Fire evacuation method based on big language model and electronic equipment

Abstract

The invention relates to the technical field of fire evacuation, in particular to a fire evacuation method based on a big language model and electronic equipment. The evacuation method comprises the following steps: acquiring fire information of a target area, and constructing a real-time combustion scene in a three-dimensional scene model; determining the comburent category and combustion intensity of the combustion area according to the real-time combustion scene; obtaining aerodynamic parameter data of the target area, determining air flow data in the three-dimensional scene model, and constructing a real-time smoke scene in the three-dimensional scene model according to the air flow data, the comburent category and the combustion intensity; determining all selectable evacuation paths according to the passable area; acquiring traffic condition data of each position in the optional evacuation path, and converting the traffic condition data into readable statements; and the readable statements are imported into the large language model, and the optimal evacuation path is selected from all the selectable evacuation paths through the large language model. Direct guidance can be provided for fire evacuation, and evacuation work can be efficiently, safely and orderly carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of fire evacuation, and more particularly, to a fire evacuation method and an electronic device based on a large language model. Background Art

[0002] With the increase in population and the increasing complexity of buildings, fire evacuation simulation has become particularly important. The existing fire evacuation methods still rely highly on on-site manual supervision and guidance. With the increase in population density and the complexity of buildings, the difficulty and complexity of fire evacuation have increased significantly. Fire evacuation has extremely high requirements for evacuation efficiency, and the level of evacuation efficiency is directly related to the safety of public life and property. The limitations of traditional evacuation methods are becoming increasingly obvious in today's complex urban environment.

[0003] In view of this, the present application is specifically proposed. Summary of the Invention

[0004] The first object of the present invention is to provide a fire evacuation method based on a large language model, which can provide direct guidance for fire evacuation, effectively improve the fire evacuation efficiency, and at the same time help reduce the risk of chaos during the evacuation process, and can promote the efficient, safe and orderly progress of the evacuation work.

[0005] The first object of the present invention is to provide an electronic device, which can provide direct guidance for fire evacuation, effectively improve the fire evacuation efficiency, and at the same time help reduce the risk of chaos during the evacuation process, and can promote the efficient, safe and orderly progress of the evacuation work.

[0006] The embodiments of the present invention are implemented as follows:

[0007] A fire evacuation method based on a large language model includes the following steps:

[0008] S1. Obtain the fire information of the target area, and construct a real-time combustion scene in the three-dimensional scene model of the target area according to the fire information.

[0009] S2. Determine the category and combustion intensity of the combustibles in the combustion area according to the real-time combustion scene.

[0010] S3. Obtain the aerodynamic parameter data of the target area, and based on the aerodynamic model, determine the air flow data in the three-dimensional scene model according to the aerodynamic parameter data and the opening and closing states of all channels in the target area, and construct a real-time smoke scene in the three-dimensional scene model according to the air flow data, the category of combustibles and the combustion intensity.

[0011] S4. Determine the passable area according to the real-time combustion scene and the real-time smoke scene, and determine all optional evacuation paths according to the passable area.

[0012] S5. Obtain the passage condition data at each position in the optional evacuation routes, and convert the passage condition data into readable statements.

[0013] S6. Import the readable statements into a large language model, and select the optimal evacuation route from all the optional evacuation routes through the large language model.

[0014] Furthermore, the method for obtaining fire information includes at least one of on-site monitoring and fire detectors.

[0015] Furthermore, S3 further includes: determining the combustion spread prediction data and the smoke spread prediction data according to the three-dimensional scene model, air flow data, type of combustibles, combustion intensity, and type of surrounding combustibles.

[0016] S4 further includes: correcting the passable area according to the combustion spread prediction data and the smoke spread prediction data.

[0017] Furthermore, the passage condition data includes at least one of the distance from the combustibles, body sensation temperature, smoke concentration, visibility, temperature of environmental objects, path length, risk of environmental object collapse, risk of environmental object explosion, risk of hypoxia, risk of body burning, width of the passage, and difficulty of passing through the passage.

[0018] Furthermore, in S4, when determining the passable area, the passage where the door in the closed state and the door body temperature is higher than the safe temperature is determined as an unpassable area.

[0019] Furthermore, in S4, when determining the passable area, the passage where the door with an opening degree satisfying at least single-person passage is located is divided into the passable area.

[0020] Furthermore, the passage condition data includes: temperature of environmental objects.

[0021] Furthermore, the fire evacuation method based on a large language model further includes the step of: repeatedly executing S1 to S6 to update the optimal evacuation route.

[0022] Furthermore, the readable statements include: sentences with a length less than or equal to the length threshold.

[0023] An electronic device, which includes: a memory and a processor.

[0024] The memory stores a computer program, and the computer program is set to execute the above-mentioned fire evacuation method based on a large language model when running.

[0025] The processor is set to execute the above-mentioned fire evacuation method based on a large language model through the computer program.

[0026] The beneficial effects of the technical solutions of the embodiments of the present invention include:

[0027] The fire evacuation method based on the large language model provided by the embodiments of the present invention can simulate real people according to the real-time fire status of the target area to determine the most suitable evacuation path. Since the large language model is used to simulate the selection process of real people on the scene, the selected optimal evacuation path is more acceptable to the public. In the face of a sudden fire situation, the optimal evacuation path is more in line with people's thinking logic and selection tendency. During the process of guiding people to evacuate along the optimal evacuation path, it can effectively reduce the probability of people having excessive stress reactions and irrational behaviors caused by external stimuli in the fire scene, and can effectively reduce the probability of chaos during the evacuation process from the perspectives of physiology and psychology, effectively improving the safety, evacuation efficiency and controllability of the evacuation process.

[0028] Generally speaking, the fire evacuation method based on the large language model provided by the embodiments of the present invention can provide direct guidance for fire evacuation, effectively improve the fire evacuation efficiency, and at the same time help reduce the risk of chaos during the evacuation process, and can promote the efficient, safe and orderly progress of the evacuation work.

[0029] The electronic device provided by the embodiments of the present invention can provide direct guidance for fire evacuation, effectively improve the fire evacuation efficiency, and at the same time help reduce the risk of chaos during the evacuation process, and can promote the efficient, safe and orderly progress of the evacuation work. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0031] Figure 1 It is a schematic flow chart of the fire evacuation method based on the large language model provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0033] Accordingly, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0034] It should be noted that like reference numerals and letters denote like items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0035] As shown in this specification and the claims, unless the context clearly dictates otherwise, words such as "a", "the", etc. do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only imply the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. A method or device may also include other steps or elements.

[0036] The flowcharts used in this specification are used to illustrate the operations performed by the system according to the embodiments of this specification. It can be understood that the operations of each step do not necessarily need to be executed precisely in sequence. Instead, the steps can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more steps can be removed from these processes.

[0037] To overcome the deficiencies in the prior art, please refer to Figure 1 , this embodiment provides a fire evacuation method based on a large language model, and this method includes the following steps:

[0038] S1. Obtain the fire information of the target area, and construct a real-time combustion scene in the three-dimensional scene model of the target area. Among them, the fire information includes: combustion position, combustion area (burning area), combustion intensity (fire size), etc. The three-dimensional scene model is constructed according to the target area to restore the actual appearance of the target area. The three-dimensional scene model also records the object material information of each position in the target area, specifically the combustible type / non-combustible type. The real-time combustion scene refers to the flame combustion situation in the current three-dimensional scene model, including the position where the fire is burning and the flame size at the corresponding position. Optionally, the fire information can be obtained through on-site monitoring and / or fire detectors.

[0039] S2. Determine the combustible category and combustion intensity of the burning area according to the real-time combustion scene. According to the combustible category and combustion intensity, the size of the smoke generated by the combustion can be determined.

[0040] S3. Obtain the aerodynamic parameter data of the target area. Based on the aerodynamic model, determine the air flow data in the three-dimensional scene model according to the aerodynamic parameter data and the opening and closing states of all channels in the target area, and construct a real-time smoke scene in the three-dimensional scene model according to the air flow data, the type of combustibles, and the combustion intensity. The specific type of the aerodynamic parameter data can be flexibly set according to actual needs, and this application does not make specific restrictions. The real-time smoke scene can reflect the influence of the current air flow situation on the smoke distribution, so as to restore the actual distribution of the smoke in the target area in the three-dimensional scene model. Optionally, the real-time smoke scene can also be corrected by combining the detection data of on-site monitoring and / or fire detectors.

[0041] S4. Determine the passable area according to the real-time combustion scene and the real-time smoke scene, and determine all optional evacuation paths according to the passable area. Exemplarily, the area where the fire exceeds the fire threshold (the fire threshold can be preset according to the actual situation), the area where the smoke concentration exceeds the concentration threshold (the concentration threshold can be preset according to the actual situation), the area where the visibility is lower than the visibility threshold (the visibility threshold can be preset according to the actual situation), the area where the body sensation temperature exceeds the body sensation temperature threshold (the body sensation temperature threshold can be preset according to the actual situation), the area where there is an explosion risk of environmental objects, and the area where the passage is too narrow do not belong to the passable area, and this is not limited thereto. The determination rule of the passable area can be further screened in combination with the actual situation.

[0042] S5. Obtain the traffic condition data of each position in the optional evacuation path, and convert the traffic condition data into readable statements. The traffic condition data can be understood as the road conditions of the optional evacuation path, including but not limited to: the distance from the combustibles, the distance from the environmental objects, the body sensation temperature, the smoke concentration, the visibility, the temperature of the environmental objects, the path length, the risk of environmental object collapse, the risk of environmental object explosion, the risk of oxygen deficiency, the risk of body burning, the spaciousness of the passage, and the difficulty of passing through the passage. The readable statement can be a sentence with a length less than or equal to the length threshold, and this is not limited thereto.

[0043] S6. Import the readable statements into the large language model, and select the optimal evacuation path from all the optional evacuation paths through the large language model. The large language model is used to simulate the path selection process of a real person on-site according to the traffic condition data in each optional evacuation path, so as to select the most suitable evacuation path as the optimal evacuation path.

[0044] With this design, when a fire occurs, the most suitable evacuation route (i.e., the optimal evacuation route) can be determined by simulating real people based on the real-time fire status in the target area. Since the large language model is used to simulate the selection process of real people on-site, the selected optimal evacuation route is more acceptable to the public. In the face of such an emergency as a fire, the optimal evacuation route is more in line with people's thinking logic and selection tendency. During the process of guiding people to evacuate along the optimal evacuation route, the probability of excessive stress reactions and irrational behaviors caused by external stimuli in the fire scene can be effectively reduced, and the probability of chaos during the evacuation process can be effectively reduced from the physiological and psychological perspectives, effectively improving the safety, evacuation efficiency, and controllability of the evacuation process.

[0045] Generally speaking, the fire evacuation method based on the large language model provided in this embodiment can provide direct guidance for fire evacuation, effectively improve the fire evacuation efficiency, and at the same time help reduce the risk of chaos during the evacuation process, and can promote the efficient, safe, and orderly progress of the evacuation work.

[0046] Optionally, S3 further includes: determining the combustion spread prediction data and the smoke spread prediction data according to the three-dimensional scene model, air flow data, combustion material category, combustion intensity, and surrounding combustible material category.

[0047] Correspondingly, S4 further includes: correcting the passable area according to the combustion spread prediction data and the smoke spread prediction data.

[0048] With this design, the passable area can be corrected in combination with the development trend of the fire, thereby reducing the risk of people being in danger again during the evacuation process, further improving the safety of the optimal evacuation route, and effectively avoiding secondary injuries.

[0049] Furthermore, in S4, when determining the passable area, the passage where the door in the closed state and the door body temperature (the door body temperature can be collected by the corresponding temperature sensor and is not limited to this) is higher than the safety temperature is determined as an unpassable area.

[0050] Furthermore, in S4, when determining the passable area, the passage where the door with an opening degree that meets at least single-person passage is located is divided into the passable area. At this time, the passage condition data may include: the distance from environmental objects and the temperature of environmental objects. When there is a passage corresponding to a door that meets at least single-person passage, the door body temperature and door frame temperature of this door will be collected as the temperature of environmental objects. Combining the opening degree of this door, the distance between the person and the door body and door frame when passing through this door can be determined. According to the distance from the door body and door frame when passing through and the temperature of the door body and door frame, it can help judge the safety level when passing through this door, thereby providing a selection reference for the selection of the most optimal evacuation route and helping to further reduce the safety risk during the evacuation process.

[0051] Optionally, the large language model-based fire evacuation method further includes the step of: after a fire occurs, repeatedly execute S1 to S6 to update the optimal evacuation path.

[0052] This embodiment also provides an electronic device, which includes: a memory and a processor. The memory stores a computer program, and the computer program is configured to execute the above-mentioned large language model-based fire evacuation method when running. The processor is configured to execute the above-mentioned large language model-based fire evacuation method through the computer program.

[0053] In summary, the large language model-based fire evacuation method provided by the embodiments of the present invention can provide direct guidance for fire evacuation, effectively improve the fire evacuation efficiency, and at the same time help reduce the risk of chaos during the evacuation process, and can promote the efficient, safe and orderly progress of the evacuation work.

[0054] The electronic device provided by the embodiments of the present invention can provide direct guidance for fire evacuation, effectively improve the fire evacuation efficiency, and at the same time help reduce the risk of chaos during the evacuation process, and can promote the efficient, safe and orderly progress of the evacuation work.

[0055] The foregoing is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A fire evacuation method based on large language models, characterized in that, It includes the following steps: S1. Obtain the fire information of the target area, and construct a real-time combustion scene in the three-dimensional scene model of the target area according to the fire information; S2. Determine the combustible category and combustion intensity of the combustion area according to the real-time combustion scene; S3. Obtain the aerodynamic parameter data of the target area, and based on the aerodynamic model, determine the air flow data in the three-dimensional scene model according to the aerodynamic parameter data and the opening and closing states of all channels in the target area, and construct a real-time smoke scene in the three-dimensional scene model according to the air flow data, the combustible category and the combustion intensity; S4. Determine the passable area according to the real-time combustion scene and the real-time smoke scene, and determine all optional evacuation paths according to the passable area; S5. Obtain the passage condition data of each position in the optional evacuation path, and convert the passage condition data into readable statements; S6. Import the readable statements into the large language model, and select the optimal evacuation path from all the optional evacuation paths through the large language model.

2. The fire evacuation method based on a large language model according to claim 1, wherein The acquisition method of the fire information includes at least one of on-site monitoring and fire detectors.

3. The fire evacuation method based on the large language model according to claim 1, wherein S3 further includes: determining the combustion spread prediction data and the smoke spread prediction data according to the three-dimensional scene model, the air flow data, the combustible category, the combustion intensity and the surrounding combustible category; S4 further includes: correcting the passable area according to the combustion spread prediction data and the smoke spread prediction data.

4. The fire evacuation method based on a large language model according to claim 1, wherein The passage condition data includes at least one of the distance from the combustible, the body feeling temperature, the smoke concentration, the visibility, the temperature of the environmental object, the path length, the risk of environmental object collapse, the risk of environmental object explosion, the risk of hypoxia, the risk of body burning, the width of the channel and the difficulty of passing through the channel.

5. The fire evacuation method based on a large language model according to claim 1, wherein In S4, when determining the passable area, the channel where the door in the closed state and the temperature of the door body is higher than the safety temperature is determined as an unpassable area.

6. The fire evacuation method based on a large language model according to claim 1, wherein In S4, when determining the passable area, the channel where the door with an opening degree satisfying at least single-person passage is located is divided into the passable area.

7. The method for fire evacuation based on a large language model according to claim 6, wherein The passage condition data includes: the temperature of the environmental object.

8. The method for fire evacuation based on large language models according to claim 1, characterized in that The fire evacuation method based on the large language model further includes the step of: repeatedly executing S1-S6 to update the optimal evacuation path.

9. The method for fire evacuation based on a large language model according to claim 1, wherein The readable statements include: sentences with a length less than or equal to the length threshold.

10. An electronic device, characterized in that, It includes: A memory and a processor; The memory stores a computer program, and the computer program is set to execute the fire evacuation method based on the large language model according to any one of claims 1-9 when running; The processor is set to execute the fire evacuation method based on the large language model according to any one of claims 1-9 through the computer program.

Images