An intelligent evacuation method and system based on online fire hazard assessment

Abstract

This invention discloses an intelligent evacuation method and system based on online fire hazard assessment, comprising the following steps: collecting fire scene information; performing noise reduction on the fire scene information based on wavelet transform; determining the fire source location information; assessing the spatial hazard of the fire scene; constructing a fire evacuation potential function; calculating the optimal evacuation direction and optimal evacuation path based on Dijkstra's algorithm; and updating the optimal evacuation direction and optimal evacuation path according to location and time. This invention effectively solves the problems of insufficient fire source information collection and processing in current fire evacuation systems, lack of real-time hazard analysis of various areas within a building during a fire, and inability to update evacuation paths according to fire development.

Description

Technical Field

[0001] This invention relates to the technical field of building fire safety evacuation, and in particular to an intelligent evacuation method and system based on online fire hazard assessment. Background Technology

[0002] As the internal structures of urban buildings become increasingly complex, rescue efforts during building fires are also becoming more challenging. To ensure the timely and safe evacuation of people to the outside of buildings in the event of a fire, the planning and upgrading of fire evacuation systems are essential. Current fire evacuation systems generally consist of fire monitoring systems and emergency evacuation guidance systems. These systems often rely solely on single-factor information related to the fire source for hazard assessment when issuing fire alarms, resulting in numerous drawbacks in practical applications, such as response delays and unclear evacuation route guidance.

[0003] Current fire evacuation systems are generally static systems, collecting only limited information in the early stages of a fire for prediction and assessment, and planning evacuation routes. As the fire progresses, the original fire scene information changes, and the evacuation plans based on the initial fire information become inapplicable, and may even lead people to dangerous areas.

[0004] Current fire evacuation systems have limited capacity for collecting and processing fire source information, lack effective integration and analysis of fire scene information, and are unable to make fire spread predictions based on monitoring data and to comprehensively plan evacuation routes in conjunction with fire source location information.

[0005] Current fire evacuation systems lack sufficient hazard analysis of different areas within a building during a fire. Since fires are constantly developing, when people at the fire scene cannot know whether different areas within the building are in a dangerous state or about to become dangerous, simple evacuation instructions can easily lead people to dangerous areas, increasing the difficulty of rescue.

[0006] The current fire evacuation system does not update the planning frequency of evacuation routes enough. Fire scenes can change rapidly, and if reasonable evacuation routes cannot be provided in a timely and effective manner, the safety of people at the fire scene cannot be effectively guaranteed. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the existing technology and provide an intelligent evacuation method and system based on online fire hazard assessment. This method can effectively solve the problems of insufficient collection and processing of fire source information in current fire evacuation systems, lack of real-time hazard analysis of various areas inside buildings during a fire, and inability to update evacuation routes according to the development of the fire. This method can effectively improve the comprehensive processing capability of existing fire evacuation systems for fire scene information and enhance the efficiency of personnel evacuation in fire scenes.

[0008] The first objective of this invention is to provide an intelligent evacuation method based on online fire hazard assessment.

[0009] The second objective of this invention is to provide an intelligent evacuation system based on online fire hazard assessment.

[0010] The first objective of this invention can be achieved by adopting the following technical solution:

[0011] An intelligent evacuation method based on online fire hazard assessment includes the following steps:

[0012] S1. Real-time acquisition of multiple multi-source fire scene information, and formation of a four-dimensional information vector sample of the time sequence of on-site combustion based on the real-time acquisition of multiple multi-source fire scene information;

[0013] S2. Based on the wavelet transform principle, the four-dimensional information vector sample obtained in step S1 is denoised to obtain effective multi-source monitoring information after denoising.

[0014] S3. Using the least squares method, the effective multi-source monitoring information after noise reduction in step S2 is used to deduce the fire source parameters and determine the fire source location information.

[0015] S4. Based on the fire source location information obtained in step S3, and in accordance with the ISO safety threat principle, conduct a dynamic assessment of the spatial state hazard of the fire scene.

[0016] S5. Based on the fire source location information, the spatial hazard level of the fire scene, and the personnel density and location information at the fire scene, calculate the evacuation risk of each evacuation area at the fire scene, and construct the fire evacuation potential function based on the evacuation risk of each evacuation area.

[0017] S6. Using the fire risk potential function as the optimal strategy principle for pedestrian movement path selection, the optimal evacuation direction and optimal evacuation path are calculated based on the Dijkstra algorithm.

[0018] S7. Starting from the position at the time of evacuation, and based on the position updated at intervals, continuously calculate and update the optimal evacuation direction and the optimal evacuation path.

[0019] Furthermore, real-time fire information monitoring equipment is used to collect information on multiple fire sources in real time. This equipment includes smoke sensors, temperature sensors, luminance meters, and video acquisition units. These devices are deployed in various areas within the building to monitor the fire situation in each area in real time. Additionally, auxiliary evacuation equipment is installed in each evacuation route and room within the building to receive information on the location of the fire source, the hazard level of the fire scene, fire scene information, building structure information, optimal evacuation direction, and optimal evacuation path.

[0020] Furthermore, the fire scene information includes one or more of the following: temperature information, carbon monoxide concentration information, oxygen concentration information, flame brightness information, and video information.

[0021] Furthermore, in step S2, the following operations are performed:

[0022] S2.1, the four-dimensional information vector sample obtained in step S1 Among them, for the signal set i represents the information type, k represents the time series, and α j Represents a spatial sequence. These represent the signals at three consecutive selected time points. Then, based on fire development dynamics, a fast fire model is used, and a time-series fire signal simulation curve is established according to the wavelet transform function form:

[0023]

[0024] in, For simulated signals; s n Let be the translation parameter in space n; k be the time series; a j The sensor number indicates the location at which it is mounted (x, y). j ,y j )((x j ,y j )∈n) on the ordinal number a j Sensors;

[0025] S2.2 Calculate the average value of the signal features over the first k time steps, and predict the value of the signal features at time k+1 based on this average value; take the average value of the signal features at three consecutive time steps t1, t2, and t3 from the first k time steps as the reference value for noise reduction processing, denoted as... This represents the noise reduction reference values ​​for three consecutive time points t1, t2, and t3, where... This represents the reference value for noise reduction processing at time t1; This represents the reference value for noise reduction processing at time t2; This represents the reference value for noise reduction processing at time t3;

[0026] S2.3. Based on the similarity between the test signal and the simulation curve, noise reduction processing is performed:

[0027]

[0028] in, express and The vector inner product; Represents vector and norm; S ij Representing two vectors Feature similarity;

[0029] Simultaneously, a threshold δ is introduced, with a range of [0,1], and we have:

[0030] S ij <δ→False(i=1,2,...,n)

[0031] S ij >δ→True(i=1,2,...,n)

[0032] When the output value is True, it means that the signal transmitted from the probe point is valid; when the output value is False, it means that the signal transmitted from the probe point is invalid and this signal data needs to be removed.

[0033] Finally, effective multi-source monitoring information after noise reduction is obtained.

[0034] Furthermore, in step S3, the following operations are performed:

[0035] S3.1 Effective multi-source monitoring information after noise reduction in step S2 Based on the least squares optimization model method, the least squares optimization function of the fire source objective function is established, namely:

[0036]

[0037] Where X represents the parameters to be inverted from the fire source, and N represents the total number of detection points;

[0038] S3.2. Using a genetic algorithm, the monitoring temperature information of N detection points is iteratively solved to calculate the fire source location information.

[0039] Furthermore, in step S4, the following operations are performed:

[0040] Based on the fire source location information obtained in step S3, and in accordance with the ISO safety threat principle, dynamic fire scene hazard sample criteria and safety sample criteria are customized to conduct dynamic fire scene spatial state hazard assessment.

[0041] The risk assessment of the spatial condition at the fire scene is as follows:

[0042]

[0043] Where, Θ i,k (t) represents the hazard level of the fire scene at time t; FC j This refers to the life safety guidelines for irritant gases that can cause disability, as recommended in ISO / TSI 13571.

[0044] Furthermore, in step S5, the following operations are performed:

[0045] S5.1 Based on the fire source location information, the hazard level of the fire scene space, and the personnel density and location information at the fire scene, calculate the evacuation risk of each evacuation area at the fire scene:

[0046]

[0047] Where Risk(x,y,t) is the fire evacuation risk distribution function; Θ i,k (t) represents the spatial hazard level of the fire scene at time t;

[0048] S5.2. Based on the evacuation risk of each evacuation area, construct the fire evacuation potential function:

[0049]

[0050] Furthermore, in step S6, the following operations are performed:

[0051] Using the fire risk potential function φ(x,y,t) as the optimal strategy principle for pedestrian path selection, Dijkstra's algorithm is employed to calculate the optimal evacuation direction and optimal evacuation path. Specifically, the principle of Dijkstra's algorithm for the fire evacuation potential function is to select the path with the minimum evacuation potential function, as detailed below:

[0052]

[0053] Where P(x,y) represents the fire hazard level of each area; φ i (x,y,z) is the fire evacuation potential function φ(x,y,z) on the i-th path; M is the total number of evacuation areas involved in the evacuation plan.

[0054] Furthermore, in step S7, the following operations are performed:

[0055] Starting from the position (x0, y0) of the auxiliary evacuation equipment used during evacuation, and based on the position (x, y) updated by the auxiliary evacuation equipment at intervals, the optimal evacuation direction and optimal evacuation path are continuously calculated and updated. According to the integration direction (x′(s), y′(s)) of the integration path l, the optimal evacuation direction and optimal evacuation path are sent to the auxiliary evacuation equipment.

[0056] The second objective of this invention can be achieved by adopting the following technical solution:

[0057] An intelligent evacuation system based on online fire hazard assessment, applied to the aforementioned intelligent evacuation method based on online fire hazard assessment, includes:

[0058] Real-time fire information monitoring equipment is deployed in various areas inside the building to monitor the occurrence of fires in various areas of the building in real time and collect information on the fire scene.

[0059] The fire source location determination module is used to deduce fire source parameters using the least squares method based on fire scene information to determine the fire source location information;

[0060] The fire hazard determination module is used to dynamically assess the spatial hazard of a fire scene based on fire source location information and the ISO safety threat principle.

[0061] The optimal evacuation direction and optimal evacuation path calculation module is used to calculate the evacuation risk of each evacuation area at the fire scene based on the fire source location information, the spatial state hazard of the fire scene, and the personnel density and personnel location information at the fire scene. It constructs the fire evacuation potential function and then uses the fire risk potential function as the optimal strategy principle for pedestrian movement path selection. Based on the Dijkstra algorithm, it calculates the optimal evacuation direction and optimal evacuation path.

[0062] The optimal evacuation direction and optimal evacuation path update module is used to continuously calculate and update the optimal evacuation direction and optimal evacuation path, starting from the position at the time of evacuation and updating the position according to the time interval.

[0063] Auxiliary evacuation equipment, including display devices and handheld devices, is used to receive information on the location of the fire source, the hazard level of the fire scene space, fire scene information, building structure information, and the optimal evacuation direction and optimal evacuation route updated in real time.

[0064] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0065] 1. This invention effectively integrates and analyzes fire scene information and fire source location information, and combines multiple key pieces of information from the fire scene to make comprehensive plans for evacuation routes.

[0066] 2. This invention enables dynamic online analysis of the danger level and evacuation risk of various areas inside a building during a fire.

[0067] 3. This invention achieves dynamic optimization of evacuation routes based on online hazard analysis of the fire scene. It plans evacuation routes based on fire information at the current moment of the fire, and collects dynamic data so that the route planning can be updated over time.

[0068] 4. This invention receives the optimal evacuation direction and path through a handheld device or display device. The path planning is timely, effective, and highly reliable, which can better ensure that personnel are evacuated to a safe area as quickly as possible. Attached Figure Description

[0069] Figure 1 This is a flowchart of the intelligent evacuation method of the present invention.

[0070] Figure 2 This is a schematic diagram illustrating the deduction of fire source parameters for location X according to the present invention.

[0071] Figure 3 This is a schematic diagram illustrating the deduction of fire source parameters for position Y according to the present invention.

[0072] Figure 4 This is a schematic diagram illustrating the evacuation path optimization of the intelligent evacuation method of the present invention.

[0073] Figure 5 This is a diagram of the operating interface of the auxiliary evacuation device of the present invention.

[0074] Figure 6 This is a schematic diagram of the intelligent evacuation system of the present invention.

[0075] Figure 7 This is a schematic diagram showing the arrangement of the real-time fire information monitoring device, display device, and handheld device in this invention. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0077] like Figures 1 to 5 As shown, this embodiment provides an intelligent evacuation method based on online fire hazard assessment, including the following steps:

[0078] S1. Use fire information real-time monitoring equipment to collect i types of multi-source fire scene information in real time, and form a four-dimensional information vector sample of the time sequence of the on-site combustion based on the i types of multi-source fire scene information collected in real time.

[0079] The real-time fire information monitoring equipment includes smoke sensors, temperature sensors, brightness meters, and video acquisition units. These devices are deployed in various areas inside the building to monitor the occurrence of fires in different areas of the building in real time.

[0080] Fire scene information includes one or more of the following: temperature information, carbon monoxide concentration information, oxygen concentration information, flame brightness information, and video information;

[0081] Meanwhile, auxiliary evacuation equipment is installed in various evacuation passages and rooms inside the building to receive information on the location of the fire source, the hazard level of the fire scene, fire scene information, building structure information, optimal evacuation direction, and optimal evacuation route.

[0082] S2. Based on the wavelet transform principle, the four-dimensional information vector samples obtained in step S1 are denoised to obtain effective multi-source monitoring information after denoising; specifically, the following operations are performed:

[0083] S2.1, the four-dimensional information vector sample obtained in step S1 Among them, for the signal set i represents the information type, k represents the time series, and α j Represents a spatial sequence. These represent the signals at three consecutive selected time points. Then, based on fire development dynamics, a fast fire model is used, and a time-series fire signal simulation curve is established according to the wavelet transform function form:

[0084]

[0085] in, For simulated signals; s n Let be the translation parameter in space n; k be the time series; a j The sensor number indicates the location at which it is mounted (x, y). j ,y j )((x j ,y j )∈n) on the ordinal number a j Sensors;

[0086] S2.2 Calculate the average value of the signal features over the first k time steps, and predict the value of the signal features at time k+1 based on this average value; take the average value of the signal features at three consecutive time steps t1, t2, and t3 from the first k time steps as the reference value for noise reduction processing, denoted as... This represents the noise reduction reference values ​​for three consecutive time points t1, t2, and t3, where... This represents the reference value for noise reduction processing at time t1; This represents the reference value for noise reduction processing at time t2; This represents the reference value for noise reduction processing at time t3;

[0087] S2.3. Based on the similarity between the test signal and the simulation curve, noise reduction processing is performed:

[0088]

[0089] in, express and The vector inner product; Represents vector and norm; S ij Representing two vectors Feature similarity;

[0090] Simultaneously, a threshold δ is introduced, with a range of [0,1], and we have:

[0091] S ij <δ→False(i=1,2,...,n)

[0092] S ij >δ→True(i=1,2,...,n)

[0093] When the output value is True, it means that the signal transmitted from the probe point is valid; when the output value is False, it means that the signal transmitted from the probe point is invalid and this signal data needs to be removed.

[0094] Finally, effective multi-source monitoring information after noise reduction is obtained.

[0095] S3. For the effective multi-source monitoring information after noise reduction in step S2, use the least squares method to extrapolate the fire source parameters, calculate the distribution probability of the fire source parameters, and determine the fire source location information; specifically, perform the following operations:

[0096] S3.1 Effective multi-source monitoring information after noise reduction in step S2 Based on the least squares optimization model method, the least squares optimization function of the fire source objective function is established, namely:

[0097]

[0098] Where X represents the parameters to be inverted from the fire source, and N represents the total number of detection points;

[0099] S3.2. Using a genetic algorithm, the monitoring temperature information of N detection points is iteratively solved to calculate the fire source location information.

[0100] S4. Based on the fire source location information obtained in step S3, and in accordance with the ISO safety threat principle, conduct a dynamic assessment of the spatial hazard level of the fire scene; specifically, perform the following operations:

[0101] Based on the fire source location information obtained in step S3, and in accordance with the ISO safety threat principle, dynamic fire scene hazard sample criteria and safety sample criteria are customized to conduct dynamic fire scene spatial state hazard assessment.

[0102] The risk assessment of the spatial condition at the fire scene is as follows:

[0103]

[0104] Where, Θi ,k (t) represents the hazard level of the fire scene at time t; FC j This refers to the life safety guidelines for irritant gases that can cause disability, as recommended in ISO / TSI 13571.

[0105] S5. Based on the fire source location information, the spatial hazard level of the fire scene, and the personnel density and location information at the fire scene, calculate the evacuation risk of each evacuation area at the fire scene, and construct the fire evacuation potential function based on the evacuation risk of each evacuation area; specifically, perform the following operations:

[0106] S5.1 Based on the fire source location information, the hazard level of the fire scene space, and the personnel density and location information at the fire scene, calculate the evacuation risk of each evacuation area at the fire scene:

[0107]

[0108] Where Risk(x,y,t) is the fire evacuation risk distribution function; Θ i,k (t) represents the spatial hazard level of the fire scene at time t;

[0109] S5.2. Based on the evacuation risk of each evacuation area, construct the fire evacuation potential function:

[0110]

[0111] S6. Using the fire risk potential function as the optimal strategy principle for pedestrian path selection, and calculating the optimal evacuation direction and path based on Dijkstra's algorithm, this serves as a navigation scheme for evacuating pedestrians, realizing intelligent evacuation path planning and guidance for buildings; specifically, the following operations are performed:

[0112] Using the fire risk potential function φ(x,y,t) as the optimal strategy principle for pedestrian path selection, Dijkstra's algorithm is employed to calculate the optimal evacuation direction and optimal evacuation path. Specifically, the principle of Dijkstra's algorithm for the fire evacuation potential function is to select the path with the minimum evacuation potential function, as detailed below:

[0113]

[0114] Where P(x,y) represents the fire hazard level of each area; φ i (x,y,z) is the fire evacuation potential function φ(x,y,z) on the i-th path; M is the total number of evacuation areas involved in the evacuation plan.

[0115] S7. Starting from the position at the time of evacuation, and updating the position based on the time interval, continuously calculate and update the optimal evacuation direction and optimal evacuation path; specifically, perform the following operations:

[0116] Starting from the position (x0, y0) of the auxiliary evacuation equipment used during evacuation, and based on the position (x, y) updated by the auxiliary evacuation equipment at intervals, the optimal evacuation direction and optimal evacuation path are continuously calculated and updated. According to the integration direction (x′(s), y′(s)) of the integration path l, the optimal evacuation direction and optimal evacuation path are sent to the auxiliary evacuation equipment.

[0117] like Figure 6 , Figure 7 As shown, this embodiment also provides an intelligent evacuation system based on online fire hazard assessment, applied to the aforementioned intelligent evacuation method based on online fire hazard assessment, including:

[0118] The real-time fire information monitoring device 1 includes one or more of a smoke sensor, a temperature sensor, a brightness meter, and a video acquisition unit. The smoke sensor, temperature sensor, brightness meter, and video acquisition unit are arranged in various areas inside the building to monitor the fire situation in various areas inside the building in real time. The device collects fire scene information in real time and transmits the information to the main controller 4 via a two-wire bus.

[0119] The fire source location determination module is used to determine the fire source location by using the least squares method to deduce fire source parameters based on fire scene information.

[0120] The fire hazard determination module is used to dynamically assess the spatial hazard of a fire scene based on fire source location information and the ISO safety threat principle.

[0121] The optimal evacuation direction and optimal evacuation path calculation module is used to calculate the evacuation risk of each evacuation area at the fire scene based on the fire source location information, the spatial state hazard of the fire scene, and the personnel density and personnel location information at the fire scene. It constructs the fire evacuation potential function and then uses the fire risk potential function as the optimal strategy principle for pedestrian movement path selection. Based on the Dijkstra algorithm, it calculates the optimal evacuation direction and optimal evacuation path.

[0122] The optimal evacuation direction and optimal evacuation path update module is used to continuously calculate and update the optimal evacuation direction and optimal evacuation path, starting from the position at the time of evacuation and updating the position according to the time interval.

[0123] The auxiliary evacuation equipment includes a handheld device 2 and a display device 3. The display device 3 is placed in the fire evacuation passage of the building, with one device every 10 meters. The handheld device 2 is placed in a visible area in each room of the building. People can obtain evacuation information nearby when a fire occurs. The display device and the handheld device are used to receive information on the location of the fire source, the spatial status and hazard level of the fire scene, fire scene information, building structure information, and the optimal evacuation direction and optimal evacuation route updated in real time.

[0124] This invention fully utilizes fire hazard information and makes comprehensive evacuation decisions based on this information, thereby forming an effective and dynamically changing decision-making scheme. Simultaneously, this invention can also provide more intuitive auxiliary evacuation equipment, enabling evacuees to more directly obtain the spatial hazard level of the fire scene and acquire more effective evacuation route plans, thereby reducing the randomness of evacuation and improving the efficiency and safety of evacuation within buildings.

[0125] In addition, this invention not only plays an important role in the evacuation of buildings, but can also be extended. For example, by setting up communication equipment, it can be connected with the fire department's fire command system to send the fire source location information and the danger level of the fire scene to the fire command system. By combining with the fire department's fire command system, it can provide fire fighting and rescue services for the fire department, thereby improving the efficiency of fire fighting and rescue.

[0126] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope disclosed in the present invention, based on the technical solution and inventive concept of the present invention, shall fall within the scope of protection of the present invention.

Claims

1. An intelligent evacuation method based on online fire hazard assessment, characterized in that, Includes the following steps: S1. Real-time acquisition of various multi-source fire scene information, and formation of a four-dimensional information vector sample of the time series of on-site combustion based on the real-time acquisition of various multi-source fire scene information. For the signal set , Indicates the type of information. Representing a time series, Represents a spatial sequence. , , These represent the signals at three consecutive selected time points; S2. Based on the wavelet transform principle, the four-dimensional information vector samples obtained in step S1 are denoised to obtain effective multi-source monitoring information after denoising. ; S3. Using the effective multi-source monitoring information after noise reduction in step S2, the least squares method is used to deduce the fire source parameters and determine the fire source location information; specifically, the following operations are performed: S3.1 Effective multi-source monitoring information after noise reduction in step S2 Based on the least squares optimization model method, the least squares optimization function of the fire source objective function is established, namely: ； in, The parameters to be inverted for the fire source This represents the total number of detection points; S3.

2. Using a genetic algorithm, the genetic algorithm is used to... The location of the fire source is calculated by iteratively solving the temperature information of each detection point; S4. Based on the fire source location information obtained in step S3, and in accordance with the ISO safety threat principle, conduct a dynamic assessment of the spatial hazard level of the fire scene; specifically, perform the following operations: Based on the fire source location information obtained in step S3, and in accordance with the ISO safety threat principle, dynamic fire scene hazard sample criteria and safety sample criteria are customized to conduct dynamic fire scene spatial state hazard assessment. The risk assessment of the spatial condition at the fire scene is as follows: ； in, for The hazard level of the fire scene space at all times; The life safety guidelines for irritant gases that can cause disability, as recommended in ISO / TSI 13571. S5. Based on the fire source location information, the spatial hazard level of the fire scene, and the personnel density and location information at the fire scene, calculate the evacuation risk of each evacuation area at the fire scene, and construct the fire evacuation potential function based on the evacuation risk of each evacuation area; specifically, perform the following operations: S5.1 Based on the fire source location information, the hazard level of the fire scene space, and the personnel density and location information at the fire scene, calculate the evacuation risk of each evacuation area at the fire scene: ； in, This is the fire evacuation risk distribution function; for The hazard level of the fire scene space at all times; S5.

2. Based on the evacuation risk of each evacuation area, construct the fire evacuation potential function: ； S6. Using the fire evacuation potential function as the optimal strategy principle for pedestrian movement path selection, the optimal evacuation direction and optimal evacuation path are calculated based on the Dijkstra algorithm. S7. Starting from the position at the time of evacuation, and based on the position updated at intervals, continuously calculate and update the optimal evacuation direction and the optimal evacuation path.

2. The intelligent evacuation method based on online fire hazard assessment according to claim 1, characterized in that, Real-time fire information monitoring equipment is used to collect data in real time. The system provides multi-source fire scene information. The real-time fire information monitoring equipment includes smoke sensors, temperature sensors, luminance meters, and video acquisition units. These devices are deployed in various areas inside the building to monitor the fire situation in different areas in real time. Additionally, auxiliary evacuation equipment is installed in various evacuation routes and rooms inside the building to receive information on the location of the fire source, the hazard level of the fire scene, fire scene information, building structure information, optimal evacuation direction, and optimal evacuation route.

3. The intelligent evacuation method based on online fire hazard assessment according to claim 1, characterized in that, The fire scene information includes one or more of the following: temperature information, carbon monoxide concentration information, oxygen concentration information, flame brightness information, and video information.

4. The intelligent evacuation method based on online fire hazard assessment according to claim 1, characterized in that, In step S2, the following operations are performed: S2.1, the four-dimensional information vector sample obtained in step S1 For the signal set , Indicates the type of information. Representing a time series, Represents a spatial sequence. , , These represent the signals at three consecutive selected time points. Then, based on fire development dynamics, a fast fire model is used, and a time-series fire signal simulation curve is established according to the wavelet transform function form. ； in, This is a simulated signal; For space Translation parameters in; It is a time series; The sensor number indicates the coordinate location it is mounted on. The serial number The sensor, ; S2.2, Before Calculation The average value of signal characteristic quantities at each time point, and the prediction based on this average value. The value of the signal characteristic at time step; take the previous value. Three consecutive moments in a time period The average value of the signal characteristics is used as the reference value for noise reduction processing, denoted as... , representing three consecutive moments The noise reduction processing reference value, among which, express Reference values ​​for noise reduction processing at any given moment; express Reference values ​​for noise reduction processing at any given moment; express Reference values ​​for noise reduction processing at any given moment; S2.

3. Based on the similarity between the test signal and the simulation curve, noise reduction processing is performed: ； in, express and The vector inner product; Represents vector and The norm; Representing two vectors Feature similarity; At the same time, a threshold is introduced. Its range is ,have: ； When the output value is True, it means that the signal transmitted from the probe point is valid; when the output value is False, it means that the signal transmitted from the probe point is invalid and this signal data needs to be removed. Finally, effective multi-source monitoring information after noise reduction is obtained. .

5. The intelligent evacuation method based on online fire hazard assessment according to claim 1, characterized in that, In step S6, the following operations are performed: fire evacuation potential function As a principle for optimal path selection for pedestrian movement, Dijkstra's algorithm is used to calculate the optimal evacuation direction and optimal evacuation path. Specifically, the principle of Dijkstra's algorithm for the fire evacuation potential function is to select the path with the minimum evacuation potential function, as detailed below: ； in, Fire hazard levels for each area; Let i be the fire evacuation potential function along the i-th path. ; This represents the total number of evacuation areas involved in the evacuation plan.

6. The intelligent evacuation method based on online fire hazard assessment according to claim 1, characterized in that, In step S7, the following operations are performed: Location of auxiliary evacuation equipment used during evacuation Starting from the location and updated according to the intervals of auxiliary evacuation equipment. The optimal evacuation direction and optimal evacuation path are continuously calculated and updated, based on the direction of integration of the integration path l. It sends the optimal evacuation direction and optimal evacuation path to the auxiliary evacuation equipment.

7. An intelligent evacuation system based on online fire hazard assessment, characterized in that, The intelligent evacuation method based on online fire hazard assessment, applied to any one of claims 1 to 6, includes: Real-time fire information monitoring equipment is deployed in various areas inside the building to monitor the occurrence of fires in various areas of the building in real time and collect information on the fire scene. The fire source location determination module is used to determine the fire source location by using the least squares method to deduce fire source parameters based on fire scene information. The fire scene spatial status hazard determination module is used to dynamically assess the fire scene spatial status hazard based on fire source location information and the ISO safety threat principle. The optimal evacuation direction and optimal evacuation path calculation module is used to calculate the evacuation risk of each evacuation area at the fire scene based on the fire source location information, the spatial state hazard of the fire scene, and the personnel density and personnel location information at the fire scene. It constructs the fire evacuation potential function and then uses the fire evacuation potential function as the optimal strategy principle for pedestrian movement path selection. Based on the Dijkstra algorithm, it calculates the optimal evacuation direction and optimal evacuation path. The optimal evacuation direction and optimal evacuation path update module is used to continuously calculate and update the optimal evacuation direction and optimal evacuation path, starting from the position at the time of evacuation and updating the position according to the time interval. Auxiliary evacuation equipment, including display devices and handheld devices, is used to receive information on the location of the fire source, the hazard level of the fire scene space, fire scene information, building structure information, and the optimal evacuation direction and optimal evacuation route updated in real time.

Images