CN117195725A - Mechanism model visual construction method, equipment and medium based on industrial data - Google Patents
Mechanism model visual construction method, equipment and medium based on industrial data Download PDFInfo
- Publication number
- CN117195725A CN117195725A CN202311170010.5A CN202311170010A CN117195725A CN 117195725 A CN117195725 A CN 117195725A CN 202311170010 A CN202311170010 A CN 202311170010A CN 117195725 A CN117195725 A CN 117195725A
- Authority
- CN
- China
- Prior art keywords
- data
- industrial data
- industrial
- mechanism model
- determining
- 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
- 230000007246 mechanism Effects 0.000 title claims abstract description 106
- 238000010276 construction Methods 0.000 title claims abstract description 31
- 230000000007 visual effect Effects 0.000 title claims abstract description 28
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 86
- 238000000034 method Methods 0.000 claims abstract description 54
- 238000004364 calculation method Methods 0.000 claims abstract description 42
- 238000012545 processing Methods 0.000 claims abstract description 33
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 230000009466 transformation Effects 0.000 claims abstract description 30
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 238000003032 molecular docking Methods 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims description 30
- 238000012800 visualization Methods 0.000 claims description 20
- 238000005516 engineering process Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 16
- 238000004519 manufacturing process Methods 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 11
- 230000001419 dependent effect Effects 0.000 claims description 10
- 230000009467 reduction Effects 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000000556 factor analysis Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 abstract description 10
- 230000008569 process Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 210000001503 joint Anatomy 0.000 description 4
- 238000003066 decision tree Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000001556 precipitation Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000013068 supply chain management Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012932 thermodynamic analysis Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method, equipment and medium for visually constructing a mechanism model based on industrial data, which are used for solving the problems that an algorithm engineer is required to program in the construction of the existing mechanism model, and the construction cost is high, the period is long and secondary development is difficult. The method comprises the following steps: docking the local database with the industrial data set to obtain industrial data; determining the data processing logic of the industrial data according to the data types to determine the arrangement sequence of a plurality of common algorithm computing components, dragging and splicing the plurality of common algorithm computing components in a visual interface, and preprocessing the industrial data; calculating industrial data subjected to dimensional transformation pretreatment based on an analysis algorithm, and extracting key features of the industrial data subjected to dimensional transformation based on a feature calculation component; and the calculation component builds a mechanism model according to key features in the dimension transformed industrial data based on a learning algorithm, and triggers the debugging mechanism model based on dragging of each configuration information in the mechanism model to complete visual construction of the mechanism model.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a mechanism model visual construction method, equipment and medium based on industrial data.
Background
The mechanism model is a calculation model based on knowledge principles such as physics and chemistry, solves and optimizes specific industrial problems by means of machine learning, engineering knowledge and expert experience, and can be used for industrial scenes such as production process optimization, trend prediction analysis, decision assistance guidance and the like. The research and development design, supply chain management, production and manufacture, quality control, operation management and other links of an industrial enterprise depend on data circulation and application, and the value contained in industrial data is converted and precipitated usually in a mechanism model form, so that the production and management process of the enterprise is enabled. Most mechanism models are typically packaged in kits and kits, components in mature commercial software, built-in thermodynamic analysis, hydrodynamic analysis, finite element simulation, circuit design simulation models, etc. of commercial software such as CAE, CAD, EDA. However, due to the problems of purchase cost, sealing performance and the like of commercial software, the vast enterprises can hardly enjoy high-quality and customized mechanism model services.
At present, the traditional code programming mode is used for constructing a mechanism model, so that an algorithm engineer is required to master and understand a great amount of industrial field knowledge such as equipment mechanism, process flow and the like, and is required to have the technical capabilities of machine learning, programming development and the like, and an industrial enterprise is generally not provided with similar cross-expert talents, so that enterprise users are difficult to independently complete the design, development and application of the mechanism model according to actual demands. The main stream of mechanism model construction mode is mainly based on the programming and development of algorithm engineer codes, and the problems of high development cost, long period and difficult secondary development are generally existed.
Disclosure of Invention
The embodiment of the application provides a method, equipment and medium for visually constructing a mechanism model based on industrial data, which are used for solving the technical problems that an algorithm engineer is required to carry out code programming in the existing mechanism model construction mode, and the construction cost is high, the period is long and secondary development is difficult.
In one aspect, an embodiment of the present application provides a method for constructing a mechanism model visualization based on industrial data, including:
based on JDBC technology and a database open interface, docking a local database with an industrial data set and acquiring industrial data in the industrial data set;
determining a data type corresponding to industrial data, and determining the arrangement sequence of a plurality of pre-packaged common algorithm computing components according to data processing logic corresponding to the data type of the industrial data; the data types of the industrial data comprise time sequence type data and image type data;
splicing a plurality of common algorithm computing components corresponding to the industrial data according to the arrangement sequence in a dragging mode in a visual interface, and preprocessing the industrial data in sequence through the spliced plurality of common algorithm computing components;
performing dimension transformation on the preprocessed industrial data based on a pre-packaged analysis algorithm computing component, and extracting key features from the industrial data after dimension transformation based on a pre-packaged feature computing component;
based on a pre-packaged learning algorithm calculation component, a corresponding mechanism model is constructed according to key features in the industrial data after dimension transformation, and based on drag trigger of each configuration information in the mechanism model, the mechanism model is debugged to complete visual construction of the mechanism model.
In one implementation manner of the present application, the interfacing the local database with the industrial data set based on JDBC technology and the database open interface, and obtaining the industrial data in the industrial data set specifically includes:
acquiring time sequence type data in an industrial scene to an internet of things center based on a sensor and production equipment preset in the industrial scene through a first preset protocol, and docking a local database with the internet of things center based on a JDBC technology and a database open interface to acquire the time sequence type data in the internet of things center;
and acquiring the image class data in the industrial scene to a data center station based on production line quality inspection equipment preset in the industrial scene through a second preset protocol, and docking a local database with the data center station based on a JDBC technology and a database open interface so as to acquire the image class data in the data center station.
In one implementation manner of the present application, the determining a data type corresponding to industrial data, and determining, according to data processing logic corresponding to the data type of the industrial data, an arrangement sequence of a plurality of pre-packaged common algorithm computing components specifically includes:
analyzing a plurality of industrial data in the industrial data set, respectively determining data types corresponding to the industrial data, and determining data processing logic corresponding to the industrial data according to the data types corresponding to the industrial data;
and determining a plurality of common algorithm computing components corresponding to the industrial data in a plurality of pre-packaged common algorithm computing components according to the data processing logic corresponding to the industrial data, and determining the arrangement sequence among the plurality of common algorithm computing components corresponding to the industrial data according to the data processing logic corresponding to the industrial data.
In one implementation manner of the present application, the preprocessing of the industrial data by the plurality of spliced common algorithm computing components sequentially includes:
reading the industrial data in the industrial data set, and analyzing the industrial data in the industrial data set to obtain a corresponding analysis result;
according to the analysis result corresponding to each industrial data, determining a data quality analysis report corresponding to the industrial data, and determining a plurality of industrial data of which the data quality accords with the quality standard in the data quality analysis report;
splitting the plurality of industrial data according to a preset splitting proportion and in a random sampling mode so as to split the plurality of industrial data into a training data set corresponding to the splitting proportion and a test data set corresponding to the splitting proportion.
In one implementation manner of the present application, the pre-packaged analysis algorithm-based computing component performs dimension transformation on the preprocessed industrial data, and specifically includes:
based on a pre-packaged analysis algorithm calculation component, carrying out data analysis on the preprocessed industrial data, and respectively determining the data dimension corresponding to each preprocessed industrial data;
wherein the analysis algorithm calculation component comprises at least: a principal component analysis algorithm calculation component, a linear discriminant analysis algorithm calculation component, and a factor analysis algorithm calculation component;
at least one piece of industrial data with the data dimension larger than a preset dimension threshold is determined, and dimension reduction processing is carried out on the at least one piece of industrial data respectively so as to realize dimension transformation of the industrial data.
In one implementation manner of the present application, the feature computing component extracts key features from the dimension transformed industrial data based on the pre-packaged features, and specifically includes:
based on a pre-packaged feature calculation component, determining feature attributes corresponding to the industrial data after dimension transformation, and determining the influence degree of each feature attribute in the industrial data on a mechanism model respectively;
and determining the characteristic weight of the corresponding characteristic attribute in the industrial data according to the influence degree of each characteristic attribute on the mechanism model, and determining at least one characteristic attribute with the characteristic weight greater than a preset weight threshold value, wherein the at least one characteristic attribute is used as a key characteristic.
In one implementation manner of the present application, after the computing component based on the pre-packaged learning algorithm builds a corresponding mechanism model according to key features in the industrial data after dimension transformation, the method further includes:
based on a pre-packaged model verification computing component, inputting independent variable industrial data in a test data set obtained by preprocessing into the mechanism model, and obtaining corresponding test dependent variable industrial data;
and comparing the similarity between the test dependent variable industrial data and preset target dependent variable industrial data, and determining the accuracy of the mechanism model according to the corresponding similarity.
In one implementation manner of the present application, after the computing component based on the pre-packaged learning algorithm builds a corresponding mechanism model according to key features in the industrial data after dimension transformation, the method further includes:
based on a cross validation method, splitting the test data set obtained by preprocessing into a preset number of mutually exclusive subsets, taking the first number of mutually exclusive subsets as training subsets, and taking the second number of mutually exclusive subsets as test subsets; wherein the preset number is the sum of the first number and the second number;
and determining test results corresponding to the first number of training subsets and the second number of test subsets, and determining generalized error values corresponding to the mechanism model according to the average value of the specified number of test results.
In another aspect, an embodiment of the present application further provides a device for constructing a mechanism model visualization based on industrial data, where the device includes:
at least one processor;
and a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform an industrial data based mechanism model visualization building method as described above.
In another aspect, embodiments of the present application also provide a non-volatile computer storage medium storing computer-executable instructions configured to:
the method for constructing the mechanism model visualization based on the industrial data.
The embodiment of the application provides a mechanism model visual construction method, equipment and medium based on industrial data, which at least comprise the following beneficial effects:
based on the pre-packaged calculation component, a materialized calculation formula, knowledge of mechanism in the industrial field and expert experience can be built into the system, so that the use requirement of an algorithm engineer is met; the local database is in butt joint with the industrial data set, so that industrial data in the industrial data set can be received, data processing logic corresponding to the industrial data is determined according to the data type corresponding to the industrial data, computing components for processing the industrial data are ordered according to the determined data processing logic, and therefore the computing components can be dragged and spliced in a visual interface according to the determined ordering, so that the industrial data can be preprocessed through the spliced computing components in sequence; the method comprises the steps of carrying out dimension change on the preprocessed industrial data, extracting key features of the industrial data after dimension change, and constructing a mechanism model according to the key features and a pre-packaged computing assembly, and debugging the mechanism model through drag trigger in a visual interface to realize visual construction of the mechanism model.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a schematic flow chart of a method for constructing a mechanism model visualization based on industrial data according to an embodiment of the present application;
fig. 2 is a schematic diagram of an internal structure of a device for visualizing and constructing a mechanism model based on industrial data according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The embodiment of the application provides a mechanism model visual construction method, equipment and medium based on industrial data, which can embed a physical and chemical calculation formula, industrial field mechanism knowledge and expert experience into a system based on a pre-packaged calculation assembly, so as to meet the use requirements of algorithm engineers; the local database is in butt joint with the industrial data set, so that industrial data in the industrial data set can be received, data processing logic corresponding to the industrial data is determined according to the data type corresponding to the industrial data, computing components for processing the industrial data are ordered according to the determined data processing logic, and therefore the computing components can be dragged and spliced in a visual interface according to the determined ordering, so that the industrial data can be preprocessed through the spliced computing components in sequence; the method comprises the steps of carrying out dimension change on the preprocessed industrial data, extracting key features of the industrial data after dimension change, and constructing a mechanism model according to the key features and a pre-packaged computing assembly, and debugging the mechanism model through drag trigger in a visual interface to realize visual construction of the mechanism model. The method solves the technical problems that the mechanism model construction mode in the prior art requires an algorithm engineer to carry out code programming, and has high construction cost, long period and difficult secondary development.
The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for constructing a mechanism model visualization based on industrial data according to an embodiment of the present application. As shown in fig. 1, the method for constructing the mechanism model visualization based on the industrial data provided by the embodiment of the application mainly comprises the following steps:
101. based on JDBC technology and database open interface, the local database is docked with the industrial data set, and industrial data in the industrial data set is obtained.
The mechanism knowledge mainly refers to the knowledge about basic principles and mechanisms in the natural science fields of machinery, physics, chemistry, biology and the like, and the knowledge about the operation principles and mechanisms of various technologies and processes, such as the principle of conservation of momentum, the principle of thermodynamics, the hydrodynamic force equation, the chemical reaction equation and the like. The acquisition path includes referring to the related books, asking the specialists in the related technical fields, and the like.
In order to develop, manage and apply the mechanism model with low cost and high efficiency, the application discloses a mechanism model visual construction method based on industrial data, which provides a complete calculation component library and comprises modules of data access, data preprocessing, feature extraction, algorithm configuration, model construction, model verification to model application and the like. Meanwhile, a zero-code and visual model arrangement development tool is provided, complex mechanism knowledge, expert experience and machine learning algorithm are packaged into calculation components, precipitation multiplexing of the knowledge, experience and algorithm is achieved, a user independently drags, concatenates, configures and debugs various calculation components according to model service requirements, visual development of a mechanism model can be achieved, after model verification is completed, the model is supported to be stored and released as a standard Restful API interface, and third-party service system calling is supported.
The expert experience mainly refers to experience class knowledge owned by professional skill staff with professional knowledge, skills and rich practice experience in the production and management process of industrial enterprises, expert experience of related fields or production procedures is generally required to be obtained by exchanging records with old master in factories and workshops or field experts, and the expert experience is packaged into an industry class function algorithm in a programming code mode and used in the step of model construction, so that the platform precipitation and modeling multiplexing of the expert experience are realized.
Specifically, in one embodiment of the present application, a server acquires time sequence type data in an industrial scene to an internet of things platform based on a sensor and a production device preset in the industrial scene through a first preset protocol, and interfaces a local database with the internet of things platform based on a JDBC technology and a database open interface to acquire the time sequence type data in the internet of things platform, and simultaneously, the server acquires image type data in the industrial scene to a data platform based on a line quality inspection device preset in the industrial scene through a second preset protocol, and interfaces the local database with the data platform based on the JDBC technology and the database open interface to acquire the image type data in the data platform.
It should be noted that, in the embodiment of the present application, the first preset protocol is as follows: OPC, mudbus, MQTT, etc., and a second preset protocol, for example: http and https protocols. Industrial data sets such as: mysql, oracle, DB2, postgre Sql, mongoDB, PI, TDengine, etc. The time sequence type data can be used for training and evaluating equipment fault diagnosis type and residual life prediction type mechanism models, and the image type data can be used for testing and evaluating image classification and target detection type image models.
102. And determining the data type corresponding to the industrial data, and determining the arrangement sequence of a plurality of pre-packaged common algorithm computing components according to the data processing logic corresponding to the data type of the industrial data.
Specifically, in one embodiment of the present application, a server analyzes a plurality of industrial data in an industrial data set, determines data types corresponding to the plurality of industrial data, determines data processing logic corresponding to the industrial data according to the data types corresponding to the industrial data, then determines a plurality of common algorithm computing components corresponding to the industrial data among a plurality of common algorithm computing components pre-packaged according to the data processing logic corresponding to the industrial data, and determines an arrangement sequence among the plurality of common algorithm computing components corresponding to the industrial data according to the data processing logic corresponding to the industrial data.
103. And splicing a plurality of common algorithm computing components corresponding to the industrial data according to the arrangement sequence in a dragging mode in the visual interface, and preprocessing the industrial data in sequence through the spliced plurality of common algorithm computing components.
Specifically, in one embodiment of the present application, a server reads industrial data in an industrial data set, parses the industrial data in the industrial data set to obtain a corresponding parsing result, further determines a data quality analysis report corresponding to the industrial data according to the parsing result corresponding to each industrial data, and determines a plurality of industrial data in the data quality analysis report, where the data quality meets a quality standard. It should be noted that, the data quality analysis report in the embodiment of the present application includes analysis results of data accuracy, uniqueness and integrity.
Then, the server splits the plurality of industrial data according to a preset splitting ratio and in a random sampling mode so as to split the plurality of industrial data into a training data set corresponding to the splitting ratio and a test data set corresponding to the splitting ratio. It should be noted that random sampling refers to forming a training data set by randomly extracting sample data from an acquired original industrial data set, in random sampling, each data sample has a selected opportunity, and the data samples are mutually independent and have no relevance, so that the representativeness of the training data can be ensured, and meanwhile, the training effect of a subsequent model construction process can be improved.
In one embodiment of the application, the drag trigger of the server user in the visual interface is used for splicing a plurality of common algorithm computing components required by the corresponding data processing logic of the industrial data according to the determined arrangement sequence, then the parameters of the computing components are configured, clicking operation is performed after the configuration is completed, the system can automatically allocate computing resources, the steps are sequentially executed according to the splicing sequence of the computing components, and finally the preprocessing process of the industrial data is completed.
104. And carrying out dimension transformation on the preprocessed industrial data based on the pre-packaged analysis algorithm computing component, and extracting key features from the industrial data after dimension transformation based on the pre-packaged feature computing component.
Specifically, in one embodiment of the present application, the server performs data analysis on the preprocessed industrial data based on a pre-packaged analysis algorithm computing component, and determines a data dimension corresponding to each of the preprocessed industrial data, respectively. It should be noted that, the analysis algorithm calculating component in the embodiment of the present application at least includes: the system comprises a principal component analysis algorithm calculation component, a linear discriminant analysis algorithm calculation component and a factor analysis algorithm calculation component.
And then, the server determines at least one piece of industrial data with the data dimension larger than a preset dimension threshold value, and performs dimension reduction processing on the at least one piece of industrial data respectively so as to realize dimension transformation of the industrial data. It should be noted that, dimensional transformation of industrial data generally refers to conversion from high-dimensional data to low-dimensional data, and from complex data to simple data; the industrial production process is quite complex, the collected data set usually comprises a large number of multidimensional features, the excessive dimensional features can prevent model searching rules, meanwhile, high correlation can exist between the features, the multiple collinearity of the data can cause weak generalization capability of the model, and the data must be subjected to dimension reduction treatment to solve the problems. The number of the characteristic attributes of the data set can be reduced through dimension reduction processing, irrelevant or redundant characteristics are removed, the number of the characteristics is reduced, the model accuracy is improved, and the model training and reasoning time is shortened. The data set is used for the subsequent model construction process after the dimension reduction treatment.
The server determines the characteristic attribute corresponding to the industrial data after dimension transformation based on the characteristic calculation component packaged in advance, and determines the influence degree of each characteristic attribute in the industrial data on the mechanism model respectively, so that the characteristic weight of the corresponding characteristic attribute in the industrial data can be determined according to the influence degree of each characteristic attribute on the mechanism model, and at least one characteristic attribute with the characteristic weight larger than a preset weight threshold is determined, so that the at least one characteristic attribute is used as a key characteristic.
It should be noted that, the basis for determining the key feature is that the feature attribute has an obvious influence on the dependent variable y, the correlation coefficient between the key feature and the input result is higher, the key feature is to extract the feature attribute most useful for the machine learning task from the original industrial data set, and the extracted key feature has the functions of reducing the dimension of the data, reducing redundancy and noise, so as to improve the generalization capability and the prediction performance of the model.
105. Based on a pre-packaged learning algorithm calculation component, a corresponding mechanism model is constructed according to key features in the dimension transformed industrial data, and based on drag trigger of each configuration information in the mechanism model, the mechanism model is debugged, so that visual construction of the mechanism model is completed.
It should be noted that the learning algorithm calculation component mainly includes two types of algorithms of supervised learning and unsupervised learning, wherein the supervised learning algorithm includes naive bayes, decision trees, gradient lifting numbers, random forests, linear regression, logistic regression, correlation vector machines, support vector machines, and BP neural network algorithms; the unsupervised learning algorithm comprises a K-Means clustering algorithm and an association rule algorithm. The algorithm is core calculation logic for the whole model construction, the core calculation logic is used in the model construction stage, the mechanism knowledge can be generally abstracted into a mathematical calculation type component, expert experience can be converted into a decision tree service recommendation type component through manual coding, the service scene faced by the mechanism model is complex, and the whole model calculation process needs to integrate the mathematical calculation type component, the decision tree service recommendation type component and the machine learning algorithm component.
In one embodiment of the application, after a server builds a corresponding mechanism model according to key features in industrial data after dimension transformation based on a pre-packaged learning algorithm calculation component, the server verifies the calculation component based on the pre-packaged model, inputs independent variable industrial data in a test data set obtained through pretreatment into the mechanism model, obtains corresponding test dependent variable industrial data, compares the similarity of the test dependent variable industrial data with preset target dependent variable industrial data, and determines the accuracy of the mechanism model according to the corresponding similarity.
In one embodiment of the application, after the server calculates the assembly based on the pre-packaged learning algorithm and constructs the corresponding mechanism model according to the key characteristics in the industrial data after dimension transformation, the test data set obtained by preprocessing can be split into a preset number of mutually exclusive subsets based on a cross-validation method, the first number of mutually exclusive subsets are used as training subsets, and the second number of mutually exclusive subsets are used as test subsets. It should be noted that, in the embodiment of the present application, the preset number is the sum of the first number and the second number, the second number selected in the present application is 1, the first number is the preset number-1, and the preset number only needs to select an appropriate value according to the actual situation.
And then, the server determines test results corresponding to the first number of training subsets and the second number of test subsets, and determines generalized error values corresponding to the mechanism model according to the average value of the specified number of test results. It should be noted that, in the embodiment of the present application, the selected specific number is 10, and those skilled in the art may determine specific numerical values of the specific number according to actual situations, which is not specifically limited in the present application.
In one embodiment of the application, the server deploying the mechanism model as an API interface includes the following operational flows: the flash is a lightweight Web application framework written by using Python, a mechanism model file obtained by training verification can be issued as an API interface by means of the flash framework built in the platform, specifically, a model service application script is created based on a model code file with training evaluation completed and introduced into the flash framework, the script designates an access path and a request mode of the model API interface, executing actions when the interface is called are written, an access IP address and a port of a model interface service are bound, and service call testing is carried out by means of a postman interface testing tool.
The above is a method embodiment of the present application. Based on the same inventive concept, the embodiment of the application also provides a mechanism model visualization construction device based on industrial data, and the structure of the mechanism model visualization construction device is shown in fig. 2.
Fig. 2 is a schematic diagram of an internal structure of a device for visualizing and constructing a mechanism model based on industrial data according to an embodiment of the present application. As shown in fig. 2, the apparatus includes:
at least one processor;
and a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to:
based on JDBC technology and database open interface, the local database is in butt joint with the industrial data set, and industrial data in the industrial data set is obtained;
determining a data type corresponding to industrial data, and determining the arrangement sequence of a plurality of pre-packaged common algorithm computing components according to data processing logic corresponding to the data type of the industrial data; the data types of the industrial data include time sequence type data and image type data;
splicing a plurality of common algorithm computing components corresponding to the industrial data according to the arrangement sequence in a dragging mode in a visual interface, and preprocessing the industrial data in sequence through the spliced plurality of common algorithm computing components;
performing dimension transformation on the preprocessed industrial data based on a pre-packaged analysis algorithm computing component, and extracting key features from the industrial data after dimension transformation based on a pre-packaged feature computing component;
based on a pre-packaged learning algorithm calculation component, a corresponding mechanism model is constructed according to key features in the dimension transformed industrial data, and based on drag trigger of each configuration information in the mechanism model, the mechanism model is debugged, so that visual construction of the mechanism model is completed.
The embodiment of the application also provides a nonvolatile computer storage medium, which stores computer executable instructions, wherein the computer executable instructions are configured to:
based on JDBC technology and database open interface, the local database is in butt joint with the industrial data set, and industrial data in the industrial data set is obtained;
determining a data type corresponding to industrial data, and determining the arrangement sequence of a plurality of pre-packaged common algorithm computing components according to data processing logic corresponding to the data type of the industrial data; the data types of the industrial data include time sequence type data and image type data;
splicing a plurality of common algorithm computing components corresponding to the industrial data according to the arrangement sequence in a dragging mode in a visual interface, and preprocessing the industrial data in sequence through the spliced plurality of common algorithm computing components;
performing dimension transformation on the preprocessed industrial data based on a pre-packaged analysis algorithm computing component, and extracting key features from the industrial data after dimension transformation based on a pre-packaged feature computing component;
based on a pre-packaged learning algorithm calculation component, a corresponding mechanism model is constructed according to key features in the dimension transformed industrial data, and based on drag trigger of each configuration information in the mechanism model, the mechanism model is debugged, so that visual construction of the mechanism model is completed.
The embodiments of the present application are described in a progressive manner, and the same and similar parts of the embodiments are all referred to each other, and each embodiment is mainly described in the differences from the other embodiments. In particular, for the apparatus and medium embodiments, the description is relatively simple, as it is substantially similar to the method embodiments, with reference to the section of the method embodiments being relevant.
The foregoing describes certain embodiments of the present application. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The devices and media provided in the embodiments of the present application are in one-to-one correspondence with the methods, so that the devices and media also have similar beneficial technical effects as the corresponding methods, and since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media are not repeated here.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. 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, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, 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.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.
Claims (10)
1. The method for constructing the mechanism model visualization based on the industrial data is characterized by comprising the following steps of:
based on JDBC technology and a database open interface, docking a local database with an industrial data set and acquiring industrial data in the industrial data set;
determining a data type corresponding to industrial data, and determining the arrangement sequence of a plurality of pre-packaged common algorithm computing components according to data processing logic corresponding to the data type of the industrial data; the data types of the industrial data comprise time sequence type data and image type data;
splicing a plurality of common algorithm computing components corresponding to the industrial data according to the arrangement sequence in a dragging mode in a visual interface, and preprocessing the industrial data in sequence through the spliced plurality of common algorithm computing components;
performing dimension transformation on the preprocessed industrial data based on a pre-packaged analysis algorithm computing component, and extracting key features from the industrial data after dimension transformation based on a pre-packaged feature computing component;
based on a pre-packaged learning algorithm calculation component, a corresponding mechanism model is constructed according to key features in the industrial data after dimension transformation, and based on drag trigger of each configuration information in the mechanism model, the mechanism model is debugged to complete visual construction of the mechanism model.
2. The method for constructing the visualization of the mechanism model based on the industrial data according to claim 1, wherein the method for constructing the visualization of the mechanism model based on the JDBC technology and the database open interface, docking the local database with the industrial data set, and obtaining the industrial data in the industrial data set, specifically comprises:
acquiring time sequence type data in an industrial scene to an internet of things center based on a sensor and production equipment preset in the industrial scene through a first preset protocol, and docking a local database with the internet of things center based on a JDBC technology and a database open interface to acquire the time sequence type data in the internet of things center;
and acquiring the image class data in the industrial scene to a data center station based on production line quality inspection equipment preset in the industrial scene through a second preset protocol, and docking a local database with the data center station based on a JDBC technology and a database open interface so as to acquire the image class data in the data center station.
3. The method for constructing the visualization of the mechanism model based on the industrial data according to claim 1, wherein the determining the data type corresponding to the industrial data and determining the arrangement sequence of a plurality of pre-packaged commonly used algorithm computing components according to the data processing logic corresponding to the data type of the industrial data specifically comprises:
analyzing a plurality of industrial data in the industrial data set, respectively determining data types corresponding to the industrial data, and determining data processing logic corresponding to the industrial data according to the data types corresponding to the industrial data;
and determining a plurality of common algorithm computing components corresponding to the industrial data in a plurality of pre-packaged common algorithm computing components according to the data processing logic corresponding to the industrial data, and determining the arrangement sequence among the plurality of common algorithm computing components corresponding to the industrial data according to the data processing logic corresponding to the industrial data.
4. The visual construction method of a mechanism model based on industrial data according to claim 1, wherein the preprocessing of the industrial data by the plurality of spliced common algorithm computing components sequentially comprises the following steps:
reading the industrial data in the industrial data set, and analyzing the industrial data in the industrial data set to obtain a corresponding analysis result;
according to the analysis result corresponding to each industrial data, determining a data quality analysis report corresponding to the industrial data, and determining a plurality of industrial data of which the data quality accords with the quality standard in the data quality analysis report;
splitting the plurality of industrial data according to a preset splitting proportion and in a random sampling mode so as to split the plurality of industrial data into a training data set corresponding to the splitting proportion and a test data set corresponding to the splitting proportion.
5. The method for constructing the mechanism model visualization based on the industrial data according to claim 1, wherein the pre-packaged analysis algorithm calculation component performs dimension transformation on the preprocessed industrial data, and specifically comprises the following steps:
based on a pre-packaged analysis algorithm calculation component, carrying out data analysis on the preprocessed industrial data, and respectively determining the data dimension corresponding to each preprocessed industrial data;
wherein the analysis algorithm calculation component comprises at least: a principal component analysis algorithm calculation component, a linear discriminant analysis algorithm calculation component, and a factor analysis algorithm calculation component;
at least one piece of industrial data with the data dimension larger than a preset dimension threshold is determined, and dimension reduction processing is carried out on the at least one piece of industrial data respectively so as to realize dimension transformation of the industrial data.
6. The method for constructing the visualization of the mechanism model based on the industrial data according to claim 1, wherein the pre-packaged feature calculation component is used for extracting key features from the industrial data after dimension transformation, and specifically comprises the following steps:
based on a pre-packaged feature calculation component, determining feature attributes corresponding to the industrial data after dimension transformation, and determining the influence degree of each feature attribute in the industrial data on a mechanism model respectively;
and determining the characteristic weight of the corresponding characteristic attribute in the industrial data according to the influence degree of each characteristic attribute on the mechanism model, and determining at least one characteristic attribute with the characteristic weight greater than a preset weight threshold value, wherein the at least one characteristic attribute is used as a key characteristic.
7. The method for constructing a mechanism model visualization based on industrial data according to claim 1, wherein after the pre-packaged learning algorithm calculation component constructs a corresponding mechanism model according to key features in the industrial data after dimension transformation, the method further comprises:
based on a pre-packaged model verification computing component, inputting independent variable industrial data in a test data set obtained by preprocessing into the mechanism model, and obtaining corresponding test dependent variable industrial data;
and comparing the similarity between the test dependent variable industrial data and preset target dependent variable industrial data, and determining the accuracy of the mechanism model according to the corresponding similarity.
8. The method for constructing a mechanism model visualization based on industrial data according to claim 1, wherein after the pre-packaged learning algorithm calculation component constructs a corresponding mechanism model according to key features in the industrial data after dimension transformation, the method further comprises:
based on a cross validation method, splitting the test data set obtained by preprocessing into a preset number of mutually exclusive subsets, taking the first number of mutually exclusive subsets as training subsets, and taking the second number of mutually exclusive subsets as test subsets; wherein the preset number is the sum of the first number and the second number;
and determining test results corresponding to the first number of training subsets and the second number of test subsets, and determining generalized error values corresponding to the mechanism model according to the average value of the specified number of test results.
9. A device for constructing a mechanism model visualization based on industrial data, the device comprising:
at least one processor;
and a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the industrial data based mechanism model visualization building method of any one of claims 1-8.
10. A non-transitory computer storage medium storing computer-executable instructions, the computer-executable instructions configured to:
the industrial data-based mechanism model visualization construction method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311170010.5A CN117195725A (en) | 2023-09-12 | 2023-09-12 | Mechanism model visual construction method, equipment and medium based on industrial data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311170010.5A CN117195725A (en) | 2023-09-12 | 2023-09-12 | Mechanism model visual construction method, equipment and medium based on industrial data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117195725A true CN117195725A (en) | 2023-12-08 |
Family
ID=89001199
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311170010.5A Pending CN117195725A (en) | 2023-09-12 | 2023-09-12 | Mechanism model visual construction method, equipment and medium based on industrial data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117195725A (en) |
-
2023
- 2023-09-12 CN CN202311170010.5A patent/CN117195725A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639344B (en) | Vulnerability detection method and device based on neural network | |
| US11645548B1 (en) | Automated cloud data and technology solution delivery using machine learning and artificial intelligence modeling | |
| CN110490304B (en) | Data processing method and device | |
| EP3701403B1 (en) | Accelerated simulation setup process using prior knowledge extraction for problem matching | |
| Mayer et al. | Ontologies for describing the context of scientific experiment processes | |
| Zhu et al. | A deep multimodal model for bug localization | |
| Rajbahadur et al. | Pitfalls analyzer: quality control for model-driven data science pipelines | |
| CN117435505A (en) | Visual generation method of performance test script | |
| WO2020146460A2 (en) | Apparatus, system and method for developing industrial process solutions using artificial intelligence | |
| CN117195725A (en) | Mechanism model visual construction method, equipment and medium based on industrial data | |
| CN115830419A (en) | Data-driven artificial intelligence technology evaluation system and method | |
| Singh et al. | Using graph neural networks to model the performance of deep neural networks | |
| US11669753B1 (en) | Artificial intelligence system providing interactive model interpretation and enhancement tools | |
| CN113139332A (en) | Automatic model construction method, device and equipment | |
| Duggan et al. | Model testing | |
| US11797577B2 (en) | Smart data warehouse for cloud-based reservoir simulation | |
| Room | Data Science | |
| US20240104429A1 (en) | Model-Agnostic System for Automatic Investigation of the Impact of New Features on Performance of Machine Learning Models | |
| CN117435901B (en) | Industrial Internet data acquisition method, system, terminal and storage medium | |
| EP3905027A1 (en) | Method and system for generating engineering designs in an engineering system | |
| Dissanayake | THE ROLE OF MACHINE LEARNING IN SERVICE ORIENTED ARCHITECTURE AND BUSINESS PROCESS IMPROVEMENT | |
| US20230368086A1 (en) | Automated intelligence facilitation of routing operations | |
| US20240104394A1 (en) | Platform for Automatic Production of Machine Learning Models and Deployment Pipelines | |
| de Proença Pinho | Workflow process definition in a Cyber Physical Production System | |
| Forell et al. | Modeling meets Large Language Models |
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 |