A cloud-edge-end intelligent collaborative system and method for cross-network secure transmission

Abstract

This invention discloses a cloud-edge-device intelligent collaborative system and method for secure cross-network transmission, comprising terminal devices, edge nodes, an edge cloud platform, and a central cloud platform: The terminal devices collect production data from oil and gas field operations and upload the data to the edge nodes; the edge nodes upload the production data to the edge cloud platform and perform AI inference and execution based on received container images or model files; the edge cloud platform preprocesses the production data based on preset cloud-edge collaborative rules and transmits the processed production data to the central cloud platform via a northbound security gateway; and deploys the received container images or model files to the corresponding edge nodes; the central cloud platform trains AI algorithms based on the production data to obtain container images or model files, which are then transmitted to the edge cloud platform via a southbound security gateway. This system enables secure sharing and intelligent collaborative scheduling of oil and gas production data.

Description

Technical Field

[0001] This invention relates to the field of control technology in the petroleum energy industry, and in particular to a cloud-edge-device intelligent collaborative system and method for cross-network secure transmission. Background Technology

[0002] As numerous domestic and international companies adopt cloud computing and edge computing technologies to enhance their Internet of Things (IoT) capabilities, oil and energy companies, in the process of promoting digital transformation, have also begun to explore the gradual integration of cloud computing and edge computing into the construction of IoT in oil and gas production. This aims to improve the timeliness of production data processing at oil and gas operation sites, accelerate the effective mining of the value of IoT source data, and enhance the efficiency of enterprise production and operation.

[0003] Currently, oil and gas field enterprises primarily rely on traditional IoT technologies to collect and monitor various production data within their dedicated production networks (LANs). Simultaneously, they directly incorporate edge computing technology into critical equipment and business activities. This involves deploying algorithm models / edge applications in a scattered, point-like fashion, combining this with real-time data from various sensors or cameras to achieve local inference computation. This approach can, to some extent, meet the needs of common operational condition diagnosis and fault early warning applications within the dedicated production network. Furthermore, enterprises can use industrial control security gateways to transmit field data unidirectionally to the office network. The data lake on the cloud platform then facilitates the exchange of production data and business information with the cloud computing center, enabling cloud computing operations. Finally, AI inference results are manually pushed back to the dedicated production network for manual model / application updates and upgrades. Summary of the Invention

[0004] To achieve secure cross-network data collaborative transmission and management, this invention provides a cloud-edge-device intelligent collaborative system and method for secure cross-network transmission.

[0005] In a first aspect, embodiments of the present invention provide a cloud-edge-device intelligent collaborative system for secure cross-network transmission, comprising terminal devices, edge nodes, an edge cloud platform, and a central cloud platform:

[0006] The terminal device is used to collect production data from oil and gas field operations and upload the production data to the edge node.

[0007] The edge node is used to upload the production data to the edge cloud platform; and to perform AI inference and execution based on the received container image or model file;

[0008] The edge cloud platform is used to preprocess the production data based on preset cloud-edge collaboration rules, and transmit the processed production data to the central cloud platform through a northbound security gateway; and to deploy the received container images or model files to the corresponding edge nodes.

[0009] The central cloud platform is used to train AI algorithms based on the production data, obtain algorithm models, and package the algorithm models into container images or model files, which are then transmitted to the edge cloud platform through a southbound security gateway.

[0010] Secondly, embodiments of the present invention provide a cloud-edge-device intelligent collaboration method for cross-network secure transmission, which may include:

[0011] Collect production data from various terminal devices at the oil and gas field operation site and upload the production data to the edge node;

[0012] The edge nodes upload the production data to the edge cloud platform;

[0013] The production data is preprocessed based on preset cloud-edge collaboration rules, and the processed production data is transmitted to the central cloud platform through the northbound security gateway.

[0014] The central cloud platform trains AI algorithms based on the production data to obtain an algorithm model, and packages the algorithm model into a container image or model file, which is then transmitted to the edge cloud platform through the southbound security gateway.

[0015] The edge cloud platform will deploy the received container image or model file to the corresponding edge node;

[0016] The edge nodes perform AI inference and execution based on the received container images or model files.

[0017] Thirdly, embodiments of the present invention provide a computer-readable storage medium storing a computer program / instruction thereon, which, when executed by a processor, implements the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described above.

[0018] Fourthly, embodiments of the present invention provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described above.

[0019] Fifthly, embodiments of the present invention provide a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described above.

[0020] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0021] This invention provides a cloud-edge-device intelligent collaborative system for secure cross-network transmission. The system includes terminal devices, edge nodes, an edge cloud platform, and a central cloud platform. Terminal devices collect production data from oil and gas field operations and upload it to edge nodes. Edge nodes transmit the data to the edge cloud platform and perform AI inference based on received container images or model files. The edge cloud platform processes the data according to preset cloud-edge collaboration rules and transmits it to the central cloud platform via a northbound security gateway, deploying the container images or model files to edge nodes. The central cloud platform trains AI algorithm models based on the production data, packages the models into container images or files, and transmits them to the edge cloud platform via a southbound security gateway. This system differs from the existing three-layer "device-edge-cloud" deployment architecture by establishing a four-layer "device-edge-edge-central cloud" deployment architecture for the heterogeneous network environment of oil and gas field enterprises. By collecting and processing production data from the oil and gas field and securely and efficiently uploading it to the cloud for AI algorithm training, it enables intelligent equipment monitoring and fault prediction. By leveraging the collaboration between edge cloud platforms and central cloud platforms, real-time data processing, intelligent inference, centralized management, and monitoring can be achieved. This supports localized data processing and real-time response, enabling real-time secure sharing of production data across regions and organizations at all levels within the group company / oil company from the edge oil and gas production private network to the central cloud platform. It also facilitates intelligent collaborative scheduling of algorithm models and business applications, thereby improving the flexibility and intelligent collaborative efficiency of various business applications in the oil and gas production IoT.

[0022] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0023] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0024] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0025] Figure 1 This is a schematic diagram of the structure of a cloud-edge-device intelligent collaborative system for cross-network secure transmission provided in an embodiment of the present invention;

[0026] Figure 2 This is an example diagram illustrating the edge node functionality provided in an embodiment of the present invention;

[0027] Figure 3 This is a schematic diagram of the structure of an edge cloud platform provided in an embodiment of the present invention;

[0028] Figure 4 This is a functional example diagram of the edge-cloud-edge collaboration module provided in an embodiment of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of the central cloud platform provided in an embodiment of the present invention;

[0030] Figure 6 This is an example diagram of the core modules of the cloud-edge-device intelligent collaborative system for cross-network secure transmission provided in an embodiment of the present invention;

[0031] Figure 7 A business process example diagram provided for embodiments of the present invention;

[0032] Figure 8 A system architecture diagram provided for embodiments of the present invention;

[0033] Figure 9 This is a flowchart illustrating the cloud-edge-device intelligent collaboration method for cross-network secure transmission provided in an embodiment of this application. Detailed Implementation

[0034] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0035] The inventors discovered that in existing technologies, oil and gas field enterprises mainly use a three-layer deployment architecture of "end-edge-cloud," relying on centralized storage on cloud platforms, cloud computing, or local edge computing models. However, the application of technologies such as autonomous decision-making by edge devices and cross-network collaborative management has limitations, resulting in technical bottlenecks such as poor compatibility between edge device installation and configuration and existing systems, dispersed deployment of edge models making maintenance and upgrades difficult, and high costs for centralized cloud governance of production data. Furthermore, the exchange of field data and business information in different network environments is not timely, and the data transmission security from the edge oil and gas production IoT to the central cloud platform also faces challenges. This makes it difficult to meet the widespread reuse and large-scale promotion of IoT intelligent applications across oil and gas field enterprises, and also fails to meet the needs of secure sharing of production data and collaborative scheduling of production command across regions and organizations. Based on this, the inventors, through further research and development, have created this invention, providing a cloud-edge-end intelligent collaborative system and method for secure cross-network transmission.

[0036] Example 1

[0037] Embodiment 1 of this invention provides a cloud-edge-device intelligent collaborative system for secure cross-network transmission, referring to... Figure 1As shown, the system includes terminal device 11, edge node 12, edge cloud platform 13, and central cloud platform 14.

[0038] Terminal device 11 is used to collect production data from oil and gas field operations and upload the production data to edge node 12.

[0039] Edge node 12 is used to upload production data to edge cloud platform 13; and to perform AI inference and execution based on received container images or model files.

[0040] Edge cloud platform 13 is used to preprocess production data based on preset cloud-edge collaboration rules, and transmit the processed production data to central cloud platform 14 through northbound security gateway; and to deploy the received container images or model files to the corresponding edge nodes.

[0041] The central cloud platform 14 is used to train AI algorithms based on production data, obtain algorithm models, and package the algorithm models into container images or model files, which are then transmitted to the edge cloud platform 13 through the southbound security gateway.

[0042] Among them, terminal equipment 11, edge node 12 and edge cloud platform 13 are located on the production network side of the oilfield / operation area enterprise, and central cloud platform 14 is located on the office network side of the group company / oil company enterprise.

[0043] This invention provides a cloud-edge-device intelligent collaborative system for secure cross-network transmission. The system includes terminal devices, edge nodes, an edge cloud platform, and a central cloud platform. Terminal devices collect production data from oil and gas field operations and upload it to edge nodes. Edge nodes transmit the data to the edge cloud platform and perform AI inference based on received container images or model files. The edge cloud platform processes the data according to preset cloud-edge collaboration rules and transmits it to the central cloud platform via a northbound security gateway, deploying the container images or model files to edge nodes. The central cloud platform trains AI algorithm models based on the production data, packages the models into container images or files, and transmits them to the edge cloud platform via a southbound security gateway. This system differs from the existing three-layer "device-edge-cloud" deployment architecture by establishing a four-layer "device-edge-edge-central cloud" deployment architecture for the heterogeneous network environment of oil and gas field enterprises. By collecting and processing production data from the oil and gas field and securely and efficiently uploading it to the cloud for AI algorithm training, it enables intelligent equipment monitoring and fault prediction. By leveraging the collaboration between edge cloud platforms and central cloud platforms, real-time data processing, intelligent inference, centralized management, and monitoring can be achieved. This supports localized data processing and real-time response, enabling real-time secure sharing of production data across regions and organizations at all levels within the group company / oil company from the edge oil and gas production private network to the central cloud platform. It also facilitates intelligent collaborative scheduling of algorithm models and business applications, thereby improving the flexibility and intelligent collaborative efficiency of various business applications in the oil and gas production IoT.

[0044] The aforementioned terminal device 11 is used to collect production data from oil and gas field operations and upload the production data to the edge node 12.

[0045] Specifically, the terminal equipment can be various sensors, instruments, cameras, robots, and other "end" side equipment in the well stations and warehouses of oil production plants / operation areas, which can realize real-time acquisition of production data, multi-protocol conversion, and command control of oil and gas field operations based on the OT (Operational Technology) technology accumulated in the oil and gas smart Internet of Things.

[0046] When this system is in operation, terminal devices can execute work tasks and collect production data according to the unified scheduling of the edge cloud platform, and transmit the collected production data and information during the operation to the edge nodes. The information during the operation includes, but is not limited to, equipment status information, operation logs, and environmental monitoring data.

[0047] In one specific embodiment, the terminal device is various sensors, and the edge node is an edge smart gateway. Production data from these sensors can be collected in real time via protocols such as Modbus / OPC-UA and uploaded to the edge smart gateway. This production data includes temperature from temperature sensors, casing pressure / oil pressure from pressure sensors, flow rate from flow sensors, voltage / current / power from electrical parameter measuring instruments, displacement / load / stroke / stroke rate from dynamometers, and real-time video stream data from on-site cameras collected via the RTSP (Real-Time Streaming Protocol).

[0048] The aforementioned edge node 12 is used to upload production data to the edge cloud platform 13; and to perform AI inference and execution based on the received container images or model files. Applications can also be deployed on edge node 12.

[0049] Specifically, edge nodes can be edge-side devices such as intelligent gateways, intelligent RTUs (Remote Terminal Units), PLCs (Programmable Logic Controllers), and edge all-in-one machines on the production network side of oilfields / operation areas. These devices employ a modular board design and software products that support various hardware, including heterogeneous hardware such as x86 and ARM, and are compatible with various artificial intelligence and big data inference computing frameworks and resources. In addition to production data uploading, they can also provide intelligent analysis and inference, node data management, production data storage, data edge governance, model application deployment, and model application reception functions, thereby simplifying oil and gas equipment access and remote deployment and maintenance, and enabling on-site data processing and analysis.

[0050] In one specific embodiment, a functional example diagram of the edge node is shown below. Figure 2As shown, it includes seven functional modules: production data upload, intelligent analysis and inference, node data management, production data storage, data edge governance, model application deployment, and model application reception. Edge data upload includes data encryption, data compression, data transmission, breakpoint resumption, data forwarding configuration (MQTT / Modbus protocols, etc.), and data synchronization mechanisms. Intelligent analysis and inference includes edge model inference, event alarm notification, real-time inference preview, video and image analysis, equipment fault early warning, and real-time equipment monitoring. Node data management includes node data source, data mapping management, metadata management, business data management, data caching (regular backup / cleanup), and data bus (data query / sharing). Production data storage includes relational data storage, file data storage, object data storage, time-series data storage, compressed data storage, and offline data storage (full / incremental coverage); data edge governance includes data cleaning, data filtering, data augmentation, data quality control, function computation, and offline autonomy; model application deployment includes data acquisition application deployment (intelligent sensing, protocol parsing, etc.), control application deployment (device control, firmware upgrades, etc.), model application deployment (machine vision models, operational condition diagnostic models, etc.), and application update rollback (application start / stop, application update, application rollback, etc.); model application reception includes application reception, application retrieval, application verification, application synchronization, application configuration, and version management.

[0051] Figure 2 The edge nodes also showcase the service runtime resources, namely the hardware and software configuration of the edge node, including neural networks (TensorFlow, Caffe, etc.), containerized runtime environments (Docker, multiple runtimes), real-time stream analysis (Flink, Spark, etc.), basic operating systems (x86, ARM, etc.), model execution (Python, R, etc.), and computing resources (NPU, GPU, etc.).

[0052] In this embodiment, edge nodes are established near the terminal devices of the production network of the oilfield / operation area, thereby extending cloud computing capabilities to the edge nodes near the terminal devices. This solves problems such as network latency, network bandwidth, and data security caused by cloud computing, processes production data collected by terminal devices in real time, reduces the latency of data transmission to the cloud, and improves the response speed of data processing.

[0053] The aforementioned edge cloud platform 13, located on the production network side of the oilfield / operation area enterprise, is used to preprocess production data based on preset cloud-edge collaboration rules, and transmit the processed production data to the central cloud platform via a northbound security gateway; and to deploy received container images or model files to the corresponding edge nodes. The structure of the edge cloud platform can be as follows: Figure 3As shown, the system includes a data acquisition module 131, a data storage module 132, an operation monitoring module 133, an edge-cloud-edge collaboration module 134, an edge mirror repository module 135, and a northbound security gateway 136. The following provides a detailed description of the data acquisition module 131, data storage module 132, operation monitoring module 133, edge-cloud-edge collaboration module 134, edge mirror repository module 135, and northbound security gateway 136:

[0054] The data acquisition module 131 is used to preprocess the production data of the edge node 12 and upload the processed production data to the central cloud platform 14 through the northbound security gateway based on the preset cloud-edge collaboration rules.

[0055] Specifically, this can involve uniformly preprocessing the production data uploaded from each edge node, including data cleaning, deduplication, protocol parsing, and data format conversion. Then, the data acquisition module uploads the processed production data to the central cloud platform according to preset cloud-edge collaboration rules. These preset cloud-edge collaboration rules are set based on the system's needs and are intelligently controlled by both the edge and central cloud-edge collaboration modules. Specifically, the preset cloud-edge collaboration rules can include real-time upload, scheduled upload, change-based upload, and event-triggered upload.

[0056] The data storage module 132 is used to store the processed production data obtained by the data acquisition module 131.

[0057] The operation monitoring module 133 is used to acquire the operation status and production data of the edge cloud platform, edge nodes and terminal devices in real time, and trigger alarms when the operation status or data is abnormal.

[0058] Specifically, the operation monitoring module can be used to monitor the operational status of the edge cloud platform, edge nodes, and terminal devices in real time, collecting various key operational indicators such as equipment load, network bandwidth, storage capacity, processing speed, and production data indicators such as oil pressure, casing pressure, temperature, liquid level, and electrical parameters. When any abnormal operational status of any device on the production network side (including terminal devices, edge nodes, and edge cloud platform) of the oilfield / operation area is detected (such as equipment failure, resource overload, network interruption, production parameter exceeding limits, or abnormal production indicators), the operation monitoring module will immediately trigger an alarm. The alarm information will be transmitted to the edge-cloud-edge collaboration module, which can then intelligently adjust cloud-edge collaboration rules and strategies based on the alarm information. It can also notify maintenance personnel to take timely measures in response to the alarm information. The operation monitoring module helps ensure stable system operation, reduce downtime, improve production efficiency, and enhance overall performance and production processes through data analysis and optimization.

[0059] The edge-cloud-edge collaboration module 134 is used to adjust the preset cloud-edge collaboration rules based on the global information of the acquired edge nodes and the central cloud platform; and to deploy the received container images or model files to the corresponding edge nodes.

[0060] Specifically, the edge-cloud-edge collaboration module is responsible for comprehensive management and monitoring of edge nodes, ensuring collaborative operation between the cloud and the edge. Its functions involve multiple aspects, mainly including node management and monitoring, data and resource scheduling, automatic adjustment of cloud-edge collaboration rules, and deployment of container images or models.

[0061] The edge-cloud-edge collaboration module is responsible for managing edge nodes, connecting them to the edge cloud platform, and ensuring their proper functioning. By monitoring the status of edge nodes in real time, including device operation, resource utilization, and health, the module can promptly detect anomalies and respond accordingly. This helps the system quickly locate problems and perform remote maintenance, reducing the impact of device failures.

[0062] The edge-cloud-edge collaboration module is also responsible for distributing container images or model files from the edge image repository module to the corresponding edge nodes, enabling these containerized applications or model files to perform local inference tasks on the edge nodes, improving response speed and decision-making efficiency. Through an automated deployment process, the edge-cloud-edge collaboration module ensures that edge nodes always run the latest version of models and applications, supporting the real-time needs of edge computing.

[0063] Meanwhile, the edge-cloud-edge collaboration module enables unified scheduling of data, models, applications, resources, and services. Regarding data, the module determines which data needs to be processed at the edge nodes and which needs to be uploaded to the central cloud platform based on the computing power and real-time requirements of each endpoint. The module also schedules and manages the execution of algorithm models, deploying appropriate container images or model files to suitable edge nodes for local inference and data processing. In this way, the system optimizes the allocation of computing resources, improves data processing efficiency, and reduces transmission latency.

[0064] Furthermore, the edge-cloud-edge collaboration module automatically adjusts the preset cloud-edge collaboration rules based on acquired global information (such as edge node load and data transmission status). When resource constraints or network bandwidth limitations are detected on certain edge nodes, the edge-cloud-edge collaboration module automatically adjusts the preset cloud-edge collaboration rules, thereby adjusting data flow or task allocation to ensure the efficient operation of the system and the rational utilization of resources. For example, on edge nodes with high loads, the system can transfer some tasks to other edge nodes to avoid overloading a single node.

[0065] In one specific embodiment, the edge-cloud-edge collaboration module utilizes the KubeEdge framework to combine oil and gas production IoT technology with cloud-native edge computing technology, serving as the core for edge computing and real-time inference operations at oil and gas work sites. A functional example diagram of the edge-cloud-edge collaboration module is shown below. Figure 4 As shown, it includes six functional modules: application model management, edge node management, edge operation monitoring, device control management, edge security authentication, and edge data access. Application model management includes application / model uploading, application / model deloading, application / model subscription, application / model distribution, application / model deployment, application service orchestration, model file distribution, application image repository, and image version synchronization. Edge node management includes automatic node registration, node configuration update, node data collection configuration, node application configuration, node model configuration, node operation monitoring, node upgrade logs, node group management, and batch node management. Edge operation monitoring includes... Point status monitoring, resource scheduling monitoring, operation and maintenance log monitoring, application service monitoring, model call monitoring, and data transmission monitoring; equipment control and management includes oil and gas physical models, equipment asset management, edge gateway management, data acquisition and configuration distribution, alarm event linkage, combined control strategies, operation status reporting, equipment alarm reporting, data acquisition reporting, remote equipment start and stop, production parameter adjustment, and firmware OTA upgrade; edge security authentication includes identity authentication, node security, data security, application security, model security, and operation and maintenance security; edge data access includes device access, data acquisition, data transmission, data parsing, data calculation, and data caching.

[0066] Edge Image Repository Module 135 is used to store container images or model files.

[0067] The Northbound Security Gateway 136 is used to transmit data from the edge cloud platform to the central cloud platform via Kafka. The data transmission process employs encrypted and decrypted messages, device token authentication, and data hierarchical security policies.

[0068] Specifically, the edge cloud platform can transmit production data to the central cloud platform through a northbound security gateway. Through one-way physical isolation technology or logical control mechanisms, the northbound security gateway can prevent the reverse flow of data, prevent network intrusion, and ensure that production data is only uploaded to the central cloud platform in the predetermined direction.

[0069] In one specific embodiment, the northbound security gateway can employ encryption protocols such as SSL / TLS to encrypt data during transmission, ensuring that data is protected from interception or tampering during transmission. Simultaneously, production data is streamed via Kafka message queues, with standardized message formats, including fields such as terminal device ID (TenantID), data value, data point variable (Tag), and timestamp, to guarantee data standardization and traceability.

[0070] Furthermore, the northbound security gateway supports defining hierarchical data security policies, setting different transmission priorities based on device type, business needs, or data importance. For critical data, priority can be given to transmission with stronger encryption strategies to ensure that critical production data can be uploaded to the central cloud platform in a timely and secure manner.

[0071] Before data is uploaded, the northbound security gateway also supports master data mapping, which establishes a correspondence between the data in the data lake module of the terminal device and the data in the central cloud platform, thereby achieving standardized connection between the production data of the edge cloud platform and the production data in the central cloud platform.

[0072] The aforementioned central cloud platform 14 is used to train AI algorithms based on massive amounts of production data, obtain algorithm models, and package the algorithm models into container images or model files, which are then transmitted to the edge cloud platform via a southbound security gateway. The structure of the central cloud platform can be as follows: Figure 5 As shown, it includes a data lake module 141, a cloud computing module 142, a central model repository module 143, a central cloud-edge collaboration module 144, and a southbound security gateway 145. The following provides a detailed description of the data lake module 141, cloud computing module 142, central image repository module 143, central cloud-edge collaboration module 144, and southbound security gateway 145:

[0073] The data lake module 141 is used to store downsampled production data uploaded by the edge cloud platform 13.

[0074] The cloud computing module 142 is used to train AI algorithms based on production data, obtain algorithm models, and package the algorithm models into container images or model files. The cloud computing module 142 includes an AI training module 1421 and a model deployment module 1422.

[0075] AI training module 1421 is used to train AI algorithms based on production data 1421 to obtain algorithm models.

[0076] Specifically, the AI ​​training module can be used to train AI algorithm models based on production data. The AI ​​training module extracts large amounts of production data from the data lake module according to business needs and preprocesses the production data, including cleaning, normalization, and feature extraction, to ensure data quality and consistency. Subsequently, the AI ​​training module builds models using deep learning frameworks (such as TensorFlow or PyTorch) and trains the models through iterative optimization algorithms (such as gradient descent). During training, the AI ​​training module supports distributed computing to accelerate the processing of large-scale data. After training is complete, the AI ​​training module validates the model, evaluates its accuracy and adaptability, and performs hyperparameter tuning or structural optimization based on the validation results to ensure that the model meets actual business needs, resulting in a well-trained algorithm model.

[0077] The model deployment module 1422 is used to package the algorithm model into a container image or model file.

[0078] Specifically, the model deployment module first combines the dependencies and configuration files required for the algorithm model to run, and then uses container tools (such as Docker) to build a container image. Using the Flask framework, it encapsulates the trained algorithm model into a model service API interface, and then exports the model file. This ensures that the algorithm model can run on edge nodes or edge cloud platforms via the container image or model file. Simultaneously, the resulting container image or model file is version-controlled, with change logs recorded, supporting algorithm model updates, rollbacks, and multi-environment compatibility. Subsequently, the model deployment module uploads the container image or model file to the central image repository module, providing support for the central cloud-edge collaboration module to distribute it to edge nodes.

[0079] The central image repository module 143 is used to store container images or model files.

[0080] The central cloud-edge collaboration module 144 is used to adjust the cloud-edge collaboration strategy based on the information obtained from the central cloud platform and the edge cloud platform; and to distribute container images or model files to the corresponding edge cloud platform; wherein, the cloud-edge collaboration strategy includes data flow strategy, computing resource allocation strategy and model / application update strategy.

[0081] Specifically, the central cloud-edge collaboration module can connect various edge cloud platforms to the central cloud platform, ensuring that each edge cloud platform can operate stably and effectively execute assigned tasks. Simultaneously, by continuously monitoring the status of the edge cloud platforms, such as overall performance, resource usage, and health, the central cloud-edge collaboration module can quickly identify potential problems and trigger corresponding early warnings or automatic repair measures, thereby ensuring the high availability and reliability of the system.

[0082] The central cloud-edge collaboration module collects runtime data from all connected edge cloud platforms, including but not limited to load status, network conditions, and data processing volume. This information provides a basis for subsequent strategy adjustments, enabling the entire system to respond flexibly to actual conditions and maintain optimal performance.

[0083] In the dynamic adjustment of the cloud-edge collaboration strategy, the central cloud-edge collaboration module intelligently optimizes the following aspects of the strategy based on the collected global information:

[0084] Data flow strategy: Based on the real-time needs and processing capabilities of each edge cloud platform, determine which data should be processed on the central cloud platform and which can be sent to the edge cloud platforms for processing. This helps reduce unnecessary data transmission and improve data processing efficiency.

[0085] Computing resource allocation strategy: Rationally plan the distribution of computing resources to ensure that each edge cloud platform receives sufficient computing power to support its task execution. When resource constraints are detected on certain platforms, the central cloud-edge collaboration module will reallocate tasks to avoid single-point overload.

[0086] Model / application update strategy: Regularly check and push the latest versions of container images, model files or applications to the corresponding edge cloud platform to ensure that these resources are always up-to-date to cope with ever-changing workloads and technical requirements.

[0087] Meanwhile, the central cloud-edge collaboration module is responsible for securely transferring container images or model files from the central image repository module to the edge image repository module of the designated edge cloud platform, so that the edge image repository module can then distribute or deploy them.

[0088] In one specific embodiment, the central cloud-edge collaboration module utilizes the KubeEdge framework to combine oil and gas production IoT technology with cloud-native edge computing technology, serving as the core of the group company / oil company's enterprise office network-side computing and real-time inference business. A functional example diagram of the central cloud-edge collaboration module is shown below. Figure 4 As shown, it includes six functional modules: application model management, edge node management, edge operation monitoring, device control management, edge security authentication, and edge data access. These functional modules have already been explained in detail above and will not be repeated here.

[0089] The Southbound Security Gateway 145 is used to transmit data from the central cloud platform to the corresponding edge cloud platform using DMZ isolation and encrypted transmission technologies.

[0090] Specifically, the southbound security gateway can use SSL / TLS encryption protocols to encrypt data, ensuring the confidentiality and integrity of the data during transmission.

[0091] Then, based on DMZ isolation technology, container images, model files, and other data in the central cloud platform are transferred to the DMZ zone of the corresponding edge cloud platform for decryption. After decryption, the container images and model files are stored in the edge image repository module.

[0092] Furthermore, the southbound security gateway integrates multi-tenant IAM authentication technology and role-based access control technology. Through identity authentication, it ensures that only authorized users or platforms can access the data resources of the central cloud platform. For example, different user roles have different access permissions; ordinary users can only view the synchronization status, while administrators can operate the synchronization and updates of models or applications, further ensuring the system's security and controllability.

[0093] To facilitate understanding of this solution by those skilled in the art, the following provides a clearer and more complete description of the cloud-edge-device intelligent collaborative system for cross-network secure transmission provided in the embodiments of this application: (Refer to...) Figure 6 The diagram shown illustrates the core modules of this system and their data transmission directions. The diagram comprises four parts: terminal, edge, edge cloud, and central cloud, corresponding to the terminal device 11, edge node 12, edge cloud platform 13, and central cloud platform 14 mentioned above, respectively. Terminal devices can be instruments, sensors, cameras, robots, etc. Edge nodes can be edge gateways, intelligent RTUs, PLCs, edge all-in-one machines, etc. The edge cloud platform includes a data acquisition module, a data storage module, an operation monitoring module, an edge cloud-edge collaboration module, an edge mirror repository module, and a northbound security gateway. The central cloud platform includes a data lake module, a cloud computing module, a central model repository module, a central mirror repository module, a central cloud-edge collaboration module, and a southbound security gateway. The cloud computing module includes an AI training module and a model deployment module.

[0094] To facilitate understanding of this solution by those skilled in the art, the business process of this system is described in a clear and complete sequence based on a specific embodiment: (Refer to...) Figure 7 As shown in the diagram, it comprises four parts: terminal, edge, edge cloud, and central cloud, corresponding to the terminal device 11, edge node 12, edge cloud platform 13, and central cloud platform 14 mentioned above, respectively. The red numbers ①-⑧ in the diagram represent the execution steps of the business process, and each step will be explained below:

[0095] The first step involves collecting production data from various terminal devices on the production site. The edge smart gateway uses protocols such as Modbus / OPC-UA to collect production data from various sensors in real time, as well as real-time video stream data from on-site cameras collected via the RTSP streaming media protocol. This data is then uploaded to the edge cloud platform via the MQTT / Modbus protocol, which is the smart Internet of Things.

[0096] The second step involves the edge cloud platform's intelligent IoT transmitting production data on demand to the intelligent IoT sharing service component in the central cloud platform via Kafka time-series data streams and a northbound security gateway. This component supports forwarding the aggregated data to other business systems via various methods such as MQTT and Kafka, and provides online configuration and management of data flow rules. Users can select specific device data for targeted forwarding as needed. The intelligent IoT sharing service component is part of the edge-cloud-edge collaboration module mentioned above.

[0097] Before production data is pushed to the data lake module, master data mapping is required. This involves establishing a mapping relationship between the collected oil and gas production equipment time-series data and the core business entities of the group's main lake or regional lake, ensuring that the data is consistent with business objectives.

[0098] Specifically, on-demand transmission can be achieved by setting the data inflow range and push frequency according to actual business needs, selecting the corresponding data source (such as object model or device) and Kafka transformation method, and then improving the specific parameter configuration according to the selected data flow method, including data transmission path, format conversion requirements, etc. After verifying that the configuration is correct, confirm the completion of rule addition and activation to achieve on-demand data push.

[0099] The third step involves the central cloud platform's intelligent IoT sharing service component standardizing the production data and uploading it to the central cloud platform's data lake module via a northbound security gateway. The transmitted data is streamed using a standardized Kafka message queue, as illustrated in the following example:

[0100] "data":[

[0101] {

[0102] "TenantID":"xx", / / Terminal device code

[0103] "Value":"xxx", / / Data value

[0104] "Tag":"xxx", / / Format (UID_METRIC), data point variable

[0105] "TimeStamp":"xxx" / / Timestamp

[0106] } ]

[0108] Among them, TenantID is used to identify the terminal device to which the data belongs; Value represents the collected data value; Tag defines the type or format of the data point variable (such as UID_METRIC); and TimeStamp records the timestamp of the data to ensure the temporal consistency and traceability of the data.

[0109] From steps four to seven, the central cloud platform completes the training of the AI ​​algorithm. After obtaining the trained algorithm model, there are three ways to deploy the model to the edge nodes:

[0110] The first approach, based on the pre-trained algorithm model source files, firstly encapsulates and deploys the model online through the central mirror repository module, and encapsulates it into an algorithm model service API interface based on the Flask framework. This supports online debugging and verification of model accuracy and input / output parameters. After confirming that there are no errors, the model files are directly exported. Then, the model files are synchronously synchronized to the edge cloud platform through the southbound security gateway. Next, the files are distributed to edge nodes through the edge cloud-edge collaboration module. Finally, the start, stop, scheduling, and monitoring of the model service are achieved by executing Python script commands on the edge nodes.

[0111] The second approach, based on the pre-trained algorithm model source files, firstly, encapsulates and deploys the model online through the central image repository module. It then encapsulates the model as an algorithm model service API interface based on the Flask framework, supporting online debugging and verification of model accuracy and input / output parameters. After confirmation, a one-click package is created as a model file. Next, based on the corresponding model service files (such as environment configuration) and the Dockerfile configuration file, the model file is packaged into a container image. Staff then download the container image from the edge cloud platform, which is then synchronized to the edge cloud via the southbound industrial control security gateway. The container image is then distributed to edge nodes through the edge cloud edge collaboration module. Finally, by executing Docker runtime scripts on the edge nodes, the start / stop scheduling and monitoring of the model service are achieved.

[0112] The third approach, based on the trained algorithm model source files and a lightweight central image model repository, firstly involves downloading or exporting the algorithm model source files to the edge cloud platform. Then, the central image model repository is manually packaged using Dockerfile to generate a container image. Next, the algorithm model source files and the central image model repository are synchronized to the edge cloud platform via a southbound security gateway. Finally, the edge cloud-edge collaboration module distributes the algorithm model source files and the container images from the central image model repository to the edge nodes. Finally, the edge model repository is started, stopped, scheduled, and monitored by executing Docker scripts on the edge nodes. The algorithm model source files are then loaded onto the edge nodes and encapsulated into an algorithm model service API interface using the Flask framework, completing the online service encapsulation and deployment on the edge side.

[0113] Step 8: After obtaining real-time production data, the edge nodes call the API interface of the model file to implement edge inference early warning.

[0114] Taking the intelligent single-well management edge application at the edge node as an example, it requires algorithm models such as operating condition diagnosis and dynamometer card trend analysis. After deploying and distributing the corresponding algorithm models to the edge node through the edge-cloud-edge collaboration module, enterprise users can access and view the real-time data intelligent monitoring and inference results of the intelligent single-well management application through the edge-cloud-edge collaboration module of the edge cloud platform, and can also view abnormal alarm data and online fault handling.

[0115] To facilitate understanding of this solution by those skilled in the art, another specific embodiment is provided here to fully illustrate the architecture of this system: Refer to Figure 8As shown in the diagram, the office network at the top refers to the enterprise office network and central cloud platform of the group company / oil company mentioned above, while the production network at the bottom refers to the production network of the oilfield / operation area mentioned above. The central cloud platform on the office network side adopts a centralized deployment method, while the edge cloud platform, edge nodes, and terminal devices on the production network side adopt a distributed and independent deployment method. The central cloud platform can achieve real-time acquisition and sharing of production data across networks, utilize historical data collected from oil and gas field production sites for AI training, and generate online or offline model images. This satisfies the centralized monitoring and unified scheduling of node resources / applications / models / data of each production network on the enterprise office network side. Each production network side can deploy the edge cloud-edge collaboration module and N edge nodes of the edge cloud platform as needed. These nodes are first registered and managed in the edge cloud-edge collaboration module, and then synchronized to the office network in real-time through cross-network data sharing, thereby satisfying the enterprise-level node management and resource scheduling needs of the group company / oil company. By unifying the scheduling of computing resources, batch upgrading of firmware, and comprehensive operational diagnostics of terminal equipment at oil and gas field operation sites, the problem of the overall performance of edge devices being unable to meet the increasing data computing and storage needs as the amount of collected data continues to increase is solved, thus meeting the needs of unattended operation and remote, efficient maintenance at the operation site.

[0116] Compared with existing technologies, the key processes and beneficial effects of this invention are as follows:

[0117] First, this system differs from the existing three-layer "end-edge-cloud" deployment architecture. Instead, it establishes a four-layer "end-edge-edge cloud-central cloud" deployment architecture for the heterogeneous network environment of oil and gas field enterprises. This addresses the problem of failing to meet the increasingly complex needs for cross-regional and cross-organizational data sharing and business collaboration. The central cloud platform's central cloud-edge collaboration module centrally manages and monitors the edge cloud platforms of various oil and gas field enterprises in a multi-tenant mode. This module enables unified management and scheduling of data from multiple oil and gas field enterprises, supporting cross-regional and cross-organizational data sharing. The edge cloud-edge collaboration module of the edge cloud platform centrally manages and monitors the edge nodes of individual oil and gas field enterprises, enabling centralized management and monitoring of internal data within each enterprise, supporting localized data processing and real-time response. This four-layer architecture enables real-time secure sharing of production data across regions and organizations at all levels within the group company / oil company, and intelligent collaborative scheduling of algorithm models and business applications between the oil and gas production private network and the cloud platform. This improves the flexibility and intelligent collaborative efficiency of various business applications in the oil and gas production IoT.

[0118] Secondly, the security measures of this system differ from existing solutions. Traditional secure transmission solutions between the oil and gas production IoT and the cloud platform only support one-way transmission from the production network to the office network and do not support reverse transmission. Furthermore, they lack an effective two-way authentication mechanism, failing to ensure the integrity and security of data during transmission. This system proposes deploying two secure gateways with different transmission directions and encryption strategies between the edge cloud platform and the central cloud platform. Each gateway should ensure that data can only be transmitted in the designated direction to prevent reverse intrusion. First, a high-speed secure channel for production data is established through the northbound secure gateway. Data from different devices is categorized and classified, and corresponding encryption transmission strategies are adopted. This enables the automatic push of real-time data collected from oil and gas field operations to the data lake module of the central cloud platform, ensuring the timeliness and security of cross-network data transmission. Then, using the southbound secure gateway, relevant data from the central cloud platform is transmitted via encryption strategies to the DMZ zone of the edge cloud platform for decryption. Simultaneously, models / applications are synchronized to the edge image repository module of the edge cloud platform in the form of container images or model files, ensuring the secure and controllable exchange of various types of information across networks.

[0119] Finally, the cloud-edge-device intelligent collaborative operating environment involved in this system differs from the semi-automatic processing method in most existing solutions, which involves manually packaging and generating container images and deploying them to edge devices via KubeEdge. This invention addresses the multi-source, heterogeneous edge devices in oil and gas field operations by building a lightweight cloud-edge-device collaborative operating environment compatible with both ARM and Linux architectures, ensuring efficient operation even on resource-constrained edge devices. Simultaneously, based on KubeEdge+Docker container technology, it standardizes the container image building process and file format to achieve an automatic packaging and update deployment mechanism for edge model / application container image systems, replacing the existing manual packaging method and improving on-site deployment and maintenance efficiency.

[0120] In this embodiment, the method constructs a cloud-edge-device intelligent collaborative system for cross-network secure transmission of information between the Internet of Things (IoT) and cloud platforms in oil and gas production. Based on the cross-network collaborative sharing needs of production and office networks in oil and gas field enterprises, a cloud-edge-device intelligent collaborative operating environment based on a cloud platform architecture is built. This system deeply integrates the IoT for oil and gas production with the cloud-native edge computing framework KubeEdge, and utilizes edge devices with higher computing power. It establishes a mapping relationship between these devices and edge nodes, and then uses these edge devices as instances for unified management. Following this, it sequentially performs secure access and device modeling, real-time data acquisition and edge autonomy, hierarchical storage of time-series data, edge application orchestration and resource scheduling, and firmware installation and upgrades. The system features lightweight design, high reliability, low latency, and high throughput, effectively improving the real-time performance of production data acquisition and operation monitoring at oil and gas operation sites, as well as the efficiency of on-site maintenance.

[0121] Meanwhile, this method adopts a hybrid cloud deployment and one-way encrypted transmission to simplify the production data synchronization and feedback process, improve data transmission efficiency and stability, and establish a secure high-speed channel for production data, enabling the automatic push of real-time production data collected by terminal devices to the data lake of the central cloud platform. Furthermore, the cloud computing center conducts AI algorithm training based on the historical production data of various equipment gathered in the data lake, generates model services, and packages them into container images for one-click deployment to edge nodes to achieve offline edge deployment. Finally, the oil and gas operation site directly transmits the real-time collected data to call the edge model service to achieve local real-time inference calculation.

[0122] This method integrates multiple technologies such as precise data acquisition and control, secure transmission links, edge-distributed intelligence, and cross-network cloud-edge collaboration to improve the level of intelligent applications in enterprise production operation monitoring, fault diagnosis and early warning, production decision optimization, and safety command and control. It can meet the needs of unmanned or minimally manned operation sites in oil and gas operations, and help enterprises reduce costs and increase efficiency.

[0123] Example 2

[0124] Based on the same inventive concept, embodiments of the present invention also provide a cloud-edge-device intelligent collaboration method for cross-network secure transmission, referring to... Figure 9 As shown, the method includes:

[0125] S101: Collect production data from various terminal devices at the oil and gas field operation site and upload the production data to the edge node;

[0126] S102: The edge node uploads the production data to the edge cloud platform;

[0127] S103: The production data is preprocessed based on preset cloud-edge collaboration rules, and the processed production data is transmitted to the central cloud platform through the northbound security gateway.

[0128] S104: The central cloud platform trains AI algorithms based on the production data to obtain an algorithm model, and packages the algorithm model into a container image or model file, and transmits it to the edge cloud platform through the southbound security gateway;

[0129] S105: The edge cloud platform will deploy the received container image or model file to the corresponding edge node;

[0130] S106: The edge node performs AI inference and execution based on the received container image or model file.

[0131] Example 3

[0132] Based on the same inventive concept, embodiments of the present invention also provide a computer-readable storage medium storing a computer program / instruction thereon, which, when executed by a processor, implements the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described in Embodiment 2 above.

[0133] Example 4

[0134] Based on the same inventive concept, this embodiment of the invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described in Embodiment 2 above.

[0135] Example 5

[0136] Based on the same inventive concept, this embodiment of the invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory. When the processor executes the computer program, it implements the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described in Embodiment 2 above.

[0137] 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 and optical storage) containing computer-usable program code.

[0138] 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.

[0139] 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.

[0140] 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.

[0141] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A cloud-edge-device intelligent collaborative system for secure cross-network transmission, characterized in that, This includes terminal devices, edge nodes, edge cloud platforms, and central cloud platforms: The terminal device is used to collect production data from oil and gas field operations and upload the production data to the edge node. The edge node is used to upload the production data to the edge cloud platform; and to perform AI inference and execution based on the received container image or model file; The edge cloud platform is used to preprocess the production data based on preset cloud-edge collaboration rules, and transmit the processed production data to the central cloud platform through a northbound security gateway. And, deploy the received container image or model file to the corresponding edge node; The central cloud platform is used to train AI algorithms based on the production data, obtain algorithm models, and package the algorithm models into container images or model files, which are then transmitted to the edge cloud platform through a southbound security gateway.

2. The system as described in claim 1, characterized in that, The edge cloud platform includes: a data acquisition module, an edge cloud-edge collaboration module, an edge mirror warehouse module, and a northbound security gateway; The data acquisition module is used to preprocess the production data of the edge nodes and upload the processed production data to the central cloud platform through the northbound security gateway based on preset cloud-edge collaboration rules; the preprocessing includes parsing, transformation and cleaning. The edge-cloud-edge collaboration module is used to adjust the preset cloud-edge collaboration rules based on the global information of the edge nodes and the central cloud platform; and to deploy the received container images or model files to the corresponding edge nodes. The edge image repository module is used to store the container image or model file; The northbound security gateway is used to enable one-way encrypted transmission of data from the edge cloud platform to the central cloud platform.

3. The system as described in claim 2, characterized in that, The edge cloud platform also includes: a data storage module and an operation monitoring module; The data storage module is used to store the processed production data obtained by the data acquisition module; The operation monitoring module is used to obtain the operating status of the edge cloud platform, the edge node and the terminal device in real time, and trigger an alarm when the operating status is abnormal.

4. The system as described in claim 2 or 3, characterized in that, The northbound security gateway is specifically used to transmit data from the edge cloud platform to the central cloud platform via Kafka, using encrypted transmission, device authentication, and data classification strategies.

5. The system as described in claim 1, characterized in that, The central cloud platform includes: a data lake module, a cloud computing module, a central image warehouse module, a central cloud-edge collaboration module, and a southbound security gateway; The data lake module is used to store production data uploaded by the edge cloud platform; The cloud computing module is used to train AI algorithms based on the production data, obtain an algorithm model, and package the algorithm model into a container image or model file. The central image repository module is used to store the container image or model file; The central cloud-edge collaboration module is used to adjust the cloud-edge collaboration strategy based on the information obtained from the central cloud platform and the edge cloud platform; and to distribute the container image or model file to the corresponding edge cloud platform; wherein, the cloud-edge collaboration strategy includes a data flow strategy, a computing resource allocation strategy, and a model / application update strategy. The southbound security gateway is used to enable directional encrypted transmission of data from the central cloud platform to the edge cloud platform.

6. The system as described in claim 5, characterized in that, The cloud computing module includes: an AI training module and a model deployment module; The AI ​​training module is used to train AI algorithms based on the production data to obtain an algorithm model; The model deployment module is used to package the algorithm model into a container image or a model file.

7. The system as described in claim 5 or 6, characterized in that, The southbound security gateway is specifically used to transmit data from the central cloud platform to the corresponding edge cloud platform using DMZ isolation technology and encrypted transmission technology.

8. A cloud-edge-device intelligent collaboration method for secure cross-network transmission, characterized in that, include: Collect production data from various terminal devices at the oil and gas field operation site and upload the production data to the edge node; The edge nodes upload the production data to the edge cloud platform; The production data is preprocessed based on preset cloud-edge collaboration rules, and the processed production data is transmitted to the central cloud platform through the northbound security gateway. The central cloud platform trains AI algorithms based on the production data to obtain an algorithm model, and packages the algorithm model into a container image or model file, which is then transmitted to the edge cloud platform through the southbound security gateway. The edge cloud platform will deploy the received container image or model file to the corresponding edge node; The edge nodes perform AI inference and execution based on the received container images or model files.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instruction is executed by the processor, it implements the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described in claim 8.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described in claim 8.

11. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the cloud-edge-device intelligent collaboration method for cross-network secure transmission as described in claim 8.

Images