A visual Internet of Things rule arrangement method, device and medium

Abstract

The application discloses a visual Internet of Things rule arrangement method and device and medium. The method comprises the following steps: converting an input natural language into a structured text; collecting original data of a device through an Internet of Things platform, and performing standardization processing on the original data to construct a rule arrangement data set; performing multi-source feature extraction and correlation capture on the structured text and the rule arrangement data set through a pre-trained BERT-GPT Internet of Things rule model to generate an initial Internet of Things rule of a target device; performing rule node and logic operator dragging on the initial Internet of Things rule through a pre-set visual rule arrangement interface to obtain a process version Internet of Things rule of the target device; performing data simulation on the process version Internet of Things rule through a pre-set simulation tool to obtain a rule execution result of the target device; and performing optimization on the rule execution result through the BERT-GPT Internet of Things rule model according to historical operation data to obtain a final version Internet of Things rule of the target device.

Description

Technical Field

[0001] This application relates to the field of Internet of Things (IoT) technology, and in particular to a method, device, and medium for visualizing IoT rule arrangement. Background Technology

[0002] In existing IoT systems, business rule orchestration is a core component for enabling device linkage, remote monitoring, and anomaly response.

[0003] There are two main types of existing IoT rule orchestration methods: one relies on manually writing JavaScript, Python, and other scripts, which requires technicians to master both programming skills and device communication protocol details, making it extremely difficult to get started; the other is a simple visual drag-and-drop tool, which simplifies the operation but still requires manual definition of logic and cannot automatically adapt to the characteristics of multiple brands and models of devices.

[0004] Currently, in industrial and smart home scenarios, there is a wide variety of device types (sensors, controllers, etc.), diverse communication protocols (MQTT, CoAP, Modbus, etc.), and frequently changing business requirements. This leads to problems with the two mainstream orchestration methods mentioned above, such as inconsistent cross-device parameter identification (e.g., temperature corresponding to Temp and temperature value), long generation cycles for complex linkage rules, and difficulty in dynamically optimizing based on device operating status. Ultimately, this results in low orchestration efficiency and poor adaptation success rate of existing rules, which not only restricts the intelligent upgrade of the system but also significantly increases operation and maintenance costs. Summary of the Invention

[0005] This application provides a visual IoT rule orchestration method, device, and medium to solve the problems of low rule orchestration efficiency and poor rule adaptation success rate in existing IoT rule orchestration methods.

[0006] The embodiments of this application adopt the following technical solutions:

[0007] On one hand, embodiments of this application provide a visual IoT rule orchestration method, which includes: converting natural language containing IoT business requirements into structured text; collecting raw data from devices through an IoT platform and standardizing the raw data to construct a rule orchestration dataset; using a pre-trained BERT-GPT IoT rule model to extract multi-source features and capture correlations between the structured text and the rule orchestration dataset to generate initial IoT rules for the target device; dragging and dropping rule nodes and logical operators on the initial IoT rules through a preset visual rule orchestration interface to obtain the process version IoT rules for the target device; using a preset simulation tool to perform data simulation and verification of the process version IoT rules to obtain the rule execution results for the target device; and optimizing the rule execution results using the BERT-GPT IoT rule model based on historical running data to obtain the final version IoT rules for the target device.

[0008] In one example, after optimizing the rule execution results using the BERT-GPT IoT rule model based on the rule execution results and the historical running data to obtain the final version of the IoT rules for the target device, the method further includes: converting the final version of the IoT rules into device running instructions and deploying them to the target device; monitoring the execution of the instructions on the target device and synchronizing the monitoring data to the cloud platform to iteratively update the BERT-GPT IoT rule model.

[0009] In one example, the initial IoT rules are dragged and dropped using a preset visual rule arrangement interface to obtain the process version IoT rules for the target device. Specifically, this includes: laying out the initial IoT rules in the middle canvas area of ​​the visual rule arrangement interface according to the logical sequence of the initial IoT rules; dragging nodes from the rule node library to the middle canvas area to replace the existing nodes of the initial IoT rules according to the multi-source features and associations of the initial IoT rules; dragging operators from the logic operator library to the middle canvas area according to the multi-source features and associations of the initial IoT rules to connect the nodes of the initial IoT rules with the dragged operators; and performing logical conflict detection on the initial IoT rules after the operators are connected. If there is no logical conflict, the initial IoT rules after the operators are connected are determined as the process version IoT rules.

[0010] In one example, the process version of IoT rules is simulated and verified using a preset simulation tool to obtain the rule execution result of the target device. Specifically, this includes: simulating the device operation scenario using the preset simulation tool to generate time-series simulation data; performing data simulation and verification on the process version of IoT rules based on the time-series simulation data to obtain the verification result; and visually displaying the verification result to obtain the rule execution result of the target device.

[0011] In one example, a pre-trained BERT-GPT IoT rule model is used to extract multi-source features and capture associations from the structured text and the rule orchestration dataset to generate initial IoT rules for the target device. Specifically, this includes: using the BERT-GPT IoT rule model to extract demand semantics, device parameter statistics, and protocol adaptation features from the structured text and rule orchestration dataset to obtain multi-source features; using a multi-head attention mechanism to capture the temporal and cross-device associations between business requirements and device parameters to obtain associations; and integrating the extracted multi-source features with the captured associations to generate initial IoT rules for the target device.

[0012] In one example, raw data from a preset type of device is collected through an IoT platform, and the raw data is standardized to construct a rule orchestration dataset. Specifically, this includes: parsing the protocols of the preset type of device through a preset device-side protocol conversion gateway to collect the metadata of the preset type of device; collecting historical operating data of the preset type of device through a preset edge node data acquisition module and historical acquisition cycle; and performing semantic mapping and missing value filling on the metadata and historical operating data to obtain the rule orchestration dataset.

[0013] In one example, converting input natural language into structured text specifically includes: performing semantic recognition and semantic segmentation on the input natural language requirement using a pre-trained NLP module to obtain semantic recognition results; extracting information from the semantic recognition results using named entity recognition technology to obtain information extraction results; information extraction includes extracting requirement type, target device, triggering conditions, and execution actions; and performing ambiguity processing and semantic completion on the information extraction results through preset interactive questions to generate structured text.

[0014] In one example, based on the historical operational data, the BERT-GPT IoT rule model is used to optimize the rule execution results to obtain the final version of the IoT rule for the target device. Specifically, this includes: inputting the rule execution results and the historical operational data into the BERT-GPT IoT rule model to obtain optimization suggestions; optimizing the rule execution results based on the optimization suggestions to obtain optimized IoT rules; and simulating and verifying the optimized IoT rules. When the simulation pass rate is greater than a preset pass rate threshold and the simulation response time is less than a preset response time threshold, the optimized IoT rule is determined as the final version of the IoT rule.

[0015] On the other hand, embodiments of this application provide a visual IoT rule orchestration device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute any of the above-mentioned visual IoT rule orchestration methods.

[0016] On the other hand, embodiments of this application provide a visual IoT rule orchestration non-volatile computer storage medium storing computer-executable instructions that can execute any of the above-mentioned visual IoT rule orchestration methods.

[0017] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects:

[0018] This application integrates AI large-scale models and natural language processing technology to automatically convert users' unstructured natural language requirements into structured text that the system can recognize. This eliminates the need for users to have programming or device protocol knowledge, lowering the operational threshold for IoT rule orchestration. Through device-side protocol conversion gateways and data standardization processing, it supports the parsing of multiple protocols such as MQTT, CoAP, and Modbus, unifying the mapping of heterogeneous parameter identifiers (such as Temp and temperature values) from different devices, significantly improving rule adaptation success rates. A pre-trained BERT-GPT IoT rule model is used for multi-source feature extraction and correlation capture, combined with a visual drag-and-drop interface for user rule adjustments, greatly improving the efficiency of rule orchestration for complex business rules. Simulation tools are used for data simulation verification and AI-driven automatic optimization, combined with iterative rule iteration based on historical operational data, significantly improving rule accuracy, shortening rule execution response time, and enhancing the efficiency and reliability of IoT rule execution. Attached Figure Description

[0019] To more clearly illustrate the technical solution of this application, some embodiments of this application will be described in detail below with reference to the accompanying drawings, in which:

[0020] Figure 1 A flowchart illustrating a visual IoT rule orchestration method provided in this application embodiment;

[0021] Figure 2 This is a schematic diagram of the structure of a visual IoT rule orchestration device provided in an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] Some embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0024] Figure 1 This is a flowchart illustrating a visual IoT rule orchestration method provided in an embodiment of this application. This method can be applied to different business domains. Certain input parameters or intermediate results in this process allow for manual intervention and adjustment to help improve accuracy.

[0025] The analysis method involved in the embodiments of this application can be implemented by a terminal device or a server, and this application does not impose any special limitations on it. For ease of understanding and description, the following embodiments are all described in detail using a server as an example.

[0026] Based on this Figure 1 The process may include the following steps:

[0027] S101: Convert natural language containing IoT business requirements into structured text; the structured text includes requirement type, target device, triggering conditions, and execution actions.

[0028] In some embodiments of this application, the conversion of natural language is the starting point of the entire IoT rule orchestration process. The core is to transform the user's unstructured natural language requirements into structured text that the system can recognize, so as to provide accurate semantic basis for subsequent rule generation.

[0029] First, the system receives IoT business requests described by users in everyday language through an interactive interface (which supports text or voice input). The requests must clearly include the type of request, the target device, the triggering conditions, and the action to be performed.

[0030] It should be noted that users can input text via keyboard or voice input (the system has a built-in speech-to-text module that supports Mandarin Chinese and common industry terms), ensuring that non-professional users (such as workshop managers and home users) can accurately express their needs without having to master technical terms.

[0031] Furthermore, the pre-trained NLP module performs semantic recognition and semantic segmentation on the input natural language requirements, identifies the user's core intent, and eliminates redundant expressions (such as colloquialisms like "please help me set it up" and other meaningless words) to obtain the semantic recognition results.

[0032] Furthermore, named entity recognition technology is used to extract information from the semantic recognition results. Specifically, for the extraction of demand types, matching is performed from preset categories (monitoring, control, early warning, data analysis), such as sending an early warning SMS corresponding to the early warning category, and turning on ventilation equipment corresponding to the control category. For the extraction of target equipment, the specific equipment involved in the demand is clearly defined, including the equipment name, number, type or location (e.g., workshop No. 1 temperature and humidity sensor, warehouse cold chain equipment, living room lighting controller), and the system supports auto-completion of fuzzy equipment names. Regarding the extraction of trigger conditions, quantitative parameters, logical relationships and thresholds are extracted, including parameter types, comparison operators, and specific values ​​(e.g., exceeding 75℃). If the user does not specify the threshold (e.g., alarm when the temperature is too high), the system automatically recommends a default threshold (e.g., 90% of the upper limit of the equipment's safe temperature) based on historical data and prompts the user for confirmation. Regarding the extraction of execution actions, the expected equipment operation and information transmission methods of the user are broken down, and the execution order of the actions is marked (e.g., turn off the equipment first, then send an SMS).

[0033] Furthermore, through pre-set interactive questions, the information extraction results are ambiguous and semantically completed to generate structured text.

[0034] This application integrates AI large-scale models with natural language processing technology to automatically transform users' unstructured natural language needs into structured text that the system can recognize, without requiring users to have programming or device protocol knowledge, thus lowering the operational threshold for IoT rule arrangement.

[0035] S102: Collect raw data from preset types of devices through an IoT platform, and standardize the raw data to construct a rule-arranged dataset; the raw data includes metadata and historical operation data.

[0036] In some embodiments of this application, while obtaining structured text, the protocol of a preset type of device is parsed through a preset device-side protocol conversion gateway to collect the metadata of the preset type of device; the gateway supports parsing multiple protocols such as MQTT, CoAP, and Modbus.

[0037] At the same time, historical operating data of preset types of devices are collected through preset edge node data acquisition modules and historical acquisition periods (such as within 6 months).

[0038] Furthermore, semantic mapping and missing value imputation are performed on metadata and historical operational data (e.g., Temp and temperature values ​​are uniformly mapped to temperature parameters, and missing values ​​are supplemented by interpolation) to obtain a rule orchestration dataset.

[0039] By using a device-side protocol conversion gateway and data standardization processing, it supports the parsing of multiple protocols such as MQTT, CoAP, and Modbus, and uniformly maps heterogeneous parameter identifiers (such as Temp and temperature values) of different devices, which greatly improves the success rate of rule adaptation.

[0040] S103: Using a pre-trained BERT-GPT IoT rule model, perform multi-source feature extraction and correlation capture on the structured text and the rule orchestration dataset to generate initial IoT rules for the target device.

[0041] In some embodiments of this application, after obtaining the rule orchestration dataset, the pre-trained BERT-GPT IoT rule model is used to extract demand semantics, device parameter statistics, and protocol adaptation features from the structured text and rule orchestration dataset to obtain multi-source features.

[0042] At the same time, through a multi-head attention mechanism, the temporal correlation between business needs and equipment parameters (such as the mapping relationship between high temperature warning and temperature parameters exceeding the threshold) and cross-device correlation (such as the linkage logic between device A overheating and device B starting) are captured to obtain the correlation relationship.

[0043] Furthermore, the extracted multi-source features are integrated with the captured correlations to generate initial IoT rules for the target device.

[0044] S104: Through a preset visual rule arrangement interface, the initial IoT rules are dragged and dropped to obtain the process version IoT rules for the target device; the visual rule arrangement interface includes a rule node library and a logic operator library.

[0045] In some embodiments of this application, after obtaining the initial IoT rules, the initial IoT rules are laid out in the middle canvas area of ​​the visual rule arrangement interface according to the logical sequence of the initial IoT rules; then, according to the multi-source features and association relationships of the initial IoT rules, nodes in the rule node library are dragged to the middle canvas area to replace the existing nodes of the initial IoT rules (such as replacing SMS notification nodes with APP push nodes), and the system automatically detects the compatibility of the new nodes with the upstream and downstream logic.

[0046] Furthermore, based on the multi-source features and relationships of the initial IoT rules, operators from the logic operator library are dragged and dropped to the intermediate canvas area to connect the nodes of the initial IoT rules with the dragged operators.

[0047] Logical conflict detection is performed on the initial IoT rules after operator connection. When a conflict is found, the problematic node is highlighted with a red border. If no logical conflict exists, the initial IoT rules after operator connection are determined as the procedural version IoT rules. Logical conflicts include parameter conflicts, logical contradictions, and timing errors.

[0048] It should be noted that the visual rule orchestration interface includes a central canvas area, a left-side resource library, and a right-side parameter configuration area. Both the rule node library and the logic operator library reside in the left-side resource library. The rule node library includes device interaction nodes, data processing nodes, action execution nodes, and process auxiliary nodes. The logic operator library includes basic logic operators, sequential logic operators, and branch logic operators.

[0049] The central canvas area uses a flowchart visualization method to display the rule logic, supports drag, zoom, and pan operations, and displays the logical relationship between nodes and operators through connecting lines, highlighting the currently operated element in real time.

[0050] After selecting a node or operator in the canvas in the parameter configuration area on the right, a form of configurable parameters will be automatically displayed. It supports interactive methods such as numerical input, drop-down selection, and slider adjustment. Some parameters provide intelligent recommended values.

[0051] By using a pre-trained BERT-GPT IoT rule model to extract multi-source features and capture correlations, and combining this with a visual drag-and-drop interface for users to adjust rules, the efficiency of rule orchestration for complex business rules is greatly improved.

[0052] S105: The process version of the IoT rules is simulated and verified using a preset simulation tool to obtain the rule execution result of the target device.

[0053] In some embodiments of this application, after obtaining the process version of IoT rules, a device operation scenario is simulated using a preset simulation tool to generate time-series simulation data; then, based on the time-series simulation data, the process version of IoT rules is verified by data simulation to obtain the data verification result.

[0054] Furthermore, visualization technology is used to visualize the execution process of the data verification results, thereby obtaining the rule execution results of the target device.

[0055] It should be noted that the time-series simulation data includes basic parameter simulation data, which is continuous parameter data such as temperature, humidity, and pressure generated according to a set trend (e.g., temperature 60-75℃ in the first 10 minutes, 75-85℃ in the first 10-20 minutes), and the data includes random fluctuations (±1℃) to approximate the measurement error of real equipment; abnormal data injection, which is to trigger abnormal parameters at a set time point (e.g., the temperature suddenly rises to 85℃ at the 12th minute) to simulate equipment failure or sudden environmental changes; and cross-device linkage data, which is to automatically generate response data of related devices when a certain device triggers a rule condition.

[0056] S106: Based on the historical operation data, optimize the rule execution results through the BERT-GPT IoT rule model to obtain the final version of the IoT rule for the target device; the optimization includes device adaptation optimization, action timing optimization, threshold setting optimization, and logic timing optimization.

[0057] In some embodiments of this application, after obtaining the rule execution result, the rule execution result and historical operation data are input into the BERT-GPT IoT rule model to obtain optimization suggestions; based on the optimization suggestions, the rule execution result is optimized to obtain optimized IoT rules; these include device adaptation optimization, action timing optimization, threshold setting optimization, and logic timing optimization. Regarding threshold adjustment optimization, for example, when the temperature reaches 80°C in the simulation, the device is close to the critical state, it is recommended to lower the threshold to 78°C to reserve reaction time; regarding logic optimization, for example, when humidity parameters are reported with a 2-second delay during multi-condition linkage, it is recommended to add fault tolerance logic that allows a ±1-second delay; regarding action timing optimization, for example, when the temperature drops slowly after the ventilation equipment is turned on, it is recommended to extend the running time to 10 minutes; regarding device adaptation optimization, for example, when virtual devices frequently go offline causing action failures, it is recommended to add logic to the rules to switch to a backup device when the device goes offline.

[0058] Furthermore, the optimized IoT rules are simulated and verified. When the simulation pass rate is greater than the preset pass rate threshold and the simulation response time is less than the preset response time threshold, the optimized IoT rules are determined as the final IoT rules.

[0059] Furthermore, the final IoT rules are converted into device operation instructions and deployed to the target device; then the instruction operation process of the target device is monitored, and the monitoring data is synchronized to the cloud platform to iteratively update the BERT-GPT IoT rule model.

[0060] It should be noted that, although the embodiments in this application are based on... Figure 1 Steps S101 to S106 will be described sequentially, but this does not mean that steps S101 and S106 must be performed in a strict order. The reason this embodiment follows this order is... Figure 1 The order in which steps S101 to S106 are described is provided to facilitate understanding of the technical solutions of the embodiments of this application by those skilled in the art. In other words, in the embodiments of this application, the order of steps S101 to S106 can be appropriately adjusted according to actual needs.

[0061] pass Figure 1This application utilizes a method that integrates large AI models and natural language processing technology to automatically convert users' unstructured natural language requirements into structured text that the system can recognize. This eliminates the need for users to possess programming or device protocol knowledge, lowering the operational threshold for IoT rule orchestration. Through device-side protocol conversion gateways and data standardization processing, it supports the parsing of multiple protocols such as MQTT, CoAP, and Modbus, unifying the mapping of heterogeneous parameter identifiers (such as Temp and temperature values) from different devices, significantly improving the rule adaptation success rate. A pre-trained BERT-GPT IoT rule model is used for multi-source feature extraction and correlation capture, combined with a visual drag-and-drop interface for users to adjust rules, greatly improving the efficiency of rule orchestration for complex business rules. Simulation tools are used for data simulation verification and AI-automated optimization, combined with iterative rules based on historical operational data, significantly improving rule accuracy, shortening rule execution response time, and enhancing the efficiency and reliability of IoT rule execution.

[0062] Figure 2 A schematic diagram of a visual IoT rule orchestration device provided in this application embodiment includes:

[0063] At least one processor; and,

[0064] A memory that is communicatively connected to at least one processor; wherein,

[0065] A visual IoT rule orchestration method is provided, in which the memory stores instructions that can be executed by at least one processor to enable at least one processor to perform any of the above-mentioned tasks.

[0066] Some embodiments of this application provide a visual IoT rule orchestration non-volatile computer storage medium storing computer-executable instructions that can execute any of the above-mentioned visual IoT rule orchestration methods.

[0067] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0068] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0069] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0070] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0071] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0072] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0073] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0074] Memory may include non-persistent storage in computer-readable media, random access memory (RAM), and non-volatile memory such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0075] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0076] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0077] The above are merely embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the technical principles of this application should fall within the protection scope of this application.

Claims

1. A visual IoT rule orchestration method, characterized in that, The method includes: The natural language containing IoT business requirements is converted into structured text; the structured text includes the requirement type, target device, triggering conditions, and execution actions. Raw data from preset types of devices is collected through an IoT platform, and the raw data is standardized to construct a rule-based orchestration dataset; the raw data includes metadata and historical operational data. The process of collecting raw data from preset types of devices through an IoT platform and standardizing the raw data to construct a rule-based orchestration dataset specifically includes: The protocol of a preset type of device is parsed through a preset device-side protocol conversion gateway in order to collect the metadata of the preset type of device; The system collects historical operating data of preset types of devices by using preset edge node data acquisition modules and historical acquisition cycles. Semantic mapping and missing value imputation are performed on the metadata and historical operation data to obtain the rule orchestration dataset; Using a pre-trained BERT-GPT IoT rule model, multi-source feature extraction and correlation capture are performed on the structured text and the rule orchestration dataset to generate initial IoT rules for the target device. The process involves using a pre-trained BERT-GPT IoT rule model to perform multi-source feature extraction and correlation capture on the structured text and the rule orchestration dataset, generating initial IoT rules for the target device. Specifically, this includes: Using the BERT-GPT IoT rule model, demand semantics, device parameter statistics, and protocol adaptation feature extraction are performed on the structured text and the rule orchestration dataset to obtain multi-source features; By using a multi-head attention mechanism, the temporal and cross-device correlations between business requirements and device parameters are captured to obtain the correlation relationships; The extracted multi-source features are integrated with the captured correlations to generate the initial IoT rules for the target device; Through a preset visual rule arrangement interface, the initial IoT rules are dragged and dropped to obtain the process version of the IoT rules for the target device; the visual rule arrangement interface includes a rule node library and a logic operator library; The process version of the IoT rules is verified by data simulation using a preset simulation tool to obtain the rule execution results of the target device. Based on the rule execution results and the historical operation data, the rule execution results are optimized using the BERT-GPT IoT rule model to obtain the final version of the IoT rules for the target device; the optimization includes device adaptation optimization, action timing optimization, threshold setting optimization, and logic timing optimization.

2. The method according to claim 1, characterized in that, After optimizing the rule execution results using the BERT-GPT IoT rule model based on the rule execution results and the historical operation data to obtain the final version of the IoT rules for the target device, the method further includes: The final version of the IoT rules is converted into device operation instructions and deployed to the target device; The operation of the target device's commands is monitored, and the monitoring data is synchronized to the cloud platform to iteratively update the BERT-GPT IoT rule model.

3. The method according to claim 1, characterized in that, The process involves dragging and dropping rule nodes and logic operators from the initial IoT rules through a preset visual rule arrangement interface to obtain the process version of the IoT rules for the target device. Specifically, this includes: Based on the logical sequence of the initial IoT rules, the initial IoT rules are laid out in the middle canvas area of ​​the visual rule arrangement interface. Based on the multi-source characteristics and relationships of the initial IoT rules, drag the nodes in the rule node library to the middle canvas area to replace the existing nodes of the initial IoT rules. Based on the multi-source features and relationships of the initial IoT rules, drag and drop operators from the logic operator library to the middle canvas area to connect the nodes of the initial IoT rules with the dragged operators. Logical conflict detection is performed on the initial IoT rules after operator connection. If no logical conflict exists, the initial IoT rules after operator connection are determined as the process version IoT rules.

4. The method according to claim 1, characterized in that, The step of using a preset simulation tool to perform data simulation and verification of the process version IoT rules to obtain the rule execution results of the target device specifically includes: The system simulates equipment operation scenarios using preset simulation tools, generating time-series simulation data. Based on the time-series simulation data, the process version of the IoT rules is verified by data simulation to obtain the verification results. The execution process of the verification results is visualized using visualization technology to obtain the rule execution results of the target device.

5. The method according to claim 1, characterized in that, The process of converting natural language containing IoT business requirements into structured text specifically includes: The semantic recognition result is obtained by performing semantic recognition and semantic word segmentation on the input natural language requirements through a pre-trained NLP module. By using named body recognition technology, information is extracted from the semantic recognition results to obtain information extraction results; the information extraction includes the extraction of demand type, target device, triggering conditions and execution actions; By using pre-set interactive questions, the information extraction results are subjected to ambiguity processing and semantic completion to generate structured text.

6. The method according to claim 1, characterized in that, The step involves optimizing the rule execution results based on the historical operational data using the BERT-GPT IoT rule model to obtain the final version of the IoT rules for the target device, specifically including: The execution results of the rules and the historical running data are input into the BERT-GPT IoT rule model to obtain optimization suggestions; Based on the optimization suggestions, the execution result of the rule is optimized to obtain the optimized IoT rule; The optimized IoT rules are simulated and verified. When the simulation pass rate is greater than the preset pass rate threshold and the simulation response time is less than the preset response time threshold, the optimized IoT rules are determined as the final IoT rules.

7. A visual IoT rule orchestration device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the visual IoT rule orchestration method according to any one of claims 1-6.

8. A visual Internet of Things (IoT) rule arrangement storage medium, storing computer-executable instructions, characterized in that, The computer-executable instructions are capable of executing the visual Internet of Things rule orchestration method described in any one of claims 1-6.

Images