Heterogeneous device adaptive access gateway, method and electronic device

Abstract

The invention provides a heterogeneous device self-adaptive access gateway, a method and an electronic device. The gateway comprises a protocol identification module used for inputting a data message of a connected device to a first large model and outputting a protocol type of the data message predicted by the first large model; the format conversion module is used for converting the data format of the data message into a target format based on the data message and the protocol type; and the processing and analyzing module is used for processing and analyzing the data message in the target format based on the data type of the data message. Through the protocol identification module, the communication protocol of the heterogeneous equipment can be automatically identified based on a large model technology, and the equipment access efficiency is improved; through the combination of the format conversion module and the processing and analysis module, the data formats of the data messages of different devices can be converted into target formats, the processing problem caused by data format differences is solved, and the efficiency and accuracy of data processing and analysis are improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a heterogeneous device adaptive access gateway, method, and electronic device. Background Art

[0002] In the field of industrial automation, with the rapid development of the Industrial Internet of Things (IIoT), more and more devices in factories require network access to enable functions such as data collection, remote monitoring, and collaborative control. However, industrial environments contain a large number of heterogeneous devices from different manufacturers, with varying communication protocols, data formats, and interface standards. Traditional industrial gateways face numerous challenges in connecting to these heterogeneous devices, such as complex protocol adaptation and difficulty analyzing data, resulting in inefficient device access.

[0003] Therefore, how to provide a gateway that can adaptively connect heterogeneous devices has become a technical problem that needs to be solved urgently in the industry. Summary of the Invention

[0004] The present application provides a heterogeneous device adaptive access gateway, method and electronic device to solve the technical problem of how to provide a gateway capable of adaptively accessing heterogeneous devices in the prior art.

[0005] In a first aspect, the present application provides a heterogeneous device adaptive access gateway, comprising: a protocol identification module, configured to input a data packet of a connected device into a first large model and output a protocol type of the data packet predicted by the first large model; the first large model is trained based on sample data packets and sample protocol types corresponding to the sample data packets; a format conversion module, configured to convert the data format of the data message into a target format based on the data message and the protocol type; A processing and analysis module is used to process and analyze the data message in the target format based on the data type of the data message.

[0006] In some embodiments, the protocol identification module is specifically used to: Collect data packets from multiple connected devices based on different types of data acquisition interfaces and different types of acquisition drivers; Inputting the data message into the first large model; Controlling the data cleaning layer of the first large model to process the data message based on a preset data cleaning rule; Controlling the first data structure feature of the data message after the feature extraction layer of the first large model extracts and processes: The protocol identification layer of the first large model is controlled to identify the protocol type of the data message based on the first data structure feature.

[0007] In some embodiments, the format conversion module is specifically configured to: Inputting the data message and the protocol type into a second large model to obtain semantic features of the data message output by the second large model; generating a data format conversion strategy corresponding to the data message based on the semantic feature; The data format of the data message is converted into the target format based on the data format conversion strategy.

[0008] In some embodiments, the protocol type includes an unknown protocol type and a protocol type stored in a database; the format conversion module is further specifically configured to: In a case where the protocol type is the unknown protocol type, inputting the data message and the protocol type into a second large model to obtain the semantic features and second data structure features of the data message output by the second large model; generating a data format conversion strategy corresponding to the data message based on the semantic feature and the second data structure feature; In which, the second data structure feature is predicted by the second large model based on a confidence assessment mechanism; the confidence assessment mechanism includes obtaining the confidence of the initial second data structure feature predicted by the second large model, and when the confidence is greater than a preset threshold, using the initial second data structure feature as the second data structure feature.

[0009] In some embodiments, the format conversion module is further specifically configured to: Verifying the completeness and data type of the data message converted into the target format; If the verification is successful, the data message in the target format is sent to the processing and analysis module.

[0010] In some embodiments, the processing and analysis module is specifically configured to: When the data message is a sensor data message, performing statistical analysis on the data message in the target format based on a sliding window algorithm to obtain statistical data of the data message in the target format, analyzing the operating status of the device based on the statistical data, and generating early warning information for the device when the operating status is abnormal; In the case that the data message is a control instruction data message, the control instruction corresponding to the control instruction data message is parsed and the control instruction is executed.

[0011] In some embodiments, the invention further includes a management configuration module, wherein the management configuration module is configured to: Provide a registration interface for the device, perform security authentication for the device, obtain device information for the device, collect and display device operation data for the device, and perform group management and permission settings for multiple devices.

[0012] In some embodiments, a communication interface module is further included, wherein the communication interface module is configured to: Configuring drivers for multiple types of communication interfaces to support communication connections for devices with multiple types of communication interfaces; In the event of a communication link failure, the failure is repaired based on a preset failure recovery mechanism.

[0013] In a second aspect, the present application provides a method for adaptive access of heterogeneous devices, including: Inputting a data message of a device connected to the gateway into a first large model, and outputting a protocol type of the data message predicted by the first large model; the first large model is trained based on sample data messages and sample protocol types corresponding to the sample data messages; converting a data format of the data message into a target format based on the data message and the protocol type; The data message in the target format is processed and analyzed based on the data type of the data message.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to implement the above method when executing the program through the computer program.

[0015] The heterogeneous device adaptive access gateway, method and electronic device provided in the embodiments of the present application can automatically identify the communication protocols of heterogeneous devices based on large model technology through the protocol identification module, thereby improving device access efficiency and reducing maintenance costs without the need for human intervention; through the combination of the format conversion module and the processing and analysis module, the data format of data messages of different devices can be converted into the target format, solving the processing difficulties caused by data format differences, and can automatically connect devices to the gateway, thereby improving the efficiency and accuracy of data processing and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] Figure 1A schematic diagram of the structure of a heterogeneous device adaptive access gateway provided in an embodiment of the present application; Figure 2 A schematic diagram of a process for adaptive access of heterogeneous devices provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0019] It should be noted that the terms "first", "second" etc. in this application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable in appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, the process, method, system, product or equipment comprising a series of steps or modules is not necessarily limited to those steps or modules clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products or equipment.

[0020] Figure 1 A schematic diagram of the structure of the adaptive access gateway for heterogeneous devices provided in the embodiment of the present application is shown as follows: Figure 1 As shown, the gateway includes a protocol identification module 110 , a format conversion module 120 and a processing and analysis module 130 .

[0021] The protocol identification module 110 is configured to input a data packet from a connected device into a first large model and output a protocol type of the data packet predicted by the first large model; the first large model is trained based on sample data packets and sample protocol types corresponding to the sample data packets; A format conversion module 120 is configured to convert the data format of the data message into a target format based on the data message and the protocol type; The processing and analysis module 130 is configured to process and analyze the data message in the target format based on the data type of the data message.

[0022] Specifically, the first model is a protocol recognition model built based on a neural network, which is used to predict the communication protocol type from a data message.

[0023] The protocol type refers to the communication protocol category to which the data packet belongs.

[0024] A data message is a structured data unit transmitted by a device. It can include data such as a protocol header, data body, and verification information, and carries application layer communication content.

[0025] Protocol Identification Module 110: Utilizes large-scale model technology to automatically identify and adapt the communication protocols of heterogeneous devices. Using a trained first large-scale model, this module rapidly analyzes the structural characteristics of data sent by devices, identifies the protocol type used, and automatically generates corresponding adaptation strategies, eliminating the need for manual configuration of protocol mapping rules.

[0026] Format Conversion Module 120: This module leverages the semantic understanding and data format conversion capabilities of the second largest model to address the differences in data formats generated by different devices. This module converts device data packets into a target format to facilitate subsequent data processing and analysis. The target format can be a standardized, uniform format or a specific format tailored to each data packet. The specific format can be customized based on the specific situation.

[0027] Because the first model may not predict the protocol type of a data packet, in which case the protocol type output by the first model is an unknown protocol type, the data packet and its protocol type (unknown protocol type) can be input into the trained second model. The second model predicts relevant information about the data structure based on the characteristics and patterns of the input data packet and its protocol type. For example, the second model can output the length, data type, possible field names, and relationships between fields of each field. Based on this information, the data format of the data packet is then converted to the target format. The second model is trained based on sample data packets, the sample protocol type corresponding to the sample data packets, the sample data structure corresponding to the sample data packets, and the sample semantic features.

[0028] Processing and Analysis Module 130: Utilizing the powerful computing and data analysis capabilities of the third model, this module processes and analyzes data packets in the target format in real time. It supports statistical analysis, trend prediction, and anomaly detection. Based on the analysis results, it generates corresponding control instructions or early warning information, providing intelligent decision-making support for industrial production processes.

[0029] According to an embodiment of the present application, any multiple modules among the protocol identification module 110, the format conversion module 120 and the processing analysis module 130 can be combined into one module for implementation, or any one of the modules can be split into multiple modules.

[0030] Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in one module.

[0031] According to an embodiment of the present application, at least one of the protocol identification module 110, the format conversion module 120 and the processing and analysis module 130 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware, or in an appropriate combination of any of them.

[0032] Alternatively, at least one of the protocol identification module 110 , the format conversion module 120 and the process analysis module 130 may be at least partially implemented as a computer program module, and when the computer program module is executed, the corresponding function may be performed.

[0033] The adaptive access gateway for heterogeneous devices provided in the embodiment of the present application can automatically identify the communication protocols of heterogeneous devices based on large model technology through the protocol identification module, thereby improving device access efficiency and reducing maintenance costs without the need for human intervention; through the combination of the format conversion module and the processing and analysis module, the data format of data messages of different devices can be converted into the target format, solving the processing difficulties caused by data format differences, and automatically connecting devices to the gateway, thereby improving the efficiency and accuracy of data processing and analysis.

[0034] It should be noted that each implementation method of the present application can be freely combined, the order can be changed, or it can be executed separately, and does not need to rely on or depend on a fixed execution order.

[0035] In some embodiments, the protocol identification module 110 is specifically configured to: Collect data packets from multiple connected devices based on different types of data acquisition interfaces and different types of acquisition drivers; Inputting data packets into the first model; The data cleaning layer of the first large model is controlled to process data messages based on preset data cleaning rules; Control the first data structure feature of the data message after the feature extraction layer of the first large model extracts and processes: The protocol identification layer of the first large model is controlled to identify the protocol type of the data message based on the first data structure feature.

[0036] Specifically, the protocol identification module 110 mainly has the following functions: 1. Data collection and processing: Application layer protocols and corresponding data packets can be collected from the internet or offline devices, such as device status information, sensor data, and control instructions. Data noise can be removed, for example, by processing duplicate data, erroneous data, or incomplete messages. Application layer protocols are labeled, including both known and unknown protocols. Known protocols require annotation of protocol names and versions. Data packets are labeled, for example, with fixed headers, variable headers, and message bodies. The message body can be labeled with information such as encoding format, data type, data length, check digits, and verification methods for use in training the first large model.

[0037] The data acquisition interface is a hardware or software interface used to adapt to different device communication methods (such as Ethernet, serial port and field bus, etc.) to realize the collection of data messages.

[0038] The acquisition driver is a software module developed for a specific communication protocol or device, responsible for extracting data packets from the device.

[0039] You can configure various data collection interfaces and develop corresponding collection drivers for different device communication methods to ensure that the gateway can efficiently and stably collect application layer protocols and corresponding data packets from the internet or offline devices. The application layer protocol defines the transmission rules and semantics of data at the application layer, while the protocol type of the data packet refers to the protocol used at the transport layer or network layer.

[0040] Multi-threading or asynchronous processing mechanisms can be used to collect relevant data to improve the efficiency of data collection and avoid affecting the performance of the overall system due to data collection delays of a single device.

[0041] After data packets are collected, they are fed into the first large model. The data cleaning layer of the first large model removes data noise, including removing duplicate data, erroneous data, and incomplete messages. The data cleaning layer can process different types of noise based on data cleaning rules.

[0042] Data cleaning rules refer to preset filtering conditions or processing logic. Detailed data cleaning rules can be set for common types of data noise (such as duplicate data, erroneous data, or incomplete messages). For example, for duplicate data, you can determine whether it is duplicate data and perform deduplication processing by comparing features such as the timestamp and message content of the data message; for erroneous data, according to the protocol specifications and data format requirements, identify data that does not meet the requirements and mark it as abnormal data for subsequent processing. For incomplete messages, try to splice the messages based on the message structure characteristics of the protocol. For example, for data messages based on the Transmission Control Protocol / Internet Protocol (TCP / IP), the segmented messages can be reassembled into complete messages based on information such as the TCP sequence number.

[0043] 2. Feature extraction function: Use the feature extraction layer of the first model to identify the length, frequency, field distribution and other characteristics of the data message, analyze its data structure characteristics, and extract the first data structure feature of the data message.

[0044] The pattern recognition capability of the first model can be used to identify repeated patterns or structures in data messages and analyze their semantic and structural characteristics.

[0045] Specific algorithms can be developed to identify recurring patterns or structures in data messages. For example, a sequential pattern mining algorithm can be used to analyze the field sequences in a data message to identify frequent patterns. For example, in the communication protocols of certain industrial devices, specific field sequences may appear repeatedly, which may represent specific control instructions or data formats.

[0046] Identified patterns are further analyzed for their structural characteristics and semantic meaning. For example, by analyzing the relationships between fields in the pattern (such as sequence and dependencies), their role in the protocol can be determined. Identified patterns can be compared with a library of known protocol patterns to determine whether they represent variations of known protocols or new protocol patterns.

[0047] The feature extraction layer of the first model uses a variety of feature extraction algorithms to comprehensively analyze features such as the length, frequency, and field distribution of data packets. For example, a sliding window algorithm is used to statistically analyze the frequency of data packets, analyzing the changing patterns of packet occurrence within different time windows. For field distribution features, the data structure characteristics of the data packets are extracted by analyzing the value distribution range and frequency of each field in the packet. Data structure features are features that reflect the data organization and inherent patterns, extracted by analyzing information such as the value distribution range and frequency of each field in the packet.

[0048] For data packet length features, we analyze not only the length of individual packets but also the distribution of packet lengths. For example, we collect statistics such as the mean and variance of packet lengths, as well as the frequency of occurrence of packets of different lengths, thus providing richer feature information for subsequent pattern recognition and model training.

[0049] After extracting the first data structure feature of the data message, the protocol type of the data message is identified through the protocol identification layer of the first large model and then marked.

[0050] If the protocol identification layer of a large model has identified the specific protocol type of the data message, the protocol name and version will be marked; the fixed header, variable header, message body, etc. of the data message will be marked, including the encoding format, data type, data length, check bit and verification method.

[0051] A database of known protocol types can be pre-established to store detailed information such as the names, versions, and message structures of various common industrial communication protocols. During the data annotation process, the protocol names and versions are automatically annotated by comparing the collected protocols with the protocol information in the database.

[0052] When it comes to protocol message annotation, the location and characteristics of each part, such as the fixed header, variable header, and message body, are clearly defined according to the protocol specification. For example, in the Modbus protocol, the fixed header includes fields such as the function code, the variable header may include information such as the device address, and the message body contains specific sensor data or control instructions.

[0053] If the protocol identification layer of the first model fails to identify the specific protocol type of the data message, a heuristic annotation method is used to preliminarily annotate the message structure of the unknown protocol and record the uncertainty of the annotation.

[0054] First, based on the first data structure features of the data message, such as its length and field distribution, we preliminarily determine its possible message structure. For example, if the message length is fixed and contains some common field features (such as a specific byte sequence), we can mark it as a possible fixed header part of an unknown protocol.

[0055] For message body annotation, you can make a preliminary judgment based on the data encoding format (such as ASCII, binary, etc.) and data type (such as integer, floating point number, string, etc.), and annotate possible encoding format, data type, data length, etc. At the same time, record the uncertainty of these annotations to facilitate further analysis and adjustment during model training or optimization.

[0056] 3. Model Training: You can select a large model suitable for processing sequence data to build the primary model, for example, the Transformer architecture or its variants. The primary model can be customized based on the characteristics of the device communication protocol data. For example, parameters such as the number of layers and hidden units in the primary model can be adjusted to accommodate the complexity of different protocol data.

[0057] The cleaned samples are divided into a training set and a validation set. The training set is used to train the first large model, and the validation set is used to evaluate the model performance of the first large model. The first large model can be trained using various optimization algorithms (such as Adam and SGD). Using the training set to train the first large model, the first large model can predict the first data structure information defined by the protocol type corresponding to the sample data packet based on the input sample data packet, and perform reverse parsing based on the first data structure information.

[0058] For example, the first large model can predict the protocol type, field distribution, field type and other information of the sample data message based on the input sample data message, compare this information with the actual protocol definition, calculate the loss function and perform backpropagation optimization.

[0059] The validation set can be used to evaluate the performance of the top model. Key evaluation metrics include protocol type recognition accuracy, data structure prediction accuracy, and model generalization. For example, by testing the top model's ability to recognize and parse different protocol types on the validation set, metrics such as the model's accuracy and recall can be calculated.

[0060] Based on the performance of the first model on the validation set, the model parameters of the first model are adjusted to optimize its accuracy and generalization ability.

[0061] For example, if the recognition accuracy of the first model is low for certain protocol types, the model performance of the first model can be improved by adjusting the hyperparameters of the first model (such as the learning rate and regularization coefficient, etc.) or further analyzing and processing the training samples (such as adding data augmentation operations, etc.).

[0062] The adaptive access gateway for heterogeneous devices provided in the embodiment of the present application can integrate functions such as multi-protocol adaptation, intelligent data cleaning, deep feature extraction, and effective identification of protocol types through the protocol identification module, thereby realizing efficient parsing of data packets of heterogeneous devices and accurate identification of protocols.

[0063] In some embodiments, the format conversion module 120 is specifically configured to: Inputting the data message and the protocol type into the second largest model to obtain the semantic features of the data message output by the second largest model; Generate data format conversion strategies corresponding to data messages based on semantic features; The data format of the data message is converted into the target format based on the data format conversion strategy.

[0064] The protocol type includes an unknown protocol type and a protocol type stored in a database; the format conversion module 120 is further specifically configured to: In the case where the protocol type is an unknown protocol type, the data message and the protocol type are input into the second largest model to obtain the semantic features and the second data structure features of the data message output by the second largest model; Generate a data format conversion strategy corresponding to the data message based on the semantic feature and the second data structure feature; Among them, the second data structure feature is predicted by the second largest model based on a confidence assessment mechanism; the confidence assessment mechanism includes obtaining the confidence of the initial second data structure feature predicted by the second largest model, and when the confidence is greater than a preset threshold, using the initial second data structure feature as the second data structure feature.

[0065] The format conversion module 120 is further specifically configured to: Verify the completeness and data type of the data message converted to the target format; If the verification is successful, the data message in the target format is sent to the processing and analysis module 130.

[0066] Specifically, the data format conversion strategy is a set of rules that converts the data format of data packets of heterogeneous devices into a target format based on semantic features and structural features.

[0067] The second largest model is a neural network model used to extract the semantic and structural features of data packets.

[0068] Semantic features are context-related information extracted from data packets.

[0069] Based on the semantic understanding capability of the second largest model, in-depth analysis can be performed on the data packets generated by different devices.

[0070] For example, for sensor data packets, the second largest model can understand the semantic meaning of the data based on the data packet's contextual information (such as sensor type, measurement object, and timestamp). For example, for temperature sensor data packets, the second largest model can determine the actual significance of the data in the industrial production process based on its measurement object (such as a specific device component) and timestamp information, thereby deriving its semantic features.

[0071] For example, for control command data packets, the second-largest model can understand the semantics of the command, including information such as the target, operation type, and parameters. For example, for a command to control motor speed, the second-largest model can identify that the target device is the motor, the operation type is speed adjustment, and the parameter is the specific speed value, thereby deriving its semantic features.

[0072] Formulate corresponding data format conversion strategies based on semantic features.

[0073] For example, for sensor data packets, they are converted into a target format based on their semantic features and data format. The target format may include fields such as device identification, data type, data value, and timestamp.

[0074] For another example, a data message of a temperature sensor is converted into a target format of {"device identifier":"temperature sensor 1","data type":"temperature","data value":25.5","timestamp":"2025-04-08 10:00:00"}.

[0075] For example, a control command data message is converted into a unified control command format, also known as the target format, based on the command's semantic characteristics and device requirements. For example, a command to control motor speed is converted into the target format of {"Device ID": "Motor 1", "Operation Type": "Adjust Speed", "Parameters": 1000}.

[0076] If the first large model fails to identify the protocol type of the data message, and the protocol type of the data message output by the first large model is an unknown protocol type, the second large model is required to further determine the second data structure feature of the data message.

[0077] The data message is input into the trained second largest model, and the second data structure feature is predicted based on the characteristics and pattern of the input data message.

[0078] The second largest model can preliminarily determine the initial second data structure features. Then, using a confidence assessment mechanism, it compares the confidence level of the initial second data structure features with a preset threshold to determine the reliability of the preliminary prediction results. If the confidence level is greater than the preset threshold, the initial second data structure features are used as the second data structure features. If the confidence level is less than or equal to the preset threshold, the initial second data structure features are marked as data for further analysis and temporarily stored pending subsequent processing or manual intervention.

[0079] According to the second data structure feature output by the second largest model, the data format of the data message of the unknown protocol type is converted into the target format.

[0080] During the format conversion process, a fault-tolerant mechanism can be used to handle possible conversion errors (such as data type mismatches and missing fields). For example, if the data type of a field is incorrectly predicted, it can be corrected based on contextual information or marked as abnormal data.

[0081] The converted data message is verified to ensure that it conforms to the target format. For example, the data integrity and data type consistency are checked. If the verification passes, the data message is sent to the processing and analysis module 130. If the verification fails, error handling is performed and recorded for subsequent analysis and improvement.

[0082] With the continuous updating and optimization of large model technology, the gateway's protocol adaptation and data analysis capabilities will continue to improve, with good scalability and upgradeability.

[0083] The heterogeneous device adaptive access gateway provided in the embodiment of the present application realizes the format standardization of data messages through in-depth analysis of the semantic and structural features of data messages by the format conversion module, combined with the confidence assessment mechanism, thereby ensuring the compatibility and processing efficiency of heterogeneous device data.

[0084] In some embodiments, the processing and analysis module 130 is specifically configured to: When the data message is a sensor data message, a statistical analysis is performed on the data message in the target format based on a sliding window algorithm to obtain statistical data of the data message in the target format. The operating status of the device is analyzed based on the statistical data, and when the operating status is abnormal, an early warning information of the device is generated; In the case that the data message is a control instruction data message, the control instruction corresponding to the control instruction data message is parsed and the control instruction is executed.

[0085] Specifically, efficient real-time data processing algorithms can be set up, combined with the powerful computing capabilities of large models, to quickly process collected device data packets. For example, for sensor data packets, a sliding window algorithm can be used to perform real-time statistical analysis on the data packets, calculating the data packet statistics. The statistical data can include statistics such as the average, maximum, and minimum values ​​of the data, so as to timely understand the operating status of the device.

[0086] For control command data packets, the system performs real-time command parsing and execution based on the semantics and execution requirements of the control command. For example, when a data packet containing a command to start or stop a device is received, the control command content is immediately parsed and sent to the corresponding device control interface to ensure a timely response.

[0087] The processing and analysis module 130 can provide various statistical analysis functions, including statistical analysis of equipment operating status, equipment failure rate, and production efficiency. For example, by analyzing sensor data packets, it can calculate the equipment's operating status (such as normal operation or failure shutdown) over different time periods and calculate indicators such as the equipment's mean time between failures (MTBF).

[0088] For statistical analysis of production efficiency, the production efficiency of the equipment can be calculated based on the production data of the equipment (such as output and production time, etc.), and compared with historical data to analyze the changing trend of production efficiency.

[0089] The process analysis module 130 may also provide trend prediction functionality.

[0090] Leveraging the time series analysis capabilities of large models, you can use them to predict trends in data packets. For example, for device temperature data, the large model can predict future device temperature changes based on historical and current data trends. Trend predictions can help identify potential device anomalies in advance and enable timely preventative measures.

[0091] For the trend forecast of production data, it is possible to predict future production output, equipment utilization and other indicators, and provide a reference basis for the company's production planning and resource allocation.

[0092] The processing and analysis module 130 can be configured with a large-scale model-based anomaly detection algorithm to perform real-time anomaly detection on the data in the device's data packets. For example, by analyzing the changing patterns of sensor data, when abnormal data fluctuations (such as a sudden temperature increase or a sharp drop in pressure) occur, the anomaly is detected in a timely manner and an early warning message is issued.

[0093] A multi-level warning mechanism is used for anomaly detection results. Warning information is categorized into different levels based on the severity of the anomaly, such as minor, moderate, and severe. Appropriate action is taken for each level of anomaly. For example, minor anomalies can be recorded in a log for subsequent analysis; severe anomalies are immediately notified to relevant personnel for resolution.

[0094] Based on the data analysis results of the data message, corresponding control instructions can be generated. For example, if the data analysis result shows that the device temperature exceeds a preset threshold, an instruction to start the cooling device is generated and sent to the corresponding device control interface to reduce the device temperature and ensure normal operation of the device.

[0095] Intelligent decision-making algorithms can be used to generate control instructions by comprehensively considering factors such as the current status of the equipment, historical data, and production requirements. For example, while considering equipment cooling, factors such as the energy consumption of the cooling equipment need to be considered to generate the optimal control instructions.

[0096] When analyzing data packets and detecting an abnormality or predicting a possible abnormality, a corresponding warning message is generated. The warning message can include detailed information about the abnormality, such as the abnormal type, abnormal device, abnormal time, and abnormal cause. For example, the warning message may be "The temperature of device 1 has risen abnormally. The current temperature is 50°C, exceeding the normal operating temperature range. The possible cause is device overload. Please check and resolve it immediately." Warning information can be sent to relevant personnel in a variety of ways, such as SMS, email, and system notification. At the same time, the warning information will be recorded in the system log to facilitate subsequent query and analysis.

[0097] The heterogeneous device adaptive access gateway provided in the embodiment of the present application can have comprehensive data processing and analysis capabilities through the processing and analysis module, and can meet the diverse data processing and analysis needs in the Internet of Things scenario.

[0098] In some embodiments, the heterogeneous device adaptive access gateway further includes a management configuration module, which is configured to: Provides a device registration interface, performs device security authentication, obtains device information, collects and displays device operation data, and performs group management and permission settings for multiple devices.

[0099] Specifically, the management and configuration module can provide centralized management functions for heterogeneous devices, including device registration, authentication, status monitoring, and remote configuration. By working in conjunction with the protocol identification module 110, it can automatically discover newly connected devices, identify their protocol types, and assign adaptation policies to them. It also supports operations such as group management and permission setting for devices, facilitating unified management and maintenance of large numbers of devices.

[0100] The management configuration module provides a device registration interface and supports multiple device registration methods, such as manual registration and automatic discovery registration. During the device registration process, basic device information is collected, including device identification, device type, device model, and device location.

[0101] Multiple authentication methods can be used, such as username / password authentication and digital certificate authentication, to authenticate the device of the connected user. This ensures that only authenticated devices can access the gateway, ensuring system security.

[0102] The management configuration module can configure the device status monitoring function by collecting real-time device operation data, which can include device operating time, device load, and device fault status. By working in conjunction with the protocol identification module 110, it can automatically identify the device operation data and convert it into the target format for storage and display.

[0103] Device operation data can be visualized using charts and dashboards to intuitively display the device's operating status. For example, a bar chart can be used to display the device's operating time, and a line chart can be used to display the device's load changes.

[0104] The management and configuration module provides remote device configuration capabilities, supporting remote modification and adjustment of device parameters. For example, users can directly modify the device's operating parameters through the system interface, including motor speed and sensor sampling frequency.

[0105] During remote configuration, the management configuration module uses a secure communication mechanism to ensure the secure transmission of configuration information. At the same time, configuration operations are recorded and audited to facilitate subsequent query and tracing.

[0106] The management configuration module can manage devices in groups. Devices can be divided into different groups based on factors such as device type, location, and function. For example, devices in the same workshop can be grouped together, and devices of the same type can be grouped together.

[0107] The management configuration module allows you to set permissions for device groups and assign different device operation permissions based on user roles and permissions. For example, ordinary operators can only perform simple start and stop operations on devices, while senior administrators can perform parameter configuration and maintenance operations on devices.

[0108] The heterogeneous device adaptive access gateway provided in the embodiment of the present application realizes efficient registration, security authentication and group control of heterogeneous devices through centralized management, dynamic permission allocation and visual monitoring of the management configuration module, thereby improving the convenience and system security of collaborative operation and maintenance of multiple devices.

[0109] In some embodiments, the heterogeneous device adaptive access gateway further includes a communication interface module, which is configured to: Configuring drivers for multiple types of communication interfaces to support communication connections for devices with multiple types of communication interfaces; In the event of a communication link failure, the failure is repaired based on the preset fault recovery mechanism.

[0110] Specifically, the driver is used to configure device drivers for multiple communication interfaces, enabling communication connections between heterogeneous devices and the gateway. The fault recovery mechanism automatically executes preset strategies (such as reconnection and switching to a backup link) to repair the connection and ensure communication continuity when a communication link fails.

[0111] The communication interface module can be configured with drivers for multiple types of communication interfaces to support multiple communication interfaces, such as drivers for Ethernet, serial port, and field bus communication interfaces, etc., to ensure stable and reliable communication connections with various types of devices.

[0112] For example, for the Ethernet communication interface, an Ethernet communication driver is developed to support multiple Ethernet communication protocols, such as TCP / IP, User Datagram Protocol / Internet Protocol (UDP / IP), etc. This ensures that the gateway can establish stable and reliable communication connections with Ethernet-based devices.

[0113] For Ethernet communication, multi-threaded or asynchronous communication mechanisms are used to improve communication efficiency. At the same time, the automatic reconnection function of Ethernet communication is supported, which can automatically re-establish the connection when a communication link fails.

[0114] For example, for the serial communication interface, a serial communication driver is developed to ensure that the gateway can communicate stably with the serial communication device and provide serial communication interface support, including RS232, RS485 and other serial communication standards.

[0115] During serial communication, data verification mechanisms such as parity check and cyclic redundancy check (CRC) are used to ensure data transmission accuracy. Furthermore, the baud rate adaptive function of serial communication is supported, which can automatically adjust the baud rate according to the device requirements.

[0116] For example, for the fieldbus communication interface, the corresponding fieldbus communication driver is developed to ensure stable communication with the fieldbus equipment and support multiple fieldbus communication protocols such as Profibus, Modbus, etc.

[0117] For fieldbus communication, a bus arbitration mechanism is used to ensure that there is no conflict when multiple devices communicate on the same bus. At the same time, the fault diagnosis function of fieldbus communication is supported, which can detect and handle communication link failures in a timely manner.

[0118] The communication interface module also has the functions of automatic detection and fault recovery of communication links to ensure the continuity and stability of data transmission.

[0119] A communication link detection mechanism can be set up to regularly check the status of the communication link. For example, by sending a heartbeat signal or detecting the transmission of data packets, it can be determined whether the communication link is normal.

[0120] Different detection methods are used for different communication interfaces. For example, for Ethernet communication, the link status can be detected by detecting the status of the TCP connection or sending UDP heartbeat packets; for serial communication, the link status can be detected by detecting the communication status of the serial port or sending a check signal.

[0121] When a communication link failure is detected, the fault recovery mechanism is automatically activated. Appropriate recovery measures are taken based on the fault type and the characteristics of the communication interface. For example, for an Ethernet communication failure, you can try to reestablish a TCP connection or reconfigure network parameters; for a serial communication failure, you can try to reopen the serial port or reset its parameters.

[0122] During the fault recovery process, fault information and recovery operations are recorded to facilitate subsequent analysis and improvement. At the same time, user-defined fault recovery mechanisms are supported, allowing customized fault recovery solutions based on user needs and device characteristics.

[0123] The heterogeneous device adaptive access gateway provided in the embodiment of the present application can support multiple communication interfaces by configuring drivers of multiple types of communication interfaces through the communication interface module, and can adapt to the access requirements of different types of devices.

[0124] The following describes a method for adaptive access of heterogeneous devices provided in an embodiment of the present application. The method for adaptive access of heterogeneous devices described below and the adaptive access gateway for heterogeneous devices described above can refer to each other.

[0125] Figure 2 A flow chart of the method for adaptive access of heterogeneous devices provided in an embodiment of the present application is shown as follows: Figure 2 As shown, the method includes step 210, step 220 and step 230. The steps of the method flow are only a possible implementation of the present application.

[0126] Step 210: Input the data message of the device connected to the gateway into the first large model to obtain the protocol type of the data message predicted by the first large model; the first large model is trained based on the sample data message and the sample protocol type corresponding to the sample data message; Step 220: Convert the data format of the data message into a target format based on the data message and protocol type; Step 230: Process and analyze the data message in the target format based on the data type of the data message.

[0127] Specifically, the heterogeneous device adaptive access method provided in the embodiment of the present application is applicable to the terminal of the heterogeneous device adaptive access gateway. The terminal can be various electronic devices with a display screen and supporting web browsing, including but not limited to servers, smart phones, tablet computers, laptops and desktop computers, etc.

[0128] The execution entity of the heterogeneous device adaptive access method provided in the embodiment of the present application is a heterogeneous device adaptive access gateway.

[0129] The adaptive access method for heterogeneous devices provided in the embodiment of the present application can automatically identify the communication protocols of heterogeneous devices based on large model technology, without the need for human intervention, improving device access efficiency and reducing maintenance costs; by converting the data format of data messages of different devices into the target format, the processing difficulties caused by data format differences are solved, and the devices can be automatically connected to the gateway, thereby improving the efficiency and accuracy of data processing and analysis.

[0130] In some embodiments, step 210 includes: Collect data packets from multiple connected devices based on different types of data acquisition interfaces and different types of acquisition drivers; Inputting data packets into the first model; The data cleaning layer of the first large model is controlled to process data messages based on preset data cleaning rules; Control the first data structure feature of the data message after the feature extraction layer of the first large model extracts and processes: The protocol identification layer of the first large model is controlled to identify the protocol type of the data message based on the first data structure feature.

[0131] In some embodiments, step 220 includes: Inputting the data message and the protocol type into the second largest model to obtain the semantic features of the data message output by the second largest model; Generate data format conversion strategies corresponding to data messages based on semantic features; The data format of the data message is converted into the target format based on the data format conversion strategy.

[0132] In some embodiments, the protocol type includes an unknown protocol type and a protocol type stored in a database, and the method for adaptive access of heterogeneous devices further includes: In the case where the protocol type is an unknown protocol type, the data message and the protocol type are input into the second largest model to obtain the semantic features and the second data structure features of the data message output by the second largest model; Generate a data format conversion strategy corresponding to the data message based on the semantic feature and the second data structure feature; Among them, the second data structure feature is predicted by the second largest model based on a confidence assessment mechanism; the confidence assessment mechanism includes obtaining the confidence of the initial second data structure feature predicted by the second largest model, and when the confidence is greater than a preset threshold, using the initial second data structure feature as the second data structure feature.

[0133] In some embodiments, step 230 includes: When the data message is a sensor data message, a statistical analysis is performed on the data message in the target format based on a sliding window algorithm to obtain statistical data of the data message in the target format. The operating status of the device is analyzed based on the statistical data, and when the operating status is abnormal, an early warning information of the device is generated; In the case that the data message is a control instruction data message, the control instruction corresponding to the control instruction data message is parsed and the control instruction is executed.

[0134] It should be noted here that the heterogeneous device adaptive access method provided in the embodiment of the present application can be executed by the above-mentioned heterogeneous device adaptive access gateway embodiment and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those of the gateway embodiment will not be described in detail here.

[0135] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 3 As shown, the electronic device may include: a processor (Processor) 310, a communication interface (Communication Interface) 320, a memory (Memory) 330, and a communication bus 340. The processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call a computer program in the memory 330 to execute the above method.

[0136] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software function module and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0137] On the other hand, an embodiment of the present application also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the methods provided in the above embodiments.

[0138] On the other hand, an embodiment of the present application further provides a processor-readable storage medium, which stores a computer program, and the computer program is used to enable the processor to execute the methods provided in the above embodiments.

[0139] The processor-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROMs, EPROMs, EEPROMs, non-volatile memories (NANDFLASH), solid-state drives (SSDs)), etc.

[0140] The device embodiments described above are merely illustrative, wherein the modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Those skilled in the art can understand and implement the present invention without inventive effort.

[0141] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A heterogeneous device adaptive access gateway, characterized in that: include: a protocol identification module, configured to input a data message of a connected device into a first large model, and output a protocol type of the data message predicted by the first large model; The first large model is obtained by training based on the sample data packets and the sample protocol types corresponding to the sample data packets; a format conversion module, configured to convert the data format of the data message into a target format based on the data message and the protocol type; A processing and analysis module is used to process and analyze the data message in the target format based on the data type of the data message.

2. The heterogeneous device adaptive access gateway according to claim 1, characterized in that: The protocol identification module is specifically used for: Collect data packets from multiple connected devices based on different types of data acquisition interfaces and different types of acquisition drivers; Inputting the data message into the first large model; Controlling the data cleaning layer of the first large model to process the data message based on a preset data cleaning rule; Controlling the first data structure feature of the data message after the feature extraction layer of the first large model extracts and processes: The protocol identification layer of the first large model is controlled to identify the protocol type of the data message based on the first data structure feature.

3. The heterogeneous device adaptive access gateway according to claim 1, characterized in that: The format conversion module is specifically used to: Inputting the data message and the protocol type into a second large model to obtain semantic features of the data message output by the second large model; generating a data format conversion strategy corresponding to the data message based on the semantic feature; The data format of the data message is converted into the target format based on the data format conversion strategy.

4. The heterogeneous device adaptive access gateway according to claim 3, characterized in that: The protocol type includes an unknown protocol type and a protocol type stored in a database; the format conversion module is further specifically used to: In a case where the protocol type is the unknown protocol type, inputting the data message and the protocol type into a second large model to obtain the semantic features and second data structure features of the data message output by the second large model; generating a data format conversion strategy corresponding to the data message based on the semantic feature and the second data structure feature; In which, the second data structure feature is predicted by the second large model based on a confidence assessment mechanism; the confidence assessment mechanism includes obtaining the confidence of the initial second data structure feature predicted by the second large model, and when the confidence is greater than a preset threshold, using the initial second data structure feature as the second data structure feature.

5. The heterogeneous device adaptive access gateway according to claim 3, characterized in that: The format conversion module is further specifically used for: Verifying the completeness and data type of the data message converted into the target format; If the verification is successful, the data message in the target format is sent to the processing and analysis module.

6. The heterogeneous device adaptive access gateway according to claim 1, characterized in that: The processing and analysis module is specifically used to: When the data message is a sensor data message, performing statistical analysis on the data message in the target format based on a sliding window algorithm to obtain statistical data of the data message in the target format, analyzing the operating status of the device based on the statistical data, and generating early warning information for the device when the operating status is abnormal; In the case that the data message is a control instruction data message, the control instruction corresponding to the control instruction data message is parsed and the control instruction is executed.

7. The heterogeneous device adaptive access gateway according to claim 1, characterized in that: It also includes a management configuration module, which is used to: Provide a registration interface for the device, perform security authentication for the device, obtain device information for the device, collect and display device operation data for the device, and perform group management and permission settings for multiple devices.

8. The heterogeneous device adaptive access gateway according to claim 1, characterized in that: It also includes a communication interface module, wherein the communication interface module is used to: Configuring drivers for multiple types of communication interfaces to support communication connections for devices with multiple types of communication interfaces; In the event of a communication link failure, the failure is repaired based on a preset failure recovery mechanism.

9. A method for adaptive access of heterogeneous devices, characterized in that: include: Inputting a data message of a device connected to the gateway into a first large model to obtain a protocol type of the data message predicted by the first large model; The first large model is obtained by training based on the sample data packets and the sample protocol types corresponding to the sample data packets; converting a data format of the data message into a target format based on the data message and the protocol type; The data message in the target format is processed and analyzed based on the data type of the data message.

10. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to execute the heterogeneous device adaptive access method according to claim 9 through the computer program.

Images