Intelligent safety production dynamic supervision and early warning system
Through an intelligent dynamic supervision and early warning system for production safety, AI automatic analysis technology and edge detection modules are used to solve the high cost and low efficiency problems caused by the reliance on human resources in traditional production safety supervision, real-time monitoring, intelligent early warning and efficient hidden danger inspection, reducing the risk of safety hazard omissions.
Patent Information
- Application Number
- CN202411968897.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-30
- Publication Date
- 2025-04-22
AI Technical Summary
Traditional safety production supervision relies on a large amount of human resources, resulting in high costs, limited real-time monitoring and data processing capabilities, and it is difficult to achieve intelligent early warning and efficient hidden danger inspection, which poses a risk of safety hazard omissions.
Design an intelligent dynamic supervision and early warning system for production safety, including behavioral analysis and identification module, scenario model construction module, summary and statistical analysis module, safety production hazard assessment module, automatic alarm module and accident traceability module. Through AI automatic analysis technology and edge detection module, real-time monitoring and intelligent early warning of the production environment are achieved.
Reduce labor costs, improve real-time and accuracy, enhance early warning capabilities, improve data processing capabilities, optimize hidden danger investigation and accident traceability, promote continuous improvement, and adapt to the needs of diverse scenarios.
Smart Images
Figure CN119863124A_ABST
Abstract
Description
Technical Field
[0001] This document relates to the field of production safety, and in particular to an intelligent dynamic supervision and early warning system for production safety. Background Art
[0002] Production safety supervision involves multiple fields, including information technology, engineering technology, and management science. By formulating safety management systems and operating procedures, investigating and controlling hidden dangers and monitoring major sources of danger, establishing early warning mechanisms, and standardizing production behaviors, each link can meet the requirements of production safety standards, promote good production conditions in terms of people, machines, materials, methods, and the environment, and continuously improve and strengthen the standardized construction of enterprise production safety. At present, traditional supervision methods often require a large amount of human resources to manage the safety of production equipment, environment, and processes, resulting in an increase in production costs. In addition, manual management methods are prone to omissions, resulting in a series of safety hazards in the production process, which staff cannot detect in time, leading to production safety accidents.
[0003] Therefore, it is necessary to provide an intelligent dynamic production safety supervision and early warning system. Summary of the invention
[0004] This manual provides an intelligent dynamic production safety supervision and early warning system to solve the problems that traditional production safety supervision relies on a large amount of human resources, resulting in high costs, limited real-time monitoring and data processing capabilities, and difficulty in achieving intelligent early warning and efficient hidden danger detection, and there is a risk of safety hazard omissions.
[0005] The present invention proposes an intelligent dynamic safety production supervision and early warning system, which includes:
[0006] A behavior analysis and recognition module is configured to detect whether the production behavior data in the current production environment image data is safe through a recognition algorithm, and if not, send an alarm message to the automatic alarm module;
[0007] The scenario model building module is configured to generate production safety supervision processes under different business scenarios for different production sites based on the scenario model;
[0008] A summary statistics analysis module is configured to summarize dynamic supervision data, which includes production safety supervision process, safety hazard data and alarm information, and send it to the production safety hazard assessment module; wherein the safety hazard data is the production environment image data when the production behavior data is unsafe;
[0009] The production safety hazard assessment module is configured to assess the enterprise's production safety supervision technology, capabilities and existing production safety hazards based on production safety supervision data.
[0010] Furthermore, the automatic alarm module is connected to the accident tracing module. When the alarm module issues an alarm message, the accident tracing module is used to trace the production environment image data when the production behavior data is unsafe, and trace the cause of the safety hazard.
[0011] Furthermore, the recognition algorithm and the scene model are arranged in an edge detection module, and the edge detection module is deployed in a video link and connected with the behavior analysis and recognition module and the scene model construction module.
[0012] Furthermore, the production behavior data in the current production environment image data is detected to be safe by using a recognition algorithm, and the method is as follows:
[0013] Construct a safe production sample, use the safe production sample as the input of the recognition algorithm, train the recognition algorithm to obtain a trained recognition algorithm, obtain the current production environment image data in real time, input it into the trained recognition algorithm, and obtain the safety hazard category, which includes safe and unsafe.
[0014] Furthermore, the production safety samples are positive and reverse pictures or video data of safety hazards existing in actual production operations.
[0015] Furthermore, the recognition algorithm is built based on YOLO and SSD network architecture.
[0016] Furthermore, the safety production supervision process can be obtained by:
[0017] Configure different supervision methods according to actual business scenarios, analyze the categories, sequences, and relationships of safety hazards in different locations based on different supervision methods, and obtain the safety supervision process in complex scenarios.
[0018] Furthermore, the automatic alarm module includes an alarm reminder on the system side, an audible and visual alarm in the production environment, and a subsequent handling mechanism for the alarm event.
[0019] Furthermore, after tracing the cause of the safety hazard, a disposal plan is generated based on the preset disposal method.
[0020] Furthermore, after evaluating the enterprise's production safety supervision technology, capabilities, and existing production safety hazards, a plan for the next step in improving production safety will be generated.
[0021] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects:
[0022] Reduce labor costs: By introducing AI automatic analysis technology, the reliance on a large number of human resources is reduced, the company's labor cost expenditure is reduced, and the monitoring and management efficiency is improved.
[0023] Improve real-time performance and accuracy: The system can conduct real-time and comprehensive identification and analysis of production environment videos, promptly discover and deal with potential safety hazards, avoid omissions and delays that may be caused by manual monitoring, and improve the accuracy and response speed of safety supervision.
[0024] Enhanced early warning capabilities: Through intelligent algorithms and model construction, prediction and early warning of potential risks are achieved, allowing companies to take preventive measures before accidents occur, thereby effectively reducing the probability of safety accidents.
[0025] Improve data processing capabilities: The system can quickly summarize and analyze a large amount of dynamic supervision data, including production safety supervision processes, safety hazard data and alarm information, providing enterprises with a basis for scientific decision-making.
[0026] Optimize hidden danger detection and accident tracing: With the help of the behavior analysis and identification module and the accident tracing module, hidden danger detection can be carried out efficiently, and the cause can be quickly traced after the accident, and reasonable disposal plans can be formulated to prevent similar incidents from happening again.
[0027] Promote continuous improvement: Through a comprehensive assessment of the enterprise's production safety status, the system can help enterprises identify weak links in safety management, guide enterprises to continuously optimize and improve safety management systems and technical means, and promote the continuous improvement of the enterprise's production safety level.
[0028] Adapt to diverse scenario requirements: Due to the configuration of flexible scenario model building modules, the system can customize corresponding supervision processes according to different production sites and business scenarios to meet the needs of safe production in various complex environments. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0030] Figure 1 A schematic diagram of the connection relationship between modules in an intelligent dynamic safety production supervision and early warning system of the present invention;
[0031] Figure 2 The figure is a schematic diagram of the working process of an intelligent dynamic monitoring and early warning system for safe production according to the present invention. DETAILED DESCRIPTION
[0032] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.
[0033] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.
[0034] like Figure 1 and Figure 2 As shown, the present invention proposes an intelligent dynamic safety production supervision and early warning system, the system comprising:
[0035] A behavior analysis and recognition module is configured to detect whether the production behavior data in the current production environment image data is safe through a recognition algorithm, and if not, send an alarm message to the automatic alarm module;
[0036] The scenario model building module is configured to generate production safety supervision processes under different business scenarios for different production sites based on the scenario model;
[0037] A summary statistics analysis module is configured to summarize dynamic supervision data, which includes production safety supervision process, safety hazard data and alarm information, and send it to the production safety hazard assessment module; wherein the safety hazard data is the production environment image data when the production behavior data is unsafe;
[0038] The safety production hazard assessment module is configured to assess the enterprise's safety production supervision technology, capabilities and existing safety production hazards based on safety production supervision data. After assessing the enterprise's safety production supervision technology, capabilities and existing safety production hazards, a plan for the next step of safety production improvement is generated to help enterprises improve the level of intelligent supervision of safety production.
[0039] Among them, the recognition algorithm and the scene model are set in the edge detection module, and the edge detection module is deployed in the video link. As the first-layer processing node of the system, it is responsible for preliminary processing of real-time image data from the camera and is connected with the behavior analysis and recognition module and the scene model construction module. The behavior analysis and recognition module receives the image data that has been preliminarily processed, uses the trained recognition algorithm to detect unsafe behaviors in the production environment, and sends the results to the automatic alarm module.
[0040] The scenario model building module is connected to the edge detection module to build corresponding business scenario models based on the needs of different production sites. Different safety production supervision processes are configured according to actual production conditions.
[0041] The method of detecting whether the production behavior data in the current production environment image data is safe is as follows:
[0042] Construct a safe production sample, use the safe production sample as the input of the recognition algorithm, train the recognition algorithm to obtain a trained recognition algorithm, obtain the current production environment image data in real time, input it into the trained recognition algorithm, and obtain the safety hazard category, which includes safe and unsafe.
[0043] The safety production samples are positive and negative pictures or video data of safety hazards in actual production operations. The recognition algorithm is built based on the YOLO and SSD network architecture.
[0044] Positive samples: Collect images and videos of safe behaviors captured under normal operating conditions.
[0045] Reverse samples: Collect images and videos showing unsafe behaviors or potential safety hazards. These samples can be obtained from historical accident records, simulated scenarios, or daily monitoring.
[0046] Use professional image annotation tools to annotate all collected data in detail. The annotation content includes but is not limited to:
[0047] The specific location of the unsafe behavior (e.g. worker not wearing a helmet, equipment failure, etc.)
[0048] The bounding box of each object ensures that the location of the target object is accurately marked.
[0049] Category labels such as "worker", "machine", "hazardous area", etc.
[0050] The automatic alarm module in the present invention is connected to the accident tracing module. When the alarm module issues an alarm message, the accident tracing module is used to trace the production environment image data when the production behavior data is unsafe, and trace the cause of the safety hazard. After tracing the cause of the safety hazard, a disposal plan is generated based on a preset disposal method.
[0051] The automatic alarm module includes alarm reminders on the system side, sound and light alarms in the production environment, and subsequent handling mechanisms for alarm events. Specifically, for alarm reminders on the system side: develop or integrate the existing notification system to ensure that it can receive alarm signals from the behavior analysis and recognition module, and send alarms to designated managers through multiple channels (such as email, text messages, and mobile application push).
[0052] For audible and visual alarms in production environments: Choose appropriate audible and visual alarms, such as high-decibel speakers, flashing lights, or other warning devices, based on the characteristics of the production environment. Make sure they can work effectively in noisy or low-light conditions. Install audible and visual alarms in each key area that needs to be monitored, ensuring that their sound and light cover the entire work area so that all employees can hear or see the alarm. Connect audible and visual alarms to automatic alarm modules so that these devices can be activated immediately when safety hazards are detected.
[0053] The safety production supervision process in the present invention is obtained by:
[0054] Configure different supervision methods according to actual business scenarios, analyze the categories, sequences, and relationships of safety hazards in different places according to different supervision methods, and obtain the safety supervision process in complex scenarios. Specifically:
[0055] Divide all production areas into several categories (such as processing areas, storage areas, office areas, etc.) according to industry standards or internal regulations, and formulate corresponding supervision strategies for each category.
[0056] Determine the importance of various risks based on historical data and expert opinions to provide a basis for subsequent resource allocation.
[0057] Build a scenario model: record the specific environmental parameters (such as temperature, humidity, noise level, etc.) and special requirements (such as explosion protection, fire protection, etc.) of each production site. Describe in detail the working steps, tools and techniques used in each production line, especially those operation links that are prone to accidents. Based on the above information, build a logical framework with multiple nodes, each node represents a specific inspection item or control point. Allow the model to automatically adjust the inspection order and focus based on real-time monitoring data to adapt to changing production conditions.
[0058] Analyze the categories and sequence of hidden dangers: refer to international or national standards, and combine the actual situation of the enterprise to define a safety hazard classification system that is applicable to the enterprise (such as mechanical injuries, electrical hazards, chemical leaks, etc.).
[0059] Coding rules: Assign a unique code to each type of hidden danger to facilitate system management and data analysis.
[0060] Determine the inspection order
[0061] Causal chain analysis: Use fishbone diagrams or other analysis tools to find out the causal relationship between hidden dangers and clarify which hidden dangers are the direct cause or indirect influencing factor of other hidden dangers.
[0062] Path optimization: Use graph theory algorithms to calculate the optimal inspection path to ensure that all important hidden danger points are covered in the shortest time.
[0063] Build a network of relationships
[0064] Relationship Mapping
[0065] Correlation Matrix: Create a correlation matrix to show the degree of mutual influence between different hazards, for example, some hazards may increase the probability of other hazards.
[0066] Impact weight: Assign a weight value to the relationship between each pair of hidden dangers, reflecting the closeness of their connection.
[0067] In summary, the system in the above embodiment can form an intelligent identification algorithm for safe production through the input of safe production sample data, configure different intelligent supervision services for the production scenarios of different businesses, match the safety supervision services in various businesses, and replace the manual supervision method to automatically identify and supervise all safety hazards, behaviors, and events in the production site, and alarm the safety hazards that appear, promptly remind management personnel, and provide a hidden danger alarm handling process to effectively solve safety hazards and reduce the occurrence of safe production accidents. At the same time, the system of this embodiment will classify and archive the safety hazards that appear in the supervision process, which is helpful for tracing the causes of safety accidents and improving the safety supervision capabilities of production. Finally, this embodiment provides a safety production technology assessment based on multiple factors such as enterprise safety production technology, capabilities, and events, which helps enterprises understand the current safety production hazard problems and improve the safety production capabilities of enterprises.
[0068] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims
1. An intelligent dynamic safety production supervision and early warning system, characterized in that: The system includes: A behavior analysis and recognition module is configured to detect whether the production behavior data in the current production environment image data is safe through a recognition algorithm, and if not, send an alarm message to the automatic alarm module; The scenario model building module is configured to generate production safety supervision processes under different business scenarios for different production sites based on the scenario model; A summary statistics analysis module is configured to summarize dynamic supervision data, which includes production safety supervision process, safety hazard data and alarm information, and send it to the production safety hazard assessment module; wherein the safety hazard data is the production environment image data when the production behavior data is unsafe; The production safety hazard assessment module is configured to assess the enterprise's production safety supervision technology, capabilities and existing production safety hazards based on production safety supervision data.
2. According to claim 1, an intelligent dynamic safety production supervision and early warning system is characterized in that: The automatic alarm module is connected to the accident tracing module. When the alarm module sends out an alarm message, the accident tracing module is used to trace the production environment image data when the production behavior data is unsafe, and trace the cause of the safety hazard.
3. According to claim 1, an intelligent dynamic safety production supervision and early warning system is characterized in that: The recognition algorithm and the scene model are arranged in an edge detection module, and the edge detection module is deployed in a video link and connected with the behavior analysis and recognition module and the scene model construction module.
4. According to claim 1, an intelligent dynamic safety production supervision and early warning system is characterized in that: The method of detecting whether the production behavior data in the current production environment image data is safe is as follows: Construct a safe production sample, use the safe production sample as the input of the recognition algorithm, train the recognition algorithm to obtain a trained recognition algorithm, obtain the current production environment image data in real time, input it into the trained recognition algorithm, and obtain the safety hazard category, which includes safe and unsafe.
5. According to claim 4, an intelligent dynamic safety production supervision and early warning system is characterized in that: The production safety samples are positive and reverse pictures or video data of safety hazards existing in actual production operations.
6. According to claim 4, an intelligent dynamic safety production supervision and early warning system is characterized in that: The recognition algorithm is built based on YOLO and SSD network architecture.
7. The intelligent dynamic safety production supervision and early warning system according to claim 1 is characterized in that: The safety production supervision process can be obtained by: Configure different supervision methods according to actual business scenarios, analyze the categories, sequences, and relationships of safety hazards in different locations based on different supervision methods, and obtain the safety supervision process in complex scenarios.
8. The intelligent dynamic safety production supervision and early warning system according to claim 1 is characterized in that: The automatic alarm module includes alarm reminders on the system side, sound and light alarms in the production environment, and a subsequent handling mechanism for alarm events.
9. The intelligent dynamic safety production supervision and early warning system according to claim 2 is characterized in that: After tracing the causes of safety hazards, a disposal plan is generated based on the preset disposal method.
10. The intelligent dynamic safety production supervision and early warning system according to claim 2 is characterized in that: After evaluating the enterprise's production safety supervision technology, capabilities, and existing production safety hazards, a plan for the next step in improving production safety is generated.