Safety production dual-prevention management and control cloud platform driven by industrial internet

Abstract

The invention discloses an industrial internet-driven safety production dual-prevention management and control cloud platform, and the platform comprises an industrial internet data collection layer which is composed of an industrial internet-of-things terminal, an edge computing gateway and a block chain evidence storage node, and is configured to collect equipment operation parameters, environment monitoring data, personnel operation records and external associated information of a production site, preprocessing, encrypting and storing the collected data; the dual-prevention core engine is used for integrating a risk grading management and control module and a hidden danger investigation and treatment module, and the risk grading management and control module is configured to output a real-time risk grade and a management and control list based on a preset risk matrix and collected data; according to the system, the industrial interconnection data acquisition layer realizes real-time acquisition, preprocessing and encrypted evidence storage of production total factor data, the problems of lagging and low reliability of traditional data are solved, and high-quality data support is provided for dual prevention.

Description

Technical Field

[0001] This invention belongs to the field of industrial safety production technology, and in particular relates to a cloud platform for dual prevention and control of safety production driven by the industrial internet. Background Technology

[0002] In existing industrial safety production management, a dual prevention mechanism has been formed, centered on risk classification and control and hazard investigation and rectification. This involves manual inspections to record equipment status, paper forms to register hazard information, and regular meetings to assess risk levels. Some enterprises have introduced local monitoring systems, equipment sensors, and basic management software to collect and store key parameters, track the progress of hazard rectification using ledger management tools, and conduct cross-departmental communication via walkie-talkies and telephones to coordinate emergency resources for handling emergencies, thus improving the standardization of safety production management to a certain extent.

[0003] Existing technologies have significant limitations in practical applications: data collection relies on manual operation, lacks real-time performance, and is prone to omissions; risk assessments are mostly static quarterly evaluations, failing to dynamically reflect changes in production site conditions; hazard identification relies on human experience, resulting in low efficiency and limited accuracy; data is stored in a fragmented manner across departments, lacking a unified cloud-based collaborative platform, leading to chaotic access control; emergency warning responses are delayed, and resource scheduling relies on manual judgment, making rapid and accurate handling difficult. These problems prevent the dual prevention mechanism from fully realizing its effectiveness and fail to meet the demands of industrial production for intelligent, collaborative, and dynamic safety management. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of existing technologies, embodiments of the present invention provide an industrial internet-driven dual prevention and control cloud platform for safe production, which solves the problems of unreliable data lag, static risk assessment, inefficient hazard management, and poor collaborative emergency response in traditional technologies.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An industrial internet-driven cloud platform for dual prevention and control of safety production includes:

[0007] Industrial Internet Data Acquisition Layer: Composed of industrial IoT terminals, edge computing gateways and blockchain evidence storage nodes, it is configured to collect equipment operating parameters, environmental monitoring data, personnel operation records and external related information in the production site in full coverage. The edge computing gateway realizes heterogeneous data protocol conversion and low-latency preprocessing, and the blockchain evidence storage nodes encrypt and store key data, providing a real-time and reliable data source for dual prevention.

[0008] The dual-prevention core engine integrates a risk grading and control module and a hazard investigation and management module. The risk grading and control module is based on a two-dimensional matrix of probability-consequence severity and collected data. It optimizes the weight of evaluation indicators through a risk model self-learning unit and visualizes the risk evolution process through a digital twin mapping unit, outputting real-time risk levels and control lists. The hazard investigation and management module receives hazard information through a combination of AI recognition and manual reporting. After the rectification closed-loop tracking sub-unit matches the responsible parties, associates historical solutions, and tracks the management status, the management results are directly fed back to the risk level adjustment, forming a "hazard management-risk downgrading" linkage mechanism.

[0009] Cloud-based collaborative management and control layer: Implemented based on cloud computing architecture, configured to allocate differentiated data access permissions through multi-tenant permission management unit, mine the correlation patterns of risks and hidden dangers through data fusion analysis unit, generate visual management and control dashboards, and support seamless integration with enterprise ERP and MES systems to improve cross-departmental collaboration efficiency;

[0010] Intelligent early warning and response module: When the risk level exceeds the threshold or the hidden danger is not addressed in time, an early warning (sound and light / SMS / pop-up) is triggered according to the classification rules. The emergency resource database is linked to generate a scheduling plan through optimization algorithms to ensure the accuracy and timeliness of emergency response.

[0011] The above-mentioned levels achieve data interaction through industrial internet protocols, forming a complete closed loop from data collection, risk assessment, hidden danger management to early warning and disposal, and promoting the upgrade of the dual prevention mechanism from manual to data-driven.

[0012] Preferably, the low-latency preprocessing of the edge computing gateway includes data compression, outlier filtering, and protocol conversion, with a processing latency of ≤50ms, ensuring that the collected data is uploaded to the cloud in real time.

[0013] Preferably, the blockchain evidence storage node performs hash encryption on key data such as risk assessment reports and hidden danger rectification records for evidence storage. The evidence storage information includes timestamps and the operating entity, realizing data immutability and traceability.

[0014] Preferably, the risk model self-learning unit processes historical accident data and hidden danger closed-loop records through a gradient descent algorithm, and dynamically optimizes the weights of evaluation indicators every quarter, with the adjustment range not exceeding ±20% of the initial value.

[0015] Preferably, the three-dimensional visualization model constructed by the digital twin mapping unit includes an overlay layer of equipment operating parameters and a risk diffusion simulation layer, which can simulate the spatial distribution changes of temperature, pressure, and gas concentration within 0-120 minutes.

[0016] Preferably, the AI ​​recognition subunit of the hazard investigation and management module adopts the YOLOv5 algorithm, the training dataset contains at least 5,000 images of industrial scene hazards, the recognition accuracy is ≥92%, and it forms a complementary verification mechanism with manually reported information.

[0017] Preferably, the multi-tenant permission management unit allocates data access permissions according to management level, operation level, and supervision level. The management level can view the global risk heat map, the operation level can only obtain the list of hidden dangers in its own area, and the supervision level can access historical audit logs.

[0018] Preferably, the tiered early warning rules of the intelligent early warning and response module are as follows: high-risk level triggers industrial site audible and visual alarms, simultaneous push notifications via SMS to management personnel and a pop-up window at the top of the platform; medium-risk level triggers work team terminal reminders; low-risk level is pushed via job-specific messages, with an early warning delivery delay of ≤10 seconds.

[0019] Preferably, the emergency resource scheduling scheme adopts an improved A* algorithm, which generates three alternative schemes by comprehensively considering resource distance, available quantity, and path congestion coefficient. The scheme update frequency is synchronized with the risk level changes in real time.

[0020] The preferred method for dual prevention and control of security production on cloud platforms includes the following steps:

[0021] S1: The industrial internet data acquisition layer collects all production element data, which is then preprocessed, encrypted, and stored before being uploaded to the cloud.

[0022] S2: The dual prevention core engine dynamically updates the risk level, and the results of hazard management are fed back to risk adjustment in real time, forming a closed loop of "hazard-risk" linkage;

[0023] S3: The cloud-based collaborative management and control layer distributes visualized data according to permissions, and the intelligent early warning and response module triggers early warnings and generates the optimal scheduling plan;

[0024] S4: Based on closed-loop data, continuously optimize the evaluation model and governance process to upgrade the dual prevention mechanism from manual to data-driven.

[0025] The technical effects and advantages of this invention's industrial internet-driven dual prevention and control cloud platform for safe production:

[0026] 1. This invention enables the industrial internet data acquisition layer to achieve real-time acquisition, preprocessing, and encrypted storage of all production element data, solving the problems of traditional data lag and low credibility, and providing high-quality data support for dual prevention.

[0027] 2. This invention, the dual prevention core engine, achieves real-time updates of risk levels through a risk and hidden danger linkage mechanism, combined with a dynamic risk matrix and digital twin simulation, solving the problem of the disconnect between static assessment and actual scenarios, and improving the accuracy of control.

[0028] 3. This invention improves the efficiency of hazard identification and timely rectification by using AI-powered hazard identification and closed-loop tracking functions. It also strengthens governance responsibility through performance-based linkage, solving the problems of missed detections and delayed rectification caused by manual inspections.

[0029] 4. This invention enables multi-level differentiated permission management and data visualization in the cloud-based collaborative management layer, and shortens emergency response time and improves cross-departmental collaboration efficiency through hierarchical early warning and optimized scheduling algorithms in the intelligent early warning and response module.

[0030] 5. This invention forms a closed loop of data-risk-hazard-disposal, promoting the upgrade of the dual prevention mechanism for safe production from manual to industrial Internet-driven, significantly reducing the accident rate and improving the level of industrial production safety management. Attached Figure Description

[0031] Figure 1 This is a flowchart illustrating the industrial internet-driven dual prevention and control cloud platform for safe production proposed in this invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0034] Example 1

[0035] refer to Figure 1 This embodiment provides an industrial internet-driven dual prevention and control cloud platform for safe production, used for specific applications in the industrial internet data acquisition layer. Specific implementation details include:

[0036] Application scenario: The ammonia synthesis unit area of ​​a large chemical plant (a high-risk area involving high-temperature and high-pressure equipment and toxic gases).

[0037] Industrial IoT terminal configuration:

[0038] IoT sensors (model: SENSOR-IT600) were deployed on 10 reactors and 20 conveying pipelines to collect vibration (range 0-50mm / s), temperature (-20~300℃), and pressure (0-10MPa) parameters at a sampling frequency of 1Hz.

[0039] Fifteen environmental monitoring terminals (model: ENV-900) were deployed in the plant area to collect ammonia concentration (0-100ppm) and dust value (0-10mg / m³), with a sampling frequency of 10 seconds / time.

[0040] Thirty workers were equipped with smart safety helmets (model: SAFE-5G), which have built-in GPS positioning (accuracy ±1m) and RFID card readers (to record the status of certified operation).

[0041] Edge computing gateway: The ECG-800 industrial edge gateway is adopted, which supports OPCUA / MQTT protocol conversion and preprocesses the collected data (such as removing abnormal jump values ​​from temperature sensors). The preprocessing latency is stable at 35ms±5ms.

[0042] Blockchain-based evidence storage: Adopting a consortium blockchain architecture (nodes include the enterprise's security department, production department, and third-party supervisor), the hazard rectification form (including signed scanned copies) and risk assessment report are hash-encrypted for evidence storage. The evidence storage timestamp is accurate to the second and cannot be tampered with.

[0043] Results: Enables real-time collection and reliable storage of all production data, providing high-quality data input for subsequent risk assessment.

[0044] Example 2

[0045] This embodiment provides an industrial internet-driven cloud platform for dual prevention and control of safety production, used for risk-level linkage of the core engine of dual prevention. Specific implementation details include:

[0046] Application scenario: Blasting operation area of ​​an open-pit mine (risk level changes dynamically).

[0047] Risk classification and control module:

[0048] Preset risk matrix: The probability dimension is divided into "highly likely (daily blasting), possible (3 times a week), occasional (1 time a month), and very rare (1 time a quarter)"; the severity of consequences dimension includes "personnel casualties, equipment damage, slope instability, and complaints from surrounding residents".

[0049] Risk model self-learning: The gradient descent algorithm is adopted, based on data from 200 blasting accidents in the past 3 years and 500 hidden danger rectification records, and the weight of the assessment indicators is adjusted every quarter (such as the weight of "slope moisture content" is increased from 15% to 20%).

[0050] Digital twin mapping: Construct a three-dimensional twin of the blasting area (based on laser scanning modeling, with an accuracy of ±5cm), overlay real-time parameters (hole mesh parameters, explosive charge, slope displacement), simulate the shock wave diffusion range under different wind directions (east wind level 3) (0-120 minutes), and output dynamic adjustment suggestions for the safety warning radius.

[0051] Linkage between the hazard identification and mitigation modules:

[0052] When a "bursting hole blockage (Level II hazard)" is detected, and the remediation is 100% complete with no new hazards within 24 hours, the risk level is automatically downgraded from "high risk" to "medium risk".

[0053] Results: The response time for risk level updates has been reduced from 4 hours for traditional manual assessment to 5 minutes, and the linkage mechanism has reduced ineffective control measures by 30%.

[0054] Example 3

[0055] This embodiment provides an industrial internet-driven cloud platform for dual prevention and control of safety production, used for AI-based hazard identification and closed-loop tracking. Specific implementation details include:

[0056] Application scenario: A car welding workshop (high incidence of mechanical injuries and electrical fire hazards).

[0057] AI-based hazard identification subunit:

[0058] Ten high-definition industrial cameras (4K resolution, 25fps) were deployed, covering the shooting range of 20 welding robot workstations.

[0059] The training dataset contains 8,000 images of potential hazards (including 25 scenarios such as "missing guardrails for robotic arms" and "damaged cables"), and is trained using the YOLOv5 algorithm, achieving a recognition accuracy of 94.3% and a response time of 220ms.

[0060] Typical recognition case: The camera captures the "emergency stop button of the welding robot being blocked", automatically marks the location and uploads it to the platform.

[0061] Rectification closed-loop tracking:

[0062] The system automatically matches the responsible work team (Welding Team 1), pushes the historical rectification plan ("Remove the obstruction and install the protective cover"), and sets a rectification deadline of 8 hours.

[0063] After the rectification is completed, the work team uploads before-and-after comparison photos, and the system automatically links them to the work team's monthly performance (accounting for 15%).

[0064] Results: Hazard identification efficiency increased by 60%, and the timeliness of rectification increased from 75% to 92%.

[0065] Example 4

[0066] This embodiment provides an industrial internet-driven cloud platform for dual prevention and control of safety production, used for cloud-based collaboration and intelligent early warning and response. Specific implementation details include:

[0067] Application scenario: Atmospheric and vacuum distillation unit area of ​​an oil refinery (requiring collaborative management by multiple departments).

[0068] Cloud-based collaborative management layer:

[0069] Multi-tenant permissions: Management level (plant manager) can view the risk heat map of the entire plant area; Operation level (shift leader) can only view the data of their own unit area; Supervisory level (emergency management bureau) can access historical audit logs.

[0070] Visual dashboard: Updated every 5 minutes, including a pie chart showing "12% of high-risk areas" and a bar chart showing "average time for hazard rectification is 4.2 hours", linked to real-time video of distillation tower No. 3 (delay ≤ 1 second).

[0071] Intelligent early warning and response:

[0072] Tiered early warning: When the temperature at the bottom of the tower exceeds 380℃ (threshold 360℃), a level 1 early warning is triggered, and within 10 seconds, it is simultaneously sent to the on-site audible and visual alarm (110dB), the plant manager's mobile phone text message, and the platform's top pop-up window.

[0073] Emergency resource dispatch: Using an improved A* algorithm, considering the fire pump room (0.8km away), the standby cooling water pumps (2 available), and the pipe gallery congestion coefficient (0.3), three plans are generated. The optimal plan is "activate cooling water pump No. 2 + fire hose connection", which is simultaneously pushed to the fire truck GPS terminal.

[0074] Results: Emergency response time was reduced from 15 minutes to 5 minutes, and resource scheduling path optimization rate reached 20%.

[0075] Example 5

[0076] This embodiment provides an industrial internet-driven cloud platform for dual prevention and control of safety production, used for dual prevention and control methods and processes in safety production. Specific implementation details include:

[0077] Application scenario: Cleanroom in an electronics factory (full-process digital management and control).

[0078] Step S1: The IoT terminal collects temperature and humidity (23±1℃, 50±5%RH), cleanliness (Class 1000), and personnel entry and exit records. After preprocessing by the edge gateway (compression rate 30%), the data is uploaded to the cloud every 60 seconds.

[0079] Step S2: When the "FFU fan failure (Level I hazard)" treatment is 100% complete and the system is operating normally for 24 consecutive hours, the risk level is downgraded from "medium risk" to "low risk".

[0080] Step S3: The cloud pushes a "risk level downgrade notification" to the workshop director, the early warning system lifts the yellow warning, and the scheduling plan (standby fan standby) is synchronized to the equipment PLC after confirmation.

[0081] Step S4: Collect 1,000 closed-loop data points weekly (such as rectification timeliness and early warning response) and generate an optimization report monthly (such as "Recommendation to add FFU fan vibration monitoring").

[0082] Results: The entire process of the dual prevention mechanism was digitized, resulting in a 40% year-on-year decrease in the number of potential hazards.

[0083] Comparative Example 1

[0084] It provides traditional safety production management methods.

[0085] Application scenario: Cleanroom in an electronics factory (without using the platform of this invention), similar to Example 5.

[0086] Data collection: Manual inspection records (twice a day), paper forms are archived, and data lag is ≥8 hours.

[0087] Risk assessment: Quarterly manual scoring results in static risk levels that remain unchanged over a long period (e.g., "low risk"), failing to reflect real-time status.

[0088] Hazard management: Rectification is notified by phone and confirmed by written signature, but the rectification closure rate is only 60%.

[0089] Emergency response: Relying on experience to allocate resources, a "FFU fan failure" once caused a 4-hour production stoppage and a loss of 200,000 yuan.

[0090] Compared with Examples 1-5 and Comparative Example 1, the core difference between the present invention and the conventional technology lies in the technical architecture and control efficiency:

[0091] In terms of data acquisition, the embodiment achieves real-time acquisition of all elements of data through industrial IoT terminals and edge computing gateways (sampling frequency 1Hz-10 seconds / time), with preprocessing latency ≤50ms, and combines blockchain notarization to ensure data credibility; the comparative embodiment relies on manual inspection (twice a day), with data lag ≥8 hours and easy tampering.

[0092] In terms of risk and hazard management, the implementation example uses a dynamic risk matrix, AI recognition (accuracy ≥ 92%), and digital twin simulation, reducing the risk level update response time to 5 minutes and the timely rectification rate of hazards to 92%; the comparison example relies on quarterly manual assessment, with the risk level remaining static and the rectification closure rate only 60%.

[0093] In terms of collaboration and early warning, the implementation example reduces the emergency response time to 5 minutes through cloud-based multi-tenant permissions, tiered early warning (delivery delay ≤ 10 seconds), and intelligent scheduling algorithms; the comparison example relies on manual notification, resulting in delayed response and blind resource scheduling.

[0094] Overall, the implementation plan achieves "real-time data, dynamic risk assessment, closed-loop governance, and intelligent collaboration" through industrial internet technology, reducing ineffective control by 30% and the annual number of potential hazards by 40% compared to the comparative plan, highlighting the effectiveness of technology integration in improving the dual prevention mechanism for safe production.

[0095] The above embodiments can be implemented in whole or in part by software, hardware, firmware or other arbitrary combinations. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0096] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0097] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0098] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the claims.

[0099] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An industrial internet-driven cloud platform for dual prevention and control of safety production, characterized in that: include: Industrial Internet Data Acquisition Layer: Composed of industrial IoT terminals, edge computing gateways and blockchain evidence storage nodes, it is configured to collect equipment operating parameters, environmental monitoring data, personnel operation records and external related information from the production site, and to preprocess and encrypt the collected data for evidence storage. Dual prevention core engine: integrates risk classification and control module and hidden danger investigation and management module. The risk classification and control module is configured to output real-time risk level and control list based on preset risk matrix and collected data. The hidden danger investigation and management module is configured to receive hidden danger information and track the management status. The two are linked through data interface. When the completion of hidden danger management reaches the preset threshold, the risk classification and control module automatically lowers the risk level of the corresponding area. Cloud-based collaborative management and control layer: Implemented based on cloud computing architecture, configured to allocate differentiated data access permissions according to user level, mine the correlation patterns of risks and hidden dangers through data fusion algorithms, generate visual management and control dashboards, and support data interaction with enterprise ERP system and MES system; Intelligent early warning and response module: configured to trigger a graded early warning when the risk level exceeds a preset threshold or the hidden danger is not addressed within the time limit, and to link the emergency resource database to generate the optimal scheduling plan; The above-mentioned levels achieve data interaction through industrial internet protocols, forming a closed-loop management and control system of "data collection - risk assessment - hidden danger management - early warning and handling".

2. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, In the industrial internet data acquisition layer, the edge computing gateway performs protocol conversion on heterogeneous device data with a preprocessing delay of ≤50ms, and the blockchain evidence storage node performs hash encryption storage on key operation data.

3. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, In the dual prevention core engine, the risk classification and control module's preset risk matrix has four levels of probability: "highly likely, possible, occasional, and very rare," and four categories of severity of consequences: "personnel injury, equipment damage, environmental impact, and economic loss." 4. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, In the dual prevention core engine, the risk model self-learning unit adopts the gradient descent algorithm. Based on the historical accident data and closed-loop data of hidden danger rectification over the past three years, the weight of the evaluation indicators is dynamically adjusted every quarter, with the adjustment range not exceeding ±20% of the initial value.

5. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, In the dual prevention core engine, the three-dimensional risk dynamic twin constructed by the digital twin mapping unit includes a three-dimensional model of the equipment, a real-time operating parameter overlay layer, and a risk diffusion simulation layer, which can simulate the spatial distribution changes of temperature, pressure, and gas concentration within 0-120 minutes.

6. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, The training dataset for the AI ​​hazard identification subunit contains at least 5,000 images of potential hazards in industrial scenarios, with an identification accuracy of ≥92% and an identification response time of ≤300ms.

7. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, The cloud-based collaborative management and control layer's visual management dashboard updates data automatically every 5 minutes, including a pie chart of risk level distribution, a bar chart of hazard rectification timeliness, and a real-time video linkage window for high-risk areas.

8. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, In the tiered early warning system of the intelligent early warning and response module, Level 1 early warnings are simultaneously pushed through industrial site audible and visual alarm devices, management personnel's mobile phone text messages, and a pop-up window at the top of the platform, with a delivery delay of ≤10 seconds; Level 2 early warnings are pushed through work team walkie-talkies and a pop-up window on the side of the platform; and Level 3 early warnings are pushed through messages from the work station terminal.

9. The industrial internet-driven dual prevention and control cloud platform for safe production as described in claim 1, characterized in that, The emergency resource scheduling plan is generated using an improved A* algorithm, which comprehensively considers resource distance, available quantity, and transportation route congestion coefficient to generate three alternative plans and mark the optimal plan. The plan update frequency is synchronized with the risk level changes.

10. A method for dual prevention and control of safety production using an industrial internet-driven dual prevention and control cloud platform as described in any one of claims 1-9, characterized in that, Includes the following steps: S1: The industrial internet data acquisition layer collects all production element data, which is preprocessed by the edge computing gateway and stored on the blockchain. Then, it is uploaded to the cloud collaborative management and control layer at a frequency of minutes. S2: The dual prevention core engine generates real-time risk levels. When the completion rate of hazard management is ≥90% and the system has been running stably for 24 hours, the risk level of the corresponding area will be automatically lowered. S3: The cloud-based collaborative management and control layer distributes data according to permissions. The intelligent early warning and response module triggers early warnings and generates a dispatch plan. After the dispatch plan is confirmed by the management personnel, it is automatically synchronized to the emergency resource terminal. S4: Collect closed-loop data weekly and input it into the risk model's self-learning unit; generate a dual prevention mechanism optimization report monthly to achieve dynamic iteration.

Images