Semantized instant assembly method and system for streaming data

Abstract

The invention provides a semantic instant assembly method and system for streaming data, and the method comprises the steps: loading a preset semantic assembly model in real time, configuring the semantic assembly model in real time according to the assembly demand of to-be-assembled streaming data, and finally carrying out the real-time operation and publishing of the to-be-assembled streaming data according to the configured semantic assembly model. Real-time assembly of the stream data is completed; according to the method, the semantic assembly model used for assembling the streaming data in real time is preset, so that a user can carry out self-defined configuration according to the assembly requirement of the user, flexible secondary model modification is realized, and meanwhile, centralized maintenance is facilitated; besides, the assembly model is used for loading, configuring and processing stream data in real time, so that hot loading can be realized, and equipment operation does not need to be interrupted; compared with existing hard coding assembly and data hierarchical assembly, the method has the advantages that more flexible and efficient data assembly and maintenance are realized, and the performance and reliability of stream data assembly are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of stream data assembly, and more specifically, to a semantic instant assembly method and system for stream data. Background Art

[0002] Stream data is a set of data sequences that arrive in an orderly, large, fast and continuous manner. Generally speaking, stream data can be regarded as a dynamic data set that grows infinitely over time. It is widely used in network monitoring, sensor networks, aerospace, meteorological measurement and control, and financial services.

[0003] Stream data assembly refers to the calculation and processing of one or more stream data such as dimension increase and clustering, and finally publishing the processing results in the human-computer interaction interface. At present, there are mainly two methods for assembling stream data:

[0004] 1) Hard-coded assembly: Calculate the stream data through hard coding, and then combine a set of related dimensionality-increasing operation results through hard coding; the disadvantage of hard-coded assembly is that it cannot be flexibly modified and can only be achieved by modifying the coding method. However, the dimensionality-increasing operation rules are distributed in different source codes, making it difficult to maintain them centrally.

[0005] 2) Data hierarchical assembly: The assembly is divided into a predefined layer and a decoration layer. The stream data is first processed and calculated in the predefined layer, and then the output results of the predefined layer are packaged in the decoration layer. However, data hierarchical assembly only supports flexible modification within the data range of the predefined layer, and the device operation needs to be interrupted during data assembly. Summary of the invention

[0006] In order to overcome the problems of the above-mentioned prior art that, when assembling stream data, the stream data cannot be flexibly modified and needs to be interrupted, the present invention provides a semantic instant assembly method and system for stream data. By presetting a semantic assembly model for real-time assembly of stream data, users can customize the configuration according to their own assembly needs, thereby achieving flexible secondary modification of the model and facilitating centralized maintenance. In addition, the assembly model loads, configures and processes stream data in real time, and can achieve hot loading without interrupting equipment operation.

[0007] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0008] A semantic instant assembly method for stream data includes the following steps:

[0009] Load the preset semantic assembly model in real time;

[0010] configuring the semantic assembly model in real time according to the assembly requirements of the flow data to be assembled, and obtaining the configured semantic assembly model;

[0011] The stream data to be assembled is calculated and published in real time according to the configured semantic assembly model to complete the real-time assembly of the stream data.

[0012] Preferably, the semantic assembly model includes at least the following fields: a value path of stream data, a mapping rule set, a calculation rule set and a publishing path.

[0013] Preferably, the value acquisition path of the stream data includes: an identifier of the stream data acquired by the semantic assembly model, a value acquisition interface type, a value acquisition data format and a sampling frequency;

[0014] The mapping rule set of the stream data includes: a plurality of mapping rules with constraint conditions, the mapping rules are used to map the stream data to instance attribute data of its assembly target;

[0015] The set of operation rules for the stream data includes: any one or more of a dimension-raising operation rule, an average operation rule, a data normalization operation rule and an aggregation operation rule; each of the operation rules is used to perform a corresponding operation on the stream data;

[0016] The publishing path of the stream data includes: the address of the assembly target of the stream data, the publishing interface type and the publishing data format.

[0017] Preferably, the method further comprises: verifying the configured semantic assembly model, and if the verification passes, performing real-time assembly of stream data according to the configured semantic assembly model;

[0018] If the verification fails, the configured semantic assembly model is corrected, and the stream data is assembled in real time according to the corrected semantic assembly model.

[0019] Preferably, the verification includes: performing lexical rule analysis, grammatical structure analysis and semantic logic analysis on the configured semantic assembly model respectively. If the three analysis results are all correct, the verification passes; otherwise, the verification fails.

[0020] Preferably, the real-time operation and publishing of the assembly flow data to be assembled according to the configured semantic assembly model includes:

[0021] Parse assembly requirements according to the configured semantic assembly model, and establish several task queues based on the assembly requirements;

[0022] The assembly requirements include relevant information on the calculation, mapping and publishing required for the flow data to be assembled;

[0023] Obtain corresponding flow data to be assembled according to the value path of the flow data, add each of the flow data to be assembled and the required operation rules to each task queue respectively and simultaneously, and allocate corresponding computing resources to each of the task queues;

[0024] Processing the computing tasks of each of the task queues in parallel to obtain a number of computing results;

[0025] Mapping each of the operation results into instance attribute data of the assembly target according to corresponding mapping rules;

[0026] All instance attribute data are aggregated and published according to the publishing path.

[0027] Preferably, the acquiring corresponding flow data to be assembled according to the value path of the flow data comprises:

[0028] Mapping the value of the to-be-assembled stream data into a hash matrix and storing the matrix;

[0029] Using a hash function to establish an index between the identifier of the stream data to be assembled and the hash matrix;

[0030] The value of the corresponding stream data to be assembled is obtained according to the identifier of the stream data to be assembled and the index between the hash matrix.

[0031] Preferably, the semantic assembly model is constructed based on a domain specific language DSL.

[0032] The present invention also provides a semantic instant assembly system for streaming data, which uses the semantic instant assembly method for streaming data, and comprises: a data acquisition module, a calculation module and a publishing module arranged in parallel, and a hot loading module connected to the data acquisition module, the calculation module and the publishing module respectively;

[0033] The data acquisition module is used to acquire the flow data to be assembled;

[0034] The computing module is used for computing stream data;

[0035] The publishing module is used for publishing streaming data;

[0036] The hot loading module is used to load and configure a preset semantic assembly model in real time, and to schedule a data acquisition module, a calculation module and a publishing module in real time according to the configured semantic assembly model to complete the real-time assembly of streaming data.

[0037] Preferably, the system further comprises a verification module for verifying the configured semantic assembly model;

[0038] The verification module is connected to the hot loading module.

[0039] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0040] The present invention provides a semantic instant assembly method and system for stream data, which loads a preset semantic assembly model in real time, configures the semantic assembly model in real time according to the assembly requirements of the stream data to be assembled, and finally performs real-time calculation and release on the stream data to be assembled according to the configured semantic assembly model, thereby completing the real-time assembly of the stream data;

[0041] The present invention presets a semantic assembly model for real-time assembly of streaming data, so that users can customize the configuration according to their own assembly needs, thereby realizing flexible secondary modification of the model and facilitating centralized maintenance. In addition, the assembly model loads, configures and processes streaming data in real time, and can realize hot loading without interrupting equipment operation. Compared with the existing hard-coded assembly and data hierarchical assembly, the present invention realizes more flexible and efficient data assembly and maintenance, and improves the performance and reliability of streaming data assembly. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A flow chart of a semantic instant assembly method for stream data provided in Example 1.

[0043] Figure 2 This is a schematic diagram of the overall process of semantic instant assembly provided in Example 2.

[0044] Figure 3 This is a structural diagram of a semantic instant assembly system for streaming data provided in Example 3. DETAILED DESCRIPTION

[0045] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0046] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0047] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0048] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0049] Example 1

[0050] like Figure 1 As shown, this embodiment provides a semantic instant assembly method for stream data, including the following steps:

[0051] S1: Load the preset semantic assembly model in real time;

[0052] S2: configuring the semantic assembly model in real time according to the assembly requirements of the flow data to be assembled, and obtaining the configured semantic assembly model;

[0053] S3: Performing real-time calculation and publishing on the stream data to be assembled according to the configured semantic assembly model to complete the real-time assembly of the stream data.

[0054] In the specific implementation process, the preset semantic assembly model is first loaded in real time. The semantic assembly model provides a general assembly template. Different templates are selected according to different stream data. The template includes multiple fields related to the assembly process. The fields are used to control specific assembly operations. The fields in the template can be customized according to assembly requirements.

[0055] In this embodiment, there is a corresponding relationship between the preset semantic assembly model and the stream data. According to the assembly requirements, the corresponding relationship can be set as follows: each stream data has an independent assembly model (one-to-one relationship), multiple stream data share an assembly model (one-to-many relationship), and multiple stream data are processed through a summary model (many-to-one relationship);

[0056] The semantic assembly model is configured in real time according to the assembly requirements of the stream data to be assembled. This step mainly configures and initializes each field in the semantic assembly model to achieve automatic real-time assembly of stream data.

[0057] Finally, the configured semantic assembly model is used to perform real-time calculation and release of the assembly stream data to complete the real-time assembly of the stream data.

[0058] This method presets a semantic assembly model for real-time assembly flow data, and users can customize the configuration according to their own assembly needs, thereby achieving flexible secondary modification of the model and facilitating centralized maintenance. In addition, the assembly model loads, configures and processes flow data in real time, and can achieve hot loading without interrupting equipment operation.

[0059] Example 2

[0060] This embodiment provides a semantic instant assembly method for stream data, including the following steps:

[0061] Load the preset semantic assembly model in real time;

[0062] configuring the semantic assembly model in real time according to the assembly requirements of the flow data to be assembled, and obtaining the configured semantic assembly model;

[0063] Verifying the configured semantic assembly model, and if the verification passes, assembling the stream data in real time according to the configured semantic assembly model;

[0064] If the verification fails, the configured semantic assembly model is corrected, and the stream data is assembled in real time according to the corrected semantic assembly model;

[0065] The real-time assembly of the stream data includes: performing real-time calculation and publishing on the stream data to be assembled;

[0066] The semantic assembly model includes at least the following fields: a value path of stream data, a mapping rule set, an operation rule set and a publishing path;

[0067] The value acquisition path of the stream data includes: an identifier of the stream data acquired by the semantic assembly model, a value acquisition interface type, a value acquisition data format and a sampling frequency;

[0068] The mapping rule set of the stream data includes: a plurality of mapping rules with constraint conditions, the mapping rules are used to map the stream data to instance attribute data of its assembly target;

[0069] The set of operation rules for the stream data includes: any one or more of a dimension-raising operation rule, an average operation rule, a data normalization operation rule and an aggregation operation rule; each of the operation rules is used to perform a corresponding operation on the stream data;

[0070] The publishing path of the stream data includes: the address of the assembly target of the stream data, the publishing interface type and the publishing data format;

[0071] The verification includes: performing lexical rule analysis, grammatical structure analysis and semantic logic analysis on the configured semantic assembly model respectively. If the three analysis results are all correct, the verification passes; otherwise, the verification fails;

[0072] The real-time operation and release of the assembly flow data includes:

[0073] Analyze assembly requirements and establish several task queues based on assembly requirements;

[0074] The assembly requirements include relevant information on the calculation, mapping and publishing required for the flow data to be assembled;

[0075] Obtain corresponding flow data to be assembled according to the value path of the flow data, add each of the flow data to be assembled and the required operation rules to each task queue respectively and simultaneously, and allocate corresponding computing resources to each of the task queues;

[0076] Processing the computing tasks of each of the task queues in parallel to obtain a number of computing results;

[0077] Mapping each of the operation results into instance attribute data of the assembly target according to corresponding mapping rules;

[0078] Aggregate all instance attribute data and publish them according to the publishing path;

[0079] The step of obtaining corresponding flow data to be assembled according to the value path of the flow data comprises:

[0080] Mapping the value of the to-be-assembled stream data into a hash matrix and storing the matrix;

[0081] Using a hash function to establish an index between the identifier of the stream data to be assembled and the hash matrix;

[0082] Obtain the value of the corresponding stream data to be assembled according to the identifier of the stream data to be assembled and the index between the hash matrix;

[0083] In this embodiment, the semantic assembly model is constructed based on a domain specific language DSL.

[0084] In the specific implementation process, Figure 2 The semantic assembly process shown in the figure first loads the preset semantic assembly model in real time. The semantic assembly model provides a general assembly template. Different templates are selected according to different stream data. The template includes multiple fields related to the assembly process. The fields are used to control specific assembly operations. The fields in the template can be customized according to assembly requirements.

[0085] In this embodiment, a pre-designed domain specific language (DSL) is used to define the semantic assembly model;

[0086] An example of a model is shown below:

[0087] The main fields (elements) defined in the model include:

[0088] 1) Data binding (input), i.e. the value path of stream data: defines how to obtain stream data from the stream data source; this field can specify the identifier of the data source, data format, sampling frequency, etc.;

[0089] 2) Aggregate: defines the rules (operators) used to perform operations on streaming data. Operators include average calculation, data normalization, aggregation operations, etc. For example, an aggregate can specify how to convert raw data into high-dimensional data for subsequent processing.

[0090] 3) Mapping rule set (properties): defines several mapping rules, which are used to map stream data to instance attribute data of its assembly target. These instance attribute data describe the data structure, dimension, type and physical meaning of a certain attribute of the assembly target. The set contains multiple mapping rules, each of which has specific definitions and constraints.

[0091] Specifically, there is the following relationship between the flow data and the mapping rule set:

[0092] The mapping rule set defines how each data point in the stream data is mapped to a structured data form; for example, a data point in the stream data can be mapped to an object, which contains multiple attributes;

[0093] This field not only provides the context for processing stream data, making the processing logic more general and modular, but also ensures that the published data is consistent and predictable, making it easier for the receiver to parse and use it;

[0094] 4) Publish Path: defines how the processed data is published to the human-machine interface; the publish path includes information such as the address of the target system, interface type, and data format;

[0095] In this embodiment, there is a corresponding relationship between the preset semantic assembly model and the stream data. According to the assembly requirements, the corresponding relationship can be set as follows: each stream data has an independent assembly model (one-to-one relationship), multiple stream data share an assembly model (one-to-many relationship), and multiple stream data are processed through a summary model (many-to-one relationship);

[0096] The semantic assembly model is configured in real time according to the assembly requirements of the stream data to be assembled. This step mainly configures and initializes each field in the semantic assembly model to achieve automatic real-time assembly of stream data.

[0097] Then check whether the definition of the configured semantic assembly model conforms to the pre-designed DSL syntax; the specific checks include:

[0098] 1) Lexical rule analysis: Scan the input DSL code and convert it into tokens to ensure that all tokens comply with the lexical rules of the DSL;

[0099] 2) Syntax structure analysis: According to the grammatical rules of DSL, the token sequence is parsed into a syntax tree to ensure that the model definition conforms to the grammatical structure of DSL;

[0100] 3) Semantic logic analysis: further verify the logical correctness of the model definition, including: confirming that all stream data identifiers have been defined before use, checking whether the operation rules and data bindings match, verifying whether the attribute configuration in the mapping rule set is correct, and confirming that the definition of the release path conforms to the expected format and specifications;

[0101] If the above three analysis results are correct, the verification passes; otherwise, the verification fails. If the verification fails, the corresponding DSL syntax errors need to be manually corrected before stream data assembly;

[0102] Finally, the verified semantic assembly model is used to perform real-time calculation and release of the assembly stream data to complete the real-time assembly of the stream data. The process is as follows:

[0103] First, the scheduling task (assembly requirement) is parsed and the semantic assembly model set m1, m2, ..., m is read. n , a total of n assembly models; the storage location can be a file system, a database or a remote server;

[0104] The stream data in this embodiment is: the state stream data of a large number of devices collected in the monitoring and control system of the new energy unit equipment, and the assembly requirements are: perform dimensionality-upgrading calculation on one or more stream data, and finally publish the calculation results in the human-computer interaction interface;

[0105] Initialize the scheduling tasks based on the parsed and verified model set, including: Task queue establishment: Establish a task queue for each model, add the stream data to be processed and the corresponding operation rules to the task queue, allocate computing resources according to the needs of the task queue, and ensure that each task can be executed on the appropriate computing unit;

[0106] In this embodiment, the stream data to be assembled are d1, d2, ..., d n Mapping storage in the form of a hash matrix in memory; set K represents the set of all stream data identifiers, set V represents the set of all stream data values, and hash function h maps the stream data identifier to the index of the hash matrix, h: K→{0,1,...v-1}, where v is the size of the set of stream data identifiers;

[0107] When the semantic assembly model obtains stream data after verification, it maps the identifier v of the stream data to the index of the hash matrix, and finally obtains the value of the stream data from the set V;

[0108] It is not absolutely necessary to map the stream data in the form of a hash matrix in memory, but in high-performance, high-concurrency, and large-scale data processing application scenarios, this approach can significantly improve the efficiency and performance of the system;

[0109] In this embodiment, there is a one-to-one correspondence between n stream data and n assembly models, and one assembly model is used to independently process one stream data;

[0110] Then read the semantic assembly models m1, m2, ..., m after verification n All the stream data identifiers in the hash matrix are obtained, and the corresponding data values ​​are taken out; through the dimension-raising operators c1, c2, ..., c n Perform the operation and obtain the result value set s1, s2, ..., s n During the operation, all operators are processed simultaneously on different computing units through a parallel processing architecture, and the operation results of each operator are summarized through the TCP data pipeline;

[0111] It is worth mentioning that a variety of data transmission channels can be used to aggregate the calculation results of each operator, not limited to TCP data pipelines. Common data transmission channels such as UDP, WebSocket and message queues can also be selected. Different transmission channels have their own advantages and disadvantages. Choosing a suitable channel depends on the specific needs and architectural design of the system.

[0112] Then, according to the mapping rules of each assembly model, each calculation result s1, s2, ..., s n Mapping to the entity attributes corresponding to the assembly target to represent the specific physical meaning and ensure that each result value correctly matches the corresponding attribute of the assembly target;

[0113] In addition, if the result value needs further conversion or formatting, necessary data conversion operations are performed to ensure that the final data conforms to the expected publishing format;

[0114] After that, all the mapped entity attribute data will be summarized and prepared for release;

[0115] The steps of publishing include: publishing path configuration: according to the publishPath definition in the model, configure the target address of the publishing path, interface type (such as REST API, message queue, etc.), data format (such as JSON, XML, etc.) and other necessary publishing parameters;

[0116] Publishing operation: publish the aggregated entity attribute data to the target system or interface through the configured publishing path; ensure the reliability and security of the publishing operation and handle possible publishing errors or exceptions;

[0117] Confirm the release at the same time: monitor the results of the release operation to confirm that the data has been successfully released. If the release fails, record the error information and retry or error handling;

[0118] After the release is completed, the stream data is assembled once;

[0119] Then determine whether the assembly model has changed. If so, load, configure, verify and read the new assembly model set again. If not, start task scheduling directly and repeat the above steps to perform real-time assembly of streaming data.

[0120] The semantic assembly model in this method integrates data binding, operation rules, mapping rules and release statements through DSL, so that non-professionals can also quickly design and flexibly modify. Users can customize the configuration according to their own assembly needs, thereby realizing flexible secondary modification of the model; at the same time, all mapping rules and operation rules are concentrated in the semantic assembly model, which facilitates the centralized maintenance of the rules; in addition, the assembly model loads, configures and processes streaming data in real time, and can realize hot loading without interrupting the operation of the equipment.

[0121] Example 3

[0122] like Figure 3 As shown, this embodiment provides a semantic instant assembly system for stream data, and applies the semantic instant assembly method for stream data described in Example 2, including: a data acquisition module 302, a calculation module 303, a publishing module 304 and a verification module 305 arranged in parallel, and a hot loading module 301 connected to the data acquisition module 302, the calculation module 303, the publishing module 304 and the verification module 305 respectively;

[0123] The data acquisition module 302 is used to acquire the flow data to be assembled;

[0124] The operation module 303 is used for the operation of stream data;

[0125] The publishing module 304 is used for publishing stream data;

[0126] The verification module 305 is used to verify the configured semantic assembly model;

[0127] The hot loading module 301 is used to load and configure the preset semantic assembly model in real time, and to schedule the data acquisition module 302, the operation module 303, the publishing module 304 and the verification module 305 in real time according to the configured semantic assembly model to complete the real-time assembly of the stream data.

[0128] In the specific implementation process, first, the hot loading module 301 loads the preset semantic assembly model in real time. The semantic assembly model provides a universal assembly template. Different templates are selected according to different stream data. The template includes multiple fields related to the assembly process. The fields are used to control specific assembly operations. The fields in the template can be customized according to assembly requirements.

[0129] Then the hot loading module 301 configures the semantic assembly model in real time according to the assembly requirements of the stream data to be assembled. This step mainly configures and initializes each field in the semantic assembly model, thereby realizing the automatic real-time assembly of the stream data;

[0130] Then, the hot loading module 301 calls the verification module 305 to verify the configured semantic assembly model. If the verification passes, the real-time scheduling data acquisition module 302 obtains the flow data to be assembled, the operation module 303 transports the flow data, and the publishing module 304 publishes the flow data according to the configured semantic assembly model, and finally completes the real-time assembly of the flow data.

[0131] If the verification fails, the configured semantic assembly model is corrected. According to the corrected semantic assembly model, the real-time scheduling data acquisition module 302 acquires the stream data to be assembled, the operation module 303 transports the stream data, and the publishing module 304 publishes the stream data, and finally completes the real-time assembly of the stream data;

[0132] In this embodiment, the hot loading module 301 is specifically a hot loading machine, the data acquisition module 302 is specifically a binder, the operation module 303 is specifically an operator, the publishing module 304 is specifically a publisher, and the verification module 305 is specifically a verifier;

[0133] This system presets a semantic assembly model for real-time assembly flow data. Users can customize the configuration according to their own assembly needs, thereby achieving flexible secondary modification of the model and facilitating centralized maintenance. In addition, the assembly model loads, configures and processes flow data in real time, and can achieve hot loading without interrupting equipment operation.

[0134] The same or similar reference numerals correspond to the same or similar components;

[0135] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0136] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A semantic instant assembly method for stream data, characterized in that: The following steps are involved: Load the preset semantic assembly model in real time; configuring the semantic assembly model in real time according to the assembly requirements of the flow data to be assembled, and obtaining the configured semantic assembly model; The stream data to be assembled is calculated and published in real time according to the configured semantic assembly model to complete the real-time assembly of the stream data.

2. A semantic instant assembly method for stream data according to claim 1, characterized in that: The semantic assembly model includes at least the following fields: a value path of stream data, a mapping rule set, an operation rule set and a publishing path.

3. A semantic instant assembly method for stream data according to claim 2, characterized in that: The value acquisition path of the stream data includes: an identifier of the stream data acquired by the semantic assembly model, a value acquisition interface type, a value acquisition data format and a sampling frequency; The mapping rule set of the stream data includes: a plurality of mapping rules with constraint conditions, the mapping rules are used to map the stream data to instance attribute data of its assembly target; The set of operation rules for the stream data includes: any one or more of a dimension-raising operation rule, an average operation rule, a data normalization operation rule and an aggregation operation rule; each of the operation rules is used to perform a corresponding operation on the stream data; The publishing path of the stream data includes: the address of the assembly target of the stream data, the publishing interface type and the publishing data format.

4. The semantic instant assembly method of stream data according to claim 1, characterized in that: The method further comprises: verifying the configured semantic assembly model, and if the verification passes, assembling the stream data in real time according to the configured semantic assembly model; If the verification fails, the configured semantic assembly model is corrected, and the stream data is assembled in real time according to the corrected semantic assembly model.

5. A semantic instant assembly method for stream data according to claim 4, characterized in that: The verification includes: performing lexical rule analysis, grammatical structure analysis and semantic logic analysis on the configured semantic assembly model respectively. If the three analysis results are all correct, the verification passes; otherwise, the verification fails.

6. A semantic instant assembly method for stream data according to claim 3, characterized in that: The real-time operation and publishing of assembly flow data according to the configured semantic assembly model includes: Parse assembly requirements according to the configured semantic assembly model, and establish several task queues based on the assembly requirements; The assembly requirements include relevant information on the calculation, mapping and publishing required for the flow data to be assembled; Obtain corresponding flow data to be assembled according to the value path of the flow data, add each of the flow data to be assembled and the required operation rules to each task queue respectively and simultaneously, and allocate corresponding computing resources to each of the task queues; Processing the computing tasks of each of the task queues in parallel to obtain a number of computing results; Mapping each of the operation results into instance attribute data of the assembly target according to corresponding mapping rules; All instance attribute data are aggregated and published according to the publishing path.

7. A semantic instant assembly method for stream data according to claim 6, characterized in that: The step of obtaining corresponding flow data to be assembled according to the value path of the flow data comprises: Mapping the value of the to-be-assembled stream data into a hash matrix and storing the matrix; Using a hash function to establish an index between the identifier of the stream data to be assembled and the hash matrix; The value of the corresponding stream data to be assembled is obtained according to the identifier of the stream data to be assembled and the index between the hash matrix.

8. A semantic instant assembly method for stream data according to any one of claims 1 to 7, characterized in that: The semantic assembly model is constructed based on a domain specific language DSL.

9. A semantic instant assembly system for streaming data, using the semantic instant assembly method for streaming data as claimed in any one of claims 1 to 8, characterized in that: include: A data acquisition module, a calculation module and a publishing module are arranged in parallel, and a hot loading module is connected to the data acquisition module, the calculation module and the publishing module respectively; The data acquisition module is used to acquire the flow data to be assembled; The computing module is used for computing stream data; The publishing module is used for publishing streaming data; The hot loading module is used to load and configure a preset semantic assembly model in real time, and to schedule a data acquisition module, a calculation module and a publishing module in real time according to the configured semantic assembly model to complete the real-time assembly of streaming data.

10. The semantic instant assembly system of stream data according to claim 9, characterized in that: The system further comprises a verification module, which is used to verify the configured semantic assembly model; The verification module is connected to the hot loading module.

Images