Campus fire monitoring method and system

Abstract

The invention relates to a campus fire monitoring method and system, and relates to the technical field of fire monitoring, and the method comprises the steps: obtaining smoke values sent by all smoke sensors in a campus and image information sent by a camera device; determining whether a fire occurs based on the smoke value and the image information; determining a fire occurrence area based on the position information of the corresponding smoke sensor and the image information sent by the fire camera after determining that the fire occurs; carrying out dynamic simulation on the fire spreading trend based on the digital twinborn model and in combination with a smoke diffusion prediction algorithm to generate a dynamic thermodynamic diagram; dividing an affected area and a safe area based on the dynamic thermodynamic diagram; a display device and a broadcasting device in the affected area are controlled to broadcast the first warning information; a display device and a broadcasting device in the safety area are controlled to broadcast second warning information; and monitoring the escape condition of people in each area and the fire spreading condition in real time, and updating the first warning information. According to the invention, personnel in different fire areas can obtain targeted information, and the fire response efficiency of the personnel is improved.

Description

Technical Field

[0001] This application relates to the technical field of fire monitoring, and in particular, to a campus fire monitoring method and system. Background Art

[0002] The building structures on campus are complex and the personnel flow is frequent. Once a fire breaks out, it is extremely likely to cause heavy casualties and property losses. Traditional campus fire monitoring means mainly rely on smoke sensors and simple alarm devices.

[0003] Although smoke sensors can detect changes in smoke concentration to a certain extent, relying solely on smoke sensors cannot accurately determine the location of the fire, the spread trend, and the affected area. How to solve the above problems is a technical problem that those skilled in the art need to overcome. Summary of the Invention

[0004] In order to at least partially solve the above technical problems, this application provides a campus fire monitoring method and system.

[0005] In a first aspect, a campus fire monitoring method provided by this application adopts the following technical solution.

[0006] A campus fire monitoring method includes: Obtaining the smoke values sent by all smoke sensors in the campus and the image information sent by the camera devices; Judging whether a fire has occurred based on the smoke values and the image information; after determining that a fire has occurred, determining the fire occurrence area based on the position information of the corresponding smoke sensor and the image information sent by the fire camera; Dynamically simulating the fire spread trend based on the digital twin model and in combination with the smoke diffusion prediction algorithm to generate a dynamic heat map; dividing the affected area and the safe area based on the dynamic heat map; Controlling the display devices and the broadcasting devices in the affected area to broadcast a first warning message; controlling the display devices and the broadcasting devices in the safe area to broadcast a second warning message; wherein, the first warning message includes a fire occurrence reminder, the distance to the fire occurrence location, and a recommended route; Real-time monitoring of the escape situation of personnel in each area and the fire spread situation to update the first warning message.

[0007] By adopting the above technical solutions, the smoke value sent by the smoke sensors on campus and the image information sent by the camera devices are obtained. Based on the smoke value and the image information, it is judged whether a fire has occurred, and the fire occurrence area is determined by using the position information of the corresponding smoke sensors and the images. With the help of the digital twin model and the smoke diffusion prediction algorithm, the fire spread trend is dynamically simulated and a dynamic heat map is generated, and then the affected area and the safe area are divided. Different warning messages are controlled and displayed on the display devices and the broadcasting devices in the affected area and the safe area respectively, and the first warning message includes a fire occurrence reminder, the distance from the fire occurrence location, and a recommended route, which can enable personnel in different areas to obtain targeted information and improve the efficiency of personnel in dealing with fires.

[0008] Optionally, dynamically simulating the fire spread trend based on the digital twin model and combining with the smoke diffusion prediction algorithm to generate a dynamic heat map includes: Inputting the smoke values sent by all the smoke sensors on campus, the image information sent by the camera devices, and the position information of the corresponding smoke sensors obtained in real time into the digital twin model of the campus to complete the initial data loading of the digital twin model; the digital twin model of the campus imports the three-dimensional space structure data of the campus and the positions of the fire-fighting facilities; the three-dimensional space structure data includes the building layout and the position information of the passage areas; the fire-fighting facilities include fire extinguishers and fire hydrants; Determine the parameters of the smoke diffusion prediction algorithm according to the material and ventilation conditions of the campus buildings; Embed the smoke diffusion prediction algorithm with set parameters into the digital twin model, and gradually simulate the diffusion process of smoke in the campus space in time series starting from the fire occurrence area; output the simulation results once at preset time intervals; the simulation results include the smoke concentration distribution in each area of the campus and the scope affected by the fire; According to the simulation results of the fire spread trend, quantify and classify the smoke concentration and fire risk level in each area of the campus into the first level, the second level, and the third level; Map the information after quantization and classification into the three-dimensional space of the digital twin model, and represent the fire risk level of each area with different colors and brightness to generate a dynamic heat map.

[0009] Optionally, determining the parameters of the smoke diffusion prediction algorithm according to the material and ventilation conditions of the campus buildings includes: Obtain the material coefficients of different areas of the campus buildings for hindering or promoting the diffusion of smoke based on the materials of the campus buildings; the materials of the campus buildings include: wall materials, door and window materials; Determine the diffusion coefficient of smoke in the campus based on the current wind speed of the campus and the material coefficients; According to the initial smoke value detected by the smoke sensors in the fire occurrence area, combined with the size and brightness of the flames in the fire scene images captured by the camera device, the smoke generation rate in the initial stage of the fire, i.e., the initial smoke source intensity, is estimated using relevant formulas in combustion science. Embed the smoke diffusion prediction algorithm with set parameters into the digital twin model, and starting from the fire occurrence area, gradually simulate the diffusion process of smoke in the campus space according to the time series, including: In the spatial grid of the digital twin model, set the initial smoke concentration of the grid corresponding to the fire occurrence area to the value calculated based on the initial smoke source intensity, and set the initial smoke concentration of the grids in other areas to 0. Starting from the initial moment, within each time step, for each spatial grid, calculate the smoke concentration of the grid at the end of the current time step based on the smoke diffusion flux between adjacent grids according to Fick's second law of diffusion; and continuously monitor the change of the flames through the camera device to dynamically adjust the smoke source intensity according to the fire development situation.

[0010] Optionally, the method for generating the recommended route includes: Perform grid processing on the campus building structure and passage areas in the digital twin model, set a unique grid identifier for each grid and endow it with a passage attribute; the passage attribute includes whether it can be passed through. Generate an initial recommended escape route based on the grid identifier where the fire occurrence area is located and the grid identifier corresponding to the safety exit. Based on the image data collected by the camera device, determine the number of people in each grid divided by the digital twin model and determine the population density of the corresponding grid based on the grid area. Trigger the route adjustment mechanism when the population density is greater than the preset density value; the route adjustment mechanism is used to generate several alternative routes and select the optimal alternative route based on the route length of the alternative routes, the distance between the route and the fire occurrence location, and the fire fighting facilities equipped around the route. Control the display devices and broadcasting devices in the areas where the population density is greater than the preset density value to broadcast the optimal alternative route.

[0011] Optionally, the route adjustment mechanism is used to generate several alternative routes and select the optimal alternative route based on the route length of the alternative routes, the distance between the route and the fire occurrence location, and the fire fighting facilities equipped around the route, including: For each alternative route, according to the position information of each node in the digital twin model, calculate the sum of the distances between all the nodes passed by the route to obtain the length of the alternative route. Taking the central grid identifier of the fire occurrence area as a reference, calculate the Euclidean distance from each node on each alternate route to the central grid of the fire occurrence area, and take the minimum value as the distance between the alternate route and the fire location; According to the location information of the fire-fighting facilities, determine the coverage range of the fire-fighting facilities around each alternate route; calculate the proportion of the coverage range of the fire-fighting facilities in the length of the alternate route to evaluate the allocation of fire-fighting facilities around the route; Calculate a comprehensive score based on considering the route length, the distance between the route and the fire location, and the allocation of fire-fighting facilities around the route, and select the route with the highest comprehensive score as the optimal alternate route.

[0012] Optionally, the method for generating the alternate route includes: S601. Take the grid identifier of the fire occurrence area as the starting node s and the grid identifier corresponding to the safety exit as the target node t; create a queue Q, a distance array d, and a predecessor node array p; the queue Q is used to store the nodes to be explored, and the nodes in the queue are sorted in ascending order of path weight; the predecessor node array p is used to record the path, and initially the predecessor node array p is empty; the distance value d of the starting node in the distance array is set to 0; S602. Add the starting node s to the queue Q; when the priority queue Q is not empty, take out the node u with the smallest path weight from the queue Q; if u is the target node t, jump to S604; for each adjacent node v of the node u, calculate the new path weight from the node u to the node v ; where, ; is the path length from the node u to the node v, obtained according to the actual distance between nodes in the digital twin model; is the average population density on the path segment uv; is the population density influence weight, which is a constant coefficient; if the new path weight from passing through the node u to the node v is less than the currently recorded distance value of the node v , then update the distance value , and set the predecessor node of the node v to u; if the node v is not in the queue Q, add it to the queue Q; S603. Repeat S602, continuously take out nodes from the queue Q for expansion until the target node t is taken out; S604. When the target node t is taken out, generate a path from the starting node s to the target node t through the predecessor node array p to obtain an alternate route.

[0013] Optionally, after dividing the affected area and the safe area based on the dynamic heat map, the method further includes: Based on the location information of the fire - dangerous area, determine the fire - fighting facilities within the area and the preset range around it in the digital - twin model; Query the status information of the fire - fighting facilities; If it is detected that a fire - fighting facility has a fault, in the visualization interface of the digital - twin model, mark the faulty fire - fighting facility with a flashing red icon.

[0014] In a second aspect, a campus fire - monitoring system provided by the present application adopts the following technical solution.

[0015] A campus fire - monitoring system includes: A first processing module, configured to: obtain the smoke values sent by all smoke sensors in the campus and the image information sent by the camera devices; A second processing module, configured to: determine whether a fire has occurred based on the smoke values and the image information; after determining that a fire has occurred, determine the fire - occurrence area based on the location information of the corresponding smoke sensors and the image information sent by the fire cameras; A third processing module, configured to: dynamically simulate the fire - spreading trend based on the digital - twin model and in combination with the smoke - diffusion prediction algorithm to generate a dynamic heat map; divide the affected area and the safe area based on the dynamic heat map; A fourth processing module, configured to: control the display devices and the broadcasting devices in the affected area to broadcast a first warning message; control the display devices and the broadcasting devices in the safe area to broadcast a second warning message; wherein, the first warning message includes a fire - occurrence reminder, the distance to the fire - occurrence location, and a recommended route; A fifth processing module, configured to: monitor the escape situation of personnel in each area and the fire - spreading situation in real - time to update the first warning message.

[0016] In summary, the technical solution of the present application has at least the following technical effects: By obtaining the smoke values sent by the smoke sensors in the campus and the image information sent by the camera devices, determine whether a fire has occurred based on the smoke values and the image information, and use the location information of the corresponding smoke sensors and the images to determine the fire - occurrence area. With the help of the digital - twin model and the smoke - diffusion prediction algorithm, dynamically simulate the fire - spreading trend and generate a dynamic heat map, and then divide the affected area and the safe area. It enables personnel in different areas to obtain targeted information and improves the efficiency of personnel in dealing with fires. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flowchart of a campus fire - monitoring method according to an embodiment of the present application; Figure 2 is a system block diagram of a campus fire - monitoring method according to an embodiment of the present application; In the figure, 201 is the first processing module; 202 is the second processing module; 203 is the third processing module; 204 is the fourth processing module; 205 is the fifth processing module. Specific implementation manner

[0018] The following will further illustrate the present application in conjunction with the attached Figure 1-2 drawings and specific embodiments: An embodiment of the present application discloses a campus fire monitoring method, including the following steps: Step 101: Obtain the smoke values sent by all smoke sensors in the campus and the image information sent by the imaging device. Specifically, a smoke sensor is a device used to detect the smoke concentration in the environment. It senses smoke particles through internal sensitive elements. When the smoke concentration reaches a certain threshold, it converts the smoke concentration information into an electrical signal and sends it out. The imaging device refers to devices such as cameras installed at different positions in the campus, which can capture the image information of the campus scene in real time. In the fire monitoring scenario, the images captured by the imaging device can be used to assist in judging whether a fire has occurred and to determine the fire occurrence area in combination with the information of the smoke sensors.

[0019] Step 102: Judge whether a fire has occurred based on the smoke value and the image information; after determining that a fire has occurred, determine the fire occurrence area based on the position information of the corresponding smoke sensor and the image information sent by the fire camera.

[0020] Step 103: Dynamically simulate the fire spread trend based on the digital twin model and in combination with the smoke diffusion prediction algorithm to generate a dynamic heat map; divide the affected area and the safe area based on the dynamic heat map. The digital twin model is a digital virtual mapping of the real physical system, which can reflect the state, behavior, and performance of the real system in real time. In campus fire monitoring, the digital twin model constructs information such as the three-dimensional space structure, building layout, passage areas, and fire-fighting facility locations of the campus, and through the interaction with real-time data, simulates the fire spread trend. The smoke diffusion prediction algorithm is an algorithm based on principles such as combustion science and fluid mechanics, combined with various factors such as campus building materials, ventilation conditions, and wind speed, to predict the smoke diffusion situation in the campus space. Embedding the smoke diffusion prediction algorithm in the digital twin model can start from the fire occurrence area and simulate the smoke diffusion process according to the time series. The dynamic heat map is a visualization graph that intuitively shows the fire danger degree of each area of the campus with different colors and brightness. It is generated by mapping the smoke concentration and fire danger degree of each area of the campus, which are quantified and classified according to the simulation results of the smoke diffusion prediction algorithm, into the three-dimensional space of the digital twin model.

[0021] Step 104: Control the display devices and broadcasting devices in the affected area to broadcast the first warning message; control the display devices and broadcasting devices in the safe area to broadcast the second warning message; wherein, the first warning message includes a reminder of the fire occurrence, the distance to the fire occurrence location, and a recommended route.

[0022] Step 105: Monitor the evacuation situation of personnel in each area and the spread of the fire in real time to update the first warning message.

[0023] Specifically, by obtaining the smoke values sent by the smoke sensors on campus and the image information sent by the camera devices, judge whether a fire has occurred based on the smoke values and image information, and use the position information of the corresponding smoke sensors and the images to determine the fire occurrence area. Dynamically simulate the fire spread trend and generate a dynamic heat map by means of a digital twin model and a smoke diffusion prediction algorithm, and then divide the affected area and the safe area. Control the display devices and broadcasting devices in the affected area and the safe area to broadcast different warning messages respectively, and the first warning message includes a reminder of the fire occurrence, the distance to the fire occurrence location, and a recommended route, which can enable personnel in different areas to obtain targeted information and improve the efficiency of personnel in dealing with fires. Monitor the evacuation situation of personnel in each area and the spread of the fire in real time to update the first warning message, which can make the evacuation guidance always adapt to the changing actual situation, further ensure that personnel can adjust the evacuation strategy in time, minimize the casualties and property losses caused by the fire, and improve the intelligence of campus fire monitoring and response.

[0024] As a specific implementation of a campus fire monitoring method, 2. Dynamically simulate the fire spread trend and generate a dynamic heat map based on a digital twin model and in combination with a smoke diffusion prediction algorithm, including: Input the smoke values sent by all the smoke sensors on campus, the image information sent by the camera devices, and the position information of the corresponding smoke sensors obtained in real time into the digital twin model of the campus to complete the initial data loading of the digital twin model; the digital twin model of the campus imports the three-dimensional space structure data of the campus and the positions of the fire-fighting facilities; the three-dimensional space structure data includes building layout and passage area position information; the fire-fighting facilities include fire extinguishers and fire hydrants; Determine the parameters of the smoke diffusion prediction algorithm according to the material and ventilation conditions of the campus buildings; Embed the smoke diffusion prediction algorithm with set parameters into the digital twin model, and gradually simulate the diffusion process of smoke in the campus space in time series starting from the fire occurrence area; output the simulation results once at preset time intervals; the simulation results include the smoke concentration distribution in each area of the campus and the scope affected by the fire; According to the simulation results of the fire spread trend, the smoke concentration and fire risk level of each area on campus are quantitatively classified into the first level, the second level, and the third level. Map the information after quantitative classification into the three-dimensional space of the digital twin model, and represent the fire risk level of each area with different colors and brightness to generate a dynamic heat map.

[0025] Specifically, the first level (high risk): When the smoke concentration C in a certain area ≥ C1 (C1 is the high-risk smoke concentration threshold set according to fire safety standards and the actual situation of the campus), or the area is within r1 meters from the fire occurrence location (r1 is the distance threshold estimated and set according to the possible rapid spread range of the fire), then this area is classified into the first level. The second level (medium risk): When C2 ≤ smoke concentration C < C1 (C2 is the set medium-risk smoke concentration threshold, C2 < C1), and the area is between r1 meters and r2 meters from the fire occurrence location (r2 is the distance threshold estimated and set according to the possible medium spread range of the fire, r2 > r1), then this area is classified into the second level. The third level (low risk): When the smoke concentration C < C2, and the area is more than r2 meters from the fire occurrence location, this area is classified into the third level.

[0026] By inputting the smoke values sent by the smoke sensors in the campus, the image information sent by the camera devices, and the position information of the corresponding smoke sensors into the digital twin model, the initial data loading is completed. Import the three-dimensional space structure data of the campus and the positions of fire-fighting facilities, know the building layout and passage areas, and can clarify the possible paths of smoke diffusion. The position information of fire-fighting facilities helps to evaluate their potential impact on fire control. Determine the parameters of the smoke diffusion prediction algorithm according to the campus building materials and ventilation conditions. For example, different wall and door / window materials have different effects on hindering or promoting smoke diffusion. Embed the algorithm with set parameters into the digital twin model, simulate the smoke diffusion process in time series starting from the fire occurrence area, and output the results at preset time intervals, which can present the fire development trend in real time. Quantitatively classify the simulation results and map them into the three-dimensional space of the digital twin model to generate a dynamic heat map, which can enable managers to intuitively and quickly understand the smoke concentration and fire risk level of each area on campus, and then quickly divide the affected areas and safe areas, providing a basis for targeted warning information broadcasting and personnel evacuation.

[0027] As a specific implementation of a campus fire monitoring method, determine the parameters of the smoke diffusion prediction algorithm according to the materials and ventilation conditions of the campus buildings, including: Obtain the material coefficients of different areas of the campus building materials that hinder or promote smoke diffusion based on the campus building materials; the campus building materials include: wall materials, door / window materials; Determine the diffusion coefficient of smoke in the campus based on the current wind speed on campus and the material coefficients; According to the initial smoke value detected by the smoke sensors in the fire-occurring area, combined with the size and brightness of the flames in the fire scene images captured by the camera device, use relevant combustion formulas to estimate the smoke generation rate at the initial stage of the fire, that is, the initial smoke source intensity; Embed the smoke diffusion prediction algorithm with set parameters into the digital twin model, and starting from the fire-occurring area, gradually simulate the diffusion process of smoke in the campus space according to the time series, including: In the spatial grid of the digital twin model, set the initial smoke concentration of the grid corresponding to the fire-occurring area to the value calculated based on the initial smoke source intensity, and set the initial smoke concentration of the grids in other areas to 0; Starting from the initial moment, within each time step, for each spatial grid, calculate the smoke concentration of the grid at the end of the current time step based on the smoke diffusion flux between adjacent grids according to Fick's second law of diffusion; and continuously monitor the flame changes through the camera device to dynamically adjust the smoke source intensity.

[0028] Specifically, obtain the material coefficient based on the materials of the walls, doors, and windows of the campus building, which can intuitively reflect the influence characteristics of different materials on smoke diffusion. For example, thick concrete walls can hinder smoke diffusion more than light glass doors and windows; external airflows have a promoting effect on smoke propagation. Combine the current wind speed on campus and the material coefficient to determine the smoke diffusion coefficient. Estimate the initial smoke source intensity using the initial smoke value of the smoke sensor and the flame characteristics captured by the camera device. In the digital twin model, set the initial smoke concentration of the grid in the fire-occurring area based on the calculated initial smoke source intensity, and set other areas to 0. Use Fick's second law of diffusion to calculate the smoke concentration of each grid within each time step based on the smoke diffusion flux between adjacent grids, realizing the simulation of the smoke diffusion process. Continuously monitor the flame changes through the camera device to dynamically adjust the smoke source intensity, and update the simulation conditions in real time, so that the simulation results can closely follow the actual changes of the fire, which helps to take response measures more timely and effectively, reducing the fire risk and losses.

[0029] As a specific implementation of a campus fire monitoring method, the method for generating the recommended route includes: Perform grid processing on the campus building structure and passage areas in the digital twin model, set a unique grid identifier for each grid and endow it with a passage attribute; the passage attribute includes whether it can be passed through; Generate an initial recommended escape route based on the grid identifier of the fire-occurring area and the grid identifier corresponding to the safety exit; Determine the number of people in each grid divided by the digital twin model based on the image data collected by the camera device and determine the personnel density of the corresponding grid based on the grid area; Trigger the route adjustment mechanism when the personnel density is greater than the preset density value; the route adjustment mechanism is used to generate several alternative routes and select the optimal alternative route based on the route length of the alternative route, the distance between the route and the fire occurrence location, and the fire-fighting facilities equipped around the route; Control the display devices and broadcasting devices in the area where the personnel density is greater than the preset density value to broadcast the optimal alternative route.

[0030] Specifically, by setting a unique identifier for each grid and assigning a traffic attribute to define which areas are accessible to personnel, feasible paths are planned when generating the initial recommended escape routes to ensure that teachers and students have a basic escape direction in case of a fire. For the possible congestion situation during the actual escape process, the personnel density in each grid is determined based on the image data collected by the camera device. When the personnel density is greater than the preset value, the route adjustment mechanism is triggered to generate several alternative routes, and the optimal alternative route is selected by comprehensively considering the route length, the distance from the fire occurrence location, and the fire-fighting facilities equipped around the route. The display devices and broadcasting devices in the area with a large personnel density are controlled to broadcast the optimal alternative route, and the adjusted escape information is timely conveyed to the personnel in the relevant area to avoid delaying the escape time due to congestion.

[0031] As one of the implementation manners of a campus fire monitoring method, the route adjustment mechanism is used to generate several alternative routes and select the optimal alternative route based on the route length of the alternative route, the distance between the route and the fire occurrence location, and the fire-fighting facilities equipped around the route, including: For each alternative route, according to the position information of each node in the digital twin model, calculate the sum of the distances between all the nodes passed by the route to obtain the length of the alternative route; Taking the center grid identifier of the fire occurrence area as a reference, calculate the Euclidean distance from each node on each alternative route to the center grid of the fire occurrence area, and take the minimum value as the distance between the alternative route and the fire occurrence location; According to the position information of the fire-fighting facilities, determine the coverage range of the fire-fighting facilities around each alternative route; calculate the proportion of the fire-fighting facilities coverage range in the length of the alternative route to evaluate the fire-fighting facilities equipped around the route; Calculate the comprehensive score based on considering the route length, the distance between the route and the fire occurrence location, and the fire-fighting facilities equipped around the route, and select the route with the highest comprehensive score as the optimal alternative route.

[0032] As one of the implementation manners of a campus fire monitoring method, the method for generating the alternative route includes: S601. Take the grid identifier of the fire occurrence area as the starting node s and the grid identifier corresponding to the safe exit as the target node t; create a queue Q, a distance array d, and a predecessor node array p. The queue Q is used to store the nodes to be explored, and the nodes in the queue are sorted in ascending order of path weight; the predecessor node array p is used to record the path, and initially the predecessor node array p is empty; the distance value d of the starting node in the distance array is set to 0. S602. Add the starting node s to the queue Q. When the priority queue Q is not empty, take out the node u with the smallest path weight from the queue Q. If u is the target node t, jump to S604. For each adjacent node v of the node u, calculate the new path weight from the node u to the node v ; where ; is the path length from the node u to the node v, obtained according to the actual distance between nodes in the digital twin model; is the average population density on the path segment uv; is the population density influence weight, which is a constant coefficient. If the new path weight from reaching the node v through the node u is less than the currently recorded distance value of the node v, then update the distance value , and set the predecessor node of the node v to u. If the node v is not in the queue Q, add it to the queue Q. S603. Repeat S602, continuously take out nodes from the queue Q for expansion until the target node t is taken out. S604. When the target node t is taken out, generate a path from the starting node s to the target node t by backtracking through the predecessor node array p to obtain an alternative route.

[0033] Specifically, determine the grid identifiers corresponding to the fire occurrence area and the safety exits, and set them as the starting node s and the target node t respectively; create a queue Q, a distance array d, and a predecessor node array p. Among them, the queue Q sorts nodes in ascending order of path weights, and in subsequent explorations, it can preferentially select better paths for expansion; the predecessor node array p is used to record the path and is initially set to be empty to prepare for subsequent path backtracking; set the distance value of the starting node to 0. Add the starting node s to the queue Q to start the exploration process. When the node u with the smallest path weight is taken out from the queue Q, if u is the target node t, an effective path search is completed, and it jumps to the path generation step. For the adjacent node v of the node u, the path length affects the escape time, and the personnel density reflects the congestion risk. When calculating the new path weight, comprehensively consider the actual path length from the node u to v and the average personnel density on the path segment uv, so that the generated path weight better meets the requirements of safe and efficient evacuation. If the new path weight is less than the currently recorded distance value of the node v, update the distance value and set the predecessor node, and at the same time add the node v to the queue Q. This process ensures that better path information can be continuously discovered and recorded during the search process. In the S603 stage, continuously repeat S602, continuously take out nodes from the queue Q for expansion until the target node t is taken out, and comprehensively search all possible paths to find the optimal path as much as possible. In the S604 stage, backtrack through the predecessor node array p to generate the path from the starting node s to the target node t to obtain the alternative route. Provide more escape options for campus personnel during a fire, and improve the success rate of personnel evacuation.

[0034] As one of the implementation manners of a campus fire monitoring method, after dividing the affected area and the safe area based on the dynamic heat map, the method further includes: According to the location information of the fire hazard area, determine the fire-fighting facilities within the area and the preset range around it in the digital twin model; Query the status information of the fire-fighting facilities; If it is detected that a fire-fighting facility is faulty, in the visualization interface of the digital twin model, mark the faulty fire-fighting facility with a flashing red icon.

[0035] Specifically, when determining the fire-fighting facilities within the area and the preset range around it in the digital twin model according to the location information of the fire hazard area and querying the status information of these fire-fighting facilities, their working conditions can be known in a timely manner. If it is detected that a fire-fighting facility is faulty, in the visualization interface of the digital twin model, marking the faulty fire-fighting facility with a flashing red icon can enable campus management personnel to intuitively discover the problematic facilities in the first time, improve the campus fire prevention and response capabilities, and minimize the losses caused by fires as much as possible.

[0036] This application also provides a campus fire monitoring system, including: The first processing module 201 is configured to: obtain the smoke values sent by all smoke sensors on campus and the image information sent by the imaging device; The second processing module 202 is configured to: determine whether a fire has occurred based on the smoke value and the image information; determine the fire occurrence area based on the location information of the corresponding smoke sensor and the image information sent by the fire camera after determining that a fire has occurred; The third processing module 203 is configured to: dynamically simulate the fire spread trend based on the digital twin model and in combination with the smoke diffusion prediction algorithm to generate a dynamic heat map; divide the affected area and the safe area based on the dynamic heat map; The fourth processing module 204 is configured to: control the display device and the broadcasting device in the affected area to broadcast a first warning message; control the display device and the broadcasting device in the safe area to broadcast a second warning message; wherein, the first warning message includes a fire occurrence reminder, the distance to the fire occurrence location, and a recommended route; The fifth processing module 205 is configured to: monitor the escape situation of personnel in each area and the fire spread situation in real time to update the first warning message.

[0037] It should be noted that: the above embodiments are only used to illustrate the present application and do not limit the technical solutions described in the present application. Although this specification has described the present application in detail with reference to the above embodiments, those of ordinary skill in the art should understand that those skilled in the technical field can still modify the present application or make equivalent substitutions, and all technical solutions and their improvements that do not depart from the spirit and scope of the present application should be covered within the scope of the claims of the present application.

Claims

1. A campus fire monitoring method, characterized in that: include: Obtain the smoke values ​​sent by all smoke sensors on campus and the image information sent by the camera device; Determine whether a fire has occurred based on smoke values ​​and image information; After determining that a fire has occurred, the fire occurrence area is determined based on the location information of the corresponding smoke sensor and the image information sent by the fire camera; Based on the digital twin model and combined with the smoke diffusion prediction algorithm, the fire spread trend is dynamically simulated to generate a dynamic heat map; the affected area and the safe area are divided based on the dynamic heat map; Control the display device and the broadcasting device in the affected area to broadcast a first warning message; control the display device and the broadcasting device in the safe area to broadcast a second warning message; wherein the first warning message includes a fire reminder, the distance to the fire location, and a recommended route; Real-time monitoring of the escape situation of personnel in each area and the spread of fire to update the first warning information.

2. A campus fire monitoring method according to claim 1, characterized in that: Based on the digital twin model and combined with the smoke diffusion prediction algorithm, the fire spread trend is dynamically simulated to generate a dynamic thermal map, including: Input the smoke values ​​sent by all smoke sensors in the campus, the image information sent by the camera device, and the location information of the corresponding smoke sensors into the digital twin model of the campus to complete the initial data loading of the digital twin model; the digital twin model of the campus imports the three-dimensional spatial structure data of the campus and the location of the fire-fighting facilities; the three-dimensional spatial structure data includes the building layout and the location information of the passage area; the fire-fighting facilities include fire extinguishers and fire hydrants; Determine the parameters of the smoke diffusion prediction algorithm based on the material and ventilation conditions of campus buildings; The smoke diffusion prediction algorithm with set parameters is embedded into the digital twin model, and the diffusion process of smoke in the campus space is simulated step by step according to the time series, starting from the fire area; the simulation results are output once at a preset interval; the simulation results include the smoke concentration distribution in each area of ​​the campus and the scope of the fire; According to the simulation results of the fire spread trend, the smoke concentration and fire danger level of each area of ​​the campus are quantitatively graded into the first level, the second level and the third level; The quantified and graded information is mapped into the three-dimensional space of the digital twin model to generate a dynamic heat map using different colors and brightness to represent the degree of fire hazard in each area.

3. A campus fire monitoring method according to claim 2, characterized in that: The parameters of the smoke diffusion prediction algorithm are determined according to the material and ventilation conditions of the campus buildings, including: Based on the materials of campus buildings, the material coefficients of the materials in different areas that hinder or promote the diffusion of smoke are obtained; the materials of the campus buildings include: wall materials, door and window materials; Determine the diffusion coefficient of smoke in the campus based on the current wind speed of the campus and the material coefficient; According to the initial smoke value detected by the smoke sensor in the fire area, combined with the size and brightness of the flame in the fire scene image taken by the camera, the smoke generation rate in the initial stage of the fire, that is, the initial smoke source intensity, is estimated using the relevant formula of combustion science; The smoke diffusion prediction algorithm with set parameters is embedded into the digital twin model. The smoke diffusion process in the campus space is simulated step by step according to the time series, starting from the fire area, including: In the spatial grid of the digital twin model, the initial smoke density of the grid corresponding to the fire area is set to the value calculated according to the initial smoke source intensity, and the initial smoke density of the grids in other areas is set to 0; Starting from the initial moment, in each time step, for each spatial grid, the smoke concentration of the grid at the end of the current time step is calculated based on Fick's second law of smoke diffusion and the smoke diffusion flux between adjacent grids; and the smoke source intensity is dynamically adjusted by continuously monitoring the flame changes through the camera device to obtain the fire development situation.

4. A campus fire monitoring method according to claim 3, characterized in that: The method for generating the recommended route includes: In the digital twin model, the campus building structure and traffic area are gridded, each grid is assigned a unique grid identifier and a traffic attribute; the traffic attribute includes whether it is passable; Generate an initial recommended escape route based on the grid identifier of the fire area and the grid identifier corresponding to the safety exit; Determine the number of people in each grid divided by the digital twin model based on the image data collected by the camera device and determine the density of people in the corresponding grid based on the area of ​​the grid; When the density of people is greater than a preset density value, a route adjustment mechanism is triggered; the route adjustment mechanism is used to generate a number of backup routes and select the best backup route based on the length of the backup routes, the distance between the routes and the fire site, and the configuration of fire-fighting facilities around the routes; The display device and the broadcasting device in the area where the density of personnel is greater than the preset density value are controlled to broadcast the optimal backup route.

5. A campus fire monitoring method according to claim 4, characterized in that: The route adjustment mechanism is used to generate several backup routes and select the best backup route based on the route length of the backup route, the distance between the route and the fire location, and the configuration of fire fighting facilities around the route, including: For each backup route, the sum of the distances between all nodes that the route passes through is calculated based on the location information of each node in the digital twin model to obtain the length of the backup route; Taking the central grid mark of the fire area as the benchmark, calculate the Euclidean distance from each node on each backup route to the central grid of the fire area, and take the minimum value as the distance between the backup route and the fire location; According to the location information of fire-fighting facilities, determine the coverage of fire-fighting facilities around each backup route; calculate the ratio of the coverage of fire-fighting facilities to the length of the backup route to evaluate the deployment of fire-fighting facilities around the route; A comprehensive score is calculated based on the length of the route, the distance between the route and the fire site, and the fire-fighting facilities around the route, and the route with the highest comprehensive score is selected as the optimal backup route.

6. A campus fire monitoring method according to claim 5, characterized in that: The method for generating the alternate route includes: S601, taking the grid identifier of the fire area as the starting node s and the grid identifier corresponding to the safety exit as the target node t; creating a queue Q, a distance array d and a predecessor node array p; the queue Q is used to store the nodes to be explored, and the nodes in the queue are sorted from small to large according to the path weight; the predecessor node array p is used to record the path, and the predecessor node array p is empty initially; the distance value d of the starting node in the distance array is set to 0; S602, add the starting node s to the queue Q; when the priority queue Q is not empty, take out the node u with the smallest path weight from the queue Q; if u is the target node t, jump to S604; for each adjacent node v of node u, calculate the new path weight from node u to node v ;in, ; is the path length from node u to node v, which is obtained according to the actual distance between nodes in the digital twin model; is the average personnel density on the path segment uv; is the weight of the impact of personnel density, which is a constant coefficient; if the weight of the new path from node u to node v is Less than the distance value of the currently recorded node v , then update the distance value , and the predecessor node of node v Set to u; if node v is not in queue Q, add it to queue Q; S603, repeat S602, and continuously take out nodes from the queue Q for expansion until the target node t is taken out; S604: When the target node t is taken out, a backup route is obtained by backtracking the predecessor node array p to generate a path from the start node s to the target node t.

7. A campus fire monitoring method according to claim 6, characterized in that: After dividing the affected area and the safe area based on the dynamic heat map, the method further includes: Based on the location information of the fire hazard area, the fire-fighting facilities in the area and the surrounding preset range are determined in the digital twin model; Query the status information of fire protection facilities; If a fault is detected in the fire-fighting facilities, the problematic fire-fighting facilities will be marked with a flashing red icon in the visual interface of the digital twin model.

8. A campus fire monitoring system, characterized in that: include: The first processing module is used to: obtain the smoke values ​​sent by all smoke sensors in the campus and the image information sent by the camera device; The second processing module is used to: determine whether a fire has occurred based on the smoke value and the image information; after determining that a fire has occurred, determine the fire occurrence area based on the location information of the corresponding smoke sensor and the image information sent by the fire camera; The third processing module is used to: dynamically simulate the fire spread trend based on the digital twin model and the smoke diffusion prediction algorithm to generate a dynamic heat map; divide the affected area and the safe area based on the dynamic heat map; The fourth processing module is used to: control the display device and the broadcasting device in the affected area to broadcast the first warning information; control the display device and the broadcasting device in the safe area to broadcast the second warning information; wherein the first warning information includes a fire reminder, a distance from the fire location, and a recommended route; The fifth processing module is used to: monitor the escape situation of personnel in each area and the spread of fire in real time to update the first warning information.

Images