Safety production intelligent supervision platform based on cloud edge collaboration and management method

Abstract

The invention discloses a safety production intelligent supervision platform and management method based on cloud edge collaboration, and relates to the technical field of industrial safety production, the platform comprises edge side equipment, an enterprise side management platform and a group side management platform, the edge side equipment is arranged in each production site and is connected with the enterprise side management platform, and the group side management platform is connected with the enterprise side management platform. An algorithm processing task is arranged through the group side management platform and is issued to the enterprise side management platform to process and analyze data collected by the edge side equipment, if an alarm condition is found, the alarm condition is automatically uploaded to the group side management platform through the cloud, and the group side management platform carries out secondary research and judgment to form an alarm closed loop. According to the invention, the problems of data dispersion, insufficient data coordination and sharing, insufficient application support and the like in the mode of the existing supervision system can be overcome, and real-time monitoring, risk early warning and emergency response to the production process are realized.

Description

Technical Field

[0001] The present invention relates to the field of industrial production safety technology, and more specifically to an intelligent production safety supervision platform and management method based on cloud-edge collaboration. Background Art

[0002] The explosives production workshop is a typical high-risk place in the military industry, with the characteristics of flammable and explosive hazards. Safety monitoring in the production process is an essential key link in the management and control of the explosives production workshop. The explosives production workshop has dangerous behaviors and illegal operations such as overstaffing in sensitive areas, key personnel leaving their posts, wearing work clothes and hats in violation of regulations, and workers dragging product boxes to cause fires. The safety liability accidents caused by this will not only cause serious losses to the state-owned economy, but also be accompanied by a large number of casualties, causing extremely adverse effects on society.

[0003] To ensure the production safety of explosives workshops, the traditional management method is to install surveillance cameras and use 24-hour manual monitoring to monitor dangers. However, this method is not only difficult to respond immediately, but also consumes a lot of manpower costs and has poor results. Driven by the Internet of Things and artificial intelligence technologies, the field of safe production is constantly developing towards digitalization and intelligence. For fire-related and dangerous production scenarios, intelligent supervision technology has been widely used in the field of safe production. By combining the video surveillance system with artificial intelligence algorithms, such as target detection and tracking, behavior recognition and other AI algorithms, the supervision system can accurately determine the number of people in high-risk places on the production site, and conduct real-time monitoring and online alarms for potential dangerous behaviors to ensure the safety of people and property during the production process.

[0004] The current mainstream intelligent supervision platform architecture relies on decentralized data collection points and centralized data processing methods, that is, edge nodes located at the production site capture video streams through monitoring equipment. At the same time, these nodes transmit the acquired video data to the cloud center, which is responsible for performing more complex data processing tasks, such as deep learning model training and large-scale data analysis.

[0005] However, the existing analysis and processing mode has the following shortcomings: First, the data is scattered in different systems and devices, resulting in data islands, which creates obstacles to information sharing and makes it difficult to share data across systems and devices, limiting the efficiency of data utilization and the possibility of collaborative work; second, due to the lack of effective data integration and analysis methods, the dispersion of data makes it difficult to ensure data consistency, and massive amounts of data cause data transmission and computing pressure on cloud centers; in addition, the existing system also has deficiencies in application support, making it difficult to adapt to the complex and changeable needs of fire-related industrial environments, such as real-time feedback of analysis results, cross-device collaborative operations, etc., making it impossible for the results of cloud analysis to be fed back to the production site in a timely manner, making it difficult to respond quickly to alarm situations, etc. These are key issues that restrict the improvement of the efficiency of the intelligent production safety supervision platform.

[0006] Therefore, how to achieve the coordination and sharing of dispersed data and improve the response speed of data analysis is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0007] In view of this, the present invention provides an intelligent safety production supervision platform architecture based on cloud-edge collaboration, which overcomes the problems of data dispersion, insufficient data coordination and sharing, and insufficient application support in the existing supervision system model, and combines cloud computing, edge computing, Internet of Things technology, big data analysis, and artificial intelligence and other advanced technologies to achieve real-time monitoring of the production process, risk warning and emergency response, especially for safety production supervision in high-risk industries involving fire.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] A smart production safety supervision platform based on a cloud-edge system, including a group-side management platform, several enterprise-side management platforms and several edge-side devices;

[0010] The enterprise-side management platform performs coding management on edge-side devices and generates master data information;

[0011] The group-side management platform orchestrates algorithm tasks and sends them to the enterprise-side management platform; the enterprise-side management platform collects and reports local organizational structure information, reports local organizational structure information and master data information to the group-side management platform, and deploys algorithm applications according to algorithm tasks, and automatically pulls data from edge-side devices; the enterprise-side management platform runs algorithm applications on the acquired data to perform alarm monitoring, reports alarm information to the group-side management platform when it is found, and stores the alarm information; the group-side management platform judges the effectiveness of the alarm based on the alarm information, generates an alarm work order based on the valid alarm information and sends it to the enterprise-side management platform, and optimizes the algorithm application based on the invalid alarm information.

[0012] Preferably, the edge-side equipment includes video surveillance equipment and voice communication equipment; wherein the video surveillance equipment is composed of a video probe and a network video recorder to form a video surveillance platform, which is used to collect and store images of personnel and equipment in the production site operation area; the voice communication equipment is used to realize real-time voice communication between experts on the group side and production site maintenance personnel on the enterprise side.

[0013] Preferably, the group-side management platform includes a group-side basic information management system, a group-side equipment master data service system, a visualization management system, a group-side AI algorithm management system, a group-side voice communication system and a group-side alarm management system; wherein the group-side basic information management system is used to receive local organizational structure information reported by the enterprise-side management platform; the group-side equipment master data service system is used to digitally archive and manage the video surveillance equipment and voice communication equipment of the edge-side equipment according to the business type, and to perform unified coding, and to provide a service interface so that the group-side management platform can obtain the required master data information through timed synchronization or subscription; the group-side AI algorithm management system is used to push the corresponding algorithm application to the enterprise-side management platform according to the algorithm requirements of the enterprise-side production equipment The AI ​​algorithm management system on the group side monitors the usage of computing resources at the same time, dynamically adjusts the algorithm task allocation, and the AI ​​algorithm management system on the group side can mark the ROI area of ​​the enterprise-side management platform, and synchronize the ROI area marking data of the acquired master data information with the group-side equipment master data service system through the http protocol; the group-side voice communication system is used to realize synchronous and low-latency voice interaction between the group side and the enterprise side; the group-side alarm management system is responsible for the configuration of alarm rules, receiving and processing various alarm information, and issuing alarm processing work orders for valid alarms to the enterprise-side alarm management system of the enterprise-side management platform, and uploading invalid alarm information to the cloud server; the visualization management system is used for visualization, which is convenient for experts to judge the effectiveness of alarm information.

[0014] Preferably, the enterprise-side management platform includes an enterprise-side basic information management system, an enterprise-side equipment master data service system, an enterprise-side AI algorithm management system, an enterprise-side voice communication system, and an enterprise-side alarm management system; wherein the enterprise-side basic information management system is used to receive local organizational structure information reported by enterprise-side users and report it to the group-side basic information management system; the enterprise-side equipment master data service system is used to digitally archive and manage edge-side equipment according to business types, and to perform unified coding, and to provide a service interface so that the enterprise-side management platform can obtain the required master data information through timed synchronization or subscription; the enterprise-side AI algorithm management system is used to receive and deploy the algorithm information issued by the group-side management platform. The method is applied to process the data collected by the edge-side devices and perform AI analysis, and the alarm information generated by the analysis results is fed back to the enterprise-side alarm management system, and the computing resource usage and algorithm operation status are regularly reported to the group-side AI algorithm management system of the group-side management platform for resource scheduling and algorithm optimization; the enterprise-side voice communication system is used to achieve synchronous, low-latency voice interaction between the group side and the enterprise side; the enterprise-side alarm management system is used to upload the alarm information generated by the analysis of the enterprise-side AI algorithm management system to the group-side alarm system of the group-side management platform, and receive the alarm processing work order issued by the group-side alarm management system of the group-side management platform, forming an alarm closed loop after eliminating dangerous hazards.

[0015] Preferably, the basic information management system on the group side and the basic information management system on the enterprise side both include a user management module, an information entry module, an information query module and an information update module; wherein the user management module is used to complete functions such as user account creation, authority allocation and identity authentication; the information entry module is used to input basic information related to production safety into the system, including organizational structure, production process, enterprise location, production line and other basic information; the information query module is used to provide information query function, allowing users to retrieve basic information based on fuzzy query and multi-condition combination; the information update module allows users to modify and update the basic information in the system.

[0016] Preferably, the group-side equipment master data service system and the enterprise-side equipment master data service system both include an equipment master data query service module and an equipment unified coding service module; wherein the equipment master data query service module is used to query and manage all master data information related to edge-side equipment, such as models and specifications of monitoring and voice communication equipment, to achieve full life cycle management of equipment; the equipment unified coding service module is used to assign a unique code to different edge-side devices to facilitate identity identification and information management, and to achieve standardized communication and data exchange between the group-side management platform and the enterprise-side management platform.

[0017] Preferably, the visualization management system includes a data visualization module, a user interaction module and a real-time monitoring module; wherein the data visualization module is used to intuitively display the collected alarm information in the form of charts, dashboards, etc.; the user interaction module is used to provide a user interface, allowing users to interact with the system, such as setting monitoring parameters, receiving alarms, initiating remote call commands, etc.; the real-time monitoring module is used to display video data from the video surveillance device of the edge side device in real time.

[0018] Preferably, the group-side AI algorithm management system includes an algorithm training module, an algorithm management module, an algorithm orchestration module, an algorithm distribution module and an algorithm scheduling module; wherein the algorithm training module is used to train and optimize algorithm applications to improve the accuracy and efficiency of AI analysis; the algorithm management module is used to register, version control, performance evaluate and security audit the algorithm, which ensures the traceability and manageability of the algorithm; the algorithm orchestration module is used for users to select video surveillance equipment for edge-side devices, select the algorithms that need to be supported, edit the time period and running time of the algorithm, implement algorithm task orchestration, obtain algorithm tasks, support users to flexibly configure algorithm operating parameters according to business needs, and improve the flexibility and adaptability of algorithm applications; the algorithm distribution module is used to distribute the trained algorithm application from the group-side management platform to the enterprise-side management platform to facilitate real-time reasoning and decision-making near the data source; the algorithm scheduling module is used to intelligently allocate computing resources to maximize the utilization and load balancing of computing resources. This module dynamically adjusts the algorithm's running priority and resource allocation according to business needs and the computing resource usage of each enterprise-side management platform to ensure the stability and efficiency of the system during high concurrency and large-scale data processing.

[0019] Preferably, the enterprise-side AI algorithm management system includes an algorithm receiving module, an algorithm execution module, and an algorithm analysis module; wherein the algorithm receiving module is used to receive the AI ​​algorithm application issued by the algorithm issuing module of the group-side AI algorithm management system; the algorithm execution module is used to execute the corresponding algorithm application according to the algorithm task received from the algorithm orchestration module, in accordance with the time period and running time specified in the algorithm task; the algorithm analysis module is used to generate alarm information and send the alarm information to the enterprise-side alarm management system during the analysis process of executing the algorithm application if there is an analysis result that meets the alarm situation.

[0020] Preferably, the group-side voice communication system and the enterprise-side voice communication system both include a voice transmission module, a call control module and an emergency broadcast module; wherein the voice transmission module is used to process the collection, encoding, transmission and decoding processes of voice signals, ensuring efficient and stable transmission of voice signals between voice communication devices of edge-side devices connected to the group-side management platform and the enterprise-side management platform, so as to ensure the real-time and clarity of communication; the call control module is used to manage the establishment, maintenance and termination of calls, including processing call requests, releasing communication resources and monitoring communication status; the emergency broadcast module is used to quickly broadcast emergency information to all or specific areas in emergency situations, such as fire, leakage or other safety accidents.

[0021] Preferably, the group-side alarm management system includes an alarm display module, an alarm service module and an alarm data management module; wherein the alarm display module is responsible for receiving and processing alarm information from the enterprise-side alarm management system, and ensuring that alarms can be handled promptly and effectively according to the implementation of preset alarm classifications, priority sorting and response strategies; the alarm service module is used to generate alarm processing work orders for alarm information reported during the inspection process, and provide release and notification services for alarm processing work orders to ensure that relevant personnel on the enterprise side can receive alarm information in a timely manner; the alarm data management module is responsible for storing and managing detailed data of alarm events corresponding to the alarm information, including information such as the time, location, type, and processing status of the alarm.

[0022] Preferably, the enterprise-side alarm management system includes an alarm information transfer module and an alarm processing module; wherein the alarm information transfer module is responsible for receiving alarm information from the enterprise-side AI algorithm management system, and reporting the alarm information to the alarm display module of the group-side alarm management system, so as to realize automatic processing of alarm events; the alarm processing module is used to receive the alarm work order dispatched by the alarm service module of the group-side alarm management system, so as to facilitate relevant personnel on the enterprise side to accurately check and rectify the location area of ​​the alarm according to the information in the alarm work order, thereby forming a complete closed-loop process of alarm processing.

[0023] Preferably, the local organizational structure information includes the affiliated enterprise, production line location, production process, and the addition or deletion of relevant equipment information

[0024] Preferably, the alarm information includes alarm type, alarm location, alarm time, alarm picture, etc.

[0025] Preferably, the group-side management platform and the enterprise-side management platform perform voice communication via the voice communication device of the connected edge-side device.

[0026] A management method for a safe production intelligent supervision platform based on cloud-edge collaboration includes the following steps:

[0027] Step 1: The enterprise-side management platform collects and reports the organizational structure information, and uniformly codes and manages the edge node devices on the enterprise side to generate master data information; after the organizational structure information is approved by the group-side management platform, the organizational structure information and master data information are synchronized to the group-side device master data service system of the group-side management platform;

[0028] Step 2: Through the group-side AI algorithm management system, select the monitoring device on the enterprise side, select the algorithm application that needs to be supported, edit the time period and running time of the application algorithm, bind the edge-side device with the algorithm application, implement algorithm task orchestration, and push the algorithm application and algorithm tasks to the enterprise-side management platform;

[0029] Step 3: The enterprise-side management platform receives the issued algorithm application, implements algorithm deployment and upgrade replacement, and automatically pulls the data collected by the edge-side device according to the algorithm task;

[0030] Step 4: The enterprise-side management platform runs the algorithm application according to the algorithm task. When an alarm message is found, it identifies whether there is an alarm picture in the alarm message. If there is an alarm picture, the enterprise-side management platform saves the alarm picture and generates an alarm picture URL address and inserts it into the alarm message, and reports the alarm information to the group-side management platform. Otherwise, the alarm information is directly reported to the group-side management platform.

[0031] Step 5: Determine the validity of the alarm information through the group-side management platform. If it is a valid alarm, send an alarm processing work order to the enterprise-side alarm system through the group-side alarm system. If it is an invalid warning, transmit the invalid alarm information to the group-side AI algorithm management system for deep learning optimization algorithm application.

[0032] Preferably, in step 1, the enterprise-side user reports the organizational structure information through the enterprise-side basic information management system of the enterprise-side management platform; the group-side management platform and other subsystems of the enterprise-side management platform, such as the alarm management system, voice communication system, and AI algorithm management system, obtain the master data information such as the relevant equipment code required by this subsystem through timed synchronization or subscription based on the service interface provided by the equipment master data service.

[0033] Preferably, the group-side AI algorithm management system of the group-side management platform in step 2 supports the listing of algorithm applications and supports binding the video surveillance equipment of the edge-side devices with supported algorithms; users select the monitoring equipment on the enterprise side, select the algorithm applications that need to be supported, edit the time period and running time of the application algorithm through the group-side AI algorithm management system, and realize algorithm task orchestration.

[0034] Preferably, step 4 also includes the enterprise-side AI algorithm management system of the enterprise-side management platform running the specified algorithm application according to the time period and running time specified in the algorithm task, and when alarm information is found, reporting the alarm information to the group-side alarm management system of the group-side management platform through the MQTT protocol; if there is an alarm picture in the alarm information, the alarm picture is stored in the enterprise-side server, and the alarm picture URL address is provided in the alarm information; after receiving the alarm information, the group-side alarm management system reads the alarm picture according to the alarm picture URL address and saves it in the group-side OSS service.

[0035] Through the above technical solutions, it can be known that compared with the prior art, the present invention discloses a safe production intelligent supervision platform and management method based on cloud-edge collaboration, including edge-side devices, enterprise-side management platforms and group-side management platforms. The edge-side devices are arranged in each production site, connected to the enterprise-side management platform, and the algorithm processing tasks are arranged through the group-side management platform, and sent to the enterprise-side management platform to process and analyze the data collected by the edge-side devices. When the enterprise-side management platform finds alarm information, the alarm information is transmitted to the group-side management platform through the MQTT protocol, and the experts conduct secondary judgment through the group-side management platform. After the effective alarm judgment, the effective alarm is transmitted to the group-side management platform, and then the group-side management platform sends the alarm information to the corresponding enterprise-side management platform, notifying the operator to rectify the alarm situation and eliminate the alarm danger. The algorithm training module that transmits invalid alarms to the group-side management platform optimizes the algorithm application to form an alarm closed loop. This system architecture combines the advantages of cloud computing and edge computing, aiming to achieve full utilization of computing resources, efficient processing and analysis, and real-time supervision feedback for dangerous alarms in fire-related and dangerous production scenarios. Beneficial effects include:

[0036] 1) Data integration. Compared with the current intelligent production safety supervision platform, the present invention uses the edge computing nodes of the enterprise-side platform to preprocess and initially analyze the data, reducing the delay and bandwidth requirements of data transmission; centralized storage and unified coding management of data are achieved through the group-side platform to ensure data consistency and accessibility, so as to obtain the common and common problems of production safety through data analysis methods; solve the problem of data dispersion on multiple production lines, reduce the use of network bandwidth, reduce data transmission costs, and improve the real-time performance of data analysis and processing;

[0037] 2) Propose a coordination and sharing mechanism. The present invention designs a cloud-edge collaborative architecture to achieve dynamic allocation of computing resources and task scheduling, thereby improving the utilization efficiency of computing resources. By establishing standardized data interfaces and communication protocols, the device data between edge nodes can be connected.

[0038] 3) Enhance application support capabilities. The present invention provides effective feedback on the analysis results of the intelligent production safety supervision platform by realizing cloud-edge collaborative alarm management and voice communication; experts determine whether the alarm information is a valid alarm through the production safety visualization platform. If it is a valid alarm, an alarm work order is sent to the enterprise-side alarm system through the group-side alarm system. If it is an invalid alarm, the invalid alarm information is transmitted to the algorithm training module of the group-side AI algorithm management system for deep learning optimization algorithm application, forming a virtuous circle for solving problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0040] Figure 1 A schematic diagram of the structure of an intelligent production safety supervision platform based on cloud-edge collaboration provided by the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The embodiment of the present invention discloses a safe production intelligent supervision platform based on a cloud-edge system, such as Figure 1 As shown, it includes a group-side management platform, several enterprise-side management platforms and several edge-side devices; the enterprise-side management platform performs coding management on the edge-side devices and generates master data information; the group-side management platform orchestrates algorithm tasks and sends them to the enterprise-side management platform; the enterprise-side management platform collects and reports local organizational structure information, reports local organizational structure information and master data information to the group-side management platform, and deploys algorithm applications according to algorithm tasks, and automatically pulls data from edge-side devices; the enterprise-side management platform performs alarm monitoring on the algorithm applications running on the acquired data, reports alarm information to the group-side management platform when it is found, and stores the alarm information; the group-side management platform judges the effectiveness of the alarm based on the alarm information, generates an alarm work order based on the valid alarm information and sends it to the enterprise-side management platform, and optimizes the algorithm application based on the invalid alarm information.

[0043] Edge-side devices include video surveillance equipment and voice communication equipment. The video surveillance equipment consists of a video probe and a network video recorder to form a video surveillance platform, which is used to collect and store images of personnel and equipment in the production site operation area. The voice communication equipment is used to realize real-time voice communication between experts on the group side and production site maintenance personnel on the enterprise side.

[0044] The group-side management platform includes the group-side basic information management system, the group-side equipment master data service system, the visualization management system, the group-side AI algorithm management system, the group-side voice communication system and the group-side alarm management system; among them, the group-side basic information management system is used to receive the local organizational structure information reported by the enterprise-side management platform; the group-side equipment master data service system is used to digitally archive the video surveillance equipment and voice communication equipment of the edge-side equipment according to the business type, and uniformly encode them, and provide a service interface so that the group-side management platform can obtain the required master data information through timed synchronization or subscription; the group-side AI algorithm management system is used to push the corresponding algorithm application to the enterprise-side AI algorithm management system of the enterprise-side management platform according to the algorithm requirements of the enterprise-side production equipment, and monitor The group-side AI algorithm management system can mark the ROI area of ​​the enterprise-side management platform and synchronize the ROI area marking data of the acquired master data information with the group-side equipment master data service system through the http protocol; the group-side voice communication system is used to complete the real-time synchronization and low-latency voice communication between the enterprise-side production site maintenance personnel and the group-side security experts through voice intercom equipment; the group-side alarm management system is responsible for the configuration of alarm rules, receiving and processing various alarm information, and issuing alarm processing work orders for valid alarms to the enterprise-side alarm management system of the enterprise-side management platform, and uploading invalid alarm information to the cloud server; the visualization management system is used for visualization, which is convenient for experts to judge the effectiveness of alarm information.

[0045] The enterprise-side management platform includes the enterprise-side basic information management system, the enterprise-side equipment master data service system, the enterprise-side AI algorithm management system, the enterprise-side voice communication system and the enterprise-side alarm management system; among them, the enterprise-side basic information management system is used to receive the local organizational structure information filled in by the enterprise-side users and report it to the group-side basic information management system; the enterprise-side equipment master data service system is used to digitally archive and manage the edge-side equipment according to the business type, and uniformly encode it, and provide a service interface so that the enterprise-side management platform can obtain the required master data information through timed synchronization or subscription; the enterprise-side AI algorithm management system is used to receive and deploy the algorithm application issued by the group-side management platform. It is used to process the data collected by the edge-side devices and perform AI analysis, and feed back the alarm information generated by the analysis results to the enterprise-side alarm management system, and regularly report the computing resource usage and algorithm operation status to the group-side AI algorithm management system of the group-side management platform for resource scheduling and algorithm optimization; the enterprise-side voice communication system is used to realize synchronous, low-latency voice interaction between the group side and the enterprise side; the enterprise-side alarm management system is used to upload the alarm information generated by the analysis of the enterprise-side AI algorithm management system to the group-side alarm system of the group-side management platform, and receive the alarm processing work order issued by the group-side alarm management system of the group-side management platform, forming an alarm closed loop after eliminating dangerous hazards.

[0046] Furthermore, the group-side management platform adopts cloud computing architecture, and the enterprise-side management platform adopts edge computing architecture. Both cloud computing architecture and edge computing architecture are divided into three layers: infrastructure layer, data layer and platform service layer, and the three layers of cloud computing architecture and edge computing architecture correspond one to one. The infrastructure layer of cloud computing architecture and edge computing architecture realizes resource collaboration, the data layer of cloud computing architecture and edge computing architecture realizes data collaboration, and the platform service layer of cloud computing architecture and edge computing architecture realizes service collaboration. Among them, resource collaboration includes: 1) Computing resource collaboration: providing virtualized computing resources, including servers, storage and network equipment, and users can flexibly configure and expand resources according to needs; 2) Storage resource collaboration; 3) Network resource collaboration: providing network connection services to ensure the security of data transmission and the stability of the system. Data collaboration includes: 1) Data storage and management: providing data storage solutions, including data collection, cleaning, storage and backup, ensuring data integrity and consistency; 2) Data analysis and processing: supporting big data processing and analysis services, including data mining, real-time analysis and predictive modeling, helping users extract valuable information from massive data; 3) Data sharing and exchange: implementing a secure data sharing and exchange mechanism, supporting cross-departmental and cross-enterprise data exchange, and providing data access control and privacy protection functions. Service collaboration includes: 1) Algorithm and model deployment: providing an algorithm and model development, testing, deployment and management environment, supporting rapid iteration and updating of machine learning and deep learning algorithms; 2) Application development and integration: providing the SDK and API required for application development, supporting developers to quickly build and integrate safety production-related applications, such as risk assessment, early warning notifications, etc.; 3) Resource scheduling and optimization: realizing dynamic allocation and management of computing resources, intelligently scheduling edge and cloud computing resources according to task requirements and system load, and optimizing system performance and response speed.

[0047] The group-side management platform is used to coordinate and control the data upload of multiple enterprise-side management platforms, summarize the data and analysis results uploaded by the enterprise-side management platforms, and optimize and update the algorithm applications sent to the enterprise-side AI algorithm management system through deep learning. The analysis results uploaded by the enterprise-side management platform are recorded and displayed to users through the visual management system.

[0048] Furthermore, both the group-side basic information management system and the enterprise-side basic information management system include a user management module, an information entry module, an information query module and an information update module; wherein, the user management module is used to complete functions such as user account creation, authority allocation and identity authentication; the information entry module is used to input basic information related to production safety into the system, including organizational structure, production process, enterprise location, production line and other basic information; the information query module is used to provide information query function, allowing users to retrieve basic information based on fuzzy query and multi-condition combination; the information update module allows users to modify and update the basic information in the system.

[0049] Furthermore, both the group-side equipment master data service system and the enterprise-side equipment master data service system include an equipment master data query service module and an equipment unified coding service module; wherein, the equipment master data query service module is used to query and manage all master data related to edge-side equipment, such as the model and specifications of monitoring and voice communication equipment, to achieve full life cycle management of equipment; the equipment unified coding service module is used to assign a unique code to different edge-side devices to facilitate identity identification and information management, and to achieve standardized communication and data exchange between the group-side management platform and the enterprise-side management platform.

[0050] Furthermore, the visualization management system includes a data visualization module, a user interaction module and a real-time monitoring module; wherein the data visualization module is used to intuitively display the collected alarm information in the form of charts, dashboards, etc.; the user interaction module is used to provide a user interface, allowing users to interact with the system, such as setting monitoring parameters, receiving alarms, initiating remote call commands, etc.; the real-time monitoring module is used to display video data from the video surveillance device of the edge side device in real time.

[0051] Furthermore, the group-side AI algorithm management system includes an algorithm training module, an algorithm management module, an algorithm orchestration module, an algorithm distribution module and an algorithm scheduling module; wherein the algorithm training module is used to train and optimize algorithm applications to improve the accuracy and efficiency of AI analysis; the algorithm management module is used to register, version control, performance evaluate and security audit the algorithm, which ensures the traceability and manageability of the algorithm; the algorithm orchestration module is used for users to select video surveillance devices for edge devices, select the algorithms that need to be supported, edit the time period and running time of the algorithm, implement algorithm task orchestration, obtain algorithm tasks, support users to flexibly configure algorithm operating parameters according to business needs, and improve the flexibility and adaptability of algorithm applications; the algorithm distribution module is used to distribute the trained algorithm application from the group-side management platform to the enterprise-side management platform to facilitate real-time reasoning and decision-making near the data source; the algorithm scheduling module is used to intelligently allocate computing resources to maximize the utilization and load balancing of computing resources. This module dynamically adjusts the algorithm's running priority and resource allocation according to business needs and the computing resource usage of each enterprise-side management platform to ensure the stability and efficiency of the system during high concurrency and large-scale data processing.

[0052] Furthermore, the enterprise-side AI algorithm management system includes an algorithm receiving module, an algorithm execution module, and an algorithm analysis module; wherein the algorithm receiving module is used to receive the AI ​​algorithm application issued by the algorithm issuing module of the group-side AI algorithm management system; the algorithm execution module is used to execute the corresponding algorithm application according to the algorithm task received from the algorithm orchestration module of the group-side AI algorithm management system, in accordance with the time period and running time specified in the algorithm task; the algorithm analysis module is used to generate an alarm message during the AI ​​analysis process of executing the algorithm application, if there is an analysis result that meets the alarm condition, and send the alarm message to the enterprise-side alarm management system.

[0053] Furthermore, the group-side voice communication system and the enterprise-side voice communication system both include a voice transmission module, a call control module and an emergency broadcast module; wherein the voice transmission module is used to process the collection, encoding, transmission and decoding processes of voice signals, ensuring efficient and stable transmission of voice signals between voice communication devices of edge-side devices connected to the group-side management platform and the enterprise-side management platform, so as to ensure the real-time and clarity of communication; the call control module is used to manage the establishment, maintenance and termination of calls, including processing call requests, releasing communication resources and monitoring communication status; the emergency broadcast module is used to quickly broadcast emergency information to all or specific areas in emergency situations, such as fire, leakage or other safety accidents.

[0054] Furthermore, the group-side alarm management system includes an alarm display module, an alarm service module and an alarm data management module; wherein the alarm display module is responsible for receiving and processing alarm information from the enterprise-side alarm management system, and ensuring that alarms can be handled promptly and effectively according to the implementation of preset alarm classifications, priority sorting and response strategies; the alarm service module is used to generate alarm processing work orders for alarm information reported during the inspection process, and provide release and notification services for alarm processing work orders to ensure that relevant personnel on the enterprise side can receive alarm information in a timely manner; the alarm data management module is responsible for storing and managing detailed data of alarm events corresponding to the alarm information, including information such as the time, location, type, and processing status of the alarm.

[0055] Furthermore, the enterprise-side alarm management system includes an alarm information transfer module and an alarm processing module; wherein the alarm information transfer module is responsible for receiving the alarm information from the enterprise-side AI algorithm management system, and reporting the alarm information to the alarm display module of the group-side alarm management system, so as to realize the automatic processing of alarm events; the alarm processing module is used to receive the alarm work order issued by the alarm service module of the group-side alarm management system, so as to facilitate the relevant personnel on the enterprise side to accurately check and rectify the location area of ​​the alarm according to the information in the alarm work order, thereby forming a complete closed-loop process of alarm processing.

[0056] Furthermore, the local organizational structure information includes the affiliated enterprise, production line location, production process, and the addition or deletion of relevant equipment information

[0057] Furthermore, the alarm information includes alarm type, alarm location, alarm time, alarm picture, etc.

[0058] Furthermore, the group-side management platform and the enterprise-side management platform perform voice communication via the voice communication device of the connected edge-side device.

[0059] On the other hand, in a specific embodiment, a management method of a safe production intelligent supervision platform based on cloud-edge collaboration includes the following steps:

[0060] S1: Enterprise users report organizational structure information through the enterprise basic information management system of the enterprise management platform, and perform unified coding management on edge devices to obtain master data information. After approval by the group management platform, it is synchronized to the group equipment master data service system of the group management platform. Other subsystems of the group management platform and the enterprise management platform, such as the alarm management system, voice communication system, and AI algorithm management system, obtain the master data information such as relevant equipment codes required by this subsystem through scheduled synchronization or subscription based on the service interface provided by the equipment master data service.

[0061] S2: The group-side AI algorithm management system of the group-side management platform supports the listing of algorithm applications and supports binding the video surveillance devices of edge-side devices with supported algorithms. Through the group-side AI algorithm management system, users select the video surveillance devices on the edge side, select the algorithm applications that need to be supported, edit the time period and running time of the application algorithm, implement algorithm task scheduling, and push the algorithm application to the enterprise-side AI algorithm management system of the enterprise-side management platform.

[0062] S3: The enterprise-side AI algorithm management system can accept the algorithm application issued by the group-side AI algorithm management system to implement algorithm deployment and upgrade replacement; according to the algorithm tasks received from the group-side AI algorithm management system, it automatically pulls the data of the relevant video streams collected by the edge-side video surveillance equipment;

[0063] S4: The enterprise-side AI algorithm management system automatically pulls the enterprise-side video stream to run the specified algorithm application according to the time period and running time specified in the algorithm task received from the group-side AI algorithm management system. When alarm information is found by running the algorithm application based on the enterprise-side video stream, the alarm information is reported to the group-side alarm management system of the group-side management platform through the MQTT protocol; if there is an alarm picture in the alarm information, enter S5, if there is no alarm picture, enter S7;

[0064] S5: Store the alarm image in a local physical storage device on the enterprise side, and provide the alarm image URL address in the MQTT alarm information;

[0065] S6: After receiving the alarm information, the alarm management system on the group side reads the high-definition image according to the alarm image URL address and saves it in the OSS service of the group side management platform;

[0066] S7: Experts determine whether it is a valid alarm based on the alarm information displayed in the visual management system of the group-side management platform. If it is a valid alarm, an alarm work order is sent to the enterprise-side alarm system through the group-side alarm system. If it is an invalid warning, the invalid alarm information is fed back to the group-side AI algorithm management system for deep learning optimization of the algorithm model.

[0067] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0068] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A safe production intelligent supervision platform based on cloud-edge system, characterized by: Including the group-side management platform, several enterprise-side management platforms and several edge-side devices; The enterprise-side management platform performs coding management on edge-side devices and generates master data information; The group-side management platform arranges algorithm tasks and sends them to the enterprise-side management platform; the enterprise-side management platform collects and reports local organizational structure information, reports local organizational structure information and master data information to the group-side management platform, and deploys algorithm applications according to algorithm tasks, automatically pulling data from edge-side devices; the enterprise-side management platform monitors the alarms of algorithm applications running on the acquired data, reports alarm information to the group-side management platform when it is found, and stores the alarm information; The group-side management platform determines the effectiveness of the alarm based on the alarm information, generates an alarm work order based on the valid alarm information and sends it to the enterprise-side management platform, and optimizes the algorithm application based on the invalid alarm information.

2. According to claim 1, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: The group-side management platform includes the group-side basic information management system, the group-side equipment master data service system, the visualization management system, the group-side AI algorithm management system, the group-side voice communication system and the group-side alarm management system; among them, the group-side basic information management system is used to receive the local organizational structure information reported by the enterprise-side management platform; the group-side equipment master data service system is used to digitally archive and manage edge-side equipment according to business types, and uniformly encode it, and provide a service interface so that the group-side management platform can obtain the required master data information through timed synchronization or subscription; the group-side AI algorithm management system is used to apply the corresponding algorithm according to the algorithm requirements of the enterprise-side production equipment. It is pushed to the enterprise management platform, and at the same time monitors the computing resource usage, dynamically adjusts the algorithm orchestration task allocation, and also performs ROI area annotation on the enterprise management platform, and synchronizes the ROI area annotation data of the acquired master data information with the group-side equipment master data service system through the http protocol; the group-side voice communication system is used to achieve synchronous and low-latency voice interaction between the group and the enterprise; the group-side alarm management system is responsible for the configuration of alarm rules, receiving and processing various alarm information, dispatching alarm processing work orders for valid alarms and sending them to the enterprise management platform, and uploading invalid alarm information to the cloud server; the visualization management system is used to visualize the alarm information; The enterprise-side management platform includes the enterprise-side basic information management system, the enterprise-side equipment master data service system, the enterprise-side AI algorithm management system, the enterprise-side voice communication system and the enterprise-side alarm management system; among them, the enterprise-side basic information management system is used to receive the local organizational structure information filled in by the enterprise-side users and report it to the group-side basic information management system; the enterprise-side equipment master data service system is used to digitally archive and manage the edge-side equipment according to the business type, and uniformly encode it, and provide a service interface so that the enterprise-side management platform can obtain the required master data information through timed synchronization or subscription; the enterprise-side AI algorithm management system is used to receive and Deploy the algorithm application issued by the group-side management platform, process the data collected by the edge-side equipment and perform AI analysis, feed back the alarm information generated by the analysis results to the enterprise-side alarm management system, and regularly report the computing resource usage and algorithm operation status to the group-side AI algorithm management system of the group-side management platform for resource scheduling and algorithm optimization; the enterprise-side voice communication system is used to achieve synchronous, low-latency voice interaction between the group side and the enterprise side; the enterprise-side alarm management system is used to upload alarm information to the group-side alarm management system, and receive the alarm processing work order issued by the group-side alarm management system, forming an alarm closed loop after eliminating dangerous hazards.

3. According to claim 2, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: Both the group-side basic information management system and the enterprise-side basic information management system include a user management module, an information entry module, an information query module and an information update module; among them, the user management module is used to complete the creation of user accounts, authority allocation and identity authentication functions; the information entry module is used to input basic information related to production safety into the system; the information query module is used to provide information query functions, allowing users to retrieve basic information based on fuzzy queries and multi-condition combinations; the information update module allows users to modify and update the basic information in the system.

4. According to claim 2, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: Both the group-side equipment master data service system and the enterprise-side equipment master data service system include an equipment master data query service module and an equipment unified coding service module; among them, the equipment master data query service module is used to query and manage all master data information related to edge-side equipment to realize equipment full life cycle management; the equipment unified coding service module is used to assign a unique code to different edge-side devices to facilitate identity identification and information management, and realize standardized communication and data exchange between the group-side management platform and the enterprise-side management platform.

5. According to claim 2, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: The visualization management system includes a data visualization module, a user interaction module and a real-time monitoring module; the data visualization module is used to intuitively display the collected alarm information in the form of charts and dashboards; the user interaction module is used to provide a user interface to allow users to interact with the system; the real-time monitoring module is used to display video data from edge devices in real time.

6. According to claim 2, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: The group-side AI algorithm management system includes an algorithm training module, an algorithm management module, an algorithm orchestration module, an algorithm delivery module, and an algorithm scheduling module; among them, the algorithm training module is used to train and optimize algorithm applications; the algorithm management module is used to register, version control, performance evaluate, and conduct security audits on algorithms; the algorithm orchestration module is used for users to select edge devices, select algorithms that need to be supported, edit the time period and running time of the algorithm, implement algorithm task orchestration, obtain algorithm tasks, and support users to flexibly configure algorithm running parameters according to business needs; the algorithm delivery module is used to deliver trained algorithm applications from the group-side management platform to the enterprise-side management platform; the algorithm scheduling module is used to intelligently allocate computing resources to maximize the utilization and load balancing of computing resources. This module dynamically adjusts the algorithm's running priority and resource allocation according to business needs and the computing resource usage of each enterprise-side management platform; The enterprise-side AI algorithm management system includes an algorithm receiving module, an algorithm execution module and an algorithm analysis module; wherein the algorithm receiving module is used to receive the algorithm application issued by the algorithm issuing module; the algorithm execution module is used to execute the corresponding algorithm application according to the algorithm task received from the algorithm orchestration module, in accordance with the time period and running time specified in the algorithm task; the algorithm analysis module is used to generate alarm information during the analysis process of executing the algorithm application, if there is an analysis result that meets the alarm situation, and send the alarm information to the enterprise-side alarm management system.

7. According to claim 2, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: Both the group-side voice communication system and the enterprise-side voice communication system include a voice transmission module, a call control module and an emergency broadcast module; the voice transmission module is used to process the collection, encoding, transmission and decoding processes of voice signals; the call control module is used to manage the establishment, maintenance and termination of calls, including processing call requests, releasing communication resources and monitoring communication status; the emergency broadcast module is used to broadcast emergency information to all or specific areas in an emergency.

8. According to claim 2, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: The group-side alarm management system includes an alarm display module, an alarm service module and an alarm data management module; the alarm display module is responsible for receiving alarm information from the enterprise-side alarm management system and processing it according to the preset alarm classification, priority sorting and response strategy; the alarm service module generates an alarm processing work order for the alarm information reported during the inspection process, and provides the release and notification services of the alarm processing work order; the alarm data management module is responsible for storing and managing the detailed data of the alarm events corresponding to the alarm information; The enterprise-side alarm management system includes an alarm information transfer module and an alarm processing module; among them, the alarm information transfer module is responsible for receiving alarm information from the enterprise-side AI algorithm management system and reporting the alarm information to the alarm display module; the alarm processing module is used to receive the alarm processing work order distributed by the alarm service module.

9. According to claim 1, a cloud-edge system-based safe production intelligent supervision platform is characterized in that: Edge-side devices include video surveillance equipment and voice communication equipment. The video surveillance equipment consists of a video probe and a network video recorder to form a video surveillance platform, which is used to collect and store images of personnel and equipment in the production site operation area. The voice communication equipment is used to realize real-time voice communication between experts on the group side and production site maintenance personnel on the enterprise side.

10. A management method for a safe production intelligent supervision platform based on cloud-edge collaboration, characterized in that: A smart production safety supervision platform based on cloud-edge collaboration as described in any one of claims 1 to 9 comprises the following steps: Step 1: The enterprise-side management platform collects and reports the organizational structure information, and uniformly codes and manages the edge node devices on the enterprise side to generate master data information; after the organizational structure information is approved by the group-side management platform, the organizational structure information and master data information are synchronized to the group-side device master data service system of the group-side management platform; Step 2: Through the group-side AI algorithm management system, select the monitoring device on the enterprise side, select the algorithm application that needs to be supported, edit the time period and running time of the application algorithm, bind the edge-side device with the algorithm application, implement algorithm task orchestration, and push the algorithm application and algorithm tasks to the enterprise-side management platform; Step 3: The enterprise-side management platform receives the issued algorithm application, implements algorithm deployment and upgrade replacement, and automatically pulls the data collected by the edge-side device according to the algorithm task; Step 4: The enterprise-side management platform runs the algorithm application according to the algorithm task. When an alarm message is found, it identifies whether there is an alarm picture in the alarm message. If there is an alarm picture, the enterprise-side management platform saves the alarm picture and generates an alarm picture URL address and inserts it into the alarm message, and reports the alarm information to the group-side management platform. Otherwise, the alarm information is directly reported to the group-side management platform. Step 5: Determine the validity of the alarm information through the group-side management platform. If it is a valid alarm, send an alarm processing work order to the enterprise-side alarm system through the group-side alarm system. If it is an invalid warning, transmit the invalid alarm information to the group-side AI algorithm management system for deep learning optimization algorithm application.

Images