Method and system for adapting industrial intelligent gateway to multiple industrial protocols

Abstract

The invention discloses a method and system for adapting an industrial intelligent gateway to multiple industrial protocols, and relates to the technical field of industrial Internet of Things edge computing, and the method comprises the steps: monitoring industrial network traffic self-learning, and constructing a protocol fingerprint database; dynamically constructing a virtual equipment twinborn body based on the fingerprint database to form a simulation environment; automatically verifying protocol configuration in a simulation environment; the application verification configuration is connected with the entity device to collect data; performing in-stream calculation processing on the data stream in real time to generate an advanced event and feature data; according to the method, automatic identification and compatibility of unknown protocols are realized, production accidents are avoided through virtual simulation verification, data value extraction is synchronously completed in the data acquisition process, and the security, reliability and data processing efficiency of protocol adaptation are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of industrial Internet of Things (IIoT) edge computing technology, and in particular to a method and system for adapting an industrial smart gateway to multiple industrial protocols. Background Technology

[0002] With the rapid development of the Industrial Internet of Things (IIoT), industrial smart gateways are playing an increasingly crucial role as a bridge connecting industrial equipment and the digital world. They not only enable interconnection between devices but also provide robust support for data acquisition, protocol conversion, and edge computing, making them one of the core devices driving the intelligent transformation of industry.

[0003] Existing industrial smart gateway technologies still suffer from numerous shortcomings in protocol adaptation. Due to the inconsistent protocol standards of devices from different manufacturers, achieving efficient compatibility between devices remains a technical challenge. This results in mainstream industrial smart gateways on the market supporting a very limited range of industrial protocols and lacking support for secondary development and extension of these protocols. Furthermore, existing technologies often require custom interface development when dealing with multiple protocols and devices, leading to low development efficiency and difficulty in handling situations with numerous services. This can easily result in complex interaction relationships, making it difficult to understand inter-system relationships and troubleshoot errors. These issues limit the application and promotion of industrial smart gateways in complex industrial environments. Summary of the Invention

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

[0005] Therefore, this invention provides a method for adapting an industrial smart gateway to multiple industrial protocols, which solves the problems of insufficient protocol compatibility, low development efficiency, and how to achieve efficient adaptation of multiple protocols in existing industrial smart gateway methods.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for adapting an industrial smart gateway to multiple industrial protocols, characterized by comprising the following steps:

[0008] Monitor industrial network traffic to learn and build an initial protocol fingerprint database;

[0009] Virtual device twins are dynamically constructed based on the initial protocol fingerprint database to form a simulation environment;

[0010] The protocol configuration to be deployed is placed in a simulation environment for automated verification to output a verified configuration;

[0011] The application has been verified to establish a connection with physical industrial equipment and collect raw data streams;

[0012] Perform real-time in-stream computation on the raw data stream to generate advanced event and feature data;

[0013] High-level event and feature data are uniformly encapsulated and distributed to upper-layer applications.

[0014] As a preferred embodiment of the industrial intelligent gateway and multi-industry protocol adaptation method described in this invention, the specific steps of monitoring industrial network traffic to learn and build an initial protocol fingerprint database are as follows:

[0015] The Gateway Data Plane Development Kit (DPDK) bypasses and captures all raw network packets flowing through the network interface;

[0016] The header and payload of the original network packet are parsed to extract static features including packet length, transmission period, and target port;

[0017] The interaction sequence patterns and timing intervals of the original network packets are analyzed to construct a communication timing feature model;

[0018] The static features are aggregated and clustered with the communication timing feature model to form an initial protocol fingerprint library representing different protocol types.

[0019] As a preferred embodiment of the industrial intelligent gateway and its adaptation method to various industrial protocols described in this invention, the specific steps of dynamically constructing a virtual device twin based on an initial protocol fingerprint database to form a simulation environment are as follows:

[0020] The initial protocol fingerprint database is analyzed to identify the protocol types and versions of various devices in the network;

[0021] Instantiate a corresponding virtual device twin for each type of identified physical device;

[0022] Configure each virtual device twin with the protocol command set, response data format and response latency characteristics to be simulated based on the protocol fingerprint;

[0023] All configured virtual device twins are connected to a virtual network to form a high-fidelity simulation environment isolated from the real network topology.

[0024] As a preferred embodiment of the industrial intelligent gateway and multi-industry protocol adaptation method described in this invention, the specific steps of placing the protocol configuration to be deployed in a simulation environment for automated verification to output the verified configuration are as follows:

[0025] Deploy manually written or imported protocol parsing scripts and configuration parameters onto the corresponding virtual device twin in the simulation environment;

[0026] The virtual device twin is driven to respond to instructions issued by the protocol parsing script based on its configured protocol behavior;

[0027] Monitor and record all interactive messages, response times, and data parsing results during script execution;

[0028] Analyze the monitoring records and select configurations that can correctly complete data reading and writing without communication errors as verified configurations.

[0029] As a preferred embodiment of the industrial intelligent gateway and multi-industry protocol adaptation method described in this invention, the application has been verified to establish a connection with physical industrial equipment and collect raw data streams. The specific steps are as follows:

[0030] The scheduler loads the verified configuration and sends it to the gateway's protocol processing engine;

[0031] The protocol processing engine establishes a secure communication link with the target entity's industrial equipment based on the configuration and initiates a handshake negotiation.

[0032] After successfully establishing a connection, poll to read device data or subscribe to device events according to the configured strategy, and receive the raw byte stream returned by the device;

[0033] The received raw byte stream undergoes preliminary frame verification and sequence reconstruction, outputting a complete raw data stream.

[0034] As a preferred embodiment of the industrial intelligent gateway and multi-industry protocol adaptation method described in this invention, the specific steps of performing real-time in-stream computation processing on the original data stream to generate advanced event and feature data are as follows:

[0035] Receive the raw data stream and execute multiple lightweight data processing tasks in parallel;

[0036] Perform data cleaning, unit conversion, and linearization on the raw values ​​in the data stream;

[0037] Based on predefined rules, the cleaned data is subjected to status judgment, threshold comparison and logical operation to generate event tags;

[0038] The values ​​of multiple related data points are aggregated and calculated in real time to generate new feature data points.

[0039] Secondly, this invention provides an industrial intelligent gateway and a system for adapting to multiple industrial protocols, including,

[0040] The traffic detection module listens to industrial network traffic to learn and build an initial protocol fingerprint database.

[0041] The twin simulation module dynamically constructs a virtual device twin based on the initial protocol fingerprint library to form a simulation environment;

[0042] The automated verification module places the protocol configuration to be deployed in a simulation environment for automated verification and outputs the verified configuration.

[0043] The secure connection module is configured to establish a connection with physical industrial equipment and collect raw data streams.

[0044] The stream computing module performs real-time in-stream computation on the raw data stream to generate advanced event and feature data;

[0045] The encapsulation and distribution module encapsulates advanced event and feature data and distributes it to upper-layer applications.

[0046] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the industrial intelligent gateway and multiple industrial protocol adaptation method as described in the first aspect of the present invention.

[0047] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the industrial intelligent gateway and multiple industrial protocol adaptation method as described in the first aspect of the present invention.

[0048] The beneficial effects of this invention are as follows: By monitoring network traffic and learning to construct a protocol fingerprint library, automatic identification and compatibility with unknown protocols are achieved, overcoming the limitations of traditional gateways that rely on predefined protocol libraries; by constructing a virtual device twin and automatically verifying protocol configurations in a simulation environment, production accidents that may be caused by directly operating physical devices are avoided, significantly improving the safety and reliability of system debugging; by embedding real-time in-stream computing capabilities at the protocol layer, cleaning, transformation, and feature extraction are completed synchronously during data acquisition, transforming raw data into high-value information, effectively reducing uplink bandwidth pressure and cloud computing load, and realizing a leap from data pass-through to edge intelligent processing. Attached Figure Description

[0049] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart of the method for adapting an industrial smart gateway to multiple industrial protocols in Example 1. Detailed Implementation

[0051] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0052] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0053] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0054] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for adapting an industrial smart gateway to multiple industrial protocols, characterized by including the following steps:

[0055] Monitor industrial network traffic to learn and build an initial protocol fingerprint database;

[0056] Virtual device twins are dynamically constructed based on the initial protocol fingerprint database to form a simulation environment;

[0057] The protocol configuration to be deployed is placed in a simulation environment for automated verification to output a verified configuration;

[0058] The application has been verified to establish a connection with physical industrial equipment and collect raw data streams;

[0059] Perform real-time in-stream computation on the raw data stream to generate advanced event and feature data;

[0060] High-level event and feature data are uniformly encapsulated and distributed to upper-layer applications.

[0061] It should be noted that,

[0062] Monitor industrial network traffic to learn and build an initial protocol fingerprint database

[0063] After the industrial intelligent gateway starts up, it first calls its underlying integrated Data Plane Development Kit (DPDK) to capture all raw network data packets flowing through its physical ports in bypass monitoring mode. Subsequently, the system parses the captured packets layer by layer, performing structured processing from Ethernet frame headers and IP packet headers to transport layer payloads, extracting static feature parameters including packet length distribution, transmission cycle patterns, destination port numbers, and fixed identifiers at specific byte offsets. Simultaneously, the system uses time series analysis algorithms to model the interaction sequence of packets and the time intervals between requests and responses, generating dynamic temporal features of communication behavior. Finally, through unsupervised clustering algorithms in machine learning, the static and dynamic temporal features are fused and analyzed to automatically classify and summarize the unique fingerprint features of different protocol types, forming a systematic initial protocol fingerprint database.

[0064] Through deep message parsing and time-series behavior modeling, the system achieves automatic identification and feature extraction of unknown and non-standard protocols, overcoming the limitations of traditional gateways that rely on manually predefined protocol rules, and providing adaptive protocol discovery capabilities for heterogeneous device access.

[0065] Virtual device twins are dynamically constructed based on the initial protocol fingerprint database to form a simulation environment.

[0066] The system parses the pre-built initial protocol fingerprint database to identify various device protocols and their version information in the current network. For each identified physical device, the gateway automatically instantiates a corresponding virtual device twin in an isolated virtualization environment. Each twin accurately simulates the protocol behavior of the real device based on the characteristics recorded in the protocol fingerprint database, including but not limited to supported command sets, data response formats, exception response modes, and communication latency characteristics consistent with the actual device. All virtual device twins are connected to a fully simulated virtual network topology, which is isolated from the real physical network, forming a high-fidelity protocol simulation test platform.

[0067] By constructing a highly realistic virtual device twin and simulation environment, a safe and isolated testing condition is provided for protocol configuration, avoiding system downtime and production risks that may be caused by directly operating physical devices, and significantly improving the safety and reliability of the debugging process.

[0068] The protocol configuration to be deployed is placed in a simulation environment for automated verification to output a verified configuration.

[0069] Engineers use the management interface to write or import protocol parsing scripts and communication parameter configurations for new devices. The system then deploys this configuration to the corresponding virtual device twin in the simulation environment. Subsequently, the driver twin responds to various commands issued by the parsing script based on its predefined protocol behavior model. The system comprehensively monitors and records all communication messages, response times, data parsing correctness, and error occurrences throughout the entire interaction process. Finally, through automated analysis of the monitoring logs, configuration items that can stably and correctly complete data reading and writing without any communication errors are selected and marked as verified configurations.

[0070] By automating the testing and verification of protocol configuration in a simulation environment, the accuracy and robustness of the configuration logic are ensured, fundamentally eliminating field equipment failures caused by configuration errors, and greatly improving the efficiency and first-time success rate of system deployment.

[0071] The application has been verified to establish a connection with physical industrial equipment and acquire raw data streams.

[0072] The system scheduler loads the verified configuration items and sends them to the gateway's protocol processing engine. Based on the configuration parameters, the protocol processing engine establishes a communication link with the target industrial device using standard industrial security mechanisms, completing the protocol handshake and session initialization. After a successful connection, the engine periodically polls to read device data points or subscribes to device events according to the configured strategy, receiving the raw byte stream data returned by the device. Subsequently, it performs basic frame integrity verification, message sequence reassembly, and error retransmission processing on the received byte stream, outputting a complete raw data stream for subsequent processing.

[0073] By establishing a reliable connection with physical devices through pre-verified secure configurations, the stability and efficiency of the data acquisition process are ensured, while reducing the risk of communication interruption due to configuration errors, thus providing continuous and complete data input for upper-layer applications.

[0074] Real-time in-stream computation is performed on the raw data stream to generate advanced event and feature data.

[0075] After receiving the raw data stream, the system utilizes its built-in stream processing engine to execute multiple lightweight data processing tasks in parallel. First, the data values ​​are cleaned to remove outliers and invalid data, and necessary unit conversions and sensor nonlinearity corrections are performed. Then, based on a predefined business rule engine, the cleaned data undergoes real-time status assessments, threshold comparisons, and multi-condition logical operations to generate corresponding event markers. Simultaneously, the system performs in-window aggregation calculations on the values ​​of multiple related data points, such as real-time averages and moving variances, generating new feature data points and extracting higher-dimensional information.

[0076] By embedding real-time stream computing capabilities into the data acquisition chain, the value extraction and knowledge extraction of data are completed at the source, significantly reducing redundant data transmission, reducing uplink bandwidth pressure and cloud processing load, and realizing the transformation from raw data to usable knowledge.

[0077] High-level event and feature data are uniformly encapsulated and distributed to upper-layer applications.

[0078] The system standardizes and encapsulates the advanced event and feature data generated by the stream processing engine according to a predefined unified data format (such as JSON). Each data entry is given a precise timestamp, data source device identifier, and data quality identifier. After encapsulation, the data is pushed to the message middleware via a lightweight messaging protocol (such as MQTT) based on pre-subscribed topic information. Finally, the message middleware reliably distributes the data to various upper-layer application systems, such as cloud-based IoT platforms or local monitoring systems.

[0079] By unifying data formats and standardizing distribution mechanisms, data supply is decoupled from the upper layer, improving system compatibility and scalability, while ensuring the reliability and consistency of end-to-end data transmission.

[0080] Specifically, the steps for monitoring industrial network traffic to learn and build an initial protocol fingerprint database are as follows:

[0081] The Gateway Data Plane Development Kit (DPDK) bypasses and captures all raw network packets flowing through the network interface;

[0082] The header and payload of the original network packet are parsed to extract static features including packet length, transmission period, and target port;

[0083] The interaction sequence patterns and timing intervals of the original network packets are analyzed to construct a communication timing feature model;

[0084] The static features are aggregated and clustered with the communication timing feature model to form an initial protocol fingerprint library representing different protocol types.

[0085] It should be noted that the Gateway Data Plane Development Kit (DPDK) bypasses and captures all raw network packets flowing through the network interface.

[0086] The industrial intelligent gateway initiates the Data Plane Development Kit (DPDK) driver, directly taking over network traffic from the physical network interface via bypass technology. The system configures the network interface card (NIC) to operate in promiscuous mode, ensuring the capture of all raw data frames flowing through the port. DPDK bypasses the operating system kernel protocol stack, efficiently reading raw packet data from the NIC's DMA ring buffer using a polling mode. Captured packets undergo initial filtering to remove physical layer redundancy before being transmitted to the user-space packet processing pipeline in a zero-copy manner. The entire process employs a multi-threaded parallel architecture, with each thread bound to an independent CPU core to process packets from a specific queue, ensuring high-throughput packet capture performance. The system records the timestamp and input port information of each packet in real time, providing a complete raw data source for subsequent analysis.

[0087] High-performance packet capture is achieved through DPDK bypass technology, ensuring the acquisition of complete communication data without packet loss in high-speed industrial network environments. This provides a real and reliable raw data foundation for protocol analysis and overcomes the shortcomings of insufficient processing capabilities of traditional operating system protocol stacks.

[0088] The header and payload of the original network packet are parsed to extract static features including packet length, transmission period, and destination port.

[0089] The system performs multi-layer protocol parsing on the captured raw packets. First, it identifies the Ethernet frame type and extracts the MAC address information. Then, it parses the IP layer protocol header to obtain the source and destination IP addresses and protocol type. For transport layer packets, it performs detailed parsing of TCP / UDP header fields, especially the destination port number and service type identifier. It also deeply analyzes the application layer payload content, identifying the identifier fields and function codes of specific protocols. The system statistically analyzes the packet length distribution characteristics of each communication flow, calculates the time interval between adjacent packets, and analyzes their periodic patterns. Simultaneously, it extracts feature parameters such as fixed identifiers, function code fields, and address mapping patterns from the packets. All extracted feature parameters are standardized and stored in a structured feature vector format.

[0090] By obtaining multi-level static feature parameters through deep message parsing, sufficient feature dimensions are provided for protocol identification, enabling the system to accurately distinguish the unique features of different protocols and laying a data foundation for the subsequent construction of a protocol fingerprint database.

[0091] Analyze the interaction sequence patterns and timing intervals of the original network packets to construct a communication timing feature model.

[0092] The system performs pattern analysis on the message interaction sequence during communication. First, it reconstructs the complete request-response dialogue process and identifies typical transaction processing patterns. Using time series analysis algorithms, it calculates the response time characteristics between each communication node, including minimum, maximum, and average response delays. A hidden Markov model is used to establish a state transition probability matrix, describing the transition characteristics of the protocol state machine. The distribution pattern of message interval times is analyzed to establish a temporal behavior model, identifying periodic and event-driven communication patterns. Simultaneously, dynamic characteristics such as error retransmission mechanisms, flow control behavior, and connection maintenance strategies are modeled to form a complete description of protocol behavior characteristics.

[0093] By capturing the dynamic interaction features of protocols through temporal behavior modeling, it is possible to effectively identify protocol variants with similar static features but different behavioral patterns, thereby enhancing the ability to distinguish complex industrial protocols and improving the accuracy and reliability of protocol identification.

[0094] The static features are aggregated and clustered with the communication time-series feature model to form an initial protocol fingerprint database representing different protocol types.

[0095] The system employs feature fusion technology to normalize and aggregate static and temporal feature vectors across multiple dimensions, forming a complete protocol feature descriptor. Principal component analysis (PCA) is used to reduce the dimensionality of high-dimensional features, retaining the most discriminative feature components. Unsupervised clustering is used to automatically classify a large number of protocol samples, and density clustering is employed to identify naturally formed protocol categories. Representative features are extracted from each cluster center to generate standardized protocol fingerprint templates. A protocol fingerprint database is established to store feature templates, protocol attributes, and identification rules for each protocol type, while also recording the statistical distribution information and confidence indices of the protocol features.

[0096] By establishing a systematic protocol fingerprint database through multi-feature fusion and intelligent clustering, the system can accurately identify known protocols and reasonably classify unknown protocols, providing industrial intelligent gateways with adaptive protocol identification capabilities and significantly improving protocol compatibility and system intelligence.

[0097] Specifically, the steps for dynamically constructing a virtual device twin based on the initial protocol fingerprint database to form a simulation environment are as follows:

[0098] The initial protocol fingerprint database is analyzed to identify the protocol types and versions of various devices in the network;

[0099] Instantiate a corresponding virtual device twin for each type of identified physical device;

[0100] Configure each virtual device twin with the protocol command set, response data format and response latency characteristics to be simulated based on the protocol fingerprint;

[0101] All configured virtual device twins are connected to a virtual network to form a high-fidelity simulation environment isolated from the real network topology.

[0102] It should be noted that the initial protocol fingerprint database is parsed to identify the protocol types and versions of various devices existing in the network.

[0103] The system loads an initial protocol fingerprint database and employs a combination of rule-based and machine learning approaches for protocol identification. First, the feature templates in the fingerprint database are indexed and optimized to establish a fast retrieval mechanism. A feature matching algorithm calculates the similarity between real-time network traffic features and the templates in the fingerprint database, using a multi-dimensional weighted scoring mechanism to determine the best-matching protocol. For each identified protocol, the system further parses its version feature field, accurately identifying the specific version number by comparing version-specific identifiers and functional support differences. Simultaneously, a device protocol list is established, recording the network address, protocol type, version number, and feature confidence score for each device. The system employs a continuous learning mechanism to provide feedback and optimization to the identification results, continuously improving the accuracy and coverage of protocol identification.

[0104] By using a deep analysis of the protocol fingerprint database to automatically identify device protocols, it can accurately distinguish the protocol types and version differences of devices from different manufacturers, providing precise protocol feature descriptions for subsequent virtualization simulation, and solving the pain point of traditional gateways requiring manual configuration of protocol information.

[0105] Instantiate a corresponding virtual device twin for each type of identified physical device.

[0106] Based on the identified device protocol information, the system invokes the virtualization engine to create a corresponding virtual twin instance for each physical device. First, it allocates independent virtualization resource space, including memory isolation areas and processor time slices. Then, it loads the device protocol stack template and initializes the corresponding protocol processing state machine according to the device type. Each virtual twin is equipped with an independent network stack and protocol processor, capable of simulating the complete communication behavior of a real device. The system configures virtual device identifiers for the twins to ensure a correspondence with the physical devices. Simultaneously, a twin management table is established to record the resource allocation status, running status, and performance indicators of each virtual instance, enabling full lifecycle monitoring and management.

[0107] By creating virtual twin instances of physical devices, a digital mapping of device behavior is achieved, providing a secure and isolated experimental environment for protocol testing and avoiding the production risks that may arise from directly operating physical devices.

[0108] Configure each virtual device twin with the protocol command set, response data format, and response latency characteristics to be simulated based on the protocol fingerprint.

[0109] The system precisely configures the protocol behavior characteristics of each virtual device twin based on detailed feature parameters extracted from the protocol fingerprint database. First, it loads the protocol instruction set mapping table and configures the command codes and function sets supported by the twin. Then, it sets the response data generation rules, including data format templates, numerical ranges, and encoding methods. Based on the timing characteristics in the protocol fingerprint, it configures request-response latency parameters, including minimum response time, maximum response time, and typical latency distribution patterns. Simultaneously, it simulates response behaviors under various abnormal conditions, such as error code returns, timeout handling, and retransmission mechanisms. Finally, it verifies the completeness and consistency of the configuration to ensure that the virtual device twin can accurately simulate the protocol behavior characteristics of a real device.

[0110] By precisely configuring the protocol behavior characteristics of virtual devices, high-fidelity device behavior simulation is achieved, providing a realistic and reliable experimental environment for protocol testing and ensuring that the test results can truly reflect the actual operating conditions.

[0111] All configured virtual device twins are connected to a virtual network to form a high-fidelity simulation environment isolated from the real network topology.

[0112] The system creates an independent virtual network environment and uses software-defined networking (SDN) technology to construct the virtual network topology. First, virtual switches and highly realistic routers are configured to establish a complete network infrastructure. Then, all configured virtual device twins are connected to the virtual network, and the connections are configured according to the actual network topology. Network isolation policies are set to ensure complete isolation between the virtual environment and the real production network. A network traffic monitoring and logging system is configured to capture and analyze communication data in the virtual environment in real time. Finally, the integrity and consistency of the simulation environment are verified to ensure that the communication behavior between virtual devices is consistent with the actual network environment, providing a reliable experimental platform for protocol testing.

[0113] By constructing a fully isolated, high-fidelity virtual simulation environment, a safe and reliable experimental platform is provided for protocol development and testing. This avoids impacting production systems while realistically simulating actual operating environments, significantly improving the efficiency and quality of protocol development.

[0114] Specifically, the step of automatically verifying the protocol configuration to be deployed in a simulation environment to output a verified configuration involves the following steps:

[0115] Deploy manually written or imported protocol parsing scripts and configuration parameters onto the corresponding virtual device twin in the simulation environment;

[0116] The virtual device twin is driven to respond to instructions issued by the protocol parsing script based on its configured protocol behavior;

[0117] Monitor and record all interactive messages, response times, and data parsing results during script execution;

[0118] Analyze the monitoring records and select configurations that can correctly complete data reading and writing without communication errors as verified configurations.

[0119] It should be noted that the manually written or imported protocol parsing scripts and configuration parameters are deployed to the corresponding virtual device twin in the simulation environment.

[0120] The system receives protocol parsing scripts and configuration files uploaded by users through the configuration management interface. First, it performs syntax checks and security verification on the scripts to ensure the absence of malicious code or syntax errors. After successful verification, the configuration distribution engine transmits the scripts and parameter files to the target virtual device twin in the simulation environment. The transmission process employs a secure encryption protocol to guarantee the integrity and confidentiality of the configuration information. On each virtual device twin, the configuration loader receives and parses the configuration file, initializing the corresponding protocol stack parameters according to the protocol type, including register mapping tables, function code definitions, and data packet formats. The system establishes an independent configuration space for each twin, ensuring that configurations across different devices are isolated. Finally, the configuration verification module performs an integrity check on the loaded configuration to ensure that all necessary parameters are correctly set and usable.

[0121] Through automated configuration distribution and verification mechanisms, the protocol parsing scripts are accurately deployed to the target virtual device, providing a correct and reliable initial environment for subsequent protocol testing and avoiding errors and inconsistencies that may arise from manual configuration.

[0122] The virtual device twin responds to instructions issued by the protocol parsing script based on its configured protocol behavior.

[0123] The simulation engine initiates a protocol test sequence, sending protocol commands to the virtual device twin according to predefined test cases. The virtual device twin parses the received command messages, identifying the command type and parameter requirements based on its configured protocol behavior characteristics. The protocol processing module generates corresponding response data based on the command content, including data returns under normal conditions and error responses under various abnormal conditions. The response generation process strictly adheres to the configured timing characteristics, simulating the processing latency and response time of a real device. For protocols requiring state maintenance, the virtual device twin maintains the corresponding session state machine to ensure state consistency across multiple interactions. Simultaneously, the system simulates various network anomalies, such as packet loss, duplication, and out-of-order delivery, testing the robustness of the protocol parsing script.

[0124] By driving a virtual device twin to simulate the protocol response behavior of a real device, a comprehensive and realistic testing environment is provided for the protocol parsing script. This environment can verify the script's processing capabilities under various normal and abnormal conditions, ensuring its reliability in real-world environments.

[0125] Monitor and record all interactive messages, response times, and data parsing results during script execution.

[0126] The system employs a distributed monitoring agent, deploying data acquisition points at key nodes throughout the simulation environment to comprehensively capture all interactive data during protocol testing. The network monitoring component captures and records all packet data flowing through the virtual network in real time, including metadata information such as source and destination addresses, port numbers, and protocol types. The performance monitoring module accurately measures the round-trip time, processing latency, and throughput metrics for each request and response. The data parsing monitor records the parsing results of each response from the protocol parsing script, including parsed data values, status information, and error codes. All monitoring data is stored in a time-series database with high-precision timestamps, and an index is created for rapid retrieval and analysis. The system also records resource usage, including CPU, memory, and network bandwidth usage metrics.

[0127] Through comprehensive monitoring and data logging, complete and traceable information about the protocol testing process was obtained, providing rich data support for subsequent analysis and problem localization, and ensuring the transparency and analyzability of the testing process.

[0128] Analyze monitoring records and select configurations that can correctly complete data read and write operations without communication errors as verified configurations.

[0129] The data analysis engine loads test data from the monitoring log database and analyzes each test run using a multi-dimensional evaluation algorithm. First, it checks for communication-level errors, including timeouts, verification failures, and sequence errors. Then, it verifies the correctness of data parsing, comparing the expected results with the actual parsed results. The performance analysis module evaluates whether the response time meets requirements and whether the throughput reaches expected metrics. The system uses machine learning algorithms to identify abnormal patterns and potential problems during the testing process. Finally, the comprehensive evaluation system scores and categorizes each test configuration according to preset pass criteria, selecting configurations that fully meet the requirements. Verified configurations are digitally signed, stored in a verified configuration library, and detailed test reports and performance analysis documents are generated.

[0130] By using automated test result analysis and configuration screening, we ensure that only fully verified and correct configurations can be used in actual production environments, which greatly improves the reliability and security of system deployment and reduces the risks and costs of on-site debugging.

[0131] Specifically, the application has been verified to establish a connection with physical industrial equipment and collect raw data streams. The specific steps are as follows:

[0132] The scheduler loads the verified configuration and sends it to the gateway's protocol processing engine;

[0133] The protocol processing engine establishes a secure communication link with the target entity's industrial equipment based on the configuration and initiates a handshake negotiation.

[0134] After successfully establishing a connection, poll to read device data or subscribe to device events according to the configured strategy, and receive the raw byte stream returned by the device;

[0135] The received raw byte stream undergoes preliminary frame verification and sequence reconstruction, outputting a complete raw data stream.

[0136] It should be noted that the scheduler loads the verified configuration and sends it to the gateway's protocol processing engine.

[0137] The system configuration manager retrieves the target device's protocol configuration information from the verified configuration library, including complete configuration data such as protocol type, communication parameters, and data point mapping table. The scheduler first verifies the digital signature and integrity checksum of the configuration file to ensure the configuration has not been tampered with. Then, based on device priority and system load, it selects the optimal protocol processing engine instance. The configuration data is sent to the target protocol processing engine via a secure transmission channel, using encryption protection during transmission. Upon receiving the configuration, the protocol processing engine immediately initializes its runtime environment, including allocating a memory buffer, establishing a protocol state machine, setting timeout parameters, and implementing a retry mechanism. The engine loads the corresponding protocol stack components according to the configuration content and establishes binding relationships with physical ports. Finally, it executes a self-check program before the configuration takes effect to ensure all parameters are loaded correctly and ready for immediate use.

[0138] By employing automated configuration loading and verification mechanisms, the system ensures that tested and reliable configurations can be accurately deployed to the runtime environment. This achieves standardization and automation of configuration management, avoids errors that may be introduced by manual operations, and improves the efficiency and reliability of system deployment.

[0139] The protocol processing engine establishes a secure communication link with the target entity's industrial equipment based on the configuration and initiates a handshake negotiation.

[0140] The protocol processing engine initializes the physical network interface based on the configuration information, setting the correct communication parameters including baud rate, data bits, stop bits, and parity. The engine first sends a device probe message to confirm the target device's online status and basic communication capabilities. Then, it initiates a complete handshake negotiation process according to the protocol specifications, including version number exchange, feature negotiation, and security parameter negotiation. For protocols requiring authentication, the engine performs an authentication process, providing the necessary credentials. During connection establishment, link quality metrics, including signal strength, bit error rate, and response latency, are monitored in real time. After a successful connection establishment, the engine maintains session state information and sets up a heartbeat mechanism to keep the connection active. Simultaneously, a link monitoring thread is started to continuously monitor the connection status and handle anomalies promptly.

[0141] By establishing standardized connection processes and secure negotiation mechanisms, the reliability and security of communication with industrial equipment are ensured. It can automatically adapt to the protocol characteristics of different devices, realizing the automation and intelligence of device connection and greatly reducing the complexity of connection configuration.

[0142] After successfully establishing a connection, poll to read device data or subscribe to device events according to the configured strategy, and receive the raw byte stream returned by the device.

[0143] The protocol processing engine generates corresponding data request commands based on the data point list and acquisition strategy defined in the configuration file. In polling mode, the engine periodically sends read data commands at set time intervals, with the command content converted to the corresponding binary format according to the protocol type. In event subscription mode, the engine registers corresponding event listeners, waiting for the device to actively report data. The received raw byte stream data is stored in a circular buffer, and the engine simultaneously records the reception timestamp and sequence number of each data packet. For large data volume transmissions, the engine implements a flow control mechanism to dynamically adjust the request frequency to prevent device overload. For critical data points, the engine implements a retransmission mechanism to ensure data integrity. Simultaneously, it provides real-time statistics on acquisition success rate, response time, and other performance metrics.

[0144] Through flexible acquisition strategies and reliable data receiving mechanisms, efficient acquisition of industrial equipment data is achieved. It can adapt to the communication characteristics of different devices, ensure data integrity and real-time performance, and provide a stable and reliable data source for subsequent data processing.

[0145] The received raw byte stream undergoes preliminary frame verification and sequence reassembly to output the complete raw data stream.

[0146] The protocol processing engine performs frame boundary detection on the received byte stream to identify the start and end positions of complete data frames. It executes cyclic redundancy check (CRC) and other verification algorithms to verify the integrity of data frames, discarding erroneous frames that fail verification. For fragmented data packets, the engine reassembles them according to sequence numbers to restore complete data content. A reordering buffer is maintained during processing to handle potentially out-of-order data packets. The engine also detects and processes duplicate data packets to ensure data uniqueness. The verified and reassembled data is converted to a unified internal format, and metadata information such as timestamps and data quality tags is added. Finally, a standardized data stream is output for use by subsequent processing modules. An anomaly detection mechanism is implemented throughout the process to promptly identify and handle persistent communication problems.

[0147] Through a robust frame processing and verification mechanism, the integrity and accuracy of the collected data are ensured. This mechanism can effectively handle common communication anomalies in industrial environments, providing high-quality and reliable data input for upper-layer applications and laying a solid foundation for data processing.

[0148] Specifically, the steps for performing real-time in-stream computation on the raw data stream to generate advanced event and feature data are as follows:

[0149] Receive the raw data stream and execute multiple lightweight data processing tasks in parallel;

[0150] Perform data cleaning, unit conversion, and linearization on the raw values ​​in the data stream;

[0151] Based on predefined rules, the cleaned data is subjected to status judgment, threshold comparison and logical operation to generate event tags;

[0152] The values ​​of multiple related data points are aggregated and calculated in real time to generate new feature data points.

[0153] It should be noted that the data stream processing engine, which receives the raw data stream and executes multiple lightweight data processing tasks in parallel, establishes multiple parallel processing pipelines, each equipped with independent memory buffers and processor resources. The raw data stream first enters the data distributor, which intelligently routes the data based on data characteristics (such as device ID, data type, timestamp, etc.), evenly distributing the data to different processing threads. Each processing thread runs on an independent CPU core, employing lock-free queues and circular buffers to achieve efficient data exchange. The system creates dedicated processing channels for each data type, including sensor data channels, status data channels, and event data channels. Each channel implements traffic monitoring and load balancing mechanisms to dynamically adjust processing resource allocation. A strict time-sequence preservation mechanism is implemented during data processing to ensure the sequential and real-time nature of data processing. Simultaneously, a processing status monitoring system is established to track the execution progress and resource consumption of each data processing task in real time.

[0154] Through a parallel data processing architecture and intelligent task allocation mechanism, high-speed data streams are processed efficiently, making full use of the computing power of multi-core processors. This significantly improves data processing throughput and real-time performance, laying a solid foundation for subsequent refined data processing.

[0155] The raw values ​​in the data stream undergo data cleaning, unit conversion, and linearization. The data cleaning module first applies a statistical outlier detection algorithm to identify and filter values ​​outside the reasonable range. A sliding window mechanism is used to detect transient data changes, combined with historical device behavior patterns for rationality verification. For missing data, appropriate interpolation or imputation strategies are selected based on the data type. The unit conversion engine loads a device-specific dimensional mapping table to convert raw data from different sources into unified international standard units. The linearization module, targeting the characteristics of nonlinear sensors, applies device calibration curves and compensation algorithms to convert raw measurements into accurate physical quantities. A quality labeling mechanism is implemented throughout the process, adding processing status and quality identifiers to each data point. A processing log is also established to record all data transformation operations and parameter adjustment history.

[0156] Through systematic data cleaning and standardization, noise and anomalies in the original data were eliminated, ensuring the reliability and consistency of data quality and providing an accurate and reliable data foundation for subsequent data analysis and decision-making.

[0157] Based on predefined rules, the cleaned data undergoes state judgment, threshold comparison, and logical operations to generate event tagging rules. The engine loads a predefined set of state judgment rules, described in a declarative language, supporting complex condition combinations and nested logic. The system implements multi-level threshold management, including normal range, warning threshold, and alarm threshold. Each threshold can be configured with a hysteresis interval to prevent state fluctuations. The logical operation module supports real-time calculation of derived variables, including rate calculation, difference calculation, and cumulative value calculation. The state machine management module tracks device state transitions, records state duration, and supports state transition condition checks. The event generator produces corresponding event tags based on the rule execution results. Each event includes a timestamp, event level, event description, and associated data point information. An event aggregation mechanism is also implemented to prevent duplicate reporting of the same event.

[0158] Through a configurable rules engine and real-time logic processing, intelligent identification and event detection of device status are achieved, enabling timely detection of anomalies and important status changes, and providing high-quality event information for upper-layer applications.

[0159] Real-time aggregation calculations are performed on the values ​​of multiple related data points to generate new feature data points.

[0160] Through real-time data aggregation and feature extraction, higher-level information features are extracted from raw data, providing rich feature inputs for data analysis and machine learning, and realizing the transformation from raw data to valuable information.

[0161] The feature calculation engine maintains a time-sliding window buffer to store historical data within a recent period. The aggregation calculation module performs various statistical operations, including real-time average, moving variance, trend slope, and peak detection. The spatial aggregation module performs correlation analysis on data from multiple measurement points on the same device to calculate overall device status indicators. The time series analysis module detects periodic patterns and trend changes, generating corresponding feature indicators. The machine learning feature extractor runs lightweight algorithms to calculate advanced features such as frequency domain features and correlation coefficients in real time. All generated feature data points have complete timestamps and quality labels and are associated with the original data. The system implements feature caching and reuse mechanisms to avoid redundant calculations and improve processing efficiency.

[0162] Specifically, the steps for uniformly encapsulating and distributing high-level event and feature data to upper-layer applications are as follows:

[0163] The advanced event and feature data output by the stream processing engine are encapsulated in a unified JSON format;

[0164] Add a timestamp, data source device identifier, and data quality tag to each piece of encapsulated data;

[0165] The encapsulated data message is pushed to the message middleware via the MQTT protocol according to the preset topic;

[0166] The message middleware reliably distributes data to cloud platforms or local monitoring systems that have subscribed to the topic.

[0167] It should be noted that the advanced event and feature data output by the stream processing engine are encapsulated in a unified JSON format.

[0168] The data encapsulation module first establishes a standardized JSON template structure, which includes both fixed and dynamic fields. Fixed fields include metadata such as protocol version, gateway identifier, and packet sequence number; dynamic fields are dynamically generated based on data type and include event data, feature data, and extended data areas. The encapsulation processor iterates through the data objects output by the stream processing engine, mapping the numerical value, type description, precision information, and other attributes of each data point to the corresponding fields of the JSON object. For complex data structures, such as arrays or nested objects, a specific JSON structure is used for standardized representation. Data compression is implemented during encoding, precision control is applied to floating-point numbers, and strings are standardized. Syntax validation and structural integrity checks are also performed to ensure that the generated JSON document conforms to standards and can be correctly parsed by various platforms. Finally, the encapsulated JSON data is optimized for size and memory alignment to improve transmission efficiency.

[0169] By encapsulating data in a unified JSON format, data standardization and structuring are achieved, ensuring data compatibility and interoperability between different systems, providing a consistent data interface for upper-layer applications, and greatly simplifying the complexity of data integration and processing.

[0170] Add a timestamp, data source device identifier, and data quality tag to each piece of encapsulated data.

[0171] The timestamp service uses a high-precision clock source to add nanosecond-level precision time stamps to each data packet, while simultaneously recording time zone information and time synchronization status. The device identification system adds complete device traceability information to each data packet based on the device registry, including device manufacturer, device model, device serial number, and network address. The data quality assessment engine performs quality assessments on each data point, generating quality labels that include confidence scores, data integrity indicators, and processing history. The label information adopts a hierarchical structure, including basic quality information and extended quality attributes. All metadata information is encapsulated along with business data and indexed for rapid retrieval. A metadata verification mechanism is also implemented to ensure that the added identification information is accurate and consistent with the data content.

[0172] By adding complete metadata information, comprehensive data traceability and quality control are achieved, providing reliable contextual information for data analysis and decision-making, and significantly improving the credibility and usability of the data.

[0173] Based on the preset topic, the encapsulated data message is pushed to the message middleware via the MQTT protocol.

[0174] The topic management system loads pre-configured topic mapping rules and determines the publication topic for each data packet based on data content, device type, and business requirements. The MQTT client connection pool manages multiple concurrent MQTT connections, with each connection implementing heartbeat maintenance and automatic reconnection mechanisms. A tiered Quality of Service (QoS) strategy is employed during message publishing: critical data uses QoS 1 or 2 to ensure reliable transmission, while ordinary data uses QoS 0 to improve transmission efficiency. Message flow control is implemented to prevent network congestion, including sending rate limiting and batch message aggregation. Message content is optimized for serialization and uses binary format to reduce transmission overhead. Simultaneously, transmission encryption and authentication are implemented to ensure data transmission security. Message queue status and transmission performance are monitored in real time, and publishing strategies are dynamically adjusted accordingly.

[0175] By utilizing the publish-subscribe model and tiered quality of service mechanism of the MQTT protocol, efficient and reliable data transmission is achieved, which can adapt to different network environments and business needs, ensuring the real-time performance and integrity of the data.

[0176] The message middleware reliably distributes data to cloud platforms or local monitoring systems that have subscribed to the topic.

[0177] The message middleware implements topic routing and message forwarding mechanisms to accurately distribute data to each receiver based on subscription relationships. The load balancer monitors the processing capacity of each subscriber, dynamically allocating message traffic to prevent overload of any single subscriber. A persistent storage system persists important messages to disk, ensuring no data loss in the event of system failure. A message acknowledgment mechanism tracks the delivery status of each message and implements retry strategies for unacknowledged messages. The subscription management service maintains a dynamic subscription relationship table, supporting dynamic joining and leaving of subscribers. Message filtering and transformation functions are implemented, providing data format conversion and content filtering services according to the needs of different subscribers. Delivery status monitoring and statistics functions are also provided, displaying message flow and system performance indicators in real time.

[0178] Through the reliable distribution mechanism of the message middleware, data multiplexing and flexible distribution are achieved, which can simultaneously meet the data needs of multiple upper-layer application systems, improve the scalability and flexibility of the system, and ensure the stability and reliability of data services.

[0179] This embodiment also provides an industrial intelligent gateway and a system for adapting to multiple industrial protocols, including:

[0180] The traffic detection module listens to industrial network traffic to learn and build an initial protocol fingerprint database.

[0181] The twin simulation module dynamically constructs a virtual device twin based on the initial protocol fingerprint library to form a simulation environment;

[0182] The automated verification module places the protocol configuration to be deployed in a simulation environment for automated verification and outputs the verified configuration.

[0183] The secure connection module is configured to establish a connection with physical industrial equipment and collect raw data streams.

[0184] The stream computing module performs real-time in-stream computation on the raw data stream to generate advanced event and feature data;

[0185] The encapsulation and distribution module encapsulates advanced event and feature data and distributes it to upper-layer applications.

[0186] This embodiment also provides a computer device applicable to the adaptation method of industrial smart gateway with multiple industrial protocols, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the adaptation method of industrial smart gateway with multiple industrial protocols as proposed in the above embodiment.

[0187] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0188] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the method for adapting an industrial intelligent gateway to multiple industrial protocols as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0189] In summary, this invention achieves automatic identification and compatibility with unknown protocols by: constructing a protocol fingerprint library through network traffic monitoring and self-learning, overcoming the limitations of traditional gateways that rely on predefined protocol libraries; avoids potential production accidents caused by direct operation of physical devices by constructing a virtual device twin and automatically verifying protocol configurations in a simulation environment, significantly improving the safety and reliability of system debugging; and transforms raw data into high-value information by embedding real-time in-stream computing capabilities at the protocol layer, simultaneously completing cleaning, transformation, and feature extraction during data acquisition, effectively reducing uplink bandwidth pressure and cloud computing load, and achieving a leap from data pass-through to edge intelligent processing.

[0190] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for adapting an industrial intelligent gateway to multiple industrial protocols, characterized in that, Includes the following steps: Monitor industrial network traffic to learn and build an initial protocol fingerprint database; Virtual device twins are dynamically constructed based on the initial protocol fingerprint database to form a simulation environment; The protocol configuration to be deployed is placed in a simulation environment for automated verification to output the verified configuration; The application has been verified to establish a connection with physical industrial equipment and collect raw data streams; Perform real-time in-stream computation on the raw data stream to generate advanced event and feature data; High-level event and feature data are uniformly encapsulated and distributed to upper-layer applications.

2. The method for adapting an industrial intelligent gateway to multiple industrial protocols as described in claim 1, characterized in that: The specific steps for monitoring industrial network traffic to learn and build an initial protocol fingerprint database are as follows: The Gateway Data Plane Development Kit (DPDK) bypasses and captures all raw network packets flowing through the network interface; The header and payload of the original network packet are parsed to extract static features including packet length, transmission period, and target port; The interaction sequence patterns and timing intervals of the original network packets are analyzed to construct a communication timing feature model; The static features are aggregated and clustered with the communication timing feature model to form an initial protocol fingerprint library representing different protocol types.

3. The method for adapting an industrial intelligent gateway to multiple industrial protocols as described in claim 2, characterized in that: The specific steps for dynamically constructing a virtual device twin based on the initial protocol fingerprint database to form a simulation environment are as follows: The initial protocol fingerprint database is analyzed to identify the protocol types and versions of various devices in the network; Instantiate a corresponding virtual device twin for each type of identified physical device; Configure each virtual device twin with the protocol command set, response data format and response latency characteristics to be simulated based on the protocol fingerprint; All configured virtual device twins are connected to a virtual network to form a high-fidelity simulation environment isolated from the real network topology.

4. The method for adapting an industrial intelligent gateway to multiple industrial protocols as described in claim 3, characterized in that: The specific steps for automatically verifying the protocol configuration to be deployed in a simulation environment to output a verified configuration are as follows: Deploy manually written or imported protocol parsing scripts and configuration parameters onto the corresponding virtual device twin in the simulation environment; The virtual device twin is driven to respond to instructions issued by the protocol parsing script based on its configured protocol behavior; Monitor and record all interactive messages, response times, and data parsing results during script execution; Analyze the monitoring records and select configurations that can correctly complete data reading and writing without communication errors as verified configurations.

5. The method for adapting an industrial intelligent gateway to multiple industrial protocols as described in claim 4, characterized in that: The application has been verified to establish a connection with physical industrial equipment and collect raw data streams. The specific steps are as follows: The scheduler loads the verified configuration and sends it to the gateway's protocol processing engine; The protocol processing engine establishes a secure communication link with the target entity's industrial equipment based on the configuration and initiates a handshake negotiation. After successfully establishing a connection, poll to read device data or subscribe to device events according to the configured strategy, and receive the raw byte stream returned by the device; The received raw byte stream undergoes preliminary frame verification and sequence reconstruction, outputting a complete raw data stream.

6. The method for adapting an industrial intelligent gateway to multiple industrial protocols as described in claim 5, characterized in that: The specific steps for performing real-time in-stream computation on the raw data stream to generate advanced event and feature data are as follows: Receive the raw data stream and execute multiple lightweight data processing tasks in parallel; Perform data cleaning, unit conversion, and linearization on the raw values ​​in the data stream; Based on predefined rules, the cleaned data is subjected to status judgment, threshold comparison and logical operation to generate event tags; The values ​​of multiple related data points are aggregated and calculated in real time to generate new feature data points.

7. The method for adapting an industrial intelligent gateway to multiple industrial protocols as described in claim 6, characterized in that: The specific steps for uniformly encapsulating and distributing high-level event and feature data to upper-layer applications are as follows: The advanced event and feature data output by the stream processing engine are encapsulated in a unified JSON format; Add a timestamp, data source device identifier, and data quality tag to each piece of encapsulated data; The encapsulated data message is pushed to the message middleware via the MQTT protocol according to the preset topic; The message middleware reliably distributes data to cloud platforms or local monitoring systems that have subscribed to the topic.

8. An industrial intelligent gateway and multiple industrial protocol adaptation system, based on the industrial intelligent gateway and multiple industrial protocol adaptation method according to any one of claims 1 to 7, characterized in that: include, The traffic detection module listens to industrial network traffic to learn and build an initial protocol fingerprint database. The twin simulation module dynamically constructs a virtual device twin based on the initial protocol fingerprint library to form a simulation environment; The automated verification module places the protocol configuration to be deployed in a simulation environment for automated verification and outputs the verified configuration. The secure connection module is configured to establish a connection with physical industrial equipment and collect raw data streams. The stream computing module performs real-time in-stream computation on the raw data stream to generate advanced event and feature data; The encapsulation and distribution module encapsulates advanced event and feature data and distributes it to upper-layer applications.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the industrial intelligent gateway and the method for adapting to multiple industrial protocols as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the industrial intelligent gateway and the method for adapting to multiple industrial protocols as described in any one of claims 1 to 7.

Images