CN113554063A - Industrial digital twin virtual and real data fusion method, system, equipment and terminal - Google Patents
Industrial digital twin virtual and real data fusion method, system, equipment and terminal Download PDFInfo
- Publication number
- CN113554063A CN113554063A CN202110713629.0A CN202110713629A CN113554063A CN 113554063 A CN113554063 A CN 113554063A CN 202110713629 A CN202110713629 A CN 202110713629A CN 113554063 A CN113554063 A CN 113554063A
- Authority
- CN
- China
- Prior art keywords
- ontology
- virtual
- data
- real data
- fusion
- 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
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 26
- 230000004927 fusion Effects 0.000 claims abstract description 107
- 238000013507 mapping Methods 0.000 claims abstract description 51
- 238000010276 construction Methods 0.000 claims abstract description 43
- 238000000034 method Methods 0.000 claims abstract description 39
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 19
- 238000009776 industrial production Methods 0.000 claims abstract description 15
- 238000000605 extraction Methods 0.000 claims description 34
- 230000006870 function Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 238000007670 refining Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 230000002085 persistent effect Effects 0.000 claims description 3
- 238000011524 similarity measure Methods 0.000 claims description 3
- 238000005266 casting Methods 0.000 description 30
- 238000005516 engineering process Methods 0.000 description 17
- 238000011160 research Methods 0.000 description 11
- 238000004519 manufacturing process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000000463 material Substances 0.000 description 5
- 238000004088 simulation Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000007499 fusion processing Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 239000007787 solid Substances 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 229910052500 inorganic mineral Inorganic materials 0.000 description 2
- 239000011707 mineral Substances 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
Abstract
The invention belongs to the technical field of industrial intelligence and discloses an industrial digital twin virtual-real data fusion method, a system, equipment and a terminal, wherein the industrial digital twin virtual-real data fusion method comprises the following steps: acquiring virtual and real data in the industrial production process; realizing the construction of a global ontology and a local ontology according to an ontology description language and an ontology construction principle; extracting semantic features of structured and unstructured virtual and real data of industrial digital twin; mapping from the local ontology to the global ontology is realized based on a similarity algorithm; and finishing the fusion of the industrial digital twin virtual and real data based on a fusion rule. The method for fusing the industrial digital twin virtual and real data provides a feasible solution for fusing the industrial digital twin virtual and real data, has higher applicability, can realize the value of the virtual and real data, promotes the sharing of the virtual and real data, reduces the repeated use of the virtual and real data, and enables the industrial digital twin to fully utilize the fused data to perform accurate prediction and reliable decision.
Description
Technical Field
The invention belongs to the technical field of industrial intelligence, and particularly relates to an industrial digital twin virtual-real data fusion method, system, equipment and terminal.
Background
At present, the core of the digital twin technology is a model and data, the model is mainly used as a carrier, the data is used as a drive, the data is subjected to fusion analysis, the behavior of a physical entity in a real environment is simulated, and valuable information in each data can be effectively utilized to really realize 'data-centered production' in an industrial production process. The industrial production process comprises the stages of product research and development, manufacturing production, equipment service and the like. For example, in the development stage, the accuracy of product design is improved through fused data; in the production phase, an optimized production decision is provided for the product; in the maintenance stage, a favorable decision basis is provided for the repair and fault prediction of the product, and the reliability and the availability of the product are effectively improved.
The data fusion in industry is called sensor fusion, and is mainly characterized in that fusion processing is carried out on data collected in sensors, and comprehensive analysis and intelligent fusion are carried out on sensor data or knowledge with different sources and semantic representation ambiguities under a certain rule, so that consistency explanation and accuracy description of a target object are obtained, and better decision information is provided for an industrial production process.
At present, data fusion technology is deeply researched at home and abroad, but relatively speaking, the research on the industrial data fusion technology at home and abroad is much earlier than at home. As early as 1970, research institutes in the united states have proposed a concept of data fusion in order to automatically detect the specific location of an enemy when the navy detects the sea area, and to provide a reliable decision for subsequent tactical formulation. Data fusion techniques are then continually being applied to a variety of fields with great success. While the research on the data fusion technology in China is relatively late, after relevant research on data fusion is carried out in the United states, a plurality of research papers related to the data fusion technology appear in the academic world from 1989. As the data fusion technology is gradually emphasized by researchers and academia, in 1990 to 1999, colleges and research units have studied the data fusion technology in various fields according to their different research directions. For example, researchers of geological mineral products begin to perform fusion analysis on geological information and remote sensing information by using a data fusion technology to improve the effect of searching for ores and minerals, positioning, identity recognition, tracking and traffic control of vehicles are realized by using the data fusion technology in the field of transportation, target recognition and tracking of multiple sensors in a digital twin body are realized by using the data fusion technology in the field of industry, and accuracy prediction of the industrial production process is improved.
As can be seen from the above studies, the development of data fusion technology has never been stopped, and both basic theory, application field, academic research and the like are the main research directions of the data fusion technology at present. The research on the data fusion technology in the industrial digital twin field is still a hot point, and as the virtual and real data in the industrial digital twin not only come from data sensed by different sensors, but also come from data in a plurality of virtual spaces, the virtual and real data comprise historical data of equipment operation, multi-source heterogeneous data such as real-time sensing data, environmental data, sensor information, virtual model information and the like, the data have different sources and different structures, so that the virtual and real data of the industrial digital twin have the characteristics of various structures, close connection and the like. Therefore, there is a need for theoretical research and technical implementation for industrial multi-isomeric data fusion.
Through the above analysis, the problems and defects of the prior art are as follows: the existing data fusion technology has the problems of multi-source isomerism and semantic isomerism of virtual and real data.
The difficulty in solving the above problems and defects is: how to provide a uniform formal expression mode for virtual and real heterogeneous data with different sources and different structures, eliminate the obstacle of multi-source heterogeneity and semantic heterogeneity in the virtual and real data fusion process, and further improve the efficiency of virtual and real data fusion is a problem existing in the current data fusion technology.
The significance of solving the problems and the defects is as follows: the ontology-based data fusion technology can solve the problems of multi-source isomerism and semantic isomerism of virtual and real data, realize the efficient application of the multi-isomerism virtual and real data in the whole life cycle process of industrial products, and provide real-time and intelligent decision optimization for industrial digital twins.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an industrial digital twin virtual-real data fusion method, system, equipment and terminal, and particularly relates to an industrial digital twin virtual-real data fusion method, system, equipment and terminal based on a body.
The invention is realized in such a way that an industrial digital twin virtual-real data fusion method comprises the following steps:
acquiring virtual and real data in an industrial production process; the method has the advantages that the concept terms of the virtual and real data are defined conveniently by introducing the industrial field library, the correctness of relation expression among concepts is ensured, and preparation is made for subsequent ontology construction.
Secondly, constructing a global ontology and a local ontology according to an ontology description language and an ontology construction principle; the established local ontology and global ontology can further reflect the essential relation and semantic relation between concepts, and the subsequent extraction of semantic features of virtual and real data and ontology mapping are facilitated.
Thirdly, extracting semantic features of structured and unstructured virtual and real data of the industrial digital twin; virtual and real data generated in an industrial production process are difficult to read by non-professional personnel without processing. Therefore, the information in the virtual and real data is usually extracted based on the concept and the instance in the local ontology, and the integrity of the data semantics can be greatly ensured by extracting the semantic features through the ontology because the local ontology expresses complete semantic information.
Step four, mapping from the local ontology to the global ontology is realized based on a similarity algorithm; the ontology is heterogeneous in different data sources or ontologies constructed by using different ontology languages, and the phenomenon can prevent sharing and interoperation of information among ontologies, so that concepts and examples describing the same object are mapped into a global ontology by adopting an ontology mapping method on the basis of eliminating inconsistency of the ontology languages, and the purpose of data communication and sharing among ontologies is realized.
And step five, completing the fusion of the industrial digital twin virtual and real data based on a fusion rule. Because the instances and attribute values describing the same object in different ontologies may be different, if the contents are not analyzed, the object description is inaccurate, and the result fed back to the user is inconsistent. Therefore, a fusion process needs to be performed on the instances of the same object in the global ontology.
Further, in step two, the implementation of the global ontology and the local ontology according to the ontology description language and the ontology construction principle includes:
(1) determining the application field and the application range of the ontology: the application range of the clearly constructed body is the industrial field;
(2) data collection and analysis: by means of collected industrial virtual and real data, concept terms of the virtual and real data are introduced into an industrial field library to ensure the correctness of relational expression among concepts;
(3) establishing an initial body: reusing or modifying the ontology meeting the condition from the ontology library or establishing an initial ontology by using an ontology description language and an ontology construction tool prot g;
(4) refining and verifying the ontology: refining and verifying the constructed ontology to ensure that the constructed ontology follows the ontology construction principle;
(5) ontology becoming and publishing: and storing the verified ontology in an ontology library in a file form.
Further, in step three, the semantic feature extraction of the structured and unstructured virtual and real data of the industrial digital twin includes:
(1) extracting semantic features of the structured data;
(2) and adopting a TF-IDF algorithm associated with the ontology to realize semantic feature extraction of the unstructured data.
Further, in step (1), extracting semantic features of the structured data includes:
1) matching table names in the database with the concept of the local ontology;
2) an acquisition example: get all records in the table and put the results into the collectionD＝{d1，d2，d3，…，dn}；
3) Instantiation: all records in the same table only correspond to one concept in a local ontology, and the data of the table is formatted and described in an xml mode;
4) and (3) persistent storage: and storing concepts obtained after semantic feature extraction is carried out on the examples into a description file, and completing semantic feature extraction on the structured data.
Further, in the step (2), the extracting semantic features of the unstructured data by using the body-associated TF-IDF algorithm includes:
1) the number of times a word w appears in a certain text is wiNext, the number of occurrences of all words in the text is tiThen, using TF-IDF method to calculate the initial weight of vocabulary w as:
2) searching the vocabulary in the local ontology, and when the vocabulary exists in the ontology, the optimized weight of the vocabulary w is represented as:
3) a weight value (W) obtained from the vocabulary W and a predetermined threshold value theta1Comparing when the weight value is larger than the threshold value theta1And taking the vocabulary as the characteristic words of the body, and putting the characteristic words into a characteristic word set V:
V＝{wi|i＝1，2，….，n}；
4) and storing the extracted feature words into a description file in an xml mode, and finishing the semantic feature extraction of the text data.
Further, in step four, the mapping from the local ontology to the global ontology based on the similarity algorithm includes:
(1) calculating the similarity between two ontology elements from the concept, relationship, attribute and example angles of the ontology according to a similarity measurement function, and comprehensively weighting the similarity of each angle to obtain a final similarity result; the similarity between two ontology elements calculated by using the similarity measure function meets the following requirements:
S(ei，ej)∈[0，1]；
when the similarity between the two ontologies is 1, the two ontology elements have completely same semantic information, and the attributes and the instance elements of the two local ontologies are mapped into a global ontology; if the similarity is 0, the semantic information between the two ontology elements is completely different.
(2) Obtaining the mapping from the local ontology to the global ontology according to the mapping process of the elements among the ontologies:
1) inputting local ontology (O)1And O2) And a global ontology (O);
2) finding elements of the local ontology and the global ontology, which have not established a mapping relation;
3) similarity calculation is carried out on the ontology elements from the aspects of concepts, attributes and examples by using a similarity measurement function, and finally, a similarity value is obtained through comprehensive weighting;
4) judging the value and the threshold value of the similarity in the body elements;
5) when the similarity value is equal to the threshold value, marking, and establishing a mapping relation for the similarity elements;
6) and repeating the steps until no element needing to establish the mapping relation exists, and converting the mapping result into a structured text form for outputting.
Further, in the fifth step, the fusion of the industrial digital twin virtual and real data is completed based on the fusion rule, which includes:
(1) single fused results
A single fused result applies to the case where the final returned result is a single numeric value or a single character. The following fusion rules are proposed for this result: an average rule, a confidence priority rule, a majority priority rule, and a weighted rule.
1) Average rule: the rule is only applicable to the case that the examples are numerical values, and for different examples describing the same object, a centralized trend expression of all data is realized, so as to reflect the average level of various data; let R be { R ═ R1，r2，r3，…，rnThe data sets from different sources are represented, and the final fusion result of the method is as follows:
2) weighting rules: when the user gives a certain weight (W) to different instances according to a specific standardi) Carrying out weighted summation on the examples of different data sources according to the weight values; let R be { R ═ R1，r2，r3，…，rnDenotes data sets of different origins, WiThe weight value given by the user to a certain data source is represented, and the final fusion result obtained by using the method is as follows:
3) confidence priority rules: the method is suitable for the case that the examples are numerical values and characters, and under the normal condition, after the data sources are sorted according to the user-defined credibility, the highest credibility is the fusion result with the most accurate result; let R be { R ═ R1，r2，r3，…，rnDenotes a data set of different origin, Conf (r)i) Representing data riThe rule is defined as:
4) majority priority rules: the method considers that the reliability of the result with a large number of occurrences in the result set is high, and assumes that R ═ R is greater1，r2，r3，…，rnDenotes a data set of different origin, the rule is defined as:
(2) multiple fused results
The multi-fusion result considers the final returned result to be a set, and the following different fusion rules are given to the result: upper and lower bound rules, or rules.
1) And (3) upper and lower limit rules: the rule contains all the results for the maximum and minimum values in the result set when rjApplying an upper and lower limit rule when the numerical example is used; let R be { R ═ R1，r2，r3，…，rnDenotes a data set of different origin, the rule is defined as:
2) or a rule: examples in the result set are single-valued or when comparison is not possible, a rule is adopted for processing data of numeric type or character type; let R be { R ═ R1，r2，r3，…，rnDenotes data sets of different origins, the rule is expressed by the following formula:
Or(R)＝{r1orr2or…orrn}。
another object of the present invention is to provide an industrial digital twinning virtual-real data fusion system using the industrial digital twinning virtual-real data fusion method, the industrial digital twinning virtual-real data fusion system comprising:
the virtual and real data acquisition module is used for acquiring virtual and real data in the industrial production process;
the ontology construction module is used for realizing the construction of a global ontology and a local ontology according to an ontology description language and an ontology construction principle;
the semantic feature extraction module is used for extracting semantic features of structured and unstructured virtual and real data of industrial digital twin;
the ontology mapping module is used for realizing the mapping from the local ontology to the global ontology based on a similarity algorithm;
and the data fusion module is used for completing the fusion of the industrial digital twin virtual and real data based on the fusion rule.
It is a further object of the invention to provide a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:
acquiring virtual and real data in the industrial production process; according to an ontology description language and an ontology construction principle, a set of ontology construction method aiming at the industrial field is provided, and global ontology and local ontology construction are realized; extracting semantic features of structured and unstructured virtual and real data of industrial digital twin;
mapping from the local ontology to the global ontology is realized based on a similarity algorithm; and finishing the fusion of the industrial digital twin virtual and real data based on a fusion rule.
Another object of the present invention is to provide an information data processing terminal for implementing the industrial digital twin virtual-real data fusion system.
By combining all the technical schemes, the invention has the advantages and positive effects that: because the data sources of the industrial digital twin virtual and real data are different, and the data sources and the semantics of the data are different in different degrees, the virtual and real data have the characteristics of various structures, close connection and the like, and effective fusion operation can be carried out only by fully analyzing the virtual and real data, so that the value of the virtual and real data is realized, the sharing of the virtual and real data is promoted, the repeated use of the virtual and real data is reduced, and the industrial digital twin can fully utilize the fused data to carry out accurate prediction and reliable decision.
The industrial digital twin virtual-real data fusion method provided by the invention has higher applicability. The method completes the construction of local ontology and global ontology in the industrial field by the provided ontology construction method, and provides a feasible solution for the fusion of industrial digital twin virtual and real data; secondly, a data semantic feature extraction method is adopted, and a uniform formal expression mode is provided for heterogeneous virtual and real data with different sources and different structures; and then, the similarity value of elements among the ontologies is calculated from multiple angles based on a similarity algorithm, so that the accuracy of ontology mapping is improved, and finally, fusion of virtual and real data is completed according to a fusion rule, so that an industrial digital twin can make full use of the fused data to perform accurate prediction and reliable decision.
The invention utilizes the ontology description language and the ontology construction method to carry out uniform formal representation on the industrial digital twin virtual and real heterogeneous data, solves the problems of multi-source heterogeneity and semantic heterogeneity of the virtual and real data, and further improves the efficiency of virtual and real data fusion. Meanwhile, the invention takes the fan hub casting process digital twin as an example for verification, realizes the unified description of virtual and real heterogeneous data in the fan hub casting process digital twin through the body-based industrial digital twin virtual and real data fusion method, completes the virtual and real data fusion of the fan hub casting process digital twin, and verifies the effectiveness of the method.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of an industrial digital twin virtual-real data fusion method provided by an embodiment of the present invention.
Fig. 2 is a schematic diagram of an industrial digital twin virtual-real data fusion method provided by an embodiment of the invention.
FIG. 3 is a block diagram of an industrial digital twin virtual-real data fusion system according to an embodiment of the present invention;
in the figure: 1. a virtual and real data acquisition module; 2. an ontology building module; 3. a semantic feature extraction module; 4. an ontology mapping module; 5. and a data fusion module.
FIG. 4 is a process diagram of an industrial digital twin body construction process provided by an embodiment of the invention.
Fig. 5 is a flow chart of semantic feature extraction of unstructured data based on ontology association provided in an embodiment of the present invention.
Fig. 6 is a flowchart of ontology mapping based on similarity calculation according to an embodiment of the present invention.
FIG. 7 is a schematic diagram of a connection between partial concepts in a global body of a hub casting process provided by an embodiment of the invention.
FIG. 8 is a view of an embodiment of a hub solid partial body portion provided in accordance with an embodiment of the present invention.
FIG. 9 is an illustration of an example of a hub phantom body portion provided in accordance with an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Aiming at the problems in the prior art, the invention provides an industrial digital twin virtual-real data fusion method, system, equipment and terminal, and the invention is described in detail with reference to the accompanying drawings.
As shown in fig. 1, the method for fusing industrial digital twin virtual-real data provided by the embodiment of the present invention includes the following steps:
s101, acquiring virtual and real data in an industrial production process;
s102, constructing a global ontology and a local ontology according to an ontology description language and an ontology construction principle;
s103, extracting semantic features of structured and unstructured virtual and real data of the industrial digital twin;
s104, mapping from the local ontology to the global ontology is realized based on a similarity algorithm;
and S105, completing fusion of the industrial digital twin virtual and real data based on a fusion rule.
The schematic diagram of the industrial digital twin virtual-real data fusion method provided by the embodiment of the invention is shown in fig. 2.
As shown in fig. 3, an industrial digital twin virtual-real data fusion system provided in an embodiment of the present invention includes:
the virtual and real data acquisition module 1 is used for acquiring virtual and real data in the industrial production process;
the ontology construction module 2 is used for realizing the construction of a global ontology and a local ontology according to an ontology description language and an ontology construction principle;
the semantic feature extraction module 3 is used for performing semantic feature extraction on the structured and unstructured virtual and real data of the industrial digital twin;
the ontology mapping module 4 is used for realizing the mapping from the local ontology to the global ontology based on a similarity algorithm;
and the data fusion module 5 is used for completing the fusion of the industrial digital twin virtual and real data based on a fusion rule.
The technical solution of the present invention will be further described with reference to the following examples.
Example 1
Aiming at the problems stated in the background technology, the invention provides an industrial digital twin virtual-real data fusion method based on a body, which adopts the following technical scheme:
(1) and acquiring virtual and actual data in the industrial production process.
(2) According to an ontology description language and an ontology construction principle, a set of ontology construction method aiming at the industrial field is provided; and the construction of a global ontology and a local ontology is realized. The construction process is shown in FIG. 4.
(2.1) determining the application field and the application range of the ontology: the application range of the clearly constructed ontology is the industrial field.
(2.2) data collection and analysis: by means of collected industrial virtual and real data, concept terms of the virtual and real data are introduced into an industrial field library to ensure the correctness of relational expression between concepts.
(2.3) establishing an initial ontology: the establishment of the initial ontology can be obtained from two aspects: on one hand, the ontology meeting the condition is reused or modified from the ontology library; another aspect is to build the initial ontology using an ontology description language and an ontology building tool prot g.
(2.4) refining and validating ontology: and refining and verifying the constructed ontology to ensure that the constructed ontology follows the ontology construction principle.
(2.5) ontology literacy and publishing: and storing the verified ontology in an ontology library in a file form so as to facilitate subsequent sharing, multiplexing and application.
(3) And performing semantic feature extraction on structured and unstructured virtual and real data in the industrial digital twin.
And (3.1) extracting semantic features of the structured data.
1) Table names in the database match the concept of the local ontology. For example: the virtual-real data N is under the field of "product line name" in the "product line information" table, and then it is mapped into the concept of "product line information" in the local ontology, and the attribute is "product line name".
2) An instance is obtained. Get all records in the table and put the result into the set D ═ D1，d2，d3，...，dn}。
3) And (6) instantiation. All records in the same table correspond to only one concept in the local ontology, and the data of the table can be formatted and described in an xml mode, as shown in table 1.
4) And (5) persistent storage. And storing the concept obtained after semantic feature extraction is carried out on the example into a description file, thus completing the semantic feature extraction on the structured data.
TABLE 1 xml formatting description
(3.2) adopting a TF-IDF (feature extraction) algorithm of ontology association to realize semantic feature extraction of the unstructured data, and as can be seen from the figure 5, the steps of realizing the semantic feature extraction of the unstructured data are as follows:
1) the number of times a word w appears in a certain text is wiNext, the number of occurrences of all words in the text is tiThen, using TF-IDF method to calculate the initial weight of vocabulary w as:
2) searching the vocabulary in the local ontology, and when the vocabulary exists in the ontology, the optimized weight of the vocabulary w can be expressed as:
3) weighting value (W and predetermined threshold value theta) obtained from vocabulary W1Comparing when the weight value is larger than the threshold value theta1Then, the vocabulary is used as the characteristic word of the ontology. And putting the words into the feature word set V.
V＝{wi|i＝1，2，…，n} (3)
4) And finally, storing the extracted feature words into a description file in an xml mode, and completing semantic feature extraction of the text data.
(4) Mapping from the local ontology to the global ontology is achieved based on a similarity algorithm.
And (4.1) calculating from multiple angles such as concepts, relations, attributes and examples of the ontology according to a similarity measurement function to obtain the similarity between two ontology elements, and then comprehensively weighting the similarity of each angle to obtain a final similarity result. The similarity between two ontology elements calculated by using the similarity measure function should satisfy the following requirements:
S(ei，ej)∈[0，1] (4)
when the similarity between the two ontologies is 1, it indicates that the two ontology elements have identical semantic information, and the attributes, examples and other elements of the two local ontologies can be mapped into the global ontology, and the mapping process is shown in fig. 6. If the similarity is 0, the semantic information between the two ontology elements is completely different.
(4.2) according to the mapping flow chart of the elements between ontologies of fig. 6, the mapping steps of the local ontology to the global ontology can be obtained as follows:
1) inputting local ontology (O)1And O2) And a global ontology (O).
2) And finding the elements which are not established with the mapping relation in the local ontology and the global ontology.
3) And performing similarity calculation on the ontology elements from multiple angles such as concepts, attributes and examples by using a similarity measurement function, and finally performing comprehensive weighting to obtain a similarity value.
4) And judging the value and the threshold value of the similarity in the body element.
5) When the similarity value is equal to the threshold value, marking is carried out, and mapping relation is established for the similarity elements.
6) And repeating the steps until no element needing to establish the mapping relation exists, and converting the mapping result into a structured text form for outputting.
(5) And finishing the fusion of the industrial digital twin virtual and real data based on a fusion rule.
(5.1) Single fusion results
A single fused result applies to the case where the final returned result is a single numeric value or a single character. The following fusion rules are proposed for this result: an average rule, a confidence priority rule, a majority priority rule, and a weighted rule.
1) Average rule: the rule can realize a centralized trend expression of all data for different examples describing the same object, and is used for reflecting various dataAverage level of (d). Let R be { R ═ R1，r2，r3，…，rnThe data sets from different sources are represented, and the final fusion result of the method is as follows:
the average rule method has certain limitation, and is only applicable to the case that the examples are numerical values, and although the final fusion result can be obtained, the importance degree of each example is not considered in the fusion process, so that the method has certain deviation on the final fusion result.
2) Weighting rules: when the user gives a certain weight (W) to different instances according to a specific standardi) The rule may sum the instances of the different data sources in a weighted manner according to the weights. Let R be { R ═ R1，r2，r3，…，rnDenotes data sets of different origins, WiThe weight value given by the user to a certain data source is represented, and the final fusion result obtained by using the method is as follows:
this method is similar to the average rule method and can only be used if the examples are numerical values, for example: the pouring time in the casting process of a certain casting is R ═ {210, 240, 265}, and the information source of the pouring time is S ═ { S ═1，s2，s3The weights given to these three information sources are 0.4, 0.35, and 0.25, respectively, and the fusion result calculated using this rule is 234.25.
3) Confidence priority rules: the method is suitable for the case that the examples are numerical values and characters, and under the normal condition, after the data sources are sorted according to the user-defined credibility, the highest credibility is the fusion result with the most accurate result. Let R be { R ═ R1，r2，r3，…，rnMean differentData set of origin, Conf (r)i) Representing data riThe reliability of (2). The rule may be defined as:
the confidence priority rule considers that it is more important for data from a more trusted source than from a less trusted source, such as: the monthly yield of a certain industrial product is R ═ {5035, 6742, 5075, 5035}, and the four information sources are S ═ { S ═ S }1，s2，s3，s4S, after the credibility of the four information sources is sorted3＞s2＞s1＞s4The final fusion result is the information source s3Corresponding result 5075. However, this method has a drawback that when data of high reliability appears only once but numerical values of low reliability appear many times, it is difficult to use data from a high-reliability information source as the most accurate data.
4) Majority priority rules: the method considers that the reliability of the result with a large number of occurrences in the result set is high, and assumes that R ═ R is greater1，r2，r3，…，rnDenotes data sets of different origins. Then the definition of the rule is:
the method is also applicable to the case where the example is a character or a numerical value, which is more general than the three rules mentioned above, such as: the pouring time in a casting process is R ═ {210, 240, 265, 240}, and the optimal pouring time can be considered to be 240(s) by using the rule. But this rule has certain drawbacks: when the results from all the information sources are inconsistent, the method can not be used, and only other rules can be used for processing.
(5.2) multiple fusion results
The final returned result is considered as a set by the multi-fusion result, and the following different fusion rules are given for the result: upper and lower bound rules, or rules.
1) And (3) upper and lower limit rules: the rule contains all the results for the maximum and minimum values in the result set. Let R be { R ═ R1，r2，r3，…，rnDenotes data sets of different origins. Then the definition of the rule is:
when r isjIn the case of numerical example, the upper and lower limit rules are applied, and assuming that the monthly yield R of a certain industrial product is {5035, 6742, 5075, 5035}, the final fusion result obtained using the rule is [5035, 6742 ]]. The rule can approximately reflect the range of the fusion result.
2) Or a rule: an instance of a general result set is a single value or a rule is applied when a comparison cannot be made. Let R be { R ═ R1，r2，r3，…，rnRepresents data sets of different origins, the rule can be expressed by the following formula:
Or(R)＝{r1orr2or…orrn} (11)
the rule can be used for processing numerical data or character data, for example, the material used in the casting process of a certain casting is R ═ { QT400-18AL, QT400}, and the result after the casting is fused by adopting or the rule is { QT400-18AL or QT400 }.
In general, any fusion rule may be used when the instance is a numerical value. However, for a single fusion result, the more used rules are a weighting rule and a reliability priority rule, and for a multi-fusion result, the upper and lower limit rules are usually used for numerical fusion. And when the example is a character, the single fusion result can adopt a majority priority rule and a credibility priority rule to perform data fusion, and the multiple fusion results can use or complete the data fusion according to the rules.
The ontology-based industrial digital twin virtual-real data fusion method provided by the invention has higher applicability. Firstly, the construction of a local ontology and a global ontology in the industrial field is completed through a proposed ontology construction method, and a feasible solution is provided for the fusion of industrial digital twin virtual and real data; secondly, semantic feature extraction is carried out on the information of the virtual and real heterogeneous data based on concepts and examples in the local ontology, then similarity values of elements among the ontology are calculated from multiple angles based on a similarity calculation method, so that the accuracy of ontology mapping is improved, and finally fusion of the virtual and real data is completed according to a fusion rule, so that industrial digital twin can make full use of the fused data to carry out accurate prediction and reliable decision.
Example 2
Referring to fig. 2, the specific implementation steps of the present invention are as follows:
step 1: and acquiring virtual and actual data of the digital twins of the fan hub casting process.
Step 2: constructing a global body and a local body of the fan hub casting process according to a body construction method and a WOL + RDFS standard body modeling language:
(2.1) describing the relation between partial concepts and attributes in the hub casting process global ontology by using an ontology modeling language, wherein the partial ontology description language of the hub casting process global ontology is shown in the table 2.
TABLE 2 OWL + RDFS ontological description results
According to an ontology construction method and an industrial manufacturing field library, an ontology editing tool (prot g) is used for extracting and establishing knowledge in the hub casting process from the concept entity, the concept attribute and the relation among the concept attributes respectively, the established global ontology can further reflect the essential relation and semantic relation among the concepts, and a foundation can be laid for sharing and reusing the knowledge in the subsequent hub casting process. The continuity between partial concepts in the hub casting process global body is shown in fig. 7.
(2.2) local body construction is carried out on the fan hub casting process in the physical space and the virtual space by adopting the body construction method and the body description language, and partial conceptual relations of the established fan hub casting process local body are shown in fig. 8 and 9.
As the fan hub casting process has digital twin data different from entities in the process of casting simulation, for example: the simulation pouring speed, the simulation pouring temperature, the simulation pouring time, the simulation mould length, the mould width and the like. Therefore, before the fan hub casting process is carried out, semantic feature extraction needs to be carried out on the virtual and real data and concepts in respective local bodies.
And step 3: extracting the language features of digital twin virtual and actual data in the fan hub casting process:
the semantic feature extraction method for the virtual and real data of the fan hub casting process digital twin is mainly characterized in that table names and attribute names in a relational database are matched with concepts and attributes in a local body, and the semantic feature extraction results of partial data in the fan hub solid body surface and the fan hub virtual body surface are shown in tables 3 and 4.
TABLE 3 semantic feature extraction Table for fan hub entity data
TABLE 4 Fan hub virtual data semantic feature extraction Table
And 4, step 4: mapping from a local body of a fan hub casting process digital twin to a hub casting process global body is realized based on a similarity algorithm;
and mapping concepts and examples describing the same object into the global ontology by a similarity calculation method according to the semantic relation extraction result. Defining a local body of a hub entity in physical space as O1The local body of the virtual hub body in the virtual space is defined as O2The global body of the hub casting process is defined as O. E.g. O1Of (1) and2the "mold length", "mold width" and "mold height" in (1) are mapped to the "size" in (O). The mapping of local to global ontologies is shown in table 5.
TABLE 5 ontology mapping table
Based on the body mapping table, attribute mapping of the hub entity table and the hub virtual body table is constructed, and the attribute mapping table is shown in table 6.
Table 6 attribute mapping table
| O1 | o2 | Mapping rules |
| hub material | simula_material | O1.hub material＝O2.simula material |
| pour_rate | simula_rate | O1.pour_rate＝O2.simula_rate |
| pour_temper | simula_temper | O1.pour_temper＝O2.simula_temper |
| pour_time | simula_time | O1.pour_time＝O2.simula_time |
| inner_area | inner_gate | O1.inner_area＝02.inner_gate |
| cross_area | cross_gate | O1.cross_area＝O2.cross_gate |
| straight_area | straight_gate | O1.straight_area＝O2.straight_gate |
| dice_size | length，width，height | O1.die_size＝O2.length+O2.width+O2.height |
And 5: completing fusion of digital twin virtual and real data of the fan hub casting process based on a fusion rule;
and (3) selecting an average rule in a single fusion result and an or rule in a multi-fusion result to complete the virtual-real data fusion of the local ontology attribute and the example according to the attribute mapping rule, wherein the fusion result is shown in a table 7.
TABLE 7 fusion results Table
By the body-based industrial digital twin virtual and real data fusion method, unified description of virtual and real heterogeneous data in the fan hub casting process digital twin is achieved, virtual and real data fusion of the fan hub casting process digital twin is completed, and effectiveness of the method is verified.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When used in whole or in part, can be implemented in a computer program product that includes one or more computer instructions. When loaded or executed on a computer, cause the flow or functions according to embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.)). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. An industrial digital twin virtual and real data fusion method is characterized by comprising the following steps:
acquiring virtual and real data in an industrial production process;
secondly, constructing a global ontology and a local ontology according to an ontology description language and an ontology construction principle;
thirdly, extracting semantic features of structured and unstructured virtual and real data of the industrial digital twin;
step four, mapping from the local ontology to the global ontology is realized based on a similarity algorithm;
and step five, completing the fusion of the industrial digital twin virtual and real data based on a fusion rule.
2. The industrial digital twin virtual-real data fusion method as claimed in claim 1, wherein in the second step, the implementation of global ontology and local ontology construction according to ontology description language and ontology construction principle includes:
(1) determining the application field and the application range of the ontology: the application range of the clearly constructed body is the industrial field;
(2) data collection and analysis: by means of collected industrial virtual and real data, concept terms of the virtual and real data are introduced into an industrial field library to ensure the correctness of relational expression among concepts;
(3) establishing an initial body: reusing or modifying the ontology meeting the condition from the ontology library or establishing an initial ontology by using an ontology description language and an ontology construction tool prot g;
(4) refining and verifying the ontology: refining and verifying the constructed ontology to ensure that the constructed ontology follows the ontology construction principle;
(5) ontology becoming and publishing: and storing the verified ontology in an ontology library in a file form.
3. The method for fusing industrial digital twin virtual-real data as claimed in claim 1, wherein in step three, the semantic feature extraction of the structured and unstructured virtual-real data of the industrial digital twin comprises:
(1) extracting semantic features of the structured data;
(2) and adopting a TF-IDF algorithm associated with the ontology to realize semantic feature extraction of the unstructured data.
4. The industrial digital twin virtual-real data fusion method as claimed in claim 3, wherein in the step (1), the semantic feature extraction of the structured data comprises:
1) matching table names in the database with the concept of the local ontology;
2) an acquisition example: get all records in the table and put the result into the set D ═ D1,d2,d3,…，dn}；
3) Instantiation: all records in the same table only correspond to one concept in a local ontology, and the data of the table is formatted and described in an xml mode;
4) and (3) persistent storage: and storing concepts obtained after semantic feature extraction is carried out on the examples into a description file, and completing semantic feature extraction on the structured data.
5. The industrial digital twin virtual-real data fusion method as claimed in claim 3, wherein in the step (2), the extraction of semantic features of unstructured data by using an ontology-associated TF-IDF algorithm comprises:
1) the number of times a word w appears in a certain text is wiNext, the number of occurrences of all words in the text is tiThen, using TF-IDF method to calculate the initial weight of vocabulary w as：
2) Searching the vocabulary in the local ontology, and when the vocabulary exists in the ontology, the optimized weight of the vocabulary w is represented as:
3) a weight value (W) obtained from the vocabulary W and a predetermined threshold value theta1Comparing when the weight value is larger than the threshold value theta1And taking the vocabulary as the characteristic words of the body, and putting the characteristic words into a characteristic word set V:
V＝{wi|i＝1,2,...,n}；
4) and storing the extracted feature words into a description file in an xml mode, and finishing the semantic feature extraction of the text data.
6. The industrial digital twin virtual-real data fusion method as claimed in claim 1, wherein in step four, the mapping from the local ontology to the global ontology based on the similarity algorithm is implemented, and the mapping comprises:
(1) calculating the similarity between two ontology elements from the concept, relationship, attribute and example angles of the ontology according to a similarity measurement function, and comprehensively weighting the similarity of each angle to obtain a final similarity result; the similarity between two ontology elements calculated by using the similarity measure function meets the following requirements:
S(ei,ej)∈[0,1]；
when the similarity between the two ontologies is 1, the two ontology elements have completely same semantic information, and the attributes and the instance elements of the two local ontologies are mapped into a global ontology; if the similarity is 0, the semantic information between the two body elements is completely different;
(2) obtaining the mapping from the local ontology to the global ontology according to the mapping process of the elements among the ontologies:
1) inputting local ontology (O)1And O2) And a global ontology (O);
2) finding elements of the local ontology and the global ontology, which have not established a mapping relation;
3) similarity calculation is carried out on the ontology elements from the aspects of concepts, attributes and examples by using a similarity measurement function, and finally, a similarity value is obtained through comprehensive weighting;
4) judging the value and the threshold value of the similarity in the body elements;
5) when the similarity value is equal to the threshold value, marking, and establishing a mapping relation for the similarity elements;
6) and repeating the steps until no element needing to establish the mapping relation exists, and converting the mapping result into a structured text form for outputting.
7. The industrial digital twin virtual-real data fusion method as claimed in claim 1, wherein in step five, the fusion of the industrial digital twin virtual-real data is completed based on the fusion rule, and the fusion method comprises the following steps:
(1) single fused results
The single fused result is suitable for the case that the final returned result is a single numerical value or a single character; the following fusion rules are proposed for this result: an average rule, a credibility priority rule, a majority priority rule and a weighting rule;
1) average rule: the rule is only applicable to the case that the examples are numerical values, and for different examples describing the same object, a centralized trend expression of all data is realized, so as to reflect the average level of various data; let R be { R ═ R1，r2，r3，...，rnThe data sets from different sources are represented, and the final fusion result of the method is as follows:
2) weighting rules: when the user gives a certain weight (W) to different instances according to a specific standardi) Carrying out weighted summation on the examples of different data sources according to the weight values; let R be { R ═ R1，r2，r3，...，rnDenotes data sets of different origins, WiThe weight value given by the user to a certain data source is represented, and the final fusion result obtained by using the method is as follows:
3) confidence priority rules: the method is suitable for the case that the examples are numerical values and characters, and under the normal condition, after the data sources are sorted according to the user-defined credibility, the highest credibility is the fusion result with the most accurate result; let R be { R ═ R1，r2，r3，...，rnDenotes a data set of different origin, Conf (r)i) Representing data riThe rule is defined as:
4) majority priority rules: the method considers that the reliability of the result with a large number of occurrences in the result set is high, and assumes that R ═ R is greater1，r2，r3，...，rnDenotes a data set of different origin, the rule is defined as:
(2) multiple fused results
The multi-fusion result considers the final returned result to be a set, and the following different fusion rules are given to the result: an upper and lower bound rule, or rules;
1) and (3) upper and lower limit rules: the rule contains all the results for the maximum and minimum values in the result set when rjApplying an upper and lower limit rule when the numerical example is used; let R be { R ═ R1，r2，r3，...，rnDenotes a data set of different origin, the rule is defined as:
2) or a rule: examples in the result set are single-valued or when comparison is not possible, a rule is adopted for processing data of numeric type or character type; let R be { R ═ R1，r2，r3，...，rnDenotes data sets of different origins, the rule is expressed by the following formula:
Or(R)＝{r1 or r2 or … or rn}。
8. an industrial digital twinning virtual-real data fusion system for implementing the industrial digital twinning virtual-real data fusion method as claimed in any one of claims 1 to 7, wherein the industrial digital twinning virtual-real data fusion system comprises:
the virtual and real data acquisition module is used for acquiring virtual and real data in the industrial production process;
the ontology construction module is used for realizing the construction of a global ontology and a local ontology according to an ontology description language and an ontology construction principle;
the semantic feature extraction module is used for extracting semantic features of structured and unstructured virtual and real data of industrial digital twin;
the ontology mapping module is used for realizing the mapping from the local ontology to the global ontology based on a similarity algorithm;
and the data fusion module is used for completing the fusion of the industrial digital twin virtual and real data based on the fusion rule.
9. A computer device, characterized in that the computer device comprises a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to carry out the steps of:
acquiring virtual and real data in the industrial production process; according to an ontology description language and an ontology construction principle, a set of ontology construction method aiming at the industrial field is provided, and global ontology and local ontology construction are realized; extracting semantic features of structured and unstructured virtual and real data of industrial digital twin;
mapping from the local ontology to the global ontology is realized based on a similarity algorithm; and finishing the fusion of the industrial digital twin virtual and real data based on a fusion rule.
10. An information data processing terminal, characterized in that the information data processing terminal is used for implementing an industrial digital twinning virtual-real data fusion system according to claim 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110713629.0A CN113554063B (en) | 2021-06-25 | 2021-06-25 | Industrial digital twin virtual-real data fusion method, system, equipment and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110713629.0A CN113554063B (en) | 2021-06-25 | 2021-06-25 | Industrial digital twin virtual-real data fusion method, system, equipment and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113554063A true CN113554063A (en) | 2021-10-26 |
| CN113554063B CN113554063B (en) | 2024-04-23 |
Family
ID=78130984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110713629.0A Active CN113554063B (en) | 2021-06-25 | 2021-06-25 | Industrial digital twin virtual-real data fusion method, system, equipment and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113554063B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114399066A (en) * | 2022-01-15 | 2022-04-26 | 中国矿业大学(北京) | Mechanical equipment predictability maintenance system and maintenance method based on weak supervision learning |
| CN115687623A (en) * | 2022-11-11 | 2023-02-03 | 北京航空航天大学 | Industrial digital twin data space construction method and system |
| CN115826935A (en) * | 2023-02-14 | 2023-03-21 | 中国航空油料集团有限公司 | Digital twin model integration system and method based on unified language middleware |
| CN116502182A (en) * | 2023-06-25 | 2023-07-28 | 金锐同创(北京)科技股份有限公司 | Method, device, equipment and medium for constructing twin platform based on mapping data |
| CN116611826A (en) * | 2023-03-24 | 2023-08-18 | 国网河北省电力有限公司雄安新区供电公司 | Multi-channel electric power payment system supporting digital currency and application thereof |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111597346A (en) * | 2020-04-15 | 2020-08-28 | 成都数联铭品科技有限公司 | Data fusion method and system based on coding |
| CN111723176A (en) * | 2019-03-18 | 2020-09-29 | 西门子股份公司 | Method for generating semantic description of synthetic interaction |
| CN111858649A (en) * | 2020-08-05 | 2020-10-30 | 哈尔滨工业大学（威海） | Heterogeneous data fusion method based on ontology mapping |
| WO2021031454A1 (en) * | 2019-08-21 | 2021-02-25 | 佳都新太科技股份有限公司 | Digital twinning system and method and computer device |
| CN112765720A (en) * | 2021-02-26 | 2021-05-07 | 黄河勘测规划设计研究院有限公司 | Water conservancy and hydropower engineering multi-source data fusion method based on BIM + GIS |
| US20210157312A1 (en) * | 2016-05-09 | 2021-05-27 | Strong Force Iot Portfolio 2016, Llc | Intelligent vibration digital twin systems and methods for industrial environments |
-
2021
- 2021-06-25 CN CN202110713629.0A patent/CN113554063B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210157312A1 (en) * | 2016-05-09 | 2021-05-27 | Strong Force Iot Portfolio 2016, Llc | Intelligent vibration digital twin systems and methods for industrial environments |
| CN111723176A (en) * | 2019-03-18 | 2020-09-29 | 西门子股份公司 | Method for generating semantic description of synthetic interaction |
| WO2021031454A1 (en) * | 2019-08-21 | 2021-02-25 | 佳都新太科技股份有限公司 | Digital twinning system and method and computer device |
| CN111597346A (en) * | 2020-04-15 | 2020-08-28 | 成都数联铭品科技有限公司 | Data fusion method and system based on coding |
| CN111858649A (en) * | 2020-08-05 | 2020-10-30 | 哈尔滨工业大学（威海） | Heterogeneous data fusion method based on ontology mapping |
| CN112765720A (en) * | 2021-02-26 | 2021-05-07 | 黄河勘测规划设计研究院有限公司 | Water conservancy and hydropower engineering multi-source data fusion method based on BIM + GIS |
Non-Patent Citations (2)
| Title |
|---|
| 苏宝程;权义宁;: "基于关系模式的XML数据存取技术研究", 内蒙古石油化工, no. 12, pages 31 - 33 * |
| 闫雪锋;赵楠;: "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, no. 03, pages 96 - 101 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114399066A (en) * | 2022-01-15 | 2022-04-26 | 中国矿业大学(北京) | Mechanical equipment predictability maintenance system and maintenance method based on weak supervision learning |
| CN115687623A (en) * | 2022-11-11 | 2023-02-03 | 北京航空航天大学 | Industrial digital twin data space construction method and system |
| CN115687623B (en) * | 2022-11-11 | 2023-04-28 | 北京航空航天大学 | Industrial digital twin data space construction method and system |
| CN115826935A (en) * | 2023-02-14 | 2023-03-21 | 中国航空油料集团有限公司 | Digital twin model integration system and method based on unified language middleware |
| CN116611826A (en) * | 2023-03-24 | 2023-08-18 | 国网河北省电力有限公司雄安新区供电公司 | Multi-channel electric power payment system supporting digital currency and application thereof |
| CN116611826B (en) * | 2023-03-24 | 2023-12-12 | 国网河北省电力有限公司雄安新区供电公司 | Multi-channel electric power payment system supporting digital currency and application thereof |
| CN116502182A (en) * | 2023-06-25 | 2023-07-28 | 金锐同创(北京)科技股份有限公司 | Method, device, equipment and medium for constructing twin platform based on mapping data |
| CN116502182B (en) * | 2023-06-25 | 2023-10-27 | 金锐同创(北京)科技股份有限公司 | Method, device, equipment and medium for constructing twin platform based on mapping data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113554063B (en) | 2024-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113554063B (en) | Industrial digital twin virtual-real data fusion method, system, equipment and terminal | |
| Hou et al. | Unstructured big data analysis algorithm and simulation of Internet of Things based on machine learning | |
| CN111428054A (en) | Construction and storage method of knowledge graph in network space security field | |
| CN112463980A (en) | Intelligent plan recommendation method based on knowledge graph | |
| CN105469204A (en) | Reassembling manufacturing enterprise integrated evaluation system based on deeply integrated big data analysis technology | |
| WO2023124191A1 (en) | Depth map matching-based automatic classification method and system for medical data elements | |
| CN112463981A (en) | Enterprise internal operation management risk identification and extraction method and system based on deep learning | |
| Xu et al. | Multilane roads extracted from the OpenStreetMap urban road network using random forests | |
| CN114647741A (en) | Process automatic decision and reasoning method, device, computer equipment and storage medium | |
| Yao et al. | A machining feature recognition approach based on hierarchical neural network for multi-feature point cloud models | |
| Bian et al. | Research on an artificial intelligence-based professional ability evaluation system from the perspective of industry-education integration | |
| CN107798137B (en) | A kind of multi-source heterogeneous data fusion architecture system based on additive models | |
| CN113254517A (en) | Service providing method based on internet big data | |
| Yin et al. | A deep natural language processing‐based method for ontology learning of project‐specific properties from building information models | |
| CN116561264A (en) | Knowledge graph-based intelligent question-answering system construction method | |
| CN106816871B (en) | State similarity analysis method for power system | |
| Lu et al. | Application of PSO-based LSTM Neural Network for Outpatient Volume Prediction | |
| CN114694098A (en) | Power grid infrastructure construction risk control method based on image recognition and knowledge graph | |
| Chang et al. | Decision-making and computational modeling of big data for sustaining influential usage | |
| Yang | RETRACTED ARTICLE: Financial Information Extraction Using the Improved Hidden Markov Model and Deep Learning | |
| CN113792160A (en) | Knowledge graph expansion and fusion method for multi-source data | |
| CN113987536A (en) | Method and device for determining security level of field in data table, electronic equipment and medium | |
| Zhang et al. | A new online field feature selection algorithm based on streaming data | |
| CN111242520A (en) | Feature synthesis model generation method and device and electronic equipment | |
| CN110737777A (en) | knowledge map construction method based on GHSOM algorithm |
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 |