Safety supervision method for automatically activating accident handling emergency plan

By automatically activating the safety supervision method of accident handling emergency plans, and using comprehensive monitoring of multiple data sources and artificial intelligence engines, automatic emergency response at the accident site is achieved, solving the problem of the lack of integration and linkage in accident handling of existing systems, and improving rescue efficiency and accident control effect.

CN119541129AInactive Publication Date: 2025-02-28DALIAN HUAYELIAN AUTOMATION

Patent Information

Application Number
CN202410490448.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-04-23
Publication Date
2025-02-28
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing safety supervision system lacks integration and linkage when an accident occurs, and cannot effectively control the level and scope of the accident, resulting in an expansion of the impact of the accident, a prolonged rescue time, and it is easy to miss the best handling time due to a single sensor failure or false alarm.

Method used

A safety supervision method is adopted to automatically activate emergency response plans for accidents. Through comprehensive monitoring of multiple data sources, personnel location, video, vital signs, equipment operation status and environmental data are collected, and data processing and equipment control are used to use artificial intelligence engines and PLC control systems to calculate the safety situation index in real time, and emergency plans are automatically activated, including measures such as opening of evacuation passages, starting fire equipment, and notification of rescue personnel.

Benefits of technology

It realizes automatic emergency response at the accident site, improves the rescue speed, controls the scope and level of the accident, reduces the impact of the accident on the enterprise, and reduces the misoperation and delays caused by human negligence.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119541129A_ABST
    Figure CN119541129A_ABST
Patent Text Reader

Abstract

The invention relates to a safety supervision method for automatically activating an accident handling emergency plan. The method comprises the following steps: acquiring field personnel position data DPOS, field video data DV, operator vital sign data DLF, equipment running state data DME and production environment state data DPE; the emergency plan processing system classifies and cleans the collected data, removes invalid data, judges whether an alarm threshold value is reached or not and generates an alarm event; dividing a safety monitoring period into a plurality of monitoring time periods T, judging the type and grade of an accident, and starting a corresponding emergency plan; the emergency plan processing system automatically records the process and time of emergency plan processing, and after the accident processing is completed, the process and time are automatically preserved and filed. The method has the advantages that multi-data-source comprehensive safety monitoring is carried out on the whole production line and the whole operation area, when an accident occurs, an emergency plan is automatically started, related personnel are quickly notified of the type and the position of the accident, the response time of emergency disposal is shortened, and the rescue time is saved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of emergency plan generation, and in particular to a safety supervision method for automatically activating an accident handling emergency plan. Background Art

[0002] The traditional safety management method uses manual supervision, which is prone to omissions due to carelessness, leading to safety accidents. It is important to use technical means to provide first aid to victims on site, evacuate on-site staff in a timely manner, control the level and scope of accidents, reduce the impact of accidents, and prevent the expansion of accidents. The demand is becoming increasingly prominent, and enterprises are in urgent need of realizing the intelligence of safety supervision.

[0003] Existing enterprise safety supervision work adopts monitoring systems, personnel positioning systems, gas sensors, equipment interlock emergency stop switches, sound and light alarms, etc. to improve the level of safety management and control. The above measures have the following shortcomings:

[0004] Each system operates independently and lacks integration and linkage, making it impossible to control the level and scope of accidents and take timely measures to reduce the impact of accidents. For example, the monitoring system is only responsible for video recording, and the hazardous gas sensor is only responsible for detecting environmental hazards. There is no linkage equipment emergency stop switch. In the event of an emergency or production accident, although the user can be reminded by the sound and light alarm, the equipment is still in operation and can only be stopped manually. There is a lack of guidance for emergency personnel to quickly arrive at the scene of the accident and rescue the injured in time, missing the opportunity to save lives; it is more likely that trapped people will miss precious escape and rescue time, causing serious accidents and further increasing the severity of the accident. In addition, it is easy to cause failures or false alarms due to a single sensor, missing the best time to deal with the accident. Summary of the invention

[0005] The purpose of the present invention is to provide a safety supervision method for automatically activating an accident handling emergency plan, conduct comprehensive monitoring of personnel, equipment, and environment within the production line from multiple data sources, and automatically activate the emergency plan when an accident occurs, thereby increasing the emergency response time, improving the rescue speed, controlling the scope and level of the accident, preventing the accident from expanding, and reducing the impact of the accident on the enterprise.

[0006] To achieve the above object, the present invention is implemented through the following technical solutions:

[0007] A safety supervision method for automatically activating an accident handling emergency plan specifically comprises the following steps:

[0008] S1. Collecting on-site personnel location data D POS , on-site video data D V , Operators’ vital signs data D LF, Equipment operation status data D ME , production environment status data D PE And upload data information to the emergency plan processing system through wired LAN / 5G wireless network;

[0009] S2. The emergency plan processing system collects the location data of the on-site personnel D POS , on-site video data D V , Operators’ vital signs data D LF , Equipment operation status data D ME , production environment status data D PE Perform classification and cleaning, remove invalid data, determine whether the alarm threshold is reached and generate an alarm event;

[0010] S3. Divide the security monitoring cycle into several monitoring periods T. If an alarm event is detected in the current detection period, calculate the security situation index SI in real time according to the number and weight of the alarm event. If the security situation index SI is lower than the set security threshold S T , then start the emergency plan processing process, determine the type and level of the accident that has occurred, and start the corresponding emergency plan;

[0011] S4. The emergency plan processing system automatically records the video footage in the supervision area during the period when the emergency plan is triggered, the audio information of the dispatch room, and the process and time of the activated emergency plan processing. After the accident is handled, it is automatically retained and archived for accident lesson analysis.

[0012] In step S1, the on-site personnel location data D POS and the vital signs data of the operators LF It comes from the personnel positioning equipment and positioning base stations worn by the operators. The positioning methods include GNSS, UWB, Bluetooth, and 4G / 5G communication base stations; on-site video data D V Derived from cameras, hard disk recorders; equipment operation status data D ME Through the PLC control system, the operating status of equipment and the operating status of safety production related equipment are collected; production environment status data D PE Data collected from fire alarms, hazardous gas sensors, smoke sensors, and temperature sensors.

[0013] In step S2, it is determined whether the alarm threshold is reached and an alarm event is generated, the contents of which are as follows:

[0014] 1) Collected on-site video data D V , after the image is detected by the artificial intelligence engine, the confidence level of the detection result is determined Vpr Is the value range within the set confidence level D? Vc Within the threshold range, the formula is as follows:

[0015] D Vpr =f(Dv),D Vpr ∈[D Vc ,1]①

[0016] In formula ①: f(x) is the image detection model used by the artificial intelligence engine, D Vc is the confidence range set by the system, 0 <D Vc <1, D Vpr is the confidence level of the test result;

[0017] The image detection models used by the AI ​​engine include: flame detection, smoke detection, and person-fall detection;

[0018] If the confidence level of the test result is D Vpr If the live video data D is within the threshold range, V Record into the system and generate alarm events;

[0019] 2) Collected on-site personnel location data D POS , mainly detects whether the X, Y coordinates enter the designated danger zone D POSda Within the scope, the right to enter the object D POSp , duration of stay POSt , when the on-site personnel location data D POS After the stay time in the danger zone exceeds the set value, the sampling value D POS Record into the system and generate alarm events;

[0020] 3) Vital signs data of operators D LF , mainly detects whether the heartbeat sampling value is within the set D LFmin , the formula is as follows:

[0021] D LF ∈[D LFmin ,D LFmax ]②

[0022] In formula ②: D LFmin is the lower limit of the heart rate detection value, and its unit is beats / minute; D LFmax is the upper limit of the heart rate detection value, and its unit is beats / minute; when D LF The sampling value exceeds D LFmin , D LFmax After the value range of LF Record into the system and generate alarm events;

[0023] 4) Equipment operation status data D ME, mainly according to the different properties of the PLC sampling points of the equipment operation status, set the corresponding threshold range, including: the furnace temperature of the heating furnace, the speed of the ventilation fan motor in the enclosed space, the opening and closing signals of the exhaust valves of toxic and harmful gases such as coal gas and ammonia, and detect whether the alarm threshold is reached;

[0024] 5) Production environment status data D PE , mainly based on the type of environmental sensor and the corresponding threshold to determine whether an alarm event occurs.

[0025] In step S3, the security situation index SI is calculated using the following formula:

[0026] SI=100-[(A V ×W V )+(A BI ×W BI )+(A LF ×W LF )+(A ME ×W ME )+(A PE ×W PE )]④

[0027] In formula ④: A V is the number of image recognition alarms in the current monitoring period, in times; A BI A is the number of dangerous area intrusion alarms during the current monitoring period, in times; LF is the number of vital sign alarms during the current monitoring period, in times; A ME It is the number of equipment operation status alarms in the current monitoring period, and its unit is times; A PE It is the number of environmental alarms in the current monitoring period, in times;

[0028] The value range of the security situation index SI is 0 to 100.

[0029] The emergency plan includes on-site disposal plan, special emergency plan and comprehensive emergency plan, the contents of which are as follows:

[0030] 1) On-site disposal plan

[0031] The system defines a production accident / violation handling plan as one in which no casualties occur and no external rescue forces are required. In actual implementation, it is used to handle the alarm events of on-site personnel operating in violation of regulations or illegally entering dangerous areas;

[0032] 2) Special emergency plan

[0033] The system defines it as a production accident disposal plan that causes death, fire, and other accidents that urgently require external rescue forces to deal with. In actual implementation, it is used to handle safety incidents such as abnormal vital signs of personnel, abnormal shutdown of main production equipment, or alarms from fire alarms;

[0034] 3) Comprehensive emergency plan

[0035] In the system, it is defined as the disposal measures that are not included in the on-site disposal plan and special emergency plan, and serves as a backup emergency plan.

[0036] 6. A safety supervision method for automatically activating an emergency plan for handling an accident according to claim 5, wherein the emergency plan automatically triggers the following single or combined plans according to the actual situation on site and the rules configured by the user:

[0037] 1) Link the sound and light alarm equipment in the factory to remind relevant operators to evacuate the dangerous area;

[0038] 2) Open emergency evacuation passages for personnel and vehicles;

[0039] 3) Start the fire emergency power supply and lighting equipment, and notify firefighters of the location and area of ​​the fire;

[0040] 4) The large screen in the dispatch room will use an electronic map to inform emergency rescue personnel of the type and location of the accident, and whether there are any injured or trapped people at the scene who need treatment;

[0041] 5) Use vibration or buzzer to remind employees wearing personnel positioning devices to evacuate as soon as possible;

[0042] 6) Through the Internet of Things linkage equipment PLC controller, interlock shutdown, stop equipment operation;

[0043] 7) When a dangerous gas leakage accident occurs, immediately start the on-site exhaust equipment.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] Comprehensive safety monitoring of the entire production line and operation area of ​​the steel enterprise from multiple data sources is carried out. By collecting data on personnel behavior, the operation of production safety equipment, and the environment, the emergency plan is automatically activated when an accident occurs, the on-site workers are evacuated, the evacuation channels are opened, the fire-fighting equipment is activated, and the type and location of the accident are quickly notified to the relevant personnel, thereby improving the response time of emergency disposal, saving precious rescue time, and preventing the escalation of accidents. The impact of corporate accidents is minimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a structural diagram of a safety supervision method for automatically activating an emergency plan for handling an accident.

[0047] Figure 2 It is a control flow diagram of a safety supervision method for automatically activating an accident handling emergency plan.

[0048] Figure 3 It is a processing flow chart for automatically activating the emergency plan for accident handling. DETAILED DESCRIPTION

[0049] The present invention is described in detail below in conjunction with the accompanying drawings, but it should be noted that the implementation of the present invention is not limited to the following embodiments.

[0050] See Figures 1 to 3 , a safety supervision method for automatically activating an accident handling emergency plan, specifically comprising the following steps:

[0051] S1. Collecting on-site personnel location data D POS , on-site video data D V , Operators’ vital signs data D LF , Equipment operation status data D ME , production environment status data D PE And upload data information to the emergency plan processing system through wired LAN / 5G wireless network;

[0052] On-site personnel location data D POS and the vital signs data of the operators LF It comes from the personnel positioning equipment and positioning base stations worn by the operators. The positioning methods include GNSS, UWB, Bluetooth, and 4G / 5G communication base stations; on-site video data D V Derived from cameras, hard disk recorders; equipment operation status data D ME Through the PLC control system, the operating status of equipment and the operating status of safety production related equipment are collected; production environment status data D PE Data collected from fire alarms, hazardous gas sensors, smoke sensors, and temperature sensors.

[0053] S2. The emergency plan processing system collects the location data of the on-site personnel D POS , on-site video data D V , Operators’ vital signs data D LF , Equipment operation status data D ME , production environment status data D PE Perform classification and cleaning, remove invalid data, determine whether the alarm threshold is reached and generate an alarm event;

[0054] Determine whether the alarm threshold is reached and generate an alarm event, the content is as follows:

[0055] 1) Collected on-site video data DV , after the image is detected by the artificial intelligence engine, the confidence level of the detection result is determined Vpr Is the value range within the set confidence level D? Vc Within the threshold range, the formula is as follows:

[0056] D Vpr =f(Dv),D Vpr ∈[D Vc ,1]①

[0057] In formula ①: f(x) is the image detection model used by the artificial intelligence engine, D Vc is the confidence range set by the system, 0 <D Vc <1, D Vpr is the confidence level of the test result;

[0058] The image detection models used by the AI ​​engine include: flame detection, smoke detection, and person-fall detection;

[0059] If the confidence level of the test result is D Vpr If the live video data D is within the threshold range, V Record into the system and generate alarm events;

[0060] 2) Collected on-site personnel location data D POS , mainly detects whether the X, Y coordinates enter the designated danger zone D POSda Within the scope, the right to enter the object D POSp , duration of stay POSt , when the on-site personnel location data D POS After the stay time in the danger zone exceeds the set value, the sampling value D POS Record into the system and generate alarm events;

[0061] 3) Vital signs data of operators D LF , mainly detects whether the heartbeat sampling value is within the set D LFmin , the formula is as follows:

[0062] D LF ∈[D LFmin ,D LFmax ] ②

[0063] In formula ②: D LFmin is the lower limit of the heart rate detection value, and its unit is beats / minute; D LFmax is the upper limit of the heart rate detection value, and its unit is beats / minute; when D LF The sampling value exceeds D LFmin , D LFmax After the value range of LF Record into the system and generate alarm events;

[0064] 4) Equipment operation status data D ME , mainly according to the different properties of the PLC sampling points of the equipment operation status, set the corresponding threshold range, including: the furnace temperature of the heating furnace, the speed of the ventilation fan motor in the enclosed space, the opening and closing signals of the exhaust valves of toxic and harmful gases such as coal gas and ammonia, and detect whether the alarm threshold is reached;

[0065] 5) Production environment status data D PE , mainly based on the type of environmental sensor and the corresponding threshold to determine whether an alarm event occurs.

[0066] S3. Divide the security monitoring cycle into several monitoring periods T. If an alarm event is detected in the current detection period, calculate the security situation index SI in real time according to the number and weight of the alarm event. If the security situation index SI is lower than the set security threshold S T , then start the emergency plan processing process, determine the type and level of the accident that has occurred, and start the corresponding emergency plan;

[0067] The processing logic is as follows:

[0068]

[0069] In formula ④: A V为 The number of image recognition alarms in the current monitoring period, in times; the image recognition alarm weight W BR ; A BI is the number of dangerous area intrusion alarms during the current monitoring period, in times; the dangerous area intrusion alarm weight W BI ; A LF is the number of vital sign alarms in the current monitoring period, in times; the vital sign alarm weight W LF ; A ME is the number of equipment operation status alarms in the current monitoring period, in times; the equipment operation status alarm weight W ME ; A PE is the number of environmental alarms in the current monitoring period, in times; the environmental alarm weight W PE ;

[0070] The value range of security situation index SI is: 0~100;

[0071] Once the safety situation index in the current time period is lower than the set safety threshold, the emergency plan processing process is initiated. According to the alarm event and accident discrimination matrix, the type and level of the accident that has occurred are judged, and the corresponding on-site disposal plan, special emergency plan, and comprehensive emergency plan are triggered;

[0072] The triggering and processing logic of the emergency plan is as follows:

[0073]

[0074] Among them, S T Is the safety threshold, the value is generally 60 T <100. AccidentMatrix accident discrimination matrix method, the input value is the number of alarms, and the type and level of the accident that has occurred are judged. According to the type of accident, it is judged whether to trigger the on-site disposal plan OnSiteProcess, or the special emergency plan SpecialProcess, or the comprehensive emergency plan ComprehensiveProcess;

[0075] The emergency plan includes on-site disposal plan, special emergency plan and comprehensive emergency plan, the contents of which are as follows:

[0076] 1) On-site disposal plan

[0077] The system defines a production accident / violation handling plan as one in which no casualties occur and no external rescue forces are required. In actual implementation, it is used to handle the alarm events of on-site personnel operating in violation of regulations or illegally entering dangerous areas;

[0078] 2) Special emergency plan

[0079] The system defines it as a production accident disposal plan that causes death, fire, and other accidents that urgently require external rescue forces to deal with. In actual implementation, it is used to handle safety incidents such as abnormal vital signs of personnel, abnormal shutdown of main production equipment, or alarms from fire alarms;

[0080] 3) Comprehensive emergency plan

[0081] In the system, it is defined as the disposal measures that are not included in the on-site disposal plan and special emergency plan, and serves as a backup emergency plan;

[0082] On-site disposal plans, special emergency plans, and comprehensive emergency plans will automatically trigger the following single or combined plans based on the actual situation on site and the rules configured by the user:

[0083] 1) Link the sound and light alarm equipment in the factory to remind relevant operators to evacuate the dangerous area;

[0084] 2) Open emergency evacuation passages for personnel and vehicles;

[0085] 3) Start the fire emergency power supply and lighting equipment, and notify firefighters of the location and area of ​​the fire;

[0086] 4) The large screen in the dispatch room will use an electronic map to inform emergency rescue personnel of the type and location of the accident, and whether there are any injured or trapped people at the scene who need treatment;​

[0087] 5) Use vibration or buzzer to remind employees wearing personnel positioning devices to evacuate as soon as possible;

[0088] 6) Through the Internet of Things linkage equipment PLC controller, interlock shutdown, stop equipment operation;

[0089] 7) When a dangerous gas leakage accident occurs, immediately start the on-site exhaust equipment.

[0090] S4. The emergency plan processing system automatically records the video footage in the supervision area during the period when the emergency plan is triggered, the audio information of the dispatch room, and the process and time of the activated emergency plan processing. After the accident is handled, it is automatically retained and archived for accident lesson analysis.

[0091] The following examples are implemented on the premise of the technical solution of the present invention, and provide detailed implementation methods and specific operation processes, but the protection scope of the present invention is not limited to the following examples. The methods used in the following examples are conventional methods unless otherwise specified.

[0092] [Example 1]

[0093] See Figure 1 In a cold rolling production plant, a safety supervision method of automatically activating the emergency plan for accident handling was adopted to establish a safety supervision system and install and deploy equipment, including the following contents:

[0094] 1) Install a personnel positioning base station at the production site, and have the workers wear positioning bracelets to collect personnel location and vital signs data. Connect to the monitoring system to collect video data, connect to the gas alarm to collect production environment data, connect to the heating furnace PLC and on-site ventilation fan motor PLC to collect equipment operation data. The above data are transmitted to the safety supervision system via a wired LAN.

[0095] 2) Collect, classify, clean and process collected data in real time in the safety supervision system to determine whether an alarm event occurs.

[0096] 3) When an alarm event is found in the system during the monitoring period, the safety index is calculated immediately. When the safety index is lower than the threshold, the type and level of the accident that has occurred are determined through the accident discrimination matrix, and the corresponding emergency plan and processing mechanism are activated;

[0097] For example: Based on the personnel positioning data, the system finds that an employee has abnormal vital signs in the gas area, and there is a gas leak alarm in the area. The system determines that the comprehensive safety situation index is below the passing line. The accident discrimination matrix combines the personnel location, vital signs, and gas alarm information to comprehensively determine that a gas leak has occurred on site, and that a person suffering from gas poisoning has fallen unconscious in the leakage area and needs rapid rescue.

[0098] 4) The safety supervision system automatically activates the emergency response mechanism and adopts a special emergency rescue plan for the gas operation area: the sound and light alarm in the gas area is activated to remind on-site personnel to evacuate as soon as possible, the personnel and vehicle evacuation passages are opened to evacuate the personnel working in the area, the exhaust fan is started to reduce the gas concentration, the gas supply valve is closed to stop the gas supply, and professional emergency rescue personnel are notified to carry out rescue with gas rescue equipment. The location and vital signs of the trapped personnel are reminded to the rescue personnel in the form of an electronic map on the large screen. At the same time, the monitoring video of the production line and the audio recording of the dispatch room are recorded and saved, and the emergency response process is recorded for archiving and accident summary.

[0099] [Example 2]

[0100] In a hot rolling mill renovation project, a safety supervision method of automatically activating the accident handling emergency plan was adopted to establish a safety supervision system for related parties and install and deploy equipment, including the following contents:

[0101] 1) Install surveillance cameras at the project site to collect video data, and install gas alarms in the gas area to collect gas concentration data. The above data are transmitted to the safety supervision system through the wired LAN. Install sound and light alarms and LED large screens on site and connect to the system through the network.

[0102] 2) The system uses image recognition and detection by the camera to detect in real time whether the construction workers are wearing safety protection equipment (for example, whether work clothes and safety helmets are worn), whether they are working in violation of regulations (for example, whether workers at heights are wearing safety belts, whether workers have entered restricted areas, and whether there is smoking in gas areas), and detect whether there are abnormal behaviors (for example, people falling to the ground).

[0103] 3) When the system detects abnormal behavior of on-site workers, such as: people fall to the ground for more than 5 minutes; immediately activate the on-site disposal plan: use sound and light alarms to remind surrounding workers to quickly rescue, and notify safety supervisors and construction team leaders of related parties by information push and text messages, and contact medical rescue personnel to quickly provide first aid to the fallen person; use electronic maps on the LED large screen to remind rescue personnel to rush to the location of the fallen person as soon as possible, and record and save the on-site surveillance video, and record the emergency handling process for archiving and future reference.

Claims

1. A safety supervision method for automatically activating an emergency plan for handling an accident, characterized in that: The specific steps include: S1. Collecting on-site personnel location data D POS , on-site video data D V , Operators’ vital signs data D LF , Equipment operation status data D ME , production environment status data D PE And upload data information to the emergency plan processing system through wired LAN / 5G wireless network; S2. The emergency plan processing system collects the location data of the on-site personnel D POS , on-site video data D V , Operators’ vital signs data D LF , Equipment operation status data D ME , production environment status data D PE Perform classification and cleaning, remove invalid data, determine whether the alarm threshold is reached and generate an alarm event; S3. Divide the security monitoring cycle into several monitoring periods T. If an alarm event is detected in the current detection period, calculate the security situation index SI in real time according to the number and weight of the alarm event. If the security situation index SI is lower than the set security threshold S T , then start the emergency plan processing process, determine the type and level of the accident that has occurred, and start the corresponding emergency plan; S4. The emergency plan processing system automatically records the video footage in the supervision area during the period when the emergency plan is triggered, the audio information of the dispatch room, and the process and time of the activated emergency plan processing. After the accident is handled, it is automatically retained and archived for accident lesson analysis.

2. A safety supervision method for automatically activating an emergency plan for handling an accident according to claim 1, characterized in that: In step S1, the on-site personnel location data D POS and the vital signs data of the operators LF It comes from the personnel positioning equipment and positioning base stations worn by the operators. The positioning methods include GNSS, UWB, Bluetooth, and 4G / 5G communication base stations; on-site video data D V Derived from cameras, hard disk recorders; equipment operation status data D ME Through the PLC control system, the operating status of equipment and the operating status of safety production related equipment are collected; production environment status data D PE Data collected from fire alarms, hazardous gas sensors, smoke sensors, and temperature sensors.

3. A safety supervision method for automatically activating an emergency plan for handling an accident according to claim 1, characterized in that: In step S2, it is determined whether the alarm threshold is reached and an alarm event is generated, the contents of which are as follows: 1) The collected on-site video data D V After the image is detected by the artificial intelligence engine, the confidence level of the detection result is determined. Vpr Is the value range within the set confidence level D? Vc Within the threshold range, the formula is as follows: D Vpr =f(Dv),D Vpr ∈[D Vc ,1] ① In formula ①: f(x) is the image detection model used by the artificial intelligence engine, D Vc is the confidence range set by the system, 0 <D Vc <1, D Vpr is the confidence level of the test result; The image detection models used by the AI ​​engine include: flame detection, smoke detection, and person-fall detection; If the confidence level of the test result is D Vpr If the threshold value is within the range, the live video data D V Record into the system and generate alarm events; 2) The collected on-site personnel location data D POS , mainly detects whether the X, Y coordinates enter the designated danger zone D POSda Within the scope, the right to enter the object D POSp , duration of stay POSt , when the on-site personnel location data D POS After the stay time in the danger zone exceeds the set value, the sampling value D POS Record into the system and generate alarm events; 3) The vital signs data D of the operator LF , mainly detects whether the heartbeat sampling value is within the set D LFmin , the formula is as follows: D LF ∈[D LFmin ,D LFmax ]②In formula ②: D LFmin is the lower limit of the heart rate detection value, and its unit is beats / minute; D LFmax is the upper limit of the heart rate detection value, and its unit is beats / minute; when D LF The sampling value exceeds D LFmin , D LFmax After the value range of LF Record into the system and generate alarm events; 4) The equipment operation status data D ME , mainly according to the different properties of the PLC sampling points of the equipment operation status, set the corresponding threshold range, including: the furnace temperature of the heating furnace, the speed of the ventilation fan motor in the enclosed space, the opening and closing signals of the exhaust valves of toxic and harmful gases such as coal gas and ammonia, and detect whether the alarm threshold is reached; 5) The production environment status data D PE , mainly based on the type of environmental sensor and the corresponding threshold to determine whether an alarm event occurs.

4. A safety supervision method for automatically activating an emergency plan for handling an accident according to claim 1, characterized in that: In step S3, the security situation index SI is calculated using the following formula: SI=100-[(A V ×W V )+(A BI ×W BI )+(A LF ×W LF )+(A ME ×W ME )+(A PE ×W PE )]④ In formula ④: A V is the number of image recognition alarms in the current monitoring period, in times; A BI A is the number of dangerous area intrusion alarms during the current monitoring period, in times; LF is the number of vital sign alarms during the current monitoring period, in times; A ME It is the number of equipment operation status alarms in the current monitoring period, and its unit is times; A PE It is the number of environmental alarms in the current monitoring period, in times; The value range of the security situation index SI is 0 to 100.

5. A safety supervision method for automatically activating an emergency plan for handling an accident according to claim 4, characterized in that: The emergency plan includes on-site disposal plan, special emergency plan and comprehensive emergency plan, the contents of which are as follows: 1) On-site disposal plan The system defines a production accident / violation handling plan as one in which no casualties occur and no external rescue forces are required. In actual implementation, it is used to handle the alarm events of on-site personnel operating in violation of regulations or illegally entering dangerous areas; 2) Special emergency plan The system defines it as a production accident disposal plan that causes death, fire, and other accidents that urgently require external rescue forces to deal with. In actual implementation, it is used to handle safety incidents such as abnormal vital signs of personnel, abnormal shutdown of main production equipment, or alarms from fire alarms; 3) Comprehensive emergency plan In the system, it is defined as the disposal measures that are not included in the on-site disposal plan and special emergency plan, and serves as a backup emergency plan.

6. A safety supervision method for automatically activating an emergency plan for handling an accident according to claim 5, characterized in that: The emergency plan described above automatically triggers the following single or combined plans based on the actual situation on site and the rules configured by the user: 1) Link the sound and light alarm equipment in the factory to remind relevant operators to evacuate the dangerous area; 2) Open emergency evacuation passages for personnel and vehicles; 3) Start the fire emergency power supply and lighting equipment, and notify firefighters of the location and area of ​​the fire; 4) The large screen in the dispatch room will use an electronic map to inform emergency rescue personnel of the type and location of the accident, and whether there are any injured or trapped people at the scene who need treatment; 5) Use vibration or buzzer to remind employees wearing personnel positioning devices to evacuate as soon as possible; 6) Through the Internet of Things linkage equipment PLC controller, interlock shutdown, stop equipment operation; 7) When a dangerous gas leakage accident occurs, immediately start the on-site exhaust equipment.

Citation Information

Patent Citations

  • Method for chemical production fire early warning and fire-fighting linkage and device thereof

    CN107221117A

  • Method and device for analyzing enterprise production hidden dangers based on big data

    CN110516941A

  • Intelligent safety management and control system for field operation of chemical enterprises

    CN112213992A

  • Unattended machine room management method, server and system based on micro-service technology

    CN112685195A

  • Intelligent construction site management system

    CN112954071A

Cited By

  • Intelligent monitoring alarm system and method for escape channel

    CN120656284A