Risk operation intelligent supervision method and system based on Internet of Things

Abstract

The invention relates to an intelligent risk operation supervision method and system based on the Internet of Things, and the method comprises the steps: obtaining the work ticket information of a power plant safety platform, and binding an RFID positioning tag to a work card and a worker in charge; associating the work card, the person in charge of work, the mobile camera and the mobile sound box to the current work ticket; the RFID positioning position is scanned in real time, the work ticket information is identified, and the working state of a worker in charge is dynamically displayed in a two-dimensional plane graph; monitoring the position association state of the worker in charge and the work card in real time, and if the separation exceeds a preset time, triggering a departure alarm; detecting violation operation of a worker in charge, and if violation occurs, triggering violation alarm; and after the risk operation is finished, closing the work ticket, and removing the association binding among all the labels. Compared with the prior art, the probability that hidden dangers such as personnel departure and illegal operation in risk operation are converted into accidents is effectively reduced, and the operation safety is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of production management technology, and in particular to an intelligent monitoring method and system for risky operations based on the Internet of Things. Background Technology

[0002] High-risk operations in coal-fired power plants involve numerous specialties and processes, such as working at heights, hot work, confined space work, high-temperature and high-pressure work, electrical work, lifting work, chemical work, coal yard work, and equipment inspection and maintenance. These operations present serious safety hazards and potential dangers, and have always been a key focus and challenge in power plant safety management.

[0003] In the early days, the supervision of industrial hazardous operations relied entirely on manual methods, including paper work permit approvals, on-site inspections, and manual record keeping. Work permits, personnel attendance, and the implementation of safety measures were all visually verified by on-site managers. The effectiveness of supervision depended heavily on the responsibility and experience of the personnel, resulting in significant subjectivity and lag, and a high accident rate.

[0004] With the advancement of industrial informatization, power plants and other enterprises have introduced the "two-ticket system" and electronic management platforms to achieve online approval, circulation, and archiving of work tickets, replacing paper records and improving process standardization. Simultaneously, some high-risk areas have begun deploying fixed video surveillance equipment, allowing for remote viewing of monitoring footage to assist in supervision. However, the technology at this stage still has significant limitations: video surveillance relies on manual monitoring, location information is separated from work ticket information, and it cannot link personnel and work status in real time.

[0005] To address the aforementioned issues, there is an urgent need for an integrated regulatory technology that can combine the Internet of Things, intelligent identification, and data linkage. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a method and system for intelligent monitoring of risk operations based on the Internet of Things.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] A method for intelligent monitoring of risky operations based on the Internet of Things, the method comprising:

[0009] Obtain work order information from the power plant safety platform and bind RFID positioning tags to the work badge and the person in charge of the work respectively;

[0010] Link the work badge, work supervisor, mobile camera, and mobile speaker to the current work order;

[0011] By using RFID locators deployed in the work area to scan the RFID location and identify work ticket information in real time, the work status of the person in charge of the work is dynamically displayed on a two-dimensional plan, and the icon color is changed according to the movement status of the work card.

[0012] The system monitors the location association between the person in charge of the work and their work badge in real time. If the separation exceeds a preset time, an alarm will be triggered by moving a speaker to indicate that the person has left their post.

[0013] The AI ​​behavior recognition algorithm of the mobile camera detects the work supervisor's illegal operation. If a violation is found, a violation alarm is triggered through the mobile speaker.

[0014] After the risky operation is completed, close the work ticket and unbind all the associated tags.

[0015] Furthermore, the RFID positioning tags are bound to the personal belongings of the person in charge of the work, including a safety helmet, goggles, and work clothes.

[0016] Furthermore, the association is achieved by uniquely encoding the proprietary IDs of the work badge, the person in charge of the work, the mobile camera, and the mobile speaker with the ID information of the work ticket.

[0017] Furthermore, if the person in charge of the work changes, the work order information is updated, and the unique IDs of the work badge and the person in charge of the work are updated, and the association between the work badge, the person in charge of the work, the mobile camera and the mobile speaker and the work order is re-established.

[0018] Furthermore, when an off-duty alarm or violation alarm is triggered, a warning message is played via a mobile speaker or a remote voice command is received from the administrator.

[0019] Furthermore, the work status of the person in charge of the work includes:

[0020] Work in progress: The positioning tags of work badges, work supervisors, mobile cameras, and mobile speakers are all identified by RFID locators deployed in the work area, and the work icon is displayed in yellow on the two-dimensional plan.

[0021] Work interrupted or temporarily terminated: The person in charge of the work takes away the work tag. Neither the person in charge of the work nor the location tag of the work tag is identified by the RFID locator deployed in the work area, but the mobile camera and mobile speaker are still in place. The work icon is displayed as gray in the two-dimensional plan.

[0022] Work ticket completion: Work ticket information is closed, the work icon disappears from the 2D plan view, and all associations are removed;

[0023] Operation Abnormality: Triggering any alarm will cause the operation icon to turn red on the 2D plan view.

[0024] Furthermore, the work order information includes the content of the risky operation project, the expected time of the risky operation, and the preset scope of the risky operation.

[0025] Furthermore, the AI ​​behavior recognition algorithm includes: a helmet wearing detection model based on YOLOv7, a firework recognition model using a SlowFast network, and a DeepSORT algorithm-based method for tracking violations and associating them with specific personnel tags.

[0026] Furthermore, during risky operations, data on the operation process is recorded in real time, including work order information, location trajectory, alarm events, and equipment usage records. Multidimensional statistical analysis is performed by region, time, and operation type to generate risk heat map reports.

[0027] An intelligent monitoring system for risky operations based on the Internet of Things, the system comprising:

[0028] Binding Module: Obtains work order information from the power plant safety platform, binds RFID positioning tags to the work badge and the person in charge of the work respectively, and associates the work badge, the person in charge of the work, the mobile camera and the mobile speaker with the current work order;

[0029] Location monitoring module: By scanning the location of RFID positioning and identifying work ticket information in real time through RFID locators deployed in the work area, the work status of the person in charge of the work is dynamically displayed on a two-dimensional plan, and the icon color is changed according to the movement status of the work card;

[0030] Alarm module: Real-time monitoring of the location association status between the person in charge of the work and their work badge. If the separation exceeds a preset time, an off-duty alarm will be triggered via a mobile speaker. The AI ​​behavior recognition algorithm of the mobile camera will detect any violations by the person in charge of the work. If a violation is detected, a violation alarm will be triggered via a mobile speaker.

[0031] Work completion module: After the risky operation is completed, close the work ticket and unbind all the associated tags.

[0032] Compared with the prior art, the beneficial effects of the present invention include:

[0033] 1. This invention constructs a precise and intelligent supervision solution covering the entire lifecycle of risky operations. Through the full-process design of work order information acquisition, RFID tag binding, real-time location monitoring, dynamic status management, multi-dimensional alarms, and work completion unbinding, it achieves full control over risky operations from preparation to completion. This solution not only solves the pain point of traditional supervision's inability to monitor in real time, but also upgrades safety supervision from passive inspection to proactive early warning through the linkage of real-time positioning and intelligent identification. This effectively reduces the probability of hidden dangers such as personnel leaving their posts and violations of regulations turning into accidents during risky operations, significantly improving operational safety. At the same time, the fully automated management of the entire process reduces the cost of manual intervention, improves supervision efficiency by more than 40%, and provides a systematic solution for safety management in high-risk scenarios such as power plants.

[0034] 2. This invention achieves strong binding between personnel location and identity by fixing RFID positioning tags to the personal items of the person in charge, such as safety helmets and goggles. This avoids the positioning failure problem caused by the separation of tags and personnel in traditional positioning methods, ensuring that the positioning data truly reflects the actual status of personnel and improving the reliability of monitoring data.

[0035] 3. This invention uses one-hot encoding to associate the unique IDs of work badges, personnel, and equipment with work ticket IDs, ensuring that each element forms a unique mapping with the work ticket. This solves the problem of information cross-contamination and confusion in multi-task parallel scenarios, enabling the system to accurately identify the status of personnel and equipment for specific tasks and improving the monitoring accuracy in complex scenarios.

[0036] 4. This invention visually presents the entire operation process status through color changes of yellow, gray, red, and disappearing icons in a two-dimensional planar diagram. Managers can quickly identify abnormal scenarios, reduce information interpretation time, improve risk response speed, and shorten the accident intervention window.

[0037] 5. This invention employs advanced algorithms such as YOLOv7, SlowFast, and DeepSORT to achieve high-precision identification of violations in multiple scenarios and accurately associate them with specific individuals, reducing false alarms and missed alarms, and improving the practicality of intelligent identification. Attached Figure Description

[0038] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0039] 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, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0040] Example 1

[0041] A method for intelligent monitoring of risky operations based on the Internet of Things, the method is as follows: Figure 1 As shown, it includes steps S1-S5.

[0042] In this invention, in high-risk operation scenarios, a 5G private network and a star-topology LoRa local area network are deployed for wireless communication between devices. The star-topology LoRa local area network has anti-interference capabilities and can operate stably in complex industrial environments. It forms a redundant communication link with the 5G private network to ensure that data is not lost.

[0043] The network layer uses an encrypted transmission protocol, and the communication data between the device and the server is encrypted with AES-256 to prevent information leakage or tampering.

[0044] Step S1: Obtain the work order information from the power plant safety platform and bind the RFID positioning tag to the work badge and the person in charge of the work respectively.

[0045] The work order information includes the content of the risky operation, the expected time of the risky operation, and the preset scope of the risky operation.

[0046] The RFID positioning tags are bound to the personal items of the person in charge of the work, including a safety helmet, goggles, and work clothes.

[0047] The RFID positioning tag is a passive design with a recognition distance of 0-10 meters, a service life of ≥3 years, and a high temperature resistance range of -40℃ to 85℃. It can be embedded in carriers such as safety helmets and work badges.

[0048] The mobile wireless camera has a resolution of 4 megapixels, supports H.265 encoding, a frame rate of 25fps, a built-in AI chip, and can run behavior recognition algorithms (such as the YOLOv5 object detection model) locally with a recognition accuracy of ≥95%. It has a battery life of ≥4 hours and supports Wi-Fi 6 and Bluetooth 5.0.

[0049] Step S2: Associate the work badge, the person in charge of the work, the mobile camera, and the mobile speaker with the current work ticket.

[0050] The association is achieved by uniquely encoding the unique IDs of the work badge, the person in charge of the work, the mobile camera, and the mobile speaker with the ID information of the work ticket.

[0051] If the person in charge of the work changes, the work order information is updated, and the unique IDs of the work badge and the person in charge of the work are updated. The association between the work badge, the person in charge of the work, the mobile camera and the mobile speaker and the work order is re-established.

[0052] Step S3: The RFID locator deployed in the work area scans the RFID location and identifies the work ticket information in real time, dynamically displays the work status of the person in charge of the work on a two-dimensional plan, and changes the icon color according to the movement status of the work card.

[0053] When deploying RFID trackers, installation locations are pre-planned using 3D modeling tools to ensure complete signal coverage. The devices have a built-in self-diagnostic module that monitors battery level and network signal strength in real time, automatically reporting any anomalies to the maintenance platform. Remote restart and parameter configuration are also supported.

[0054] The work status of the person in charge includes:

[0055] Work in progress: The positioning tags of work badges, work supervisors, mobile cameras, and mobile speakers are all identified by RFID locators deployed in the work area, and the work icon is displayed in yellow on the two-dimensional plan.

[0056] Work interrupted or temporarily terminated: The person in charge of the work takes away the work tag. Neither the person in charge of the work nor the location tag of the work tag is identified by the RFID locator deployed in the work area, but the mobile camera and mobile speaker are still in place. The work icon is displayed as gray in the two-dimensional plan.

[0057] Work ticket completion: Work ticket information is closed, the work icon disappears from the 2D plan view, and all associations are removed;

[0058] Operation Abnormality: Triggering any alarm will cause the operation icon to turn red on the 2D plan view.

[0059] Step S4: Monitor the location association status between the person in charge of the work and their work badge in real time. If the separation exceeds a preset time, trigger an off-duty alarm via a mobile speaker. Detect any violations by the person in charge of the work through the AI ​​behavior recognition algorithm of the mobile camera. If a violation is detected, trigger a violation alarm via a mobile speaker.

[0060] AI behavior recognition algorithms include: a helmet-wearing detection model based on YOLOv7, a fireworks recognition model using a SlowFast network, and a DeepSORT-based algorithm for tracking violations and associating them with specific personnel tags.

[0061] When training the model, transfer learning technology is used to train the model using historical monitoring data from the factory. The model inference speed reaches 30 frames per second, the false alarm rate is <3%, and online updates of algorithm parameters are supported.

[0062] The algorithm identifies violations including not wearing a safety helmet, not wearing a safety belt, smoking, unauthorized stacking in hot work areas, exceeding the limit for the number of people in confined spaces, throwing objects from heights, abnormal equipment operation, falls, intrusion into dangerous areas, failure to control open flames according to regulations, disorderly placement of tools, and unauthorized operation of equipment.

[0063] When an off-duty alarm or violation alarm is triggered, a mobile speaker is used to play a warning message or receive remote instructions from the administrator.

[0064] Step S5: After the risk operation is completed, close the work ticket and unbind all the associated tags.

[0065] During risky operations, data such as work order information, location trajectory, alarm events, and equipment usage records are recorded in real time. Multidimensional statistical analysis is performed by region, time, and operation type to generate risk heat map reports.

[0066] The following are practical application examples based on the above methods:

[0067] 1. Equipment is linked to work order.

[0068] The system integrates with the safety platform's two-ticket system to automatically retrieve "work tickets" and "operation tickets" for operations with significant or higher risks. Operators enter their dedicated work badge ID into the system and bind this ID to an RFID positioning tag; the work badge must be affixed to a prominent location at the work site. The positioning tag worn by the person in charge of the work is activated and associated with the work badge ID; if the person in charge changes, the system automatically updates the association; a camera / speaker is also linked.

[0069] 2. Visual monitoring of construction operations

[0070] The RFID locator continuously scans the work area, identifying the positions of work badges, responsible personnel tags, and cameras / speakers in real time. Work icons are dynamically displayed on the system's two-dimensional plan view, with icon status changing according to construction progress: In the middle of a row: the icon is yellow, indicating normal work progress; Interrupted / Ended: after the responsible personnel take away the work badge, the icon turns gray, indicating work is paused; Work ticket completed: after the work ticket is closed, the icon is removed from the plan view and all bindings are released; Additional functions: clicking the icon allows viewing work ticket details, responsible personnel information, and real-time access to camera footage.

[0071] 3. Security monitoring and early warning

[0072] The system continuously compares the positions of the work badge and the person in charge label. If the two are separated for more than a preset time (e.g., 15 minutes), the system will pop up an alert window and push a notification to the mobile terminal of the management personnel. The system calls the smart power plant AI algorithm (such as safety helmet detection, smoke and fire recognition, etc.) to analyze the real-time video from the camera. If a violation is detected, the system marks the violation and triggers an alarm, while playing a warning voice through the on-site speakers (supporting remote verbal intervention by management personnel).

[0073] 4. Historical Data Management and Analysis

[0074] Completely record work order information, responsible person's location, equipment usage records and alarm events for each operation; perform statistical analysis by region, time, operation type and other dimensions to generate reports such as operation frequency, number of participants, and cumulative duration, and visualize heat maps of high-risk areas to help optimize safety management strategies.

[0075] 5. Process Closed Loop and Iteration

[0076] Anomaly Handling: Alarm events are automatically linked to historical records for subsequent accountability and process improvement.

[0077] Dynamic updates: When work order information changes (such as extension or replacement of responsible person), the system synchronizes and updates the binding relationship in real time.

[0078] Data-driven: Historical analysis results are fed back to the two-vote approval process to help assess risk levels and resource allocation.

[0079] Through the above process, the system realizes full lifecycle management from job preparation and on-site monitoring to post-event analysis, effectively improving the safety control efficiency of high-risk operations.

[0080] Example 2

[0081] This embodiment discloses an IoT-based intelligent monitoring system for risk operations, based on Embodiment 1 above. The system includes:

[0082] Binding Module: Obtains work order information from the power plant safety platform, binds RFID positioning tags to the work badge and the person in charge of the work respectively, and associates the work badge, the person in charge of the work, the mobile camera and the mobile speaker with the current work order;

[0083] Location monitoring module: By scanning the location of RFID positioning and identifying work ticket information in real time through RFID locators deployed in the work area, the work status of the person in charge of the work is dynamically displayed on a two-dimensional plan, and the icon color is changed according to the movement status of the work card;

[0084] Alarm module: Real-time monitoring of the location association status between the person in charge of the work and their work badge. If the separation exceeds a preset time, an off-duty alarm will be triggered via a mobile speaker. The AI ​​behavior recognition algorithm of the mobile camera will detect any violations by the person in charge of the work. If a violation is detected, a violation alarm will be triggered via a mobile speaker.

[0085] Work completion module: After the risky operation is completed, close the work ticket and unbind all the associated tags.

[0086] For details regarding the above modules, please refer to the relevant descriptions and effects in Example 1 for further understanding.

[0087] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for intelligent monitoring of risky operations based on the Internet of Things, characterized in that, The method includes: Obtain work order information from the power plant safety platform and bind RFID positioning tags to the work badge and the person in charge of the work respectively; Link the work badge, work supervisor, mobile camera, and mobile speaker to the current work order; By using RFID locators deployed in the work area to scan the RFID location and identify work ticket information in real time, the work status of the person in charge of the work is dynamically displayed on a two-dimensional plan, and the icon color is changed according to the movement status of the work card. The system monitors the location association between the person in charge of the work and their work badge in real time. If the separation exceeds a preset time, an alarm will be triggered by moving a speaker to indicate that the person has left their post. The AI ​​behavior recognition algorithm of the mobile camera detects the work supervisor's illegal operation. If a violation is found, a violation alarm is triggered through the mobile speaker. After the risky operation is completed, close the work ticket and unbind all the associated tags.

2. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, The RFID positioning tags are bound to the personal items of the person in charge of the work, including a safety helmet, goggles, and work clothes.

3. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, The association is achieved by uniquely encoding the unique IDs of the work badge, the person in charge of the work, the mobile camera, and the mobile speaker with the ID information of the work ticket.

4. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, If the person in charge of the work changes, the work order information is updated, and the unique IDs of the work badge and the person in charge of the work are updated. The association between the work badge, the person in charge of the work, the mobile camera and the mobile speaker and the work order is re-established.

5. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, When an off-duty alarm or violation alarm is triggered, a mobile speaker is used to play a warning message or receive remote instructions from the administrator.

6. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, The work status of the person in charge includes: Work in progress: The positioning tags of work badges, work supervisors, mobile cameras, and mobile speakers are all identified by RFID locators deployed in the work area, and the work icon is displayed in yellow on the two-dimensional plan. Work interrupted or temporarily terminated: The person in charge of the work takes away the work tag. Neither the person in charge of the work nor the location tag of the work tag is identified by the RFID locator deployed in the work area, but the mobile camera and mobile speaker are still in place. The work icon is displayed as gray in the two-dimensional plan. Work ticket completion: Work ticket information is closed, the work icon disappears from the 2D plan view, and all associations are removed; Operation Abnormality: Triggering any alarm will cause the operation icon to turn red on the 2D plan view.

7. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, The work order information includes the content of the risky operation project, the expected time of the risky operation, and the preset scope of the risky operation.

8. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, The AI ​​behavior recognition algorithm includes: a helmet wearing detection model based on YOLOv7, a firework recognition model using a SlowFast network, and a DeepSORT algorithm to track violations and associate them with specific personnel tags.

9. The intelligent monitoring method for risky operations based on the Internet of Things according to claim 1, characterized in that, During risky operations, data such as work order information, location trajectory, alarm events, and equipment usage records are recorded in real time. Multidimensional statistical analysis is performed by region, time, and operation type to generate risk heat map reports.

10. A risk operation intelligent monitoring system based on the Internet of Things, characterized in that, The system includes: Binding Module: Obtains work order information from the power plant safety platform, binds RFID positioning tags to the work badge and the person in charge of the work respectively, and associates the work badge, the person in charge of the work, the mobile camera and the mobile speaker with the current work order; Location monitoring module: By scanning the location of RFID positioning and identifying work ticket information in real time through RFID locators deployed in the work area, the work status of the person in charge of the work is dynamically displayed on a two-dimensional plan, and the icon color is changed according to the movement status of the work card; Alarm module: Real-time monitoring of the location association status between the person in charge of the work and their work badge. If the separation exceeds a preset time, an off-duty alarm will be triggered via a mobile speaker. The AI ​​behavior recognition algorithm of the mobile camera will detect any violations by the person in charge of the work. If a violation is detected, a violation alarm will be triggered via a mobile speaker. Work completion module: After the risky operation is completed, close the work ticket and unbind all the associated tags.

Images