Safety production emergency side cloud cooperation system based on industrial internet

Abstract

The invention relates to a safety production emergency side cloud collaboration system based on an industrial internet. The system comprises a plurality of cloud sub-platforms and an emergency toolbox. The emergency toolbox comprises an emergency toolbox on the side and an emergency toolbox on the end side. The data acquisition module is specifically used for acquiring an operation total environment data set through access of accessory sensor equipment in an emergency tool box at the end side and a production field sensor; the side emergency toolbox performs data preprocessing on the job total environment data set according to the received knowledge downloaded by the side cloud collaborative management to obtain a to-be-trained data set; and uploading the information to a cloud end through a side cloud cooperative communication protocol, so that the learning model optimizes iterative training, and outputting an emergency management scheme. And the cloud sub-platform configures related items of a side emergency tool box corresponding to the cloud sub-platform according to the emergency management scheme, the related items are issued to the side through side-cloud collaborative management, and the side downloads the related items from the mirror image warehouse and operates the related items. According to the method, the emergency response capability of safety production in an industrial internet environment can be improved.

Description

Technical Field

[0001] The present invention relates to the field of industrial Internet technology, and in particular to an industrial Internet-based safe production emergency edge-cloud collaborative system. Background Art

[0002] With the development of the industrial Internet, the requirements for safe production are increasing. Traditional emergency management systems are difficult to cope with complex and changing production environments and massive data processing needs. In recent years, the overall situation of production safety in my country has continued to improve, but various types of accident hazards and safety risks are intertwined and superimposed, and are prone to occur frequently. The industrial Internet is an effective way to improve safe production. Promoting the integration of the industrial Internet and safe production is a key measure to improve the inherent safety level of industrial production. The rapid development of the industrial Internet platform has laid a technical foundation for the intelligentization of safe production. The technology combining edge computing and cloud computing has gradually become a research hotspot. Combined with intelligent sensing, edge computing, artificial intelligence and other technologies, it realizes the collection and analysis of all-factor data of safe production, the rapid perception and real-time monitoring of safety incidents, the use of safe production industrial models to predict risk situations, and the use of risk assessment models to comprehensively analyze and evaluate production situations and safety accidents.

[0003] However, the traditional emergency management system mainly relies on manual judgment and disposal, adopts microservice architecture for model deployment and optimization, and uses traditional distributed computing for data processing, which cannot effectively respond to safety hazards in production. At the same time, data privacy protection, poor data flow efficiency caused by unstable communication networks between edge and cloud, and edge and cloud collaboration are all issues worth studying. Summary of the invention

[0004] Based on this, it is necessary to provide an industrial Internet-based safe production emergency edge-cloud collaborative system that can improve the efficiency and safety of emergency response to address the above-mentioned technical problems.

[0005] A safe production emergency edge-cloud collaborative system based on the industrial Internet, the system is built based on cloud native technology, and models and applications are deployed from the cloud to the edge in a mirrored manner, including: several cloud sub-platforms and emergency toolboxes. The emergency toolboxes include: emergency toolboxes on the edge and emergency toolboxes on the terminal.

[0006] The emergency toolbox on the edge is used for uplink data transmission from the edge to the cloud. The steps are as follows: obtain the full-environment data set of the operation through the connection between the accessory sensor equipment in the emergency toolbox on the edge and the production site sensors. After the emergency toolbox on the edge performs data preprocessing on the full-environment data set of the operation according to the knowledge transmitted by the edge-cloud collaborative management, it calculates the processed data locally to obtain the data set to be trained. The data set to be trained is uploaded to the cloud through the edge-cloud collaborative communication protocol so that the learning model on the cloud is optimized and iterated, and the data set to be trained is trained according to the optimized learning model to output the emergency management plan.

[0007] The emergency toolbox on the end side is used for the downlink transmission of knowledge from the cloud to the edge side. The steps are as follows: each cloud sub-platform configures the relevant items of the edge emergency toolbox corresponding to the cloud sub-platform according to the emergency management plan, and the relevant items are sent to the edge side through edge-cloud collaborative management. The edge emergency toolbox on the edge side is downloaded from the image repository so that the end side can run directly. Among them, the relevant items include: configuration parameters, applications, and operation models.

[0008] In one embodiment, the cloud sub-platform includes: an Internet of Things platform, a big data platform, a deep learning platform, a safe production industrial model management engine, and a one-stop safe production application development platform. The emergency toolbox deployed on the end side includes environmental perception devices, traffic analyzers, asset scanners, and edge gateways for each production operation link to support the collection and analysis of data transmitted in the industrial network. The emergency toolbox deployed on the edge side includes: an adaptation module, an edge computing unit, and a status tool. Among them, the adaptation module includes: multi-operating system adaptation, multi-architecture adaptation, and multi-chip adaptation. The edge computing unit includes: a data acquisition module, a data synchronization module, a data computing module, and an AI inference module. The status tool is used to determine the working mode according to the service mode of the edge side, and to retrieve the data and computing resources of the industrial Internet cloud platform according to the working status. The working modes include: stand-alone mode, cluster mode, container mode, and process mode.

[0009] A safety production emergency edge-cloud collaboration method based on industrial Internet, the method comprising:

[0010] The steps of uplink data transmission from the edge to the cloud are as follows: obtain the full-environment data set of the operation through the connection between the accessory sensor device in the emergency toolbox on the edge and the production site sensor. The emergency toolbox on the edge performs data preprocessing on the full-environment data set of the operation according to the knowledge transmitted by the edge-cloud collaborative management in the cloud, and then calculates the processed data locally to obtain the data set to be trained. The data set to be trained is uploaded to the cloud through the edge-cloud collaborative communication protocol, so that the learning model on the cloud is optimized and iterated, and the data set to be trained is trained according to the optimized learning model to output the emergency management plan.

[0011] The steps for transmitting knowledge downward from the cloud to the edge are as follows: each cloud sub-platform configures the relevant items of the emergency toolbox on the edge corresponding to the cloud sub-platform according to the emergency management plan. The relevant items are sent to the edge through edge-cloud collaborative management. The edge emergency toolbox on the edge is downloaded from the image repository so that the edge can run directly.

[0012] The above-mentioned industrial Internet-based safety production emergency edge-cloud collaborative system, first of all, significantly improves the communication efficiency through the edge-cloud collaborative management architecture. The emergency toolbox on the edge can collect the full environment data set of the work site in real time through the built-in sensor equipment, and perform data preprocessing locally. This local computing capability not only reduces the bandwidth pressure required to upload the original data to the cloud, but also efficiently transmits the training data set through the edge-cloud collaborative communication protocol, making data transmission faster and more stable. Secondly, the system deploys models and applications through cloud native technology and mirroring, effectively enhancing the collaborative ability of models and applications. In the cloud, the learning model is continuously iterated and optimized, and the optimized model is sent to the edge emergency toolbox through the mirror warehouse. This mode makes model updates flexible and fast, ensuring that the edge devices always run the latest emergency management plan, thereby greatly improving the accuracy and timeliness of emergency response. In addition, the system has been designed and optimized in terms of data privacy protection. Through data preprocessing and local computing of edge devices, sensitive data does not need to be uploaded directly to the cloud. This method reduces the risk of data leakage while achieving efficient data transmission and enhances data privacy protection capabilities. Finally, the system achieves the dynamic adjustment of knowledge downlink transmission and parameter configuration from cloud to edge through edge-cloud collaboration, so that each emergency scenario can obtain targeted optimization solutions. This mechanism significantly improves the response speed of emergency events, ensuring that in emergency situations, edge-side devices and terminal-side devices can quickly and accurately execute the best emergency plan, thereby effectively ensuring industrial production safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a structural block diagram of a safety production emergency edge-cloud collaborative system based on the industrial Internet in an embodiment;

[0014] Figure 2 A flowchart of industrial safety production data, models, and application iterative optimization steps in one embodiment;

[0015] Figure 3 A schematic diagram of the technical architecture of an emergency toolbox in one embodiment;

[0016] Figure 4 A schematic diagram of an emergency tool box product in one embodiment;

[0017] Figure 5 A network topology diagram of an emergency toolbox in an embodiment;

[0018] Figure 6 A schematic diagram of a big data platform system architecture in one embodiment;

[0019] Figure 7 A schematic diagram of a deep learning platform system architecture in one embodiment;

[0020] Figure 8 A schematic diagram of the architecture of a safe production industrial model management engine in one embodiment;

[0021] Fig. 9 A schematic diagram of a system architecture for developing and building a safe production application in one embodiment;

[0022] Fig.10 A flowchart of an edge-cloud collaboration method for emergency production safety based on the industrial Internet in one embodiment DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0024] In one embodiment, Figure 1 As shown, a safety production emergency edge-cloud collaborative system based on the industrial Internet is provided. The system is built based on cloud native technology and deploys models and applications from the cloud to the edge in a mirrored manner, including: several cloud sub-platforms and emergency toolboxes. The emergency toolboxes include: an emergency toolbox on the edge and an emergency toolbox on the terminal.

[0025] The emergency toolbox on the edge is used for uplink data transmission from the edge to the cloud. The steps are as follows: obtain the full-environment data set of the operation through the connection between the accessory sensor equipment in the emergency toolbox on the edge and the production site sensors. After the emergency toolbox on the edge performs data preprocessing on the full-environment data set of the operation according to the knowledge transmitted by the edge-cloud collaborative management, it calculates the processed data locally to obtain the data set to be trained. The data set to be trained is uploaded to the cloud through the edge-cloud collaborative communication protocol so that the learning model on the cloud is optimized and iterated, and the data set to be trained is trained according to the optimized learning model to output the emergency management plan.

[0026] The emergency toolbox on the end side is used for the downlink transmission of knowledge from the cloud to the edge side. The steps are as follows: each cloud sub-platform configures the relevant items of the edge emergency toolbox corresponding to the cloud sub-platform according to the emergency management plan, and the relevant items are sent to the edge side through edge-cloud collaborative management. The edge emergency toolbox on the edge side is downloaded from the image repository so that the end side can run directly. Among them, the relevant items include: configuration parameters, applications, and operation models.

[0027] In the above-mentioned industrial Internet-based safety production emergency edge-cloud collaborative system, first of all, the system significantly improves the communication efficiency through the edge-cloud collaborative management architecture. The emergency toolbox on the edge can collect the full environment data set of the work site in real time through the built-in sensor equipment, and perform data preprocessing locally. This local computing capability not only reduces the bandwidth pressure required to upload the original data to the cloud, but also efficiently transmits the training data set through the edge-cloud collaborative communication protocol, making the data transmission faster and more stable. Secondly, the system deploys models and applications through cloud native technology and mirroring, which effectively enhances the collaborative ability of models and applications. In the cloud, the learning model is continuously iterated and optimized, and the optimized model is sent to the edge emergency toolbox through the mirror warehouse. This mode makes model updates flexible and fast, ensuring that the edge devices always run the latest emergency management plan, thereby greatly improving the accuracy and timeliness of emergency response. In addition, the system has been designed and optimized in terms of data privacy protection. Through data preprocessing and local computing of edge devices, sensitive data does not need to be uploaded directly to the cloud. This method reduces the risk of data leakage while achieving efficient data transmission and enhances data privacy protection capabilities. Finally, the system achieves the dynamic adjustment of knowledge downlink transmission and parameter configuration from cloud to edge through edge-cloud collaboration, so that each emergency scenario can obtain targeted optimization solutions. This mechanism significantly improves the response speed of emergency events, ensuring that in emergency situations, edge-side devices and terminal-side devices can quickly and accurately execute the best emergency plan, thereby effectively ensuring industrial production safety.

[0028] In one embodiment, the cloud sub-platform includes: an Internet of Things platform, a big data platform, a deep learning platform, a safe production industrial model management engine, and a one-stop safe production application development platform. The emergency toolbox deployed on the end side includes environmental perception devices, traffic analyzers, asset scanners, and edge gateways for each production operation link to support the collection and analysis of data transmitted in the industrial network. The emergency toolbox deployed on the edge side includes: an adaptation module, an edge computing unit, and a status tool. Among them, the adaptation module includes: multi-operating system adaptation, multi-architecture adaptation, and multi-chip adaptation. The edge computing unit includes: a data acquisition module, a data synchronization module, a data computing module, and an AI inference module. The status tool is used to determine the working mode according to the service mode of the edge side, and to retrieve the data and computing resources of the industrial Internet cloud platform according to the working status. The working modes include: stand-alone mode, cluster mode, container mode, and process mode.

[0029] In one of the embodiments, the Internet of Things platform is an integrated platform for integrating device access, device management, data security communication, message subscription, message forwarding and data services. The big data platform is used to process the heterogeneous data and real-time data uploaded to the cloud by the emergency toolbox on the edge. The deep learning platform is used to unify the data format of the operation data collected by the application toolbox and build a shared database. According to the shared database, the terminal monitoring AI device is used to perform risk analysis and generate an early warning plan. The safe production industrial model management engine is used to use the industrial model stored in the cloud to comprehensively process the full-environment data set of the operation uploaded to the cloud by the edge test, and to orchestrate the deployment and operation of the industrial model through the container orchestration tool. A one-stop safe production application development platform is used to support the development, testing and deployment of safe production applications in the emergency toolbox on the edge through the full-process development and experimental testing platform on the cloud.

[0030] In one of the embodiments, the cloud sub-platform and the emergency toolbox, as well as each cloud sub-platform, are deployed through the edge-cloud collaborative management module. Edge-cloud collaborative management includes: node management, model management, and application management. Node management is used to build an edge computing framework by deploying edge models and applications, and to manage the emergency toolbox on the edge and collaborate with the cloud. Application management is used to deploy safe production applications and operating environments for the emergency toolbox on the edge, and to manage the life cycle of multiple safe production applications on edge nodes.

[0031] In one of the embodiments, the emergency toolbox on the side is also used to synchronize real-time information and generate emergency decisions with the cloud system through digital twin technology.

[0032] In one of the embodiments, edge-cloud collaborative management: functions include node management, model management, application management, etc. (Edge) Node Management: refers to the management of edge devices such as emergency toolboxes, which build edge computing capabilities by deploying edge models and applications, processing data accessed by terminal devices, and collaborating with the cloud. (Edge) Model Management: Mainly includes model distribution, model deployment, model optimization, model configuration, model update, etc. (Edge) Application Management: The edge-side emergency toolbox provides an application deployment and operating environment, and manages and schedules the life cycle of multiple applications on this node; the cloud side provides an application development and testing environment, as well as application life cycle management capabilities, and distributes applications to the emergency toolbox.

[0033] In one embodiment, if Figure 2 As shown, an industrial safety production data, model, and application iterative optimization step is provided, which specifically includes the following contents:

[0034] 1. Data collaboration: The cloud-side cloud collaboration platform accepts relevant data uploaded by the edge-side emergency toolbox, and provides storage, analysis, and value mining of massive data. It supports the controllable and orderly flow of data between the edge and the cloud, forming a complete data flow path, and efficiently and low-costly manage the life cycle and mine the value of data. The edge node is responsible for data collection, pre-processing and simple analysis of the original data according to the model or business rules, and then uploads the results and related data to the cloud; the cloud can store, analyze, and mine the value of massive data. Data collaboration between the edge and the cloud enables data to flow in an orderly manner between the edge and the cloud, thus forming a complete data flow path, which is convenient for subsequent data life cycle management and value mining. Data collaboration mainly includes three aspects:

[0035] 1) Edge processing and cloud analysis: The edge device data is collected by the terminal device and uploaded to the edge node. The edge node can filter out a large amount of redundant or invalid data through data preprocessing, screen out key data related to the business, and upload it to the cloud as needed, which can effectively reduce the consumption of network bandwidth, storage resources and computing resources. The cloud further analyzes the key data and completes the complex safety color production business scenario requirements through the intelligent applications deployed in the cloud.

[0036] 2) Privacy computing and efficient collaboration: Device data is generated at the edge. Due to the particularity of edge data, data collaboration needs to consider three aspects: first, data security and user privacy; second, data can be persisted at the edge and obtained on demand in the cloud. Third, it is because of the network problem in the environment where the edge device is located. When the edge cloud network jitters or fails, the data sender should cache the sent data and continue to send it after the network is restored. In order to solve the problem of efficient data transmission, KCP protocol communication is used. Based on the congestion control algorithm-Fast Congestion Control (FCC), by monitoring changes in network status, the sending rate is intelligently adjusted to ensure that a low latency level can be maintained even when the network is congested. At the same time, KCP also uses a prediction algorithm based on time difference to estimate the network status in the future, prepare for response in advance, and further improve the response speed of the system. The clever combination of these algorithms enables KCP to always maintain efficient and stable performance in a complex and changing network environment.

[0037] 3) Data aggregation: Usually the cloud is connected to many edge nodes, and different edge nodes may be in different business scenarios. For example, the enterprise safety production central control system is connected to a large number of cameras, and the same scene has multiple cameras at different angles. The cloud obtains edge data and aggregates it in the cloud, which can realize cross-regional and cross-system multi-dimensional spatiotemporal data fusion and collaborative analysis, and realize comprehensive safety production management and control.

[0038] 2. Model collaboration: mainly includes model delivery, model deployment, model optimization, model configuration, model update, etc. The safety production industrial model is trained and optimized and iterated in the cloud, and then delivered to the edge for deployment. The reasoning process of the safety production industrial model is placed at the edge close to the terminal device to meet the high computing and low latency requirements of edge industrial intelligent deep learning. At the same time, the IoT terminal devices collected by the edge continuously generate data that needs to be analyzed in real time using AI technologies such as deep learning or used to train deep learning models. Through data collaboration, it is uploaded to the cloud to achieve the optimization and iteration of the cloud industrial intelligent model. The safety production model has a wide range of application value in terms of quality inspection, target positioning, safety hazard detection and identification, etc. The use of the model requires processes such as data preprocessing, hyperparameter configuration, model training and evaluation, and multi-terminal model deployment, and finally realizes the deployment and application of the industrial model on the edge. On the edge side, the industrial model training tool will provide users with an efficient and convenient visual operation interface, complete localized one-stop model development, and quickly deploy and implement business online.

[0039] 3. After the model development is completed, build the image and push it to the image warehouse, use the model deployment function of the edge-cloud collaborative system to complete the automated deployment configuration and realize automated edge-side deployment.

[0040] 4. Application collaboration: The emergency toolbox on the edge side provides application deployment and operation environment, and manages and schedules the life cycle of multiple applications on this node; the cloud side provides application development and testing environment, as well as application life cycle management capabilities, and sends applications to the emergency toolbox. Application collaboration is the unified management of business applications by the cloud, deploying applications on nodes according to the needs of edge-cloud business scenarios, and upgrading business applications through edge-cloud operations. For example, when deploying face recognition applications on edge nodes, face image information is collected by the terminal and uploaded to the edge node for local processing, including preprocessing, rule matching, and artificial intelligence analysis; the collected face information is marked with edge node and other tag information and uploaded to the cloud, where the face recognition model is inferred and optimized, and the optimized model is packaged into an image and uploaded to the image warehouse. The edge node can complete the update of the local application by pulling the image to achieve a closed data loop. The ultimate goal of edge-cloud business collaboration is to optimize the deployment of business applications and achieve the best distribution of computing. In combination with specific business characteristics (such as data security, latency requirements, etc.), business processing tasks are deployed to edge nodes on demand according to user request distribution, terminal device data characteristics, edge node's own computing load and storage capacity, and edge node requests are accurately dispatched to the corresponding edge nodes, improving resource utilization and service experience, and achieving the best balance between computing efficiency, user experience and data security.

[0041] It is worth mentioning that this solution realizes edge-cloud collaboration and integrated management, solves the shortcomings of existing technologies, can effectively improve the emergency response capabilities of safe production in the industrial Internet environment, and has high innovation and practical value.

[0042] In one of the embodiments, the emergency toolbox: the bottom layer of the toolbox adopts a distributed microservice architecture to support dynamic business expansion and horizontal expansion. The core services of the toolbox are composed of modules such as behavior analysis, IoT monitoring, intelligent early warning, video aggregation, asset identification, target responsibility, risk management, hidden danger investigation, education and training, rules and regulations, laws and regulations, emergency management, accidents, and accident cases. The hardware layer comprehensively collects the working environment data by applying a variety of sensor devices and network traffic collection equipment, and configures the edge gateway to achieve data aggregation and function expansion, support the retrieval of data and computing resources of the industrial Internet cloud platform, and realize the powerful service capabilities of the toolbox on the edge side. The toolbox contains 10 types of working environment sensing devices such as portable temperature sensors, humidity sensors, pressure detectors, and liquid level detectors. At the same time, the built-in flow analyzer, asset scanner, and edge gateway support the collection and analysis of sensor data and flow data in the industrial network. In addition, considering the working site environment, it is equipped with a variety of practical devices such as mobile power supplies, emergency lighting, and micro scanners. The technical architecture is as follows Figure 3 , product form such as Figure 4 , Network topology diagram as follows Figure 5 .

[0043] In one embodiment, the big data platform, the system architecture is as follows Figure 6 Big data analysis refers to the process of integrating, analyzing, interpreting, and presenting large-scale, heterogeneous (including unstructured and structured) data to extract valuable information and knowledge. In practice, big data analysis usually uses important technologies such as computer science, statistics, machine learning, and data mining to process and analyze massive amounts of data.

[0044] In one embodiment, the deep learning platform, the system architecture is as follows Figure 7 :The deep learning platform is based on the real-time data collection and on-site video monitoring of the safety production emergency toolbox, unifies and standardizes the data format, builds a shared database, and realizes real-time query, trend query and threshold warning functions of important operating parameters of production equipment and units; and through AI intelligent analysis of on-site monitoring videos, it can realize real-time warning of the operating behavior of front-line employees and danger identification of the surrounding environment.

[0045] In one embodiment, the industrial model management platform, system architecture Figure 8:The Industrial Model Management Engine for Safe Production is an open operating system that supports the intelligent retrieval and precise calling of industrial models for safe production, targeting the needs of precipitation, dissemination, reuse and value creation of industrial knowledge such as industrial technology, experience and principles. It uses DevOps, microservices, containerization and other technologies to achieve model precipitation, platform development and autonomous breakthroughs of industrial technology, and empowers enterprises and industrial Internet platforms through networking. The orchestration of model services is achieved through container orchestration tools, which improves model deployment efficiency and reduces model operation and maintenance costs. Through the issuance of edge collaborative decision-making models, problems such as high latency, low efficiency and high cost when using cloud models are solved, and edge-side data is used to learn, train, optimize and upgrade cloud models.

[0046] In one embodiment, a one-stop safety production application development platform, system architecture Fig. 9 :The safe production application development and construction platform provides an industry low-code development platform and BI large-screen visual design tools to help users quickly build applications, and achieve continuous integration and deployment of cloud-native applications through the full-process development and experimental testing platform on the cloud. The microservice governance platform provides unified governance of the platform's microservices. The safe production applications developed based on this platform include but are not limited to safety risk perception management, major hazard source management, etc., and support the provision of safe production application services for group companies, large and medium-sized enterprises, and industrial parks.

[0047] In one embodiment, Fig.10 As shown, a safety production emergency edge-cloud collaboration method based on the industrial Internet is provided. The specific steps are as follows:

[0048] Step 1002, the data uplink transmission step from the edge to the cloud is: through the connection between the accessory sensor device in the emergency toolbox on the edge and the production site sensor, the full operation environment data set is obtained. After the emergency toolbox on the edge performs data preprocessing on the full operation environment data set according to the knowledge transmitted by the edge-cloud collaborative management in the cloud, the processed data is calculated locally to obtain the data set to be trained. The data set to be trained is uploaded to the cloud through the edge-cloud collaborative communication protocol, so that the learning model on the cloud is optimized and iterated, and the data set to be trained is trained according to the optimized learning model to output the emergency management plan.

[0049] Step 1004, the steps for transmitting knowledge downward from the cloud to the edge are as follows: each cloud sub-platform configures the relevant items of the emergency toolbox on the edge corresponding to the cloud sub-platform according to the emergency management plan, and the relevant items are sent to the edge through edge-cloud collaborative management. The edge emergency toolbox on the edge is downloaded from the image repository so that the edge can run directly.

[0050] It should be understood that although Figure 2 , Fig.10The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0051] For the specific limitations of the industrial Internet-based production safety emergency edge-cloud collaboration system, please refer to the limitations of the industrial Internet-based production safety emergency edge-cloud collaboration method above, which will not be repeated here. Each module in the above-mentioned industrial Internet-based production safety emergency edge-cloud collaboration system can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0052] Those skilled in the art will understand that Figure 1 , 3 -The structure shown in FIG. 9 is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0053] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0054] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0055] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A safety production emergency edge-cloud collaborative system based on the industrial Internet, characterized in that: The system is built based on cloud native technology and deploys models and applications from the cloud to the edge in a mirrored manner, including: several cloud sub-platforms and emergency toolboxes; the emergency toolboxes include: emergency toolboxes on the edge and emergency toolboxes on the terminal; The emergency toolbox on the edge is used for uplink transmission of data from the edge to the cloud in the following steps: obtaining a full-environment data set of the operation through access to the accessory sensor device in the emergency toolbox on the end and the production site sensor; the emergency toolbox on the edge performs data preprocessing on the full-environment data set of the operation according to the knowledge received from the edge-cloud collaborative management, and then locally calculates the processed data to obtain a data set to be trained; uploading the data set to be trained to the cloud through the edge-cloud collaborative communication protocol, so that the learning model on the cloud is optimized and iterated, and the data set to be trained is trained according to the optimized learning model, and an emergency management plan is output; The emergency toolbox on the terminal side is used for the steps of downlink transmission of knowledge from the cloud to the edge side: each of the cloud sub-platforms configures the relevant items of the edge emergency toolbox corresponding to the cloud sub-platform according to the emergency management plan, and the relevant items are sent to the edge side through the edge-cloud collaborative management, and the edge emergency toolbox on the edge side is downloaded from the image repository so that the terminal side can run directly; wherein, the relevant items include: configuration parameters, applications, and operation models.

2. The system according to claim 1, characterized in that The cloud sub-platform includes: Internet of Things platform, big data platform, deep learning platform, safe production industrial model management engine and one-stop safe production application development platform; The emergency toolbox deployed on the end side includes environmental sensing devices, flow analyzers, asset scanners, and edge gateways for each production operation link to support the collection and analysis of data transmitted in the industrial network; The emergency toolbox deployed on the edge includes: an adaptation module, an edge computing unit and a status tool; wherein the adaptation module includes: multi-operating system adaptation, multi-architecture adaptation and multi-chip adaptation; the edge computing unit includes: a data acquisition module, a data synchronization module, a data computing module and an AI inference module; the status tool is used to determine the working mode according to the service mode of the edge, and to retrieve the data and computing resources of the industrial Internet cloud platform according to the working status; The working modes include: stand-alone mode, cluster mode, container mode and process mode.

3. The system according to claim 1, characterized in that The IoT platform is an integrated platform for integrating device access, device management, data security communication, message subscription, message forwarding and data services; The big data platform is used to process the heterogeneous data and real-time data uploaded to the cloud by the emergency toolbox on the side; The deep learning platform is used to unify the data format of the operation data collected by the application toolbox and build a shared database. According to the shared database, the terminal monitors the AI ​​device to perform risk analysis and generate an early warning plan; The safe production industrial model management engine is used to use the industrial model stored in the cloud to comprehensively process the full-environment data set of the operation uploaded to the cloud by the edge test, and to orchestrate the deployment and operation of the industrial model through the container orchestration tool; The one-stop safe production application development platform is used to support the development, testing and deployment of safe production applications in the emergency toolbox on the side through the cloud-based full-process development and experimental testing platform.

4. The system according to claim 3, characterized in that The cloud sub-platform and the emergency toolbox, as well as each of the cloud sub-platforms, are deployed through an edge-cloud collaborative management module; The edge-cloud collaborative management includes: node management, model management and application management; The node management is used to build an edge computing framework by deploying edge models and applications, and to manage the emergency toolbox on the edge and collaborate with the cloud; The application management is used to deploy safe production applications and operating environments to the emergency toolbox on the edge, and to manage the life cycles of multiple safe production applications on the edge nodes.

5. The system according to claim 3, characterized in that The emergency toolbox on the side is also used to synchronize real-time information and generate emergency decisions with the cloud system through digital twin technology.

6. A safety production emergency edge-cloud collaboration method based on the industrial Internet, characterized in that: A system as claimed in any one of claims 1 to 5, comprising: The steps of uplink data transmission from the edge to the cloud are as follows: the whole operation environment data set is obtained by connecting the accessory sensor device in the emergency toolbox on the edge side to the production site sensor; the emergency toolbox on the edge side performs data preprocessing on the whole operation environment data set according to the knowledge transmitted by the edge-cloud collaborative management in the cloud, and then calculates the processed data locally to obtain the data set to be trained; the data set to be trained is uploaded to the cloud through the edge-cloud collaborative communication protocol, so that the learning model on the cloud is optimized and iterated, and the data set to be trained is trained according to the optimized learning model, and an emergency management plan is output; The steps for transmitting knowledge downward from the cloud to the edge are as follows: each of the cloud sub-platforms configures the relevant items of the emergency toolbox on the edge corresponding to the cloud sub-platform according to the emergency management plan, and the relevant items are sent to the edge through edge-cloud collaborative management. The edge emergency toolbox on the edge is downloaded from the image repository so that the edge can run directly.

Images