A Cross-Platform and Cross-System Configurable Data Interaction Method and System

Abstract

The present invention discloses a cross-platform and cross-system configurable data interaction method and system, which relates to the field of data integration and processing. It includes using a bimodal metadata derivation model to process a comprehensive data set to obtain an initial mapping rule table, establishing a global mapping model through federated learning, inputting the initial mapping rule table into the global mapping model to obtain an optimized mapping rule table, associating the target system source system data with the twin body to form a virtual-real fusion data stream, and converting the virtual-real fusion data stream into a target format and encrypting it for transmission to the target system. The present invention greatly improves the efficiency and flexibility of rule generation through automated derivation and priority optimization. Especially in a dynamically changing heterogeneous environment, it can quickly adapt to changes in field semantics and load requirements, thereby enhancing the real-time nature of data interaction.

Description

Technical Field

[0001] The present invention relates to the field of data integration and processing, and particularly to a configurable data interaction method and system across platforms and systems. Background Art

[0002] With the rapid development of informatization and digital technologies, cross-platform and cross-system data interaction has become an important research direction in fields such as cloud computing, the Internet of Things, industrial Internet, and intelligent manufacturing. Early related technologies mainly relied on static data exchange protocols to achieve data transfer between heterogeneous systems through predefined interfaces.

[0003] However, there are still many deficiencies in the current popular implementation of cross-platform and cross-system data interaction. For example, traditional methods basically rely on static mapping rules or manual configuration, making it difficult to cope with the frequent changes in heterogeneous data source formats or dynamic adjustments of business scenarios, resulting in low configuration efficiency and inability to optimize in real time. Moreover, existing digital twin technologies can synchronize physical devices with digital models, but when it comes to cross-system data fusion, they often lack dynamic verification of the reliability of mapping rules, and are prone to introducing incorrect results due to data semantic deviations or outliers. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a configurable data interaction method across platforms and systems to solve the problem of difficulty in coping with dynamic adjustments of heterogeneous data sources or business scenarios, resulting in low configuration efficiency and inability to optimize in real time.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a configurable data interaction method across platforms and systems, which includes:

[0008] Real-time access to raw data from multiple heterogeneous data sources, obtaining the metadata, traffic characteristics, and semantic analysis results of the raw data, and combining them into a comprehensive data set;

[0009] Processing the comprehensive data set using a bimodal metadata derivation model to obtain an initial mapping rule table;

[0010] Establishing a global mapping model through federated learning, inputting the initial mapping rule table into the global mapping model, and obtaining an optimized mapping rule table;

[0011] Associating the source system data of the target system with the twin body to form a virtual-real fusion data stream, and converting the virtual-real fusion data stream into a target format and encrypting it for transmission to the target system;

[0012] The target system receives data and returns feedback, and optimizes according to the feedback results.

[0013] As a preferred solution of the cross-platform and cross-system configurable data interaction method described in the present invention, wherein: the semantic analysis result refers to generating a similarity score by performing semantic analysis on the field names of the metadata through the BERT model;

[0014] Integrate the metadata, traffic characteristics, and similarity scores of the original data into a comprehensive data set in JSON format.

[0015] As a preferred solution of the cross-platform and cross-system configurable data interaction method described in the present invention, wherein: use a dual-modal metadata derivation model to process the comprehensive data set to obtain an initial mapping rule table, specifically including the following steps,

[0016] The dual-modal metadata derivation model performs in-depth analysis of field semantics based on the metadata and similarity scores to generate preliminary mapping candidates, processes the traffic characteristics, determines the matching priorities of the fields in the metadata, and generates a field priority table;

[0017] Fuse the preliminary mapping candidates with the field priority table to generate an initial mapping rule table.

[0018] As a preferred solution of the cross-platform and cross-system configurable data interaction method described in the present invention, wherein: establish a global mapping model through federated learning, input the initial mapping rule table into the global mapping model, and obtain an optimized mapping rule table, specifically including the following steps,

[0019] The federated learning framework distributes the initial mapping rule table to each training node, and each training node trains an MLP model based on local data;

[0020] Upload encrypted parameters to the asynchronous parameter server. After the server receives the parameters of each training node, it aggregates them to generate a global mapping model;

[0021] Input the initial mapping rule table into the global mapping model and output an optimized mapping rule table.

[0022] As a preferred solution of the cross-platform and cross-system configurable data interaction method described in the present invention, wherein: the formation of the virtual-real fusion data stream specifically includes the following steps,

[0023] Load the optimized mapping rule table through the digital twin mapping engine, extract real-time data from the source system, associate it with the twin data, and bind fields through a two-way dynamic mapping mechanism to form a virtual-real fusion data stream.

[0024] As a preferred solution of the cross-platform and cross-system configurable data interaction method of the present invention, wherein: the virtual-real fusion data is converted into a target format and encrypted and transmitted to the target system, which specifically includes the following steps,

[0025] Use weighted decision-making technology to generate a comprehensive credibility score for the virtual-real fusion data stream based on historical mapping records, real-time semantic analysis, and business priorities;

[0026] Set a confidence threshold according to the historical mapping record, compare the comprehensive credibility score with the confidence threshold, and if it is lower than the confidence threshold, trigger the process of re-deriving the mapping rule;

[0027] Convert the virtual-real fusion data stream into a format recognizable by the target system, and attach a check code and transmit it to the target system.

[0028] As a preferred solution of the cross-platform and cross-system configurable data interaction method of the present invention, wherein: the target system receives the data and returns feedback, and optimizes according to the feedback result, which specifically includes the following steps,

[0029] The target system parses the received virtual-real fusion data stream, generates an interaction feedback, and transmits it back to the adaptive mapping engine through the data bus;

[0030] The adaptive mapping engine adjusts the current mapping rule table according to the interaction feedback, incrementally trains and updates the global mapping model through federated learning, synchronously adjusts the state of the digital twin, and switches the strategy to optimize the transmission.

[0031] In a second aspect, the present invention provides a cross-platform and cross-system configurable data interaction system, including,

[0032] An acquisition module that accesses raw data in real time from multiple heterogeneous data sources, obtains the metadata, traffic characteristics, and semantic analysis results of the raw data, and combines them into a comprehensive data set;

[0033] A processing module that uses a bimodal metadata derivation model to process the comprehensive data set to obtain an initial mapping rule table;

[0034] An optimization module that establishes a global mapping model through federated learning, inputs the initial mapping rule table into the global mapping model, and obtains an optimized mapping rule table;

[0035] A transmission module that associates the source system data of the target system with the digital twin according to the optimized mapping rule table, forms a virtual-real fusion data stream, and converts the virtual-real fusion data stream into a target format and encrypts and transmits it to the target system;

[0036] An optimization module, the target system receives the data and returns feedback, and optimizes according to the feedback result.

[0037] The beneficial effects of the present invention are as follows: By using a bimodal model to automatically deduce field mapping relationships, the introduction of a priority table ensures that key fields are processed first when resources are limited, providing a preliminary rule framework for cross-system interaction. Through automated deduction and priority optimization, the efficiency and flexibility of rule generation are significantly improved. Especially in a dynamically changing heterogeneous environment, it can quickly adapt to changes in field semantics and load requirements, thereby enhancing the real-time nature of data interaction. In addition, by using digital twin technology to integrate source system data with the state of physical devices, the bidirectional mapping mechanism supports the dynamic synchronization of data between the virtual and the real, providing a unified interaction data stream for the target system and achieving real-time consistency of virtual and real data. This can effectively support the monitoring and decision-making of complex systems, thus significantly enhancing the integrity and practical value of data interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0039] Figure 1 It is a flowchart of the cross-platform and cross-system configurable data interaction method in Embodiment 1.

[0040] Figure 2 It is a flowchart of bimodal processing in Embodiment 1.

[0041] Figure 3 It is a federated learning training diagram in Embodiment 1.

[0042] Figure 4 It is a flowchart of virtual-real fusion in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0044] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0045] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not all refer to the same embodiment, nor is it a separate or selectively mutually exclusive embodiment with other embodiments.

[0046] Embodiment 1, referring to Figure 1 、 Figure 2 、 Figure 3 and Figure 4 , is the first embodiment of the present invention. This embodiment provides a cross-platform and cross-system configurable data interaction method, including the following steps:

[0047] S1. Real-time access the raw data from multiple heterogeneous data sources, obtain the metadata, traffic characteristics, and semantic analysis results of the raw data, and combine them into a comprehensive data set.

[0048] Specifically, it includes the following steps:

[0049] Use the OPCUA protocol to collect raw data in real time from multiple heterogeneous data sources (such as a trading system running on Windows, an Internet of Things sensor on Linux, and a cloud analysis platform). The OPCUA protocol is the Open Platform Communications Unified Architecture.

[0050] The adapter traverses the output interfaces of the data sources, reads the field names, such as TransactionID, DeviceID; reads the data types, such as strings, floating-point numbers; and the constraint conditions, such as length limits, value ranges; and generates metadata entries for each field, attaching source identifiers, such as source1, for subsequent traceability and processing;

[0051] The monitoring component samples the transmission activities of the data sources every second through embedded probes, calculates the query rate per second and the average packet size, and uses a sliding window algorithm to smooth the processing results to generate stable traffic characteristic values, such as query rate per second = 200, packet size = 1KB. To reflect the load status and transmission characteristics of the data sources.

[0052] Input all the field names into the BERT model. The BERT model calculates the semantic vector distance between fields based on word embedding technology and generates a similarity score (range 0-1). For example, TransactionID and DeviceID obtain a relatively high score because of their close semantics. This analysis process is limited to the field name level to avoid processing the complete data to reduce the computational overhead.

[0053] Use a data serialization tool to organize metadata, traffic characteristics, and similarity scores into a nested structure. For example, place metadata under the "sources" key, traffic characteristics under the "traffic" key, and similarity scores under the "semantics" key. Generate the final comprehensive dataset.

[0054] Furthermore, it provides a multi-dimensional input basis for cross-platform and cross-system data interaction. Metadata defines the structural characteristics of data, traffic characteristics reflect the real-time operating state, and semantic analysis reveals the potential associations between fields. The combination of the three lays a comprehensive data support for the subsequent derivation and optimization of mapping rules. Especially in dynamically changing scenarios, it ensures the comprehensiveness and adaptability of the interaction process.

[0055] S2. Use a dual-modal metadata derivation model to process the comprehensive dataset to obtain an initial mapping rule table.

[0056] Specifically, it includes the following steps:

[0057] S2.1. Call the BERT model through the semantic analysis module of the dual-modal derivation engine, take the field name and similarity score as inputs, and combine with a predefined context dictionary (including cross-system terms. For example, "ID" represents identification fields such as TransactionID and DeviceID, and "Value" represents numerical fields such as Amount and Temperature). Recalculate the semantic relevance between fields through the word embedding technology of the BERT model, considering the consistency of field types during the analysis. For example, string-type fields are more easily matched to generate preliminary mapping candidates such as TransactionID to DeviceID or Amount to Temperature. And assign an initial relevance weight to each pair of candidates.

[0058] S2.2. Simultaneously process the traffic characteristics in the comprehensive dataset through the traffic characteristics module of the dual-modal derivation engine. First, read the query rate and packet size, and judge the load status according to the preset load evaluation rules. For example, a query rate < 100 is low load, 100 - 300 is medium load, and > 300 is high load (specific rules can be customized according to the usage scenario and situation). Then, according to the pre-configured business importance rules, such as fields related to amount have a higher priority than identification fields (specific rules can be customized according to the usage scenario and situation), adjust the field priorities in combination with the load status. For example, mark Amount as a high-priority field, and generate a field priority table, where each field in the table is attached with a priority coefficient (range 0 - 1).

[0059] S2.3. Dual - mode Derivation Engine Operation Rule Synthesizer. The synthesizer aligns the preliminary mapping candidates with the field priority table, and eliminates the candidate pairs with a priority coefficient lower than 0.3 and a correlation weight lower than 0.7 (specific values can be customized according to their own needs and business requirements). Subsequently, an initial weight is assigned to each retained mapping rule. The initial weight can be obtained by fusing the correlation weight and the priority coefficient, or can be freely set according to experience, historical data, and business requirements. Then, sort by weight to form an initial mapping rule table, ensuring that high - priority fields are reflected first.

[0060] Preferably, compared with traditional static rule configuration or single - semantic analysis methods, the present invention significantly improves the intelligence and flexibility of mapping rule generation through dual - mode fusion, avoiding the inefficiency of manual intervention and the limitations of single - dimension analysis. The introduction of the priority mechanism further optimizes resource allocation, enabling the rule table to dynamically adapt to business needs and loads, thereby improving the initial accuracy and execution efficiency of data interaction and providing high - quality input for subsequent optimization.

[0061] S3. Establish a global mapping model through federated learning, input the initial mapping rule table into the global mapping model, and obtain an optimized mapping rule table.

[0062] Specifically, it includes the following steps:

[0063] The federated learning framework runs the distribution manager, which broadcasts the initial mapping rule table to each training node in JSON format through the TLS protocol. The distribution manager verifies the reception status of each training node. If a training node does not respond, it is marked as offline and a log is recorded to prepare a unified starting point for distributed training.

[0064] S3.1. Each training node loads local private data, such as the historical records of the trading system and the real - time readings of sensors; uses the mapping pairs in the initial rule table as supervision signals to train the MLP model (Multi - Layer Perceptron) to adjust the rule weights. Set the number of iterations, and use the mean - square error loss function to update the MLP model parameters to generate local optimized parameters. The local optimized parameters of each node are also the node parameters. The MLP model is used as the mapping model because of its light - weight and high efficiency compared to deep networks.

[0065] S3.2. The node runs the encryption module, uses differential privacy technology to encrypt the local optimized parameters to generate irreversible parameter update values. The encrypted local optimized parameters are uploaded through the TLS channel, and a 5 - second timeout limit is set (to meet real - time requirements, the specific value can be customized according to personal needs and business). If the upload fails, it will be retried once to ensure that the server receives the contributions of each training node.

[0066] S3.3. The server runs an aggregator to decrypt the received local optimization parameters, and uses a weighted average algorithm to fuse the local optimization parameters of each training node, removing outliers, such as parameters that deviate from the mean by two standard deviations, to generate global mapping model parameters.

[0067] S3.4. The global mapping model loads the initial rule table, recalculates the weight of each rule based on the trained parameters. For example, the weight from TransactionID to DeviceID may be adjusted from the initial value to a new value, and the rule table is sorted in descending order of weights. Verify the rule consistency, such as checking for conflicting mappings. If problems are found, roll back to the previous version of the parameters, and finally output the optimized mapping rule table.

[0068] Preferably, compared with the traditional centralized training method, the present invention avoids the risk of privacy leakage caused by uploading raw data through federated learning, and significantly improves data security by only transmitting encrypted parameters. At the same time, the distributed training and asynchronous aggregation mechanism make full use of the computing power of each node, accelerating the rule optimization speed, enabling the optimized mapping rule table to more accurately adapt to the complex requirements of the multi-source heterogeneous environment, thereby improving the global accuracy and real-time adaptability of data interaction.

[0069] S4. Associate the source system data of the target system with the digital twin to form a virtual-real fusion data stream, and convert the virtual-real fusion data stream into a target format and encrypt it for transmission to the target system.

[0070] Specifically, it includes the following steps.

[0071] The digital twin mapping engine calls the source system interface through the preset adapter module, parses the latest business data, and stores it in the temporary buffer. Align the buffer data with the source fields in the optimized mapping rule table, and mark it as pending processing if the field is missing to ensure that the data matches the mapping rules.

[0072] The digital twin mapping engine starts a two-way dynamic mapping mechanism. First, read the mapping pairs in the mapping rule table, extract the source system field values from the buffer, extract the corresponding attribute values from the digital twin, and bind the field pairs through a matcher. The two-way dynamic mapping mechanism supports two-way synchronization, which means that when the source system data is updated, it is synchronized to the digital twin, and when the digital twin data changes, it is feedback to the source system, thereby generating a fusion data stream containing virtual-real associations and recording the binding relationship.

[0073] Generate a comprehensive credibility score based on historical mapping records, real-time semantic analysis, and business priorities (used to quantify the criticality of fields in the business. For example, the field Amount is more important than DeviceID, and the business priorities can be freely set according to business requirements and individual needs). The comprehensive credibility score is based on weighted calculation, and the weight distribution can be 0.4 for history, 0.4 for real-time, and 0.2 for business. Specifically, the weight settings can be customized according to the usage scenario and actual situation. The comprehensive credibility score is used to evaluate the reliability of each mapping rule.

[0074] Set a confidence threshold based on historical mapping records, such as 0.7. If the comprehensive credibility score of a certain rule is lower than the confidence threshold (for example, due to outliers causing a decrease in reliability or other reasons), it is marked as unreliable and sent back to the step that triggered the mapping rule for re-derivation; if it is higher than the confidence threshold, the mapping rule is retained. And record the adjustment process in the log to ensure traceability, so as to obtain the verified virtual-real fusion data stream.

[0075] The digital twin mapping engine loads the template of the target system, and fills the virtual-real fusion data into the target system template through a format converter. For example, if the target system template is XML, the virtual-real fusion data is converted into XML format. After the conversion is completed, a verification code is generated and attached to the virtual-real fusion data in the converted format, and then transmitted to the target system.

[0076] Preferably, the traditional method is single-item data synchronization or using a mapping method without a verification method. However, the present invention significantly improves the integrity and reliability of virtual-real data fusion through bidirectional mapping and confidence evaluation, avoiding interaction failures caused by data deviation. Moreover, the encrypted transmission and verification mechanism ensures data security, enabling the fusion data stream to efficiently adapt to the diverse requirements of the target system, thereby enhancing the practicality and stability of the interaction process.

[0077] S5. The target system receives the data and returns feedback, and optimizes according to the feedback result.

[0078] Specifically, it includes the following steps.

[0079] After the target system receives the transmitted data, it first verifies the integrity of the data through the verification code. After confirming the integrity, it reads the transmitted data, extracts the field values, and compares them with the expected format of the target system to generate an interaction feedback. The interaction feedback content includes the interaction status (such as success or failure), processing delay, and exception information, and then the interaction feedback is sent back.

[0080] S5.1. Adjust the mapping rule table according to the feedback from the target system (such as "Temperature out of range"). For example, if the feedback indicates that the Temperature is out of range, increase the weight of Temperature, synchronously update the global mapping model, and output the updated mapping rule table and the digital twin mapping engine.

[0081] Define a set of states, such as including normal policies and exception recovery policies. When the interaction feedback of the target system shows an anomaly, switch to the exception recovery policy, prioritize the transmission of key field data, and record the state changes to optimize the transmission efficiency.

[0082] This embodiment also provides a cross-platform and cross-system configurable data interaction system, including:

[0083] A collection module that accesses raw data in real time from multiple heterogeneous data sources, obtains the metadata, traffic characteristics, and semantic analysis results of the raw data, and combines them into a comprehensive data set;

[0084] A processing module that uses a bimodal metadata derivation model to process the comprehensive data set to obtain an initial mapping rule table;

[0085] An optimization module that establishes a global mapping model through federated learning, inputs the initial mapping rule table into the global mapping model, and obtains an optimized mapping rule table;

[0086] A transmission module that correlates the source system data of the target system with the twin body according to the optimized mapping rule table to form a virtual-real fusion data stream, converts the virtual-real fusion data stream into a target format, and encrypts and transmits it to the target system;

[0087] An optimization module that the target system receives the data and returns feedback, and performs optimization according to the feedback results.

[0088] In summary, the present invention: automatically derives the field mapping relationship using a bimodal model, the introduction of the priority table ensures that key fields are processed first when resources are limited, provides a preliminary rule framework for cross-system interaction, greatly improves the efficiency and flexibility of rule generation through automated derivation and priority optimization, especially in a dynamically changing heterogeneous environment, can quickly adapt to changes in field semantics and load requirements, thereby enhancing the real-time nature of data interaction; in addition, by using digital twin technology to fuse the source system data with the physical device state, the bidirectional mapping mechanism supports the dynamic synchronization of data between the virtual and the real, provides a unified interaction data stream for the target system, realizes the real-time consistency of virtual and real data, can effectively support the monitoring and decision-making of complex systems, and thus significantly improves the integrity and practical value of data interaction.

[0089] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and all of them should be covered by the scope of the claims of the present invention.

Claims

1. A cross-platform and cross-system configurable data interaction method, characterized by: include, Access raw data from multiple heterogeneous data sources in real time, obtain metadata, traffic characteristics, and semantic analysis results of the raw data, and combine them into a comprehensive data set; The bimodal metadata inference model is used to process the comprehensive dataset to obtain the initial mapping rule table; Establish a global mapping model through federated learning, input the initial mapping rule table into the global mapping model, and obtain an optimized mapping rule table; Associate the source system data with the twin to form a virtual-real fusion data stream, convert the virtual-real fusion data stream into the target format and encrypt and transmit it to the target system; The target system receives data and returns feedback, and performs optimization based on the feedback results; The bimodal metadata inference model is used to process the comprehensive data set to obtain the initial mapping rule table, which specifically includes the following steps: The semantic analysis result refers to the semantic analysis of the field name of the metadata through the BERT model to generate a similarity score; The bimodal metadata inference model deeply analyzes the field semantics based on the metadata and similarity scores, generates preliminary mapping candidates, processes the traffic characteristics, determines the matching priority of the fields in the metadata, and generates a field priority table; Screen and sort the preliminary mapping candidates based on the field priority table to generate an initial mapping rule table; The associating source system data with the twin specifically includes the following steps: The digital twin mapping engine starts a bidirectional dynamic mapping mechanism, reads the mapping pairs in the optimized mapping rule table, extracts the source system field values ​​from the buffer, extracts the corresponding attribute values ​​from the twin, and binds the field pairs through the matcher.

2. The cross-platform and cross-system configurable data interaction method according to claim 1, characterized in that: A comprehensive dataset in JSON format that integrates the metadata, traffic characteristics, and similarity scores of the original data.

3. The cross-platform and cross-system configurable data interaction method according to claim 2, characterized in that: A global mapping model is established through federated learning, and the initial mapping rule table is input into the global mapping model to obtain an optimized mapping rule table, which specifically includes the following steps: The federated learning framework distributes the initial mapping rule table to each training node, and each training node trains the MLP model based on local data; Upload the encrypted parameters to the asynchronous parameter server. After receiving the parameters of each training node, the server aggregates and generates a global mapping model. The initial mapping rule table is input into the global mapping model, and the optimized mapping rule table is output.

4. The cross-platform and cross-system configurable data interaction method according to claim 3, characterized in that: The forming of the virtual-real fusion data stream specifically includes the following steps: The optimized mapping rule table is loaded through the digital twin mapping engine, real-time data is extracted from the source system, associated with the twin data, and the fields are bound through a two-way dynamic mapping mechanism to form a virtual-real fusion data flow.

5. The cross-platform and cross-system configurable data interaction method according to claim 4, characterized in that: The virtual-reality fusion data stream is converted into a target format and encrypted and transmitted to the target system, specifically including the following steps: Use weighted decision-making technology to generate a comprehensive credibility score for virtual-real fusion data streams based on historical mapping records, real-time semantic analysis, and business priorities; A confidence threshold is set based on historical mapping records, and the comprehensive credibility score is compared with the confidence threshold. If it is lower than the confidence threshold, the process of re-deriving mapping rules is triggered; Convert the virtual-reality fusion data stream into a format that the target system can recognize, and attach a checksum to transmit it to the target system.

6. The cross-platform and cross-system configurable data interaction method according to claim 5, characterized in that: The target system receives data and returns feedback, and performs optimization based on the feedback results, which includes the following steps: The target system parses the received virtual-reality fusion data stream, generates interactive feedback, and transmits it back to the adaptive mapping engine through the data bus; The adaptive mapping engine adjusts the current mapping rule table according to the interactive feedback, updates the global mapping model through incremental training of federated learning, synchronously adjusts the digital twin state, and switches the strategy to optimize the transmission.

7. A cross-platform and cross-system configurable data interaction system, based on the cross-platform and cross-system configurable data interaction method according to any one of claims 1 to 6, characterized in that: include, The collection module accesses raw data from multiple heterogeneous data sources in real time, obtains metadata, traffic characteristics, and semantic analysis results of the raw data, and combines them into a comprehensive data set; The processing module processes the comprehensive dataset using the bimodal metadata inference model to obtain an initial mapping rule table; The optimization module establishes a global mapping model through federated learning, inputs the initial mapping rule table into the global mapping model, and obtains an optimized mapping rule table; The transmission module associates the target system source system data with the twin according to the optimized mapping rule table to form a virtual-real fusion data stream, and converts the virtual-real fusion data stream into the target format and encrypts it for transmission to the target system; In the optimization module, the target system receives data and returns feedback, and performs optimization based on the feedback results.

Images