasonable interval can be repaired through a sequence mean value so as to increase the integrity and accuracy of the data.
(6) Adjacent repair. For a single data value which is not easy to cause severe change in a short time and is in a reasonable interval, when the data is in a defect or abnormal state, the adjacent compliance data is taken as repair.
(7) And (5) predicting the medicine quality. And carrying out a parameter-optimized medicine quality prediction model on support vector machine regression (SVR) based on an Adaptive Genetic Algorithm (AGA), and filling the missing data set through the acquired information data of materials, the parameter data of a production process and the quality parameters of medicines obtained in a unit section. The SVR is based on the concept of statistical regression, and the prediction capability is stronger than that of a BP neural network, an RBF neural network and linear regression under the condition of less sample data; the AGA has faster searching capability and the capability of jumping out a local optimal solution by changing a genetic operator and a mutation operator along with the updating of iteration algebra, so that the problem of SVR model parameter selection can be solved;
s2, constructing a multi-element heuristic Bayesian neural network algorithm model based on parameters and calculation modes in the correlation model of various production data and traditional Chinese medicine product quality, and solving the target prediction problem of a multi-factor driven relationship, wherein the detailed process is as follows:
aiming at the problems that the model complexity cannot be controlled and the data fitting is excessive in the actual use process of a standard neural network model, the generalization capability of the standard neural network in the actual use process is poor, the Bayesian neural network is used, a probability layer is contained in a hidden layer, and the weight value accords with the probability distribution requirement. With the addition of the probability layer, the Bayesian neural network has the ability to analyze uncertainties. Under the condition of determining the input vector, the output can be determined with different output probabilities by the probability layer unit. A bayesian neural network may be defined as an integrated network comprising a plurality of sub-networks, wherein there is a certain correlation between different sub-networks, all sub-networks being optimized simultaneously by different training procedures. When the Bayesian neural network is used for carrying out the association model simulation of the medicine quality, a plurality of forward propagation processes are used for ensuring that simulation results are obtained from different sub-networks, so that the Bayesian neural network can obtain a remarkable regularization effect, and the problem of excessive fitting possibly generated in the standard neural network can be prevented. The Bayesian neural network is obtained according to the BP neural network principle and Bayesian rules, and the Bayesian neural network training process needs to pass through 3 reasoning links.
(1) H and X are used for respectively representing Bayesian neural network model and medicine quality historical sample data, alpha and beta are used for respectively representing Bayesian neural network hyper-parameters, and posterior probability of the weight w is calculated, so that the weight w meeting the upper limit of the posterior probability is obtained up 。
(2) And combining the medicine quality historical sample data X, determining the posterior probability of the super parameter, and updating alpha and beta.
(3) And determining the model with the maximum posterior probability by comparing the saliency of different models, thereby obtaining the optimal Bayesian neural network.
Based on the reasoning, under the condition that the prior probability and likelihood are set to meet the exponential distribution, the posterior distribution of the Bayesian neural network weight is obtained as follows:
to obtain the optimal weight w up The upper limit value of the delay probability needs to be obtained, namelyThe lower limit of the total error. The monte carlo numerical integration method of updating weights is mainly used at present. In order to update the super-parameters, the Bayesian neural network needs to continuously iterate and train the drug quality historical sample data in the training process, so as to optimize the weight and determine the optimal Bayesian neural network model. In the training process, the complexity of the Bayesian neural network model structure is reduced by updating the super parameters, so that the problem of excessive fitting is avoided;
the information in the system database is used as a training set of an algorithm model, and the data volume is increased to avoid the occurrence of the over-fitting phenomenon;
s3, deploying a trained algorithm on a traditional Chinese medicine production management platform, taking material information data and real-time production data as input, and outputting quality prediction values of intermediate products and finished products obtained in a unit workshop section;
s4, combining an internal control quality standard of a target traditional Chinese medicine product as a basis, and if the predicted quality is not within a specified range, carrying out reverse dynamic adjustment on production parameters in the following specific manner:
the K-Means cluster analysis method is introduced to preprocess the observed data, cluster the observed data into sets, and use the cluster center point as the representative of the data, the K-Means cluster analysis method is an unsupervised classification algorithm, and the data sets of n samples are assumed:
X＝{x 1 ,x 2 ,…,x n }
the algorithm targets clustering the dataset into k clusters c= { C 1 ,c 2 ,…,c k Firstly, randomly selecting k initial centroids in a sample, and comparing and calculating the distance d between the sample point and each centroid i ＝‖x-μ i ‖ 2 Then, the sample points are marked into the nearest clusters; then, according to the sample points marked in each cluster, the cluster center mu is recalculated i The process is repeated until the cluster center converges.
The sum of squares of total errors E is:
wherein mu i Is cluster C i Is a centroid of (c).
Wherein, the liquid crystal display device comprises a liquid crystal display device,
is cluster C i The number of samples in (a). The function can be used for evaluating the value of k, selecting different k values to respectively calculate E values, and comparing the E change among the different k values, if the E value is reduced rapidly from the beginning, the current k value is the optimal k value. The same distance calculation method is used by a K nearest neighbor algorithm (KNeaeest Neighbors KNN) classification algorithm and K-Means, a point x is selected in the space, the distance between the point x and each known point in the plane is calculated, K points closest to the point x are selected, and the category of the x is judged according to the labels of the K points closest to the point x. And judging the predicted quality by taking the medicine production quality described by combining with the medicine monitoring standard as a basis, and if the predicted quality is not in a specified range, carrying out reverse dynamic adjustment on the medicine production parameters until the medicine quality in a unit working section meets the requirement. If the predicted quality is within the specified range, the drug information meeting the production requirement is input into a system database in batch, and is used as a data basis for the intelligent feedback tuning module until the quality of intermediate products and finished products in the unit working section meets the requirement;
s5, inputting the traditional Chinese medicine production information data meeting the production requirements into a system database in batches, and taking the traditional Chinese medicine production information data as a data basis provided for the intelligent feedback tuning module;
s6, feeding the expert opinion in the expert optimizing system back to the traditional Chinese medicine product production management platform and the data display platform, updating the data display in real time, providing decision support for a manager, and constructing an intelligent traditional Chinese medicine product production mechanism and system:
in order to realize the automatic integration and update of the expert opinion in the expert optimizing system fed back to the traditional Chinese medicine production management platform and the data display platform, the data quality problem caused in the data transmission is avoided, and an index automatic update workflow is designed. The technical process can realize the integration and storage of expert tuning data across heterogeneous databases, complex library table structures and various coding rules and meets different index types and customization conditions:
(1) And (5) coding mapping. The method mainly solves the problem that the coding rules of the identification codes of the independent systems are different, and can realize the unification of the coding rules of all the systems functionally by establishing a coding mapping table, thereby improving the access and interoperation efficiency of the databases of the independent systems;
(2) And (5) expanding the index. Aiming at the problems of inconsistent progress of expert tuning systems, newly-built systems in the future, newly-increased index types and changing requirements, a standard interface of an extensible data source and the index types is provided, and the real requirements of continuous change of management indexes are met;
(3) And (5) establishing an index and warehousing. The index establishment and storage are the final purposes of information integration, the required data are obtained from each independent system by means of links such as code mapping, index expansion, heterogeneous database access and the like, the received data are converted into standard format data conforming to an index table structure by means of a data preprocessing link of an index establishment module, and then the accurate and efficient storage of the index is realized by means of a parallel storage algorithm;
(4) And (5) updating the timing. The method aims at reducing manual operation, realizing automatic index updating and warehousing operation when no person value is on the spot, setting different updating frequencies for different index types according to actual production requirements, updating data display in real time, increasing the weight influence of medicine proportion, realizing the maximum conversion of medicine effect of a user on the premise of ensuring safety quality, and constructing an intelligent traditional Chinese medicine product production mechanism and system.
The invention has the beneficial effects that: the invention discloses a traditional Chinese medicine production optimization system based on artificial intelligence, which is characterized in that a MySQL database module, a traditional Chinese medicine production data analysis module and an intelligent feedback optimization module which are constructed by programming applications based on Python, java and C languages under a computer platform of Linux, windows and IOS operation systems are established by taking traditional Chinese medicine production data acquisition and a relational database as the basis, the prediction quality is judged by taking the information data of required materials, the function relation between the parameter data of a production process and the quality parameter data of intermediate products obtained in a unit section, the comprehensive function affecting the quality of products is obtained by taking a multi-data fusion mode, a medicine quality prediction model with parameter optimization is carried out by an AGA-SVR method, the medicine quality prediction is carried out by taking the information data of the acquired materials, the parameter data of the production process and the quality parameters of medicines obtained in the unit section as the basis, and carrying out inverse dynamic adjustment on the medicine production parameters by combining with the target traditional Chinese medicine product, and a K-Means cluster analysis method is introduced with the medicine production quality as the basis, and if the predicted quality is not within a specified range, and the inverse dynamic adjustment is carried out on the medicine quality until the medicine quality is required in the unit section. And feeding back expert opinions in the expert optimizing system to the traditional Chinese medicine production management platform and the data display platform, and updating data display in real time so as to provide decision support and intelligent optimizing of the production process for a manager on the premise of ensuring the safety quality, thereby constructing an ecological intelligent traditional Chinese medicine product production mechanism and system. . Aiming at the industrial problems of low control level, unstable quality, difficult timely feedback of production process information data and the like in the traditional Chinese medicine production process, the invention combines a cloud platform database, a heuristic multi-objective optimization algorithm and a user feedback system to realize full-period dynamic scheduling and optimization of the traditional Chinese medicine production based on production data, artificial intelligence technology and man-machine interaction, ensure the quality of the produced product and provide powerful guarantee for the health of people.
The present invention also provides a computer readable storage medium having stored therein at least one instruction that is loaded and executed by a processor to implement the above-described method. The computer readable storage medium may be, among other things, ROM, random access memory, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. The instructions stored therein may be loaded by a processor in the terminal and perform the methods described above.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The traditional Chinese medicine production optimizing system based on artificial intelligence is characterized by comprising a traditional Chinese medicine production data acquisition and database module, a traditional Chinese medicine production data analysis module and an intelligent feedback optimizing module; the operation steps of the tuning system are as follows:
s1, taking a traditional Chinese medicine production as an application target, collecting production data of the whole period of traditional Chinese medicine production based on a traditional Chinese medicine product formula and a production process, uniformly storing the production data into a traditional Chinese medicine production management system database, and establishing a correlation model of various production data and traditional Chinese medicine product quality;
s2, constructing a multi-element heuristic artificial intelligent algorithm based on parameters and calculation modes in a correlation model of various production data and traditional Chinese medicine product quality, taking information in a system database as a training set of an algorithm model, and increasing data quantity to avoid the occurrence of a fitting phenomenon;
s3, deploying a trained algorithm on a traditional Chinese medicine production management platform, taking material information data and real-time production data as input, and outputting quality prediction values of intermediate products and finished products obtained in a unit workshop section;
s4, combining an internal control quality standard of a target traditional Chinese medicine product as a basis, and if the predicted quality is not in a specified range, carrying out reverse dynamic adjustment on production parameters until the quality of intermediate products and finished products in a unit working section meets the requirement;
s5, inputting the traditional Chinese medicine production information data meeting the production requirements into a system database in batches, and taking the traditional Chinese medicine production information data as a data basis provided for the intelligent feedback tuning module;
s6, feeding the expert opinion in the expert optimizing system back to the traditional Chinese medicine product production management platform and the data display platform, updating the data display in real time, providing decision support for a manager, and constructing an intelligent traditional Chinese medicine product production mechanism and system.
2. The artificial intelligence-based traditional Chinese medicine production optimization system according to claim 1, wherein the step of collecting production data of a whole production cycle of a traditional Chinese medicine product in step S1 mainly comprises the following steps: raw and auxiliary materials, intermediate product material information data, parameter data of a production process technology and quality parameters of medicines obtained in a unit section.
3. The system of claim 1, wherein the database of the system of S1 is based on a relational database, and comprises MySQL database constructed by one type of programming application in Python, java, and C languages under a computer platform based on Linux, windows, and IOS operating systems.
4. The system for preparing and optimizing Chinese medicine based on artificial intelligence according to claim 1, wherein the step S1 is to build a correlation model of various production data and Chinese medicine product quality, based on claim 2, to build information data of required materials, and to obtain a comprehensive function affecting Chinese medicine product quality based on a data fusion mode, wherein the detailed process is as follows:
(1) Parallel timing traversal: after data is input, multi-thread verification is carried out on a plurality of batches of data according to time sequence according to the data category and the relativity among the data; (2) index range: information data of most of required materials, parameter data of production process and quality parameter data of medicines obtained in unit working section; (3) Laiyida criterion: the information data of most materials, the parameter data of the production process and the quality parameter data of the medicines obtained in the unit working section approximately obey the statistical characteristics of normal distribution, and the time sequence data can be checked by the Laida criterion; (4) sequence means: the single data values which are missing or deviate from a reasonable interval can be repaired through a sequence mean value, so that the integrity and the accuracy of the data are improved; (5) adjacent repair: for a single data value which is not easy to cause severe change in a short time and deviates from a reasonable interval, when the data is in a missing state or abnormal state, adjacent compliance data is taken as repair; (6) drug quality prediction: based on a self-adjusting genetic algorithm, carrying out parameter-optimized medicine quality prediction model on regression of the support vector machine, and filling the missing data set through the acquired information data of materials, the parameter data of the production process and the quality parameters of medicines obtained in a unit section; using the fitness function as an adaptation condition index of the medicine quality in the algorithm under a given condition; the influence of the self adaptability of the cross variation probability and the computational complexity is balanced through a segmentation threshold value thresh, and the probability of high-adaptability individual loss is reduced through reducing the cross variation probability:
wherein p is cross Indicating the probability of loss of individuals with high fitness, fit max Fit, the maximum fitness of the current population min For the minimum fitness of the current population, fit (i) is the fitness of individual i, thresh is the probability segmentation threshold.
5. The system for optimizing traditional Chinese medicine production based on artificial intelligence according to claim 1, wherein the multivariate heuristic artificial intelligence algorithm in S2 is a bayesian neural network algorithm model, and can process the target prediction problem of the multi-factor driven relationship, and the detailed process is as follows:
(1) H and X are used for respectively representing Bayesian neural network model and medicine quality historical sample data, alpha and beta are used for respectively representing Bayesian neural network hyper-parameters, and posterior probability of the weight w is calculated, so that the weight w meeting the upper limit of the posterior probability is obtained up The method comprises the steps of carrying out a first treatment on the surface of the (2) Combining with medicine quality historical sample data X, determining the posterior probability of the superparameter, updating alpha and beta, and corresponding expected E D And E is W The method comprises the steps of carrying out a first treatment on the surface of the (3) Determining a model with the maximum posterior probability by comparing the saliency of different models, thereby obtaining an optimal Bayesian neural network, and setting the prior probability z M Under the condition that (alpha, beta) and likelihood both meet the exponential distribution, the posterior distribution of the Bayesian neural network weight taking the posterior probability deviation M (w) into consideration is obtained as follows:
and acquiring upper limit update super parameters through Bayesian reasoning and determining an optimal Bayesian neural network model.
6. The system for optimizing traditional Chinese medicine production based on artificial intelligence according to claim 1, wherein the information in the system database in S2 is used as a training set of an algorithm model, and mainly refers to sub-factor data related to quality of traditional Chinese medicine products, and mainly comprises: information data of required materials, parameter data of production process and quality parameter data of intermediate products obtained in unit working section.
7. The system of claim 1, wherein the step S3 is to deploy the trained algorithm to a traditional Chinese medicine production management platform, the platform comprises a lightweight management platform at the internet web page end, an application program set at the computer end, custom software at the mobile phone end, and front-end operation management interfaces and back-end control services related to the platforms.
8. The artificial intelligence-based traditional Chinese medicine production optimization system according to claim 1, wherein the specific way of reversely dynamically adjusting the production parameters of the medicines in the S4 is as follows:
clustering the drug production parameter data into k clusters so that the sum of squares of errors from sample points to the centers of the clusters is minimum; randomly selecting k initial centroids in a sample, and comparing and calculating the distance d between a sample point and each centroid i ＝‖x-μ i ‖ 2 Marking sample points into the nearest cluster; then, according to the sample points marked in each cluster, the cluster center mu is recalculated i Repeating the process until the cluster center converges; and then adopting the contour coefficient to evaluate the clustering effect, and determining the optimal cluster number.
9. The system of claim 1, wherein the intelligent feedback optimization module in S5 is based on man-machine interaction under expert evaluation algorithm, and the feedback mode includes one or more of a web service platform and a WeChat applet platform.
10. The artificial intelligence based traditional Chinese medicine production optimizing system according to claim 1, wherein in the step S6, expert opinions in the expert optimizing system are fed back to a traditional Chinese medicine production management platform and a data display platform, data display is updated in real time, decision support is provided for a manager, an ecological intelligent traditional Chinese medicine product production mechanism and system are constructed, and integration of expert optimizing data is performed by combining a cross-heterogeneous database, a complex library table structure, various coding rules and customized conditions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211566056.4A CN116187507A (en) | 2022-12-07 | 2022-12-07 | Traditional chinese medicine production system of adjusting based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211566056.4A CN116187507A (en) | 2022-12-07 | 2022-12-07 | Traditional chinese medicine production system of adjusting based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116187507A true CN116187507A (en) | 2023-05-30 |
Family
ID=86431595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211566056.4A Pending CN116187507A (en) | 2022-12-07 | 2022-12-07 | Traditional chinese medicine production system of adjusting based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116187507A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117193025A (en) * | 2023-11-07 | 2023-12-08 | 矿冶科技集团有限公司 | Control method and device of dosing machine, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104412247A (en) * | 2012-07-24 | 2015-03-11 | 通用电气公司 | Systems and methods for improving control system reliability |
| CN112348360A (en) * | 2020-11-06 | 2021-02-09 | 华润三九（枣庄）药业有限公司 | Chinese medicine production process parameter analysis system based on big data technology |
| CN112529023A (en) * | 2019-09-18 | 2021-03-19 | 上海钛空猫智能科技有限公司 | Configured artificial intelligence scene application research and development method and system |
| CN112749849A (en) * | 2021-01-19 | 2021-05-04 | 华东理工大学 | Integrated learning online prediction method for key parameters of continuous catalytic reforming process |
-
2022
- 2022-12-07 CN CN202211566056.4A patent/CN116187507A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104412247A (en) * | 2012-07-24 | 2015-03-11 | 通用电气公司 | Systems and methods for improving control system reliability |
| CN112529023A (en) * | 2019-09-18 | 2021-03-19 | 上海钛空猫智能科技有限公司 | Configured artificial intelligence scene application research and development method and system |
| CN112348360A (en) * | 2020-11-06 | 2021-02-09 | 华润三九（枣庄）药业有限公司 | Chinese medicine production process parameter analysis system based on big data technology |
| CN112749849A (en) * | 2021-01-19 | 2021-05-04 | 华东理工大学 | Integrated learning online prediction method for key parameters of continuous catalytic reforming process |
Non-Patent Citations (6)
| Title |
|---|
| 佘星源等: "南京禄口机场低能见度天气统计分析及判断模型", 气象科学, pages 662 - 663 * |
| 余长慧等: "海量水文管理数据信息整合与更新机制研究", 水利信息化, pages 6 - 11 * |
| 尹华政等: "基于多技术多数据融合的水电环境监测管理平台建设", 水力发电, pages 3 - 4 * |
| 张铮等: "基于改进多目标自适应遗传算法的机器人路径规划", 西安理工大学学报, pages 6 * |
| 朱爱华等: "基于贝叶斯神经网络的农田分区灌溉需水量模拟", 农业工程, pages 79 - 80 * |
| 李金昌: "中药生产工艺的智能优化研究", 中国优秀硕士学位论文全文数据库信息科技辑, pages 9 - 11 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117193025A (en) * | 2023-11-07 | 2023-12-08 | 矿冶科技集团有限公司 | Control method and device of dosing machine, electronic equipment and storage medium |
| CN117193025B (en) * | 2023-11-07 | 2024-02-02 | 矿冶科技集团有限公司 | Control method and device of dosing machine, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Namadchian et al. | A new meta-heuristic algorithm for optimization based on variance reduction of guassian distribution | |
| US7340440B2 (en) | Hybrid neural network generation system and method | |
| CN103745273A (en) | Semiconductor fabrication process multi-performance prediction method | |
| Mukhopadhyay et al. | An interactive approach to multiobjective clustering of gene expression patterns | |
| CN112987666B (en) | Power plant unit operation optimization regulation and control method and system | |
| CN110571792A (en) | Analysis and evaluation method and system for operation state of power grid regulation and control system | |
| CN112100940A (en) | Method and device for predicting primary stretching technological parameters of optical fiber preform | |
| CN111898867B (en) | Airplane final assembly production line productivity prediction method based on deep neural network | |
| Kulkarni et al. | Dynamic grouping of parts in flexible manufacturing systems—a self-organizing neural networks approach | |
| CN110264079A (en) | Hot-rolled product qualitative forecasting method based on CNN algorithm and Lasso regression model | |
| CN116187507A (en) | Traditional chinese medicine production system of adjusting based on artificial intelligence | |
| Tembusai et al. | K-nearest neighbor with K-fold cross validation and analytic hierarchy process on data classification | |
| CN104732067A (en) | Industrial process modeling forecasting method oriented at flow object | |
| CN115481727A (en) | Intention recognition neural network generation and optimization method based on evolutionary computation | |
| US7206772B2 (en) | Method of operating a plant and a control system for controlling the same | |
| CN109697531A (en) | A kind of logistics park-hinterland Forecast of Logistics Demand method | |
| CN112990584A (en) | Automatic production decision system and method based on deep reinforcement learning | |
| CN117455070A (en) | Traditional Chinese medicine production data management system based on big data | |
| CN111144569A (en) | Yield improvement applicable model optimization method based on genetic algorithm | |
| CN113743461B (en) | Unmanned aerial vehicle cluster health degree assessment method and device | |
| CN115600102A (en) | Abnormal point detection method and device based on ship data, electronic device and medium | |
| Sadi-Nezhad et al. | A new fuzzy clustering algorithm based on multi-objective mathematical programming | |
| CN117539520B (en) | Firmware self-adaptive upgrading method, system and equipment | |
| CN112070200A (en) | Harmonic group optimization method and application thereof | |
| US11810644B2 (en) | System and method for genomic association |
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 |