Early warning processing method based on danger source safety monitoring

Abstract

The invention relates to the technical field of computer language processing, in particular to an early warning processing method based on danger source safety monitoring. Through video monitoring, video networking, AI calculation, big data analysis and system fusion technologies, a method for automatically monitoring, early warning and handling major hazard sources is provided for chemical enterprises and parks, and the technical problems that the major hazard sources of the chemical enterprises are not accurately identified, early warning is not timely, and traceability is not based on afterwards are solved. Early discovery and early warning of major hazard sources of chemical enterprises and chemical industrial parks are realized, an emergency disposal process is started in time, the safety risk prevention and control capability of the chemical industrial enterprise parks is provided, and the occurrence frequency of safety accidents is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of computer language processing, and in particular to an early warning disposal method based on safety monitoring of dangerous sources. Background Art

[0002] There are many hazardous chemicals involved in the production process of chemical enterprises, and the risk of accidents is high. Once an environmental pollution incident occurs, it will have a serious impact on the surrounding ecological environment and the safety of people's lives and property. Therefore, strengthening the overall level of real-time monitoring of major hazardous sources in chemical parks and improving the park's safety risk prevention and control capabilities have become issues that need to be addressed.

[0003] Under the active promotion of national and government policies at all levels, the information construction of hazardous chemicals monitoring and early warning has achieved certain phased results, but there are still certain shortcomings. It is urgent to establish a set of information-based disposal methods for safety early warning and disposal of major hazardous sources. Summary of the invention

[0004] The present invention discloses an early warning and disposal method based on safety monitoring of hazardous sources. Through video monitoring, visual network, AI computing, big data analysis and system fusion technology, a method for automatic monitoring, early warning and disposal of major hazardous sources is provided for chemical enterprises and industrial parks, so as to solve the technical pain points and difficult problems of inaccurate identification of major hazardous sources, untimely early warning and unfounded subsequent tracing of sources in chemical enterprises, and realize early discovery, early warning and timely initiation of emergency disposal procedures for major hazardous sources in chemical enterprises and chemical parks, so as to provide chemical enterprises and industrial parks with safety risk prevention and control capabilities and reduce the frequency of safety accidents.

[0005] A method for early warning and disposal based on safety monitoring of hazardous sources comprises the following steps: S1. Use Hikvision video surveillance equipment to obtain monitoring video images of hazardous chemical sites, use the visual network platform to collect and store image data, and build the Xuanyuan large-scale model AI computing center to analyze and calculate the collected video images, identify the safety risk information in the images, convert the existing hazard source risk information into text, and transmit it to the early warning and disposal platform in the form of language and text, and then automatically alarm on the early warning platform, start the hazard source early warning and disposal plan, and realize automatic calling of the system to call the emergency duty personnel in time for on-site disposal; S2. The monitoring equipment uses Hikvision video surveillance cameras, and builds a visual network video image acquisition platform based on the camera video protocol, thereby realizing video data acquisition, protocol conversion, and storage, and providing the collected images for transmission to the AI ​​center by developing an API interface; S3. The visual networking platform is based on the built-in encoding format of Hikvision video equipment to build a visual networking video image acquisition platform, combined with video protocol conversion technology, to achieve video IT system playback through protocol conversion, and to achieve real-time data conversion and storage; S4. The AI ​​center builds the Industrial Vision Xuanyuan large-scale AI computing center, develops image recognition algorithm models based on the image files collected and converted by the visual network, and conducts a large amount of hazard source data annotation training, so as to achieve accurate identification and calculation of major hazard signal risks, and then discover the safety hazard signals in the video files, and convert and transmit the signals to achieve early identification of safety risks of hazard sources; S5. The model training is to obtain a large number of hazardous source safety risk images for data annotation based on AI computing power, and to conduct large-scale centralized training of the algorithm model through the annotated images. Through the standards and calculations of massive data, the accuracy of the computing model recognition is achieved, and the computing model is further put into practical use; S6. The automatic handling platform is based on GIS maps to develop an emergency command system on the map, which includes a series of layer information such as emergency alarms, dangerous source signal sites, monitoring video points, emergency material points, emergency rescue roads, and emergency communication equipment, so as to display all information through one map; the system automatically connects emergency materials, dangerous source safety risk information, emergency handling plan information, and emergency communication equipment, so as to realize automatic identification of warnings, automatic call-up of communication equipment for emergency duty personnel communication, automatic reporting and handling of emergency events, and automatic matching of emergency plans to provide emergency handling plans. Duty personnel can quickly realize early warning and handling through the emergency plan process, reduce unnecessary process delays, and reduce the incidence of safety accidents; S7. The emergency response process is based on GIS maps, and a software emergency response platform is developed. Artificial intelligence technology, big data technology, and digital twin technology are used to realize automatic association of the entire emergency response process and intelligent handling. First, the automatic matching and rapid positioning capabilities of artificial intelligence are used to realize accurate identification of dangerous signals; secondly, emergency plan information is screened and emergency duty personnel information is queried through big data analysis to provide a basis for emergency communication quick dialing. Digital twin technology is used again to mark information such as various dangerous signal sites, emergency material locations, and emergency rescue roads. Multiple layers are displayed through one picture to realize the interconnection between locations, materials, and roads, providing decision support for the automatic alarm handling process, realizing automatic system judgment, automatic warning, and automatic handling, reducing personnel participation, improving alarm handling efficiency, and reducing the frequency of dangerous occurrences; S8. The artificial intelligence technology is used to develop image recognition, intelligent matching, and precise positioning algorithms. Through data annotation, the corresponding algorithms are trained to achieve higher requirements for algorithm calculation accuracy. Through the application of the trained algorithms, the scene application of the automatic disposal process is carried out to achieve automatic matching and rapid positioning of the disposal process. At the same time, high-precision recognition of video images is achieved. Through the application of artificial intelligence technology, the early warning disposal process is liberated and the accuracy and timeliness of the emergency disposal process are improved. S9. The danger is eliminated by using the above technology to achieve rapid positioning and automatic disposal. The disposal process is automatically evaluated, recorded, archived, and knowledge base data is formed. At the same time, the capture of video equipment and the computing power of artificial intelligence technology in the above process are used to automatically archive the on-site disposal operations, improve the emergency plan disposal process and disposal operations; and automatically review and summarize the disposal results to form a summary report. The above process completes the entire early warning disposal process and realizes automatic closed-loop disposal of the entire process.

[0006] Beneficial effects of the present invention: A method for early warning and disposal based on safety monitoring of hazardous sources, the method includes hardware equipment, computing power, model training and connection to an emergency system; the automatic early warning disposal method realizes video equipment data collection and transmission through IoT devices, realizes model training, and then realizes accurate identification of video data. After discovering dangerous signals, it promptly initiates the plan and disposal, automatically notifies the processing personnel, completes on-site disposal of hazardous sources, and relieves the emergency, so as to achieve early discovery and early disposal of hazardous sources. The computing power construction realizes real-time collection of device video data, protocol processing, and data upload through IoT acquisition technology and binding of video devices, develops AI video recognition model capabilities, realizes high-precision identification of video data, and can quickly extract hazardous source signals to realize alarm disposal of hazardous sources. The model training develops an AI video model training center to collect a large amount of video data for training AI video recognition capabilities and improve model recognition accuracy. The emergency disposal establishes an early warning automatic disposal system, integrates GIS maps, communication equipment and other technical contents, and develops a GIS-based automatic disposal system. Figure 1 The handling and dispatching capabilities shown in the picture involve one-click dispatch of emergency materials, one-click calling of on-duty personnel, one-click start of emergency rescue and other early warning automatic handling process technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 This is a schematic diagram of an application of an embodiment of the present invention. DETAILED DESCRIPTION

[0008] To solve the above problems, the present invention provides an early warning disposal method based on safety monitoring of dangerous sources. The overall idea is as follows: Figure 1 The specific steps of the technical method are as follows S1-S9.

[0009] S1. Use Hikvision video surveillance equipment to obtain monitoring video images of hazardous chemical sites, use the visual network platform to collect and store image data, and build the Xuanyuan large model AI computing center to analyze and calculate the collected video images, identify the safety risk information in the images, convert the existing hazardous source risk information into text, and transmit it to the early warning and disposal platform in the form of language and text, and then automatically alarm on the early warning platform, start the hazardous source early warning and disposal plan, and realize automatic calling of the system to call the emergency duty personnel in time for on-site disposal.

[0010] S2. Furthermore, the device uses Hikvision video surveillance cameras and builds a visual network video image acquisition platform based on the camera video protocol, thereby realizing video data acquisition, protocol conversion, and storage, and providing the collected images for transmission to the AI ​​center by developing an API interface.

[0011] S3. Furthermore, the visual networking platform builds a visual networking video image acquisition platform based on the built-in encoding format of Hikvision video equipment, studies video protocol conversion technology, realizes video IT system playback through protocol conversion, and realizes real-time data conversion and storage.

[0012] S4. Furthermore, the AI ​​center builds the Industrial Vision Xuanyuan large-model AI computing center, develops image recognition algorithm models based on image files collected and converted by the visual network, and conducts a large amount of hazard source data annotation training, so as to achieve accurate identification and calculation of major hazard signal risks, and then discover safety hazard signals in video files, and convert and transmit the signals to achieve early identification of safety risks of hazardous sources.

[0013] S5. Furthermore, the model training obtains a large number of hazardous source safety risk images for data annotation based on AI computing power, and conducts large-scale centralized training of the above algorithm model through the annotated images. Through the standards and calculations of massive data, the accuracy of the above computing power model recognition is achieved, and the computing power model is further put into practical use.

[0014] S6. Further, the automatic handling platform, based on GIS map, develops an emergency command system on the map, including a series of layer information such as emergency alarm, hazard signal site, monitoring video point, emergency material point, emergency rescue road, emergency communication equipment, etc., so as to display all information through one map. Through the system automatically connecting emergency materials, hazard safety risk information, emergency handling plan information, and emergency communication equipment, the system can automatically identify warnings, automatically call up communication equipment for emergency duty personnel to communicate, automatically report and handle emergency events, and automatically match emergency plans to give emergency handling plans. Duty personnel can quickly realize warnings and disposal through the emergency plan process, reduce unnecessary process delays, and reduce the incidence of safety accidents.

[0015] S7. Furthermore, in the emergency response process, based on GIS maps, a software emergency response platform is developed, which utilizes artificial intelligence technology, big data technology, and digital twin technology to realize automatic association of the entire emergency response process and intelligent handling. First, the automatic matching and rapid positioning capabilities of artificial intelligence are utilized to realize accurate identification of dangerous signals. Secondly, emergency plan information is screened and emergency duty personnel information is queried through big data analysis to provide a basis for emergency communication quick dialing. Digital twin technology is used again to mark information such as various dangerous signal sites, emergency material locations, and emergency rescue roads. Multiple layers are displayed through one picture to realize the interconnection between locations, materials, and roads, providing decision support for the automatic alarm handling process, realizing automatic system judgment, automatic warning, and automatic handling, reducing personnel participation, improving alarm handling efficiency, and reducing the frequency of dangerous events.

[0016] S8. Furthermore, the artificial intelligence technology is used to develop image recognition, intelligent matching, and precise positioning algorithms. Through data annotation, the corresponding algorithms are trained to achieve higher requirements for algorithm calculation accuracy. Through the application of the trained algorithms, scene applications of the automatic disposal process are carried out to achieve automatic matching and rapid positioning of the disposal process. At the same time, high-precision recognition of video images is achieved. Through the application of artificial intelligence technology, the early warning disposal process is liberated, and the accuracy and timeliness of the emergency disposal process are improved.

[0017] S9. Further, the danger is eliminated, and the above technology is used to achieve rapid positioning and automatic disposal. The disposal process is automatically evaluated, recorded, archived, and knowledge base data is formed. At the same time, the capture of video equipment and the computing power of artificial intelligence technology in the above process are used to complete the automatic archiving of on-site disposal operations, improve the emergency plan disposal process, disposal operations, etc. The disposal results are automatically reviewed and summarized to form a summary report. The entire early warning disposal process is completed through the above process, and the entire process is automatically closed-loop, reducing human participation, reducing operational errors in the disposal process, and making the early warning disposal more intelligent, automated, and efficient.

[0018] like Figure 2 As shown in the figure, the method for monitoring and handling hazardous sources based on equipment, computing power, AI models, and automatic handling includes the following steps: Based on the safety monitoring, early warning and disposal requirements of hazardous sources, select the corresponding video surveillance equipment for on-site installation and debugging.

[0019] Through IoT data collection, video data can be uploaded to the storage resource pool and computing resource pool in real time; Through the model training center, video recognition capabilities are trained through the large amount of video data collected to provide recognition accuracy; through the computing power pool, the collected data is used to identify the risk signals of hazardous sources using the trained video recognition model; through risk signal identification, the hazardous source signals of dangerous situations are judged to be dangerous, and they are automatically extracted, such as the control system to actively alarm; the early warning automatic disposal system, when receiving the alarm signal, automatically calculates and activates the corresponding emergency plan to realize the activation of the plan; the system automatically determines to start emergency communication according to the risk level, sends a notification to the relevant responsible person, and reports to the relevant person in charge to handle the dangerous incident on the spot; through the actual market processing warning identification, it is determined whether the danger signal has been lifted, and at the same time, the relevant personnel are notified to report the results of the safety incident disposal (incident assessment, incident disposal, incident analysis) to eliminate safety risks.

[0020] The technical points of the present invention are: 1. Reduce the incidence of safety incidents: The present invention uses technologies such as video surveillance, AI video recognition, model training, communication equipment integration, one-click automatic calling, and emergency material positioning to achieve beneficial effects such as active alarm, automatic disposal, and timely resolution of hazardous sources, providing users with early detection, early alarm, and early disposal of hazardous source safety incidents, reducing the frequency of hazardous source safety incidents and improving enterprise production safety.

[0021] 2. Improve work efficiency and accuracy: The present invention is based on the early warning disposal method of dangerous source safety monitoring. The beneficial effect is to realize the automatic monitoring, autonomous inspection, automatic alarm, disposal and other technologies of dangerous sources, and solve the problem that the traditional chemical plant dangerous source monitoring method requires manual inspection, which is inefficient and prone to errors. It realizes the automated management of dangerous source monitoring and improves work efficiency and accuracy.

[0022] 3. Reduce management costs: The present invention is based on the early warning and disposal method of hazardous source safety monitoring. The beneficial effect is that through the technical combination of equipment + AI computing power + emergency system + communication + personnel, it can achieve all-weather, indiscriminate real-time monitoring of hazardous sources, and solve the problem that the traditional chemical plant hazardous source monitoring method requires a lot of manpower and material resources, and cannot be monitored and warned in real time, thereby reducing the waste of manpower and material resources and reducing management costs for enterprises.

[0023] Glossary Emergency communications: including emergency telephones, video conversations, satellite phones, smart broadcasting and other communication equipment.

[0024] AI algorithm: Algorithm warehouse in the field of AI artificial intelligence technology, mainly involving the construction of video and image recognition algorithms.

Claims

1. A method for early warning and disposal based on safety monitoring of hazardous sources, characterized in that The steps include: S1. Use Hikvision video surveillance equipment to obtain monitoring video images of hazardous chemical sites, use the visual network platform to collect and store image data, and build the Xuanyuan large-scale model AI computing center to analyze and calculate the collected video images, identify the safety risk information in the images, convert the existing hazard source risk information into text, and transmit it to the early warning and disposal platform in the form of language and text, and then automatically alarm on the early warning platform, start the hazard source early warning and disposal plan, and realize automatic calling of the system to call the emergency duty personnel in time for on-site disposal; S2. The monitoring equipment uses Hikvision video surveillance cameras, and builds a visual network video image acquisition platform based on the camera video protocol, thereby realizing video data acquisition, protocol conversion, and storage, and providing the collected images for transmission to the AI ​​center by developing an API interface; S3. The visual networking platform is based on the built-in encoding format of Hikvision video equipment to build a visual networking video image acquisition platform, combined with video protocol conversion technology, to achieve video IT system playback through protocol conversion, and to achieve real-time data conversion and storage; S4. The AI ​​center builds the Industrial Vision Xuanyuan large-scale AI computing center, develops image recognition algorithm models based on the image files collected and converted by the visual network, and conducts a large amount of hazard source data annotation training, so as to achieve accurate identification and calculation of major hazard signal risks, and then discover the safety hazard signals in the video files, and convert and transmit the signals to achieve early identification of safety risks of hazard sources; S5. The model training is to obtain a large number of hazardous source safety risk images for data annotation based on AI computing power, and to conduct large-scale centralized training of the algorithm model through the annotated images. Through the standards and calculations of massive data, the accuracy of the computing model recognition is achieved, and the computing model is further put into practical use; S6. The automatic handling platform is based on GIS maps to develop an emergency command system on the map, which includes a series of layer information such as emergency alarms, hazard signal sites, surveillance video points, emergency material points, emergency rescue roads, and emergency communication equipment, so as to display all information through one map; the system automatically connects emergency materials, hazard safety risk information, emergency handling plan information, and emergency communication equipment, so as to realize automatic identification of warnings, automatic call-up of communication equipment for emergency duty personnel communication, automatic reporting and handling of emergency events, and synchronous automatic matching of emergency plans to provide emergency handling plans. Duty personnel can quickly realize early warning and handling through the emergency plan process; S7. The emergency response process is based on GIS maps, and a software emergency response platform is developed. Artificial intelligence technology, big data technology, and digital twin technology are used to realize automatic association of the entire emergency response process and intelligent handling. First, the automatic matching and rapid positioning capabilities of artificial intelligence are used to realize accurate identification of dangerous signals; secondly, emergency plan information is screened and emergency duty personnel information is queried through big data analysis to provide a basis for emergency communication quick dialing. Digital twin technology is used again to mark information such as various dangerous signal sites, emergency material locations, and emergency rescue roads. Multiple layers are displayed through one picture to realize the interconnection between locations, materials, and roads, provide decision support for the automatic alarm handling process, realize system automatic judgment, automatic warning, automatic handling, and reduce personnel participation; S8. The artificial intelligence technology is used to develop image recognition, intelligent matching, and precise positioning algorithms. Through data annotation, the corresponding algorithms are trained to achieve higher requirements for algorithm calculation accuracy. Through the application of trained algorithms, the scene application of the automatic treatment process is carried out to achieve automatic matching and rapid positioning of the treatment process, and at the same time, high-precision recognition of video images is achieved; S9. The danger is eliminated by using the above technology to achieve rapid positioning and automatic disposal. The disposal process is automatically evaluated, recorded, archived, and knowledge base data is formed. At the same time, the capture of video equipment and the computing power of artificial intelligence technology in the above process are used to automatically archive the on-site disposal operations, improve the emergency plan disposal process and disposal operations; and automatically review and summarize the disposal results to form a summary report. The above process completes the entire early warning disposal process and realizes automatic closed-loop disposal of the entire process.

Images