CN114638160A - Knowledge service method for complex equipment digital twin model - Google Patents
Knowledge service method for complex equipment digital twin model Download PDFInfo
- Publication number
- CN114638160A CN114638160A CN202210250088.7A CN202210250088A CN114638160A CN 114638160 A CN114638160 A CN 114638160A CN 202210250088 A CN202210250088 A CN 202210250088A CN 114638160 A CN114638160 A CN 114638160A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- model
- data
- digital twin
- complex equipment
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a knowledge service method for a complex equipment digital twin model, and belongs to the technical field of complex equipment knowledge engineering. The method aims at the knowledge application requirements in the digital twin of complex equipment, and discloses a knowledge classification body model construction method for complex equipment, knowledge mapping based on a complex equipment element structure tree, a knowledge service mechanism method for twin model dynamic calling, a knowledge service algorithm for realizing digital twin model dynamic calling, a dynamic request of real-time physical information data of equipment nodes for knowledge, and support for equipment abnormity prediction, fault diagnosis, performance prediction and the like according to the principles of loose coupling and extensible quoting of knowledge and a digital twin model.
Description
Technical Field
The invention relates to the technical field of knowledge engineering of complex equipment.
Background
Products in the fields of electric power, engineering machinery, rail transit and the like are typical complex equipment which are safely and reliably used under long life cycle and complex working conditions, the research and development process is a complex system engineering, and the knowledge engineering is an important means for improving the design, manufacturing, operation and maintenance efficiency of complex products and constructing an optimized iterative system of the products. However, due to the complex structure and composition, long life cycle and many involved personnel and professional fields of the products, the whole life cycle knowledge of the complex equipment has dispersity, diversity and inconsistency, and challenges are brought to acquisition, fusion and reuse of the knowledge.
Chinese patent publication No. CN201911163211.6 discloses a digital twin-driven complex equipment fault prediction method and a digital twin-driven complex equipment fault prediction device, which comprises the steps of establishing a digital twin body, calibrating, generating fault data, training a fault prediction model and applying. The method only aims at fault prediction, and a fault prediction model and a digital twin model are tightly coupled, so that the dynamic evolution of a fault prediction algorithm and the digital twin model cannot be effectively supported.
The invention provides a knowledge service method for a complex equipment digital twin model, which comprises the steps of constructing a complex equipment knowledge classification body model, mapping knowledge based on a complex equipment element structure tree, designing a knowledge service mechanism for realizing dynamic calling of the digital twin model, realizing dynamic request of real-time physical information data of equipment nodes to knowledge, and providing support for equipment abnormity prediction, fault diagnosis, performance prediction and the like.
Disclosure of Invention
The invention aims to provide a knowledge service method for a complex equipment digital twin model, which can effectively solve the problems of knowledge dispersity, diversity and immediacy of complex equipment and realize efficient organization and intelligent service of knowledge.
The purpose of the invention is realized by the following technical scheme: 1. a knowledge service method facing a complex equipment digital twin model comprises five steps:
step one, knowledge source and knowledge acquisition. According to the knowledge acquisition means, the full-life-cycle knowledge of the complex equipment is divided into two categories, namely the knowledge integrated with the existing business system and the knowledge generated by data mining, wherein the knowledge can be dynamically acquired by a data increment integration tool, and the knowledge can be further divided into the following categories: the method comprises the following steps of continuously acquiring knowledge of a data analysis processing model, association rule knowledge contained in data and external knowledge obtained through data crawling, and continuously acquiring the knowledge through dynamic registration of the data analysis processing model, operation of a data mining algorithm and knowledge crawling.
Step two: extracting knowledge characteristics and classifying and representing. The method comprises the steps of researching knowledge characteristics and knowledge service elements, dividing knowledge required to be called by a digital twin model into standard knowledge, rule knowledge, metadata (meta-model) knowledge and data analysis processing model knowledge according to the characteristics and the service elements, abstracting the knowledge, establishing ontology models of the knowledge, and uniquely identifying and quoting the knowledge through knowledge identification, wherein the data analysis processing model knowledge operates independently, so that an operation container of the data analysis processing model knowledge is used as a necessary characteristic during characteristic extraction.
And step three, mapping the knowledge based on the equipment structure tree. Aiming at the diversity of types and the diversity of knowledge of a complex equipment system, a subsystem, a component and a part, a complex equipment product meta-structure tree is constructed, a meta-structure tree body is mainly used for rapidly indexing a digital twin model and knowledge, the problem that each node has different knowledge of a complex equipment product, a system, a subsystem, a component and a part multi-level node is solved, for standard knowledge, rule knowledge, metadata (meta-model) knowledge and data analysis processing model knowledge, the relation between the meta-structure tree body and the meta-structure tree node and the multiplexing relation of specific equipment example nodes are firstly established from a body layer, the bodies are aligned on the basis of the meta-structure tree node, knowledge data are mapped and aligned on the basis of structure tree node codes, and the reference examples can be traced through the knowledge.
And step four, knowledge service mechanism. Aiming at standard knowledge, rule knowledge, metadata (meta-model) knowledge and data-driven analysis processing model knowledge, a knowledge service mechanism model based on an equipment structure tree node and a digital twin model is respectively constructed, dynamic calling relations among the meta-structure tree node, the data twin model, model parameters, an accessed physical information data source and various kinds of knowledge are configured, and loose coupling calling relation information between the knowledge and the digital twin model is established. Wherein: the standard class knowledge is used for verifying the validity of certain parameter real-time data of the digital twin model, and the rule class knowledge and the metadata class knowledge are used for calling the execution rule of the digital twin model algorithm; and the data analysis processing model class knowledge dynamically calls the service container through the time window and the physical information data.
And step five, knowledge service algorithm of physical information data and configuration drive. Aiming at an actual scene, acquiring tree node codes of example equipment and digital twin model codes, inquiring various knowledge reference relations, acquiring knowledge required by the current scene, loading the knowledge into a memory, establishing twin parameters or algorithm calling realization algorithms for standard knowledge and rule knowledge, establishing communication with model container examples through REST requests for data analysis and processing models, and acquiring and analyzing operation results through parameter data and container address requests to realize visual interaction with the models.
The invention has the following beneficial effects:
the invention provides a knowledge service method for a digital twin model of complex equipment, which is used for carrying out classification research on the knowledge of the complete life cycle of the complex equipment, constructing ontology models of various kinds of knowledge, mapping the knowledge based on tree nodes of an equipment element structure and designing a knowledge service mechanism for dynamic calling of the twin model, and designing and realizing a dynamic calling service algorithm of standard knowledge, rule knowledge, metadata (meta-model) knowledge and data analysis processing model knowledge on the basis, thereby realizing the loose coupling of the knowledge and the twin model and facilitating the dynamic expansion of the knowledge and the twin model. And the real-time knowledge service driven by physical information data is provided for equipment abnormity prediction, fault diagnosis, performance prediction and other services.
Drawings
FIG. 1 is a knowledge map of a knowledge source of a particular complex equipment based on a meta-structure tree in accordance with the present invention;
FIG. 2 is a diagram of a service mechanism for a particular class of standard knowledge in accordance with the present invention;
FIG. 3 is a diagram of the service mechanism of the rule class knowledge embodied in the invention;
FIG. 4 is a diagram of a service mechanism embodying knowledge of the metadata (meta-model) class of the present invention;
FIG. 5 is a service mechanism diagram of the data-driven analysis processing model class knowledge embodying the present invention;
FIG. 6 is a flowchart of a specific standard class knowledge dynamic calling algorithm of the present invention;
FIG. 7 is a second flowchart of the dynamic standard class knowledge invocation algorithm of the present invention;
FIG. 8 is one of the flow charts of the rule class knowledge dynamic calling algorithm of the present invention;
FIG. 9 is a second flowchart of the dynamic rule-based knowledge calling algorithm of the present invention;
FIG. 10 is a flowchart of a dynamic invocation algorithm of a data analysis processing model according to an embodiment of the present invention;
FIG. 11 is a second flowchart of the dynamic invocation algorithm of the data analysis processing model according to the present invention.
Detailed Description
In order to make the technical solutions in the present invention more clearly understood by those skilled in the art, the technical solutions in the embodiments of the present invention are further described below with reference to the drawings in the embodiments of the present invention.
1. FIG. 1 is a meta-structure tree based knowledge map of a knowledge source of a specific complex equipment according to the present invention, which is described in detail below:
according to the knowledge acquisition means, the full-life-cycle knowledge of the complex equipment is divided into two categories, namely the knowledge integrated with the existing business system and the knowledge generated by data mining, wherein the knowledge can be dynamically acquired by a data increment integration tool, and the knowledge can be further divided into the following categories: the method comprises three types of knowledge of a data analysis processing model, knowledge of association rules contained in data, and external knowledge obtained through data crawling, and comprises dynamic registration of the data analysis processing model, operation of a data mining algorithm and knowledge crawling. Because the complex equipment product has a complex structure, the knowledge organization adopts the meta-structure tree to map the knowledge, each knowledge corresponds to 1 or more meta-structure tree nodes, each meta-structure tree node can correspond to a plurality of knowledge, and the knowledge can be navigated and indexed by the equipment structure tree nodes.
2. FIG. 2 is a service mechanism of the standard class knowledge of the present invention, which is described in detail as follows:
(1) the characteristics are as follows: the criteria may be expressed by criteria terms, which are described by parameters and expressions and value ranges:
StandItemSet＝{SItemi(encoding, name, parameter set, expression, value range, method) | i ═ 1,2, … }, where the parameters may be 1 or more, and for the standard of the text type, the expression and value range may be empty, and the parameters are directly described by the method.
(2) The service mechanism is as follows: the standard class knowledge service can be divided into two steps: step 1-reference to standard terms: the digital twin model needs to reference a standard item through parameter mapping, and establishes a mapping AddRefer (@ SItem) between the standard item and the twin modeli,@TreeItemID,@TweenModelID,@Paraj'Stand'), persisting the mapping relationship in a standref, wherein: @ SItemiRepresents a standard term, @ TreeIteID represents a current equipment structure tree node, @ TwenModeld represents a digital twin model ID, @ ParajThe expression parameter, 'Stand' expresses that the knowledge reference type is a standard term. Step 2-dynamic invocation of the twin model on the standard: when the digital twin model works, the value range is judged according to the actual parameter value calculation expression value, whether the return meets the standard or not is judged, and different execution expressions are called to return the result.
3. FIG. 3 is a service mechanism of rule class knowledge according to the present invention, which is described in detail as follows:
(1) the characteristics are as follows: expressions and corresponding logical representations can be determined and executed by conditions: rule set ═ Rulei(encoding, name, parameter set, decision execution set (conditional expression, execution expression)) | i ═ 1,2, … }. Wherein: determining whether a set of expressions is a conditional expression or an executionWhen the logic pair formed by the line expressions is executed, the logic pair is analyzed into an IF conditional expression 1THEN execution expression 1; IF conditional expression 2THEN performs expressions 2, …
(2) The service mechanism is as follows: the rule-class knowledge service can be divided into two steps: step 1-rule reference: the digital twin model refers to a Rule item through parameter mapping, and establishes mapping AddRefer (@ Rule) between the Rule item and the twin modeli,@TreeItemID,@TweenModelID,@algorij,@Parak[]'Rule'), persisting the mapping relationships in a ruleref, wherein: @ RuleiRepresenting rules, where @ TreeItmID represents the current equipment structure tree node, where @ TwennModel ID represents the digital twin model ID, and where @ algorijExpress algorithm @ Parak[]Parameter values that the algorithm passes to the rules are represented, 'Rule' representing that the knowledge reference type is a Rule. Step 2-dynamic calling of twin model to rules: when the digital twin model works, different execution expressions are called to return a result according to an actual parameter calling rule and a judgment execution set (a conditional expression and an execution expression).
4. FIG. 4 is a service mechanism of metadata (meta-model) knowledge according to the present invention, which is described in detail as follows:
(1) the characteristics are as follows: such knowledge may describe a specification in terms of elements and relationships between elements: MetaModelSet ═ MetaMi(encoding, name, element set, relationship between element sets) | i ═ 1,2, …) } the meta-model can be changed by changing the relationship between elements and elements, thereby changing the associated model.
(2) The service mechanism is as follows: the service object is a product model, the configuration is called and instantiated through the algorithm of the model, the configuration is generally completed when the model is initialized, and the configured attribute value is interactively analyzed by the model. Its configuration information is passed through AddRefer (@ MetaM)iAnd @ treeleteid @ tweenodeld, 'Meta'), persisting the mapping relationship in MetaRefer, wherein: @ MetaMiRepresenting metadata encoding, @ treeleteideld representing the current equipment structure tree node, @ tweenodeld representing the digital twin model ID, 'Meta' representing the knowledge reference type as metadata.
5. FIG. 5 is a service mechanism of the data-driven analysis processing model class knowledge according to the present invention, which is described in detail as follows:
(1) the characteristics are as follows: such data-driven analytical processing models can be expressed as: DD (DD) with high heat dissipating capacityModelSet＝{DDModeli(code, name, tagged dataset, run container, input dataset, return value) | i ═ 1,2, … }. Wherein: the labeled data set can be empty for unsupervised machine learning or self-defining algorithms, the operation container is the operation program and the environment depended on by the model, the input data set is a part physical data set organized according to time windows, and the return value is a processed value, such as returning a classification result for a classification model and returning a calculation value for a calculation model.
(2) The service mechanism is as follows: the model is a reusable model of the same type of parts and components in the same field obtained through clustering, and is characterized in that a service container is packaged and stream data is accessed and configured, multi-instantiation service of the model is realized, high-frequency big data of an equipment field can be subjected to parallel scheduling of the model through caching, a distributed flow engine and multi-instance multi-task execution, and the model is generally used together with metadata. This type of knowledge service can be divided into two steps: step 1-reference of analytical processing model: the digital twin model needs to establish the mapping AddRefer (@ DDModel) between the model and the analysis processing modeli,@TreeItemID, @TweenModelID,@Parak[],TimeRate,@Metak'DModel'), persisting the mapping relationships in DModelRefer, wherein: @ DDModeliRepresents an analysis processing model ID, @ TreeIteID represents a current equipment structure tree node, @ TwenModeld represents a digital twin model ID, @ Parak[]The input parameter value is shown when the analysis processing model is called, and the TimeRate is the set time frequency; @ MetakRepresenting the processing metadata corresponding to the different model return values. Step 2-dynamic calling of twin model to rules: and when the digital twin model works, calling a data analysis processing model according to the data set of the actual parameters as parameters.
6. Fig. 6 and 7 are flowcharts of the standard class knowledge dynamic calling algorithm of the present invention, which are specifically described as follows:
the algorithm is divided into two parts, namely a parameter real-time data processing thread is obtained and initialized according to the current scene, and the threads are processed according to a set time window to obtain parameter data and dynamically call the standard. In order to improve the execution efficiency of the algorithm and reduce the read-write times of a disk, the algorithm loads the rules quoted by the current model into a memory at one time.
FIG. 6 dynamic call initialization of twin model to standard
Triggering conditions are as follows: trigger when entering digital twin scene @ TreeIteID @ TwenModeld @
Precondition: standard item reference configured
Transaction processing procedure
FIG. 7 Standard @ SItem driven by the thread execution parameter @ Para real-time dataiDynamic invocation
Triggering conditions are as follows: the thread @ Para is started and executed according to the parameter @ Para time configuration interval
Precondition: the @ Para thread has been launched
Thread execution process
7. Fig. 8 and 9 are flowcharts of the rule-class knowledge dynamic calling algorithm of the present invention, which are specifically described as follows:
the algorithm is divided into two parts, namely a real-time data processing thread for calling the rule by the algorithm is obtained and initialized according to the current scene, and the threads are processed according to the set time window to call the algorithm, and the rule is called according to the actual calculated value in the algorithm execution process. In order to improve the execution efficiency of the algorithm and reduce the read-write times of a disk, the algorithm loads the rules quoted by the current model into a memory at one time.
FIG. 8 initialization of rule dynamic Call Algorithm execution Environment by twin model
Triggering conditions are as follows: trigger when entering digital twin scene @ TreeIteID @ TwenModeld @
Precondition: rule item reference configured
Transaction processing procedure
FIG. 9 the Algor by the thread execution AlgorithmiDynamic calling Rule @ Rulei
The triggering condition is as follows: @ algoriThread starts, according to the algorithm @ algoriTime allocation interval execution
Precondition: @ algoriHas started
Thread execution process
8. Fig. 10 and fig. 11 are flowcharts of the data analysis processing model knowledge dynamic calling algorithm of the present invention, which are specifically described as follows:
the algorithm is divided into two parts, namely a calling thread for acquiring and initializing a data analysis processing model according to a current scene, and two parts, namely each thread acquires real-time data, requests the model through an RESTful interface, and calculates a return result through inputting a data set.
FIG. 10 data analysis processing model call management initialization algorithm
The triggering condition is as follows: trigger when entering digital twin scene @ TreeIteID @ TwenModeld @
Precondition: analytics processing model reference configured
Initialization process
FIG. 11 sub-algorithm 3-2: analysis processing model calling thread @ DDModeliExecuting an algorithm
Triggering conditions are as follows: the thread is started and requests @ DDModel according to a set time intervali
Precondition: request @ DDModeliHas started
Thread execution process
Claims (6)
1. A knowledge service method facing a complex equipment digital twin model comprises the following steps:
step one, knowledge source and knowledge acquisition
The method comprises the following steps of dividing the full life cycle knowledge of the complex equipment into two categories, namely the knowledge integrated with the existing service system and the knowledge generated by data mining, wherein the former is dynamically acquired by a data increment integration tool, and the latter is further divided into three categories: the method comprises the following steps that data analysis processing model knowledge, association rule knowledge contained in data, external knowledge obtained through data crawling are continuously obtained through dynamic registration of a data analysis processing model, operation of a data mining algorithm and data crawling;
step two: knowledge feature extraction and classification representation
Dividing knowledge to be called of the digital twin model into standard knowledge, rule knowledge, metadata knowledge and data analysis processing model knowledge according to characteristics and service elements; respectively abstracting the knowledge, establishing ontology models of various kinds of knowledge, and enabling the ontology models to become the uniquely identifiable and referenceable knowledge through knowledge identification; the data analysis processing model knowledge operates independently, so that an operating container of the data analysis processing model knowledge is reserved as a necessary feature during feature extraction;
step three, knowledge mapping based on complex equipment structure tree
Aiming at the diversity of types and the diversity of knowledge of a complex equipment system, a subsystem, a component and a part, constructing a meta-structure tree of a complex equipment product, wherein a meta-structure tree body is used for rapidly indexing a digital twin model and knowledge, solving the problem that the complex equipment product, the system, the subsystem, the component and the part have multi-level nodes, and each node has different knowledge problems;
step four, knowledge service mechanism
Respectively constructing a knowledge service mechanism model based on a complex equipment structure tree node and a digital twin model aiming at standard class knowledge, rule class knowledge, metadata or meta model class knowledge and data-driven analysis processing model class knowledge, configuring dynamic calling relations among the meta structure tree node, the data twin model, model parameters, an accessed physical information data source and various kinds of knowledge, and establishing loose coupling calling relation information between the knowledge and the digital twin model; wherein: the standard class knowledge carries out real-time data validity verification service aiming at a certain parameter of the digital twin model, and the rule class knowledge and the metadata class knowledge call execution rules aiming at the digital twin model algorithm; dynamically calling the operation container of the data analysis processing model type knowledge through a time window and physical information data;
step five, knowledge service algorithm driven by physical information data and configuration
Aiming at an actual scene, acquiring node codes of a complex equipment structure tree and digital twin model codes in an example, inquiring various knowledge reference relations, acquiring knowledge required by the current scene, and loading the knowledge to a corresponding running container memory; establishing twin parameters or algorithm call for standard knowledge and rule knowledge; for the data analysis processing model, communication with the model operation container instance is established through the REST request, and the operation result is obtained and analyzed through the parameter data and the operation container address request, so that the visual interaction with the model is realized.
2. The knowledge service method facing the complex equipment digital twin model as claimed in claim 1, wherein: the dynamic acquisition process of the first step specifically includes:
(1) knowledge of existing business system integration: researching the full life cycle knowledge of the complex equipment, researching the extraction methods of the general standard, rule, metadata knowledge and design example knowledge of the existing system, and establishing an incremental knowledge extraction method based on the Kettle middleware;
(2) knowledge obtained by the data mining method: the knowledge is divided into three types, namely data analysis processing model knowledge, association rule knowledge contained in the data, and external knowledge obtained through data crawling, and the classification is data crawling or continuously acquired knowledge based on script framework knowledge through dynamic registration of the data analysis processing model, operation of a data mining algorithm.
3. The knowledge service method facing the complex equipment digital twin model as claimed in claim 1, wherein: step two, respectively abstracting the knowledge and establishing ontology models of various kinds of knowledge respectively comprises the following steps:
(1) extracting and representing standard class knowledge characteristics: the criteria may be expressed by criteria terms, which are described by parameters and expressions and value ranges: StandItemSet ═ SItem ═i(code, name, parameter set, expression, value range, method) | i ═ 1,2, … }; the parameters are selected to be 1 or more, the expression and the value range can be null for the text type standard, and the parameters are directly described through a method;
(2) rule class knowledge feature extraction and representation: expressions and corresponding logical representations can be determined and executed by conditions: rule set ═ Rulei(encoding, name, parameter set, decision execution set (conditional expression, execution expression))/i ═ 1,2, … }, where: judging that the execution set is a logic pair consisting of a conditional expression and an execution expression, and analyzing the logic pair into an IF conditional expression 1THEN execution expression 1 during execution; IF stripExpression 2THEN performs expression 2, …;
(3) metadata or meta-model knowledge feature extraction and representation: this type of knowledge is described by elements and relationships between elements, which are normalized as follows: MetaModelSet { MetaMi (coding, name, element set, relationship between elements) | i ═ 1,2, … }, and changing the meta-model by changing the elements and the relationship between the elements, thereby changing the associated model;
(4) extracting and representing the knowledge characteristics of the data analysis processing model class: the data analysis processing model generally operates independently depending on a deep learning framework or a certain environment, and a digital twin model is dynamically called by a data request and expressed as: DDModel set ═ DDModeli(code, name, tagged dataset, run container, input dataset, return value)/i ═ 1,2, … }; wherein: the labeled data set can be empty for unsupervised machine learning or self-defining algorithms, the operation container is the environment on which the model operates the program and the input data set is a part physical data set organized according to time windows, and the return value is a processed value, such as a classification result for a classification model and a calculation value for a calculation model.
4. The knowledge service method facing the complex equipment digital twin model as claimed in claim 1, characterized in that: step three, establishing the relationship between the ontology layer and the meta-structure tree node and the multiplexing relationship between the ontology layer and the specific equipment instance node specifically comprises the following steps:
(1) building a complex equipment element structure tree: aiming at each product type, constructing a meta-structure tree of the product according to the hierarchy of the product, the system, the subsystem, the component and the part, taking the node codes of the meta-structure tree as unique identifiers, and having one-to-many self-association among nodes, namely each node corresponds to a father node, each father node can have a plurality of child nodes, and the child nodes are associated through the father node codes;
(2) establishing mapping between knowledge and complex equipment element structure tree nodes: for various kinds of knowledge, mapping is needed according to the node codes of the meta-structure tree, each item of knowledge corresponds to 1 or more meta-structure tree nodes, each meta-structure tree node corresponds to multiple items of knowledge, and the knowledge is navigated and indexed through the complex equipment structure tree nodes.
5. The knowledge service method facing the complex equipment digital twin model as claimed in claim 1, wherein: the step four, the construction of the knowledge service mechanism model specifically comprises the following steps:
(1) standard class knowledge service mechanism: the method comprises the steps that a standard item is quoted through parameter mapping when a complex equipment digital twin model is constructed, when the model works, an expression value is calculated according to an actual parameter value to judge a value range, and whether the value range meets the standard or not is returned; the user inquires the index standard and the standard item content interactively, and can also check whether the index standard and the standard item content meet the standard through parameter value input;
(2) rule class knowledge service mechanism: the service objects are also a digital twin model and a user, wherein the calling of the twin model to the rule is usually that after a model algorithm and parameters are quoted, the model is directly called through the algorithm when working, and a corresponding execution expression is called according to the parameter values and the judgment expression values to execute and return the result; the user can generally only inquire the content of the rule according to the rule code or the name;
(3) metadata or meta-model class knowledge service mechanisms: the metadata or the meta-model service object is a product model, and is called and instantiated for configuration through an algorithm of the model;
(4) the data analysis processing model class knowledge service mechanism is as follows: the model is a reusable model of the same type of parts in the same field obtained through clustering, and is characterized in that service container encapsulation and streaming data access configuration based on time window circulation are adopted, the requirements of different environments are met through time window configuration, multi-instantiation service of the model is realized, and parallel scheduling of the model is performed on high-frequency big data of a complex equipment field through caching, a distributed flow engine and multi-instance multi-task execution.
6. The knowledge service method facing the complex equipment digital twin model as claimed in claim 1, wherein: the fifth step specifically comprises:
(1) the standard class knowledge dynamic calling service implementation method comprises the following steps:
first, a reference to establish a standard term: the complex equipment digital twin model references a standard item through parameter mapping, and establishes mapping between the standard item and the twin model, wherein a data structure corresponding to the mapping relation is as follows: AddRefer (@ SItem)i，@TreeItemID，@TweenModelID，@Paraj'Stand'), persisting the mapping relationship in a StandRefer, where @ ParajDenotes the jth parameter in the digital twin model refers to @ SItemi(ii) a Second, dynamic invocation of the twin model to the standard: when the complex equipment digital twin model works, the expression value is calculated according to the actual parameter value to judge the value range, whether the value range meets the standard or not is returned, and different execution expressions are called to return the result;
(2) the method for realizing the service by dynamically calling the rule knowledge comprises the following steps:
first, a reference to establish a rule: the complex equipment digital twin model firstly refers to a rule item through parameter mapping, and establishes mapping between the rule item and the twin model, wherein a data structure corresponding to the mapping relation is as follows: AddRefer (@ Rule)i，@TreeItemID，@TweenModelID，@algorij，@Parak[]'Rule'), persisting the mapping relationships in a ruleref; in this expression, @ Parak[]The term of]The representing parameter has a plurality of parameters and is an array structure for storing parameter sets; the subscript k denotes the kth parameter reference Rule @ Rule in the digital twin modeli；
Second, dynamic invocation of rules by the twin model: when the complex equipment digital twin model works, calling rules according to actual parameters, executing expressions according to a judgment execution set, namely a conditional expression, and calling different execution expressions to return results;
(3) the metadata or meta-model knowledge dynamic calling service implementation method comprises the following steps: calling and instantiating configuration through an algorithm of the model, generally completing configuration during model initialization, and interactively analyzing configured attribute values by the model; the data structure of the configuration information is as follows: AddRefer (@ MetaM)iAnd @ TreeItmID, @ TwenModeldID, 'Meta') will mapThe relation persistence is stored in MetaRefer and is analyzed and executed by other algorithms;
(4) the data analysis processing model class knowledge dynamic calling service implementation method comprises the following steps: first, a reference for establishing an analysis processing model is: the complex equipment digital twin model requires to establish a mapping AddRefer (@ DDModel) between the complex equipment digital twin model and an analysis processing modeli，@TreeItemID，@TweenModelID，@Parak[]，TimeRate，@Metak'DModel'), persisting the mapping in DModelRefer; and secondly, when the complex equipment digital twin model works, calling a data analysis processing model according to a data set of actual parameters as parameters, establishing a thread initialization algorithm called by the analysis processing model based on the current scene, and establishing a thread program to collect data according to a time window and perform REST request processing between the thread program and the analysis processing model to return an analysis result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210250088.7A CN114638160B (en) | 2022-05-11 | 2022-05-11 | Knowledge service method for complex equipment digital twin model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210250088.7A CN114638160B (en) | 2022-05-11 | 2022-05-11 | Knowledge service method for complex equipment digital twin model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114638160A true CN114638160A (en) | 2022-06-17 |
| CN114638160B CN114638160B (en) | 2023-04-07 |
Family
ID=81948685
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210250088.7A Active CN114638160B (en) | 2022-05-11 | 2022-05-11 | Knowledge service method for complex equipment digital twin model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114638160B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114936294A (en) * | 2022-06-28 | 2022-08-23 | 北京龙智数科科技服务有限公司 | Automatic construction method and device of knowledge graph |
| CN114943185A (en) * | 2022-07-13 | 2022-08-26 | 北京航空航天大学 | Nondestructive acquisition method for twin data of complex equipment under extreme working conditions |
| CN115114268A (en) * | 2022-06-30 | 2022-09-27 | 北京亚控科技发展有限公司 | Tissue future state twinning method, device and equipment |
| CN116243670A (en) * | 2023-03-03 | 2023-06-09 | 南京理工大学 | Digital twinning-based electromechanical product assembly process dynamic model construction method |
| CN117829862A (en) * | 2024-03-04 | 2024-04-05 | 贵州联广科技股份有限公司 | Interconnection-based data source tracing method and system |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090187444A1 (en) * | 2007-05-11 | 2009-07-23 | Yefim Zhuk | Service knowledge map |
| CN111008502A (en) * | 2019-11-25 | 2020-04-14 | 北京航空航天大学 | Fault prediction method for complex equipment driven by digital twin |
| CN111159793A (en) * | 2020-01-20 | 2020-05-15 | 兰州理工大学 | Digital twin five-dimensional model based 3D printer modeling method and model system |
| CN111367918A (en) * | 2020-03-05 | 2020-07-03 | 西南交通大学 | Method for generating subway vehicle maintenance BOM based on EBOM |
| CN112200317A (en) * | 2020-09-28 | 2021-01-08 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Multi-modal knowledge graph construction method |
| CN112258094A (en) * | 2020-11-27 | 2021-01-22 | 西南交通大学 | Subway train performance evaluation system construction method based on digital twinning |
| CN112800668A (en) * | 2021-01-20 | 2021-05-14 | 西安科技大学 | Digital twin-based complex product operation, maintenance and manufacturing integrated overall model |
| CN113126580A (en) * | 2021-03-24 | 2021-07-16 | 东北大学 | Assembly control method and device based on digital twins and computer equipment |
| CN113221338A (en) * | 2021-04-25 | 2021-08-06 | 南京邮电大学 | Digital twin-based complex equipment system model verification method and system |
| WO2021227325A1 (en) * | 2020-05-13 | 2021-11-18 | 中国电子科技集团公司第十四研究所 | Digital twin-based production process simulation and optimization method |
| CN113761971A (en) * | 2020-06-02 | 2021-12-07 | 中国人民解放军战略支援部队信息工程大学 | Method and device for constructing target knowledge graph of remote sensing image |
-
2022
- 2022-05-11 CN CN202210250088.7A patent/CN114638160B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090187444A1 (en) * | 2007-05-11 | 2009-07-23 | Yefim Zhuk | Service knowledge map |
| CN111008502A (en) * | 2019-11-25 | 2020-04-14 | 北京航空航天大学 | Fault prediction method for complex equipment driven by digital twin |
| CN111159793A (en) * | 2020-01-20 | 2020-05-15 | 兰州理工大学 | Digital twin five-dimensional model based 3D printer modeling method and model system |
| CN111367918A (en) * | 2020-03-05 | 2020-07-03 | 西南交通大学 | Method for generating subway vehicle maintenance BOM based on EBOM |
| WO2021227325A1 (en) * | 2020-05-13 | 2021-11-18 | 中国电子科技集团公司第十四研究所 | Digital twin-based production process simulation and optimization method |
| CN113761971A (en) * | 2020-06-02 | 2021-12-07 | 中国人民解放军战略支援部队信息工程大学 | Method and device for constructing target knowledge graph of remote sensing image |
| CN112200317A (en) * | 2020-09-28 | 2021-01-08 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Multi-modal knowledge graph construction method |
| CN112258094A (en) * | 2020-11-27 | 2021-01-22 | 西南交通大学 | Subway train performance evaluation system construction method based on digital twinning |
| CN112800668A (en) * | 2021-01-20 | 2021-05-14 | 西安科技大学 | Digital twin-based complex product operation, maintenance and manufacturing integrated overall model |
| CN113126580A (en) * | 2021-03-24 | 2021-07-16 | 东北大学 | Assembly control method and device based on digital twins and computer equipment |
| CN113221338A (en) * | 2021-04-25 | 2021-08-06 | 南京邮电大学 | Digital twin-based complex equipment system model verification method and system |
Non-Patent Citations (6)
| Title |
|---|
| 丁浩宸等: "基于本体的油茶中文知识图谱构建与应用", 《世界林业研究》 * |
| 孙立: "面向边缘智能产业的工业大数据服务研究", 《信息与电脑(理论版)》 * |
| 李浩等: "数据驱动的复杂产品智能服务技术与应用", 《中国机械工程》 * |
| 江海凡等: "基于Agent的制造***建模与仿真研究", 《航空制造技术》 * |
| 管文玉等: "基于文献计量的数字孪生研究可视化知识图谱分析", 《计算机集成制造***》 * |
| 闫雪锋等: "复杂产品数字孪生的多维信息统一融合设计研究", 《机械设计与制造工程》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114936294A (en) * | 2022-06-28 | 2022-08-23 | 北京龙智数科科技服务有限公司 | Automatic construction method and device of knowledge graph |
| CN114936294B (en) * | 2022-06-28 | 2024-04-16 | 北京龙智数科科技服务有限公司 | Automatic knowledge graph construction method and device |
| CN115114268A (en) * | 2022-06-30 | 2022-09-27 | 北京亚控科技发展有限公司 | Tissue future state twinning method, device and equipment |
| CN115114268B (en) * | 2022-06-30 | 2023-08-11 | 北京亚控科技发展有限公司 | Method, device and equipment for organizing future state twinning |
| CN114943185A (en) * | 2022-07-13 | 2022-08-26 | 北京航空航天大学 | Nondestructive acquisition method for twin data of complex equipment under extreme working conditions |
| CN114943185B (en) * | 2022-07-13 | 2022-09-30 | 北京航空航天大学 | Nondestructive acquisition method for twin data of complex equipment under extreme working conditions |
| CN116243670A (en) * | 2023-03-03 | 2023-06-09 | 南京理工大学 | Digital twinning-based electromechanical product assembly process dynamic model construction method |
| CN117829862A (en) * | 2024-03-04 | 2024-04-05 | 贵州联广科技股份有限公司 | Interconnection-based data source tracing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114638160B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114638160B (en) | Knowledge service method for complex equipment digital twin model | |
| Diba et al. | Extraction, correlation, and abstraction of event data for process mining | |
| CN107315776B (en) | Data management system based on cloud computing | |
| US8756237B2 (en) | Scalable distributed processing of RDF data | |
| CN110134671B (en) | Traceability application-oriented block chain database data management system and method | |
| CN111627552B (en) | Medical streaming data blood-edge relationship analysis and storage method and device | |
| CN103425740B (en) | A kind of material information search method based on Semantic Clustering of internet of things oriented | |
| Giunchiglia et al. | Approximate structure-preserving semantic matching | |
| WO2019179408A1 (en) | Construction of machine learning model | |
| US20130030852A1 (en) | Associative Memory-Based Project Management System | |
| AU2012244271B2 (en) | Associative memory-based project management system | |
| CN111813870A (en) | Machine learning algorithm resource sharing method and system based on unified description expression | |
| Patel et al. | A novel approach to discover ontology alignment | |
| Sartipi et al. | A user‐assisted approach to component clustering | |
| CN116301760B (en) | Application Design System for Software Development | |
| Wang et al. | A Web-based CBR knowledge management system for PC troubleshooting | |
| CN112199287B (en) | Cross-project software defect prediction method based on enhanced hybrid expert model | |
| CN110471646B (en) | Method for realizing complex program logic through manual configuration | |
| Pandey et al. | Data modeling and performance analysis approach of big data | |
| Shahzad et al. | Automated Generation of Graphs from Relational Sources to Optimise Queries for Collaborative Filtering | |
| US7257455B1 (en) | System and method for collecting and representing knowledge using task-method-knowledge with structure-behavior-function in a computer system | |
| US20200401877A1 (en) | Maintaining master data using hierarchical classification | |
| KR100806523B1 (en) | Semantic information based grid management system and method for grid computing | |
| Polyvyanyy | Process querying: methods, techniques, and applications | |
| Pastorello Jr et al. | A Mechanism for Propagation of Semantic Annotations of Multimedia Content. |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |