Multi-parameter comprehensive fire-fighting monitoring system based on Internet of Things

Abstract

The invention discloses a multi-parameter comprehensive fire-fighting monitoring system based on the Internet of Things, relates to the technical field of fire-fighting supervision, and effectively reduces the risk of fire misjudgment caused by single-parameter monitoring. According to the method, the multi-parameter visual scene model is established, the real-time smoke concentration and the real-time hot streak are marked at each position of the multi-parameter visual scene model, the multi-parameter visual scene model is divided into a plurality of local fire-fighting supervision scenes, and the fire-fighting supervision time points are set; a plurality of fire-fighting detection areas are marked in the multi-parameter visual scene model, a fire-fighting abnormal area is positioned by acquiring a smoke traceability vector and real-time thermal stripes, and an equipment trigger decision is generated and executed according to the position of the fire-fighting abnormal area in the multi-parameter visual scene model until the fire-fighting abnormal area does not exist.

Description

Technical Field

[0001] The present invention relates to the technical field of fire control, and in particular to a multi-parameter comprehensive fire control monitoring system based on the Internet of Things. Background Art

[0002] With the continuous development of science and technology, intelligent fire protection systems are increasingly widely used in fire warning, fire control and fire rescue. However, when a fire occurs, traditional fire protection systems often rely on human monitoring and manual operation, and have problems such as slow response, incomplete information, and decentralized decision-making, which cannot meet the high requirements of modern fire safety.

[0003] Especially in complex and changeable fire scenes, the traditional single sensor data collection method is difficult to provide comprehensive and accurate fire information, resulting in inaccurate fire judgment and untimely emergency response, which increases the risk of fire hazards. For this reason, a multi-parameter comprehensive fire monitoring system based on the Internet of Things is provided. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a multi-parameter comprehensive fire monitoring system based on the Internet of Things.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A multi-parameter comprehensive fire monitoring system based on the Internet of Things, comprising a cloud management terminal, wherein the cloud management terminal is communicatively connected to an application scenario visualization module, a fire anomaly detection module, and a fire decision-making planning module;

[0007] The application scenario visualization module is provided with a scenario data acquisition unit and a scenario visualization unit;

[0008] The scene data acquisition unit is used to set up several composite data acquisition devices to collect multiple real-time scene data of fire scenes;

[0009] The scene visualization unit is used to generate a regional visualization model within the data perception range of each composite data acquisition device according to various real-time scene data, and then splice each regional visualization model to obtain a multi-parameter visualization scene model according to the spatial position of each composite data acquisition device in the fire scene, and mark each position of the multi-parameter visualization scene model with real-time smoke concentration and real-time thermal pattern;

[0010] The fire anomaly detection module is used to divide the multi-parameter visualization scene model into multiple local fire supervision scenes, and set fire supervision time points. According to the real-time smoke concentration and real-time thermal pattern of the local fire supervision scenes between the fire supervision time points, several fire detection areas are marked in the multi-parameter visualization scene model, and then the fire anomaly area is located by obtaining the smoke tracing vector and the real-time thermal pattern;

[0011] The firefighting decision-making planning module is used to generate and execute equipment triggering decisions according to the location of the firefighting abnormality area in the multi-parameter visualization scene model until there is no firefighting abnormality area.

[0012] Furthermore, the process of the composite data acquisition device collecting multiple real-time scene data of the fire scene includes:

[0013] Several composite data acquisition devices are installed in the fire scene, and the same data upload cycle is set for each composite data acquisition device. Whenever a data upload cycle starts, each composite data acquisition device obtains the infrared video data, optical video data and smoke concentration change curve within its data perception range, and at the same time calls the laser sensor to emit laser signals to the entire data perception range, thereby generating a three-dimensional laser signal spectrum.

[0014] Furthermore, the process of establishing the regional visualization model includes:

[0015] The scene data set is synchronized in real time with the scene visualization unit, wherein the scene visualization unit first establishes a regional optical three-dimensional model and a regional infrared three-dimensional model within the data sensing range of each composite data acquisition device according to the optical video data and the infrared video data;

[0016] A regional object distribution model within the data sensing range of the corresponding composite data acquisition device is established according to the three-dimensional laser signal spectrum, and the central point is fitted with the corresponding position of the composite data acquisition device in the three regional models, and the regional optical three-dimensional model, the regional infrared three-dimensional model and the regional object distribution model are overlapped to obtain a regional visualization model;

[0017] The regional visualization model is composed of a number of scene objects, and each scene object is covered with a real-time thermal pattern, and each thermal pattern is marked with the real-time temperature value of each part of the corresponding scene object.

[0018] Furthermore, the process of establishing the multi-parameter visualization scene model includes:

[0019] Several matching image blocks are set at the edge of each regional visualization model, and the matching image blocks at the edges of the regional visualization models of adjacent composite data acquisition devices are matched with each other. All regional visualization models are spliced ​​according to the matching results to obtain a multi-parameter visualization scene model under the corresponding data upload cycle. The real-time smoke concentration at each location is marked according to the location of each composite data acquisition device and the corresponding smoke concentration change curve.

[0020] Furthermore, the process of setting the fire detection area according to the real-time smoke concentration and the real-time thermal pattern includes:

[0021] According to the distribution of houses in the fire scene, the multi-parameter visualization scene model is divided into several local fire supervision scenes, the smoke concentration alarm threshold is set, and the temperature alarm threshold is set for various scene objects respectively, and the data upload cycle is divided into several fire supervision time points. Then, each time a fire supervision time point starts, the fire anomaly detection module determines whether the real-time smoke concentration of each local fire supervision scene is greater than or equal to the smoke concentration alarm threshold, or determines whether there is a part of the scene object whose real-time temperature value is greater than or equal to the corresponding temperature alarm threshold;

[0022] According to the judgment results, the local fire supervision scene with fire hazards is split into several fire detection areas of the same size, and the local fire supervision scenes in the corresponding time period between the fire supervision time points when there were no fire hazards recently and when fire hazards currently appear are integrated to obtain the corresponding dynamic local fire supervision scene.

[0023] Furthermore, the process of obtaining the smoke source tracing vector includes:

[0024] Set several traceability time nodes in the corresponding time segment in milliseconds. Starting from the last traceability time node, select the fire detection area in the dynamic local fire supervision scene where the real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold, and record it as the initial fire detection area;

[0025] Then select the fire detection area whose real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold at the penultimate tracing time node, and generate the smoke diffusion vector between the corresponding tracing time nodes according to each selected fire detection area and its nearest initial fire detection area;

[0026] Repeat the above process of generating smoke diffusion vectors to obtain smoke diffusion vectors between each tracing time node;

[0027] Establish a three-dimensional coordinate system, map the smoke diffusion vectors of each tracing time node into the same three-dimensional coordinate system, and count down from the last pair of tracing time nodes in chronological order, and add each smoke diffusion vector to the smoke diffusion vector in the previous pair of tracing time nodes, wherein the smoke diffusion vectors to be added are adjacent and the angle between them is within 180 degrees;

[0028] After the smoke diffusion vector between the penultimate tracing time node and the penultimate tracing time node is added, the smoke diffusion vector after the vector addition is then added to the smoke diffusion vector between the penultimate tracing time node until the first tracing time node, thereby obtaining the smoke tracing vector.

[0029] Furthermore, the fire detection area associated with the smoke source tracing vector is recorded as a fire abnormality area, and when a scene object has a real-time temperature value greater than or equal to the corresponding temperature alarm threshold, the fire detection area associated with the scene object is recorded as a fire abnormality area.

[0030] Furthermore, the process of establishing equipment triggering decision according to the fire abnormality area includes:

[0031] According to the location of the fire abnormality area in the multi-parameter visualization scene model, the fire-fighting equipment closest to the fire abnormality area is linked, and the corresponding equipment triggering decision is generated according to the abnormality type corresponding to the fire abnormality area;

[0032] When the abnormal type corresponding to the fire abnormality area is abnormal smoke concentration, the equipment triggers the decision to call the smoke exhaust system and fire extinguishing device to eliminate the fire abnormality in the fire abnormality area;

[0033] If the abnormality type corresponding to the fire abnormality area is abnormal device temperature, the device trigger decision is to turn off the power of the corresponding scene object and call the fire extinguishing device to eliminate the fire abnormality in the fire abnormality area;

[0034] When the device triggers the decision execution, the fire anomaly detection module re-judges whether there is a fire anomaly area in the multi-parameter visualization scene model at the next fire supervision time point;

[0035] If there is a fire decision-making planning module that generates a device trigger decision again and executes it, if it is judged that there is a fire abnormality area at the same location for three consecutive fire supervision time points, a fire abnormality alarm will be generated to the fire scene management personnel until there is no fire abnormality area in the multi-parameter visualization scene model.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention can monitor multiple environmental parameters of the fire scene in real time through a composite data acquisition device, providing comprehensive and accurate fire information, and to a certain extent avoiding the risk of misjudgment of fire caused by single parameter monitoring. At the same time, the regional visualization model generated by multi-parameter data can intuitively display the actual situation of the fire scene, improving the efficiency of decision-making.

[0038] 2. The present invention divides the multi-parameter visualization scene model into multiple local fire supervision scenes and sets fire supervision time points, thereby improving the efficiency of locating the source of the fire and the fire response speed while realizing real-time monitoring of the abnormal fire area. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0041] like Figure 1 As shown, a multi-parameter comprehensive fire monitoring system based on the Internet of Things includes a cloud management terminal, and the cloud management terminal is communicatively connected to an application scenario visualization module, a fire anomaly detection module, and a fire decision-making planning module;

[0042] The application scenario visualization module is provided with a scenario data acquisition unit and a scenario visualization unit;

[0043] The scene data acquisition unit is used to set up several composite data acquisition devices to collect multiple real-time scene data of fire scenes;

[0044] The scene visualization unit is used to generate a regional visualization model within the data perception range of each composite data acquisition device according to various real-time scene data, and then splice each regional visualization model to obtain a multi-parameter visualization scene model according to the spatial position of each composite data acquisition device in the fire scene, and mark each position of the multi-parameter visualization scene model with real-time smoke concentration and real-time thermal pattern;

[0045] The fire anomaly detection module is used to divide the multi-parameter visualization scene model into multiple local fire supervision scenes, and set fire supervision time points. According to the real-time smoke concentration and real-time thermal pattern of the local fire supervision scenes between the fire supervision time points, several fire detection areas are marked in the multi-parameter visualization scene model, and then the fire anomaly area is located by obtaining the smoke tracing vector and the real-time thermal pattern;

[0046] The firefighting decision-making planning module is used to generate and execute equipment triggering decisions according to the location of the firefighting abnormality area in the multi-parameter visualization scene model until there is no firefighting abnormality area.

[0047] Further, the working principle of the present invention is described below by way of examples:

[0048] Installing several composite data acquisition devices in the fire scene, wherein the composite data acquisition devices have built-in infrared sensors, optical cameras, smoke sensors and laser sensors;

[0049] It should be noted that the composite data acquisition devices are randomly distributed at various locations of the fire scene, and the data perception radius of each composite data acquisition device is the same, and there are cross-overlapping parts between the data perception radius of the composite data acquisition devices at adjacent spatial locations;

[0050] The same data upload cycle is set for each composite data acquisition device. Whenever a data upload cycle starts, each composite data acquisition device obtains infrared video data, optical video data and smoke concentration change curve within its data sensing range, and at the same time, calls the laser sensor to emit a laser signal to the entire data sensing range, thereby generating a three-dimensional laser signal spectrum.

[0051] When the data upload cycle ends, each composite data acquisition device will generate infrared video data, optical video data, three-dimensional laser signal spectrum and smoke concentration change curve and upload them to the scene data acquisition unit;

[0052] When each composite data acquisition device uploads data, the scene data acquisition unit integrates all the data to generate a scene data set.

[0053] Further, the scene data acquisition unit synchronizes the scene data set with the scene visualization unit in real time, and the scene visualization unit first establishes a regional optical three-dimensional model and a regional infrared three-dimensional model within the data perception range of each composite data acquisition device according to the optical video data and the infrared video data;

[0054] Due to the penetrability of laser signals, the composite data acquisition device can collect scene data within the data sensing range through laser signals, which cannot be collected by optical cameras, such as cables, water pipes and other scene objects embedded in the wall;

[0055] Then, a regional object distribution model within the data sensing range of the corresponding composite data acquisition device is established according to the three-dimensional laser signal spectrum. Since the regional optical three-dimensional model, the regional infrared three-dimensional model and the regional object distribution model all correspond to the same composite data acquisition device, there must be a correlation between the three.

[0056] Then, the central point is fitted at the corresponding position of the composite data acquisition device in the three regional models, and the regional optical three-dimensional model, the regional infrared three-dimensional model and the regional object distribution model are overlapped to obtain a regional visualization model;

[0057] The regional visualization model is composed of several scene objects, and each scene object is covered with real-time thermal patterns, and each thermal pattern is marked with the real-time temperature value of each part of the corresponding scene object, wherein the scene objects include walls, scene electrical components, cables in the walls, water pipes, fire-fighting equipment, etc.

[0058] Furthermore, since there are cross-overlapping parts between the data perception radii of the composite data acquisition devices at adjacent spatial positions, there are common parts between the corresponding generated regional visualization models of the adjacent composite data acquisition devices, and then a number of matching image blocks are set at the edge position of each regional visualization model, and the matching image blocks at the edges of the regional visualization models of the adjacent composite data acquisition devices are matched with each other;

[0059] If the two matching image blocks are the same, set the same label for the corresponding two matching image blocks, otherwise do nothing;

[0060] When the matching image blocks of two regional visualization models match each other, the two regional visualization models are spliced ​​according to the locations of the matching image blocks with the same annotations. After the splicing operation is completed between all the current regional visualization models, the multi-parameter visualization scene model under the corresponding data upload cycle is obtained, and the real-time smoke concentration of each position in the multi-parameter visualization scene model is annotated according to the location of each composite data acquisition device and the corresponding smoke concentration change curve.

[0061] Further, every time the application scene visualization module generates a multi-parameter visualization scene model under a data upload cycle, the multi-parameter visualization scene model is synchronously sent to the fire anomaly detection module;

[0062] The fire anomaly detection module divides the multi-parameter visual scene model into n local fire supervision scenes according to the distribution of houses in the fire scene, and sets numbers a1, a2, a3, ..., a4, a5, a6, a7, a8, a9, a10, a111, a12, a13, ..., a14, a15, a16, a17, a18, a19, a20, a21, a31, a19, a21, a32, a10, a111, a12, a13, ... ...3, ..., a10, a111, a1 n , where n is an integer greater than 0;

[0063] Set the smoke concentration alarm threshold and set the temperature alarm threshold for various scene objects respectively, divide the data upload cycle into several fire supervision time points, and then each time a fire supervision time point starts, the fire anomaly detection module determines whether the real-time smoke concentration of each local fire supervision scene is greater than or equal to the smoke concentration alarm threshold, or determines whether there is a part of the scene object whose real-time temperature value is greater than or equal to the corresponding temperature alarm threshold;

[0064] If, between a pair of fire supervision time points, there is no local fire supervision scene where the real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold, and there is no scene object where the real-time temperature value is greater than or equal to the corresponding temperature alarm threshold, then it is judged that there is no fire hazard between the current fire supervision time points;

[0065] If a local fire supervision scene occurs where the real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold, it is determined that the corresponding local fire supervision scene has a fire hazard, and then the source is traced back according to the time node corresponding to the current fire supervision time point to find the most recent fire supervision time point where there is no fire hazard;

[0066] The local fire supervision scene with fire hazards is divided into several fire detection areas of the same size, and the local fire supervision scenes in the corresponding time period between the fire supervision time point when there was no fire hazard recently and the fire hazard currently appearing are integrated to obtain the corresponding dynamic local fire supervision scene;

[0067] Set several traceability time nodes in the corresponding time segment in milliseconds. Starting from the last traceability time node, select the fire detection area in the dynamic local fire supervision scene where the real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold, and record it as the initial fire detection area;

[0068] Then select the fire detection area whose real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold at the penultimate tracing time node, and generate the smoke diffusion vector between the corresponding tracing time nodes according to each selected fire detection area and its nearest initial fire detection area;

[0069] Repeat the above process of generating smoke diffusion vectors to obtain smoke diffusion vectors between each tracing time node;

[0070] Establish a three-dimensional coordinate system, map the smoke diffusion vectors of each tracing time node into the same three-dimensional coordinate system, and count down from the last pair of tracing time nodes in chronological order, and add each smoke diffusion vector to the smoke diffusion vector in the previous pair of tracing time nodes, wherein the smoke diffusion vectors to be added are adjacent and the angle between them is within 180 degrees;

[0071] After the smoke diffusion vector between the last tracing time node and the second last tracing time node is added, the smoke diffusion vector after the vector addition is added to the smoke diffusion vector between the third last tracing time node until the first tracing time node, thereby obtaining the smoke tracing vector, and recording the fire detection area associated with the smoke tracing vector as the fire abnormal area;

[0072] When a scene object with a real-time temperature value greater than or equal to the corresponding temperature alarm threshold appears, the fire detection area associated with the scene object is recorded as a fire abnormality area;

[0073] Whenever a fire abnormality area appears at a fire supervision point, the location of the fire abnormality area is marked in the multi-parameter visualization scene model, and then the multi-parameter visualization scene model is synchronously sent to the fire decision-making planning module.

[0074] Furthermore, the firefighting decision-making planning module links the firefighting equipment closest to the firefighting abnormality area according to the location of the firefighting abnormality area in the multi-parameter visualization scene model, and generates a corresponding equipment triggering decision according to the abnormality type corresponding to the firefighting abnormality area;

[0075] When the abnormal type corresponding to the fire abnormality area is abnormal smoke concentration, the equipment triggers the decision to call the smoke exhaust system and fire extinguishing device to eliminate the fire abnormality in the fire abnormality area;

[0076] If the abnormality type corresponding to the fire abnormality area is abnormal device temperature, the device trigger decision is to turn off the power of the corresponding scene object and call the fire extinguishing device to eliminate the fire abnormality in the fire abnormality area;

[0077] When the device triggers the decision execution, the fire anomaly detection module re-judges whether there is a fire anomaly area in the multi-parameter visualization scene model at the next fire supervision time point;

[0078] If there is a fire decision-making planning module that generates a device trigger decision again and executes it, if it is judged that there is a fire abnormality area at the same location for three consecutive fire supervision time points, a fire abnormality alarm will be generated to the fire scene management personnel until there is no fire abnormality area in the multi-parameter visualization scene model.

[0079] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A multi-parameter comprehensive fire monitoring system based on the Internet of Things, including a cloud management terminal, characterized in that: The cloud management terminal is communicatively connected to an application scenario visualization module, a fire anomaly detection module, and a fire decision-making planning module; The application scenario visualization module is provided with a scenario data acquisition unit and a scenario visualization unit; The scene data acquisition unit is used to set up several composite data acquisition devices to collect multiple real-time scene data of fire scenes; The scene visualization unit is used to generate a regional visualization model within the data perception range of each composite data acquisition device according to various real-time scene data, and then splice each regional visualization model to obtain a multi-parameter visualization scene model according to the spatial position of each composite data acquisition device in the fire scene, and mark each position of the multi-parameter visualization scene model with real-time smoke concentration and real-time thermal pattern; The fire anomaly detection module is used to divide the multi-parameter visualization scene model into multiple local fire supervision scenes, and set fire supervision time points. According to the real-time smoke concentration and real-time thermal pattern of the local fire supervision scenes between the fire supervision time points, several fire detection areas are marked in the multi-parameter visualization scene model, and then the fire anomaly area is located by obtaining the smoke tracing vector and the real-time thermal pattern; The firefighting decision-making planning module is used to generate and execute equipment triggering decisions according to the location of the firefighting abnormality area in the multi-parameter visualization scene model until there is no firefighting abnormality area.

2. According to the multi-parameter comprehensive fire monitoring system based on the Internet of Things in claim 1, it is characterized in that: The process of the composite data acquisition device collecting multiple real-time scene data of fire scenes includes: Several composite data acquisition devices are installed in the fire scene, and a data upload cycle is set for each composite data acquisition device. When a data upload cycle starts, each composite data acquisition device obtains the infrared video data, optical video data and smoke concentration change curve within its data perception range, and at the same time calls the laser sensor to emit a laser signal to the entire data perception range, thereby generating a three-dimensional laser signal spectrum.

3. According to the multi-parameter comprehensive fire monitoring system based on the Internet of Things in claim 2, it is characterized in that: The process of establishing the regional visualization model includes: Establishing a regional optical three-dimensional model and a regional infrared three-dimensional model within the data sensing range of each composite data acquisition device according to the optical video data and the infrared video data; A regional object distribution model within the data sensing range of the corresponding composite data acquisition device is established according to the three-dimensional laser signal spectrum, and the central point is fitted with the corresponding position of the composite data acquisition device in the three regional models, and the regional optical three-dimensional model, the regional infrared three-dimensional model and the regional object distribution model are overlapped to obtain a regional visualization model; The regional visualization model is composed of a number of scene objects, and each scene object is covered with a real-time thermal pattern, and each thermal pattern is marked with the real-time temperature value of each part of the corresponding scene object.

4. According to the multi-parameter comprehensive fire monitoring system based on the Internet of Things as described in claim 3, it is characterized in that: The process of establishing the multi-parameter visualization scene model includes: Several matching image blocks are set at the edge of each regional visualization model, and the matching image blocks at the edges of the regional visualization models of adjacent composite data acquisition devices are matched with each other. All regional visualization models are spliced ​​according to the matching results to obtain a multi-parameter visualization scene model under the corresponding data upload cycle. According to the smoke concentration change curve of the location of each composite data acquisition device, the real-time smoke concentration of each position of the multi-parameter visualization scene model is marked.

5. The multi-parameter comprehensive fire monitoring system based on the Internet of Things according to claim 4 is characterized in that: The process of setting up fire detection areas based on real-time smoke concentration and real-time thermal patterns includes: The multi-parameter visualization scene model is divided into several local fire supervision scenes, and the smoke concentration alarm threshold is set, and the temperature alarm threshold is set for various scene objects respectively. The data upload cycle is divided into several fire supervision time points. Then, each time a fire supervision time point starts, it is judged whether the real-time smoke concentration of each local fire supervision scene is greater than or equal to the smoke concentration alarm threshold, or whether there is a part of the scene object whose real-time temperature value is greater than or equal to the corresponding temperature alarm threshold; According to the judgment results, the local fire supervision scene with fire hazards is split into several fire detection areas of the same size, and the local fire supervision scenes in the corresponding time period between the fire supervision time points when there were no fire hazards recently and when fire hazards currently appear are integrated to obtain the corresponding dynamic local fire supervision scene.

6. The multi-parameter comprehensive fire monitoring system based on the Internet of Things according to claim 5 is characterized in that: The process of obtaining the smoke source tracing vector includes: Set several traceability time nodes in the corresponding time segment in milliseconds. Starting from the last traceability time node, select the fire detection area in the dynamic local fire supervision scene where the real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold, and record it as the initial fire detection area; Then select the fire detection area whose real-time smoke concentration is greater than or equal to the smoke concentration alarm threshold at the penultimate tracing time node, and generate the smoke diffusion vector between the corresponding tracing time nodes according to each selected fire detection area and its nearest initial fire detection area; Establish a three-dimensional coordinate system, map the smoke diffusion vectors of each tracing time node into the same three-dimensional coordinate system, count down from the last pair of tracing time nodes in chronological order, and add each smoke diffusion vector to the smoke diffusion vector in the previous pair of tracing time nodes; After the smoke diffusion vector between the penultimate tracing time node and the penultimate tracing time node is added, the smoke diffusion vector after the vector addition is then added to the smoke diffusion vector between the penultimate tracing time node until the first tracing time node, thereby obtaining the smoke tracing vector.

7. The multi-parameter comprehensive fire monitoring system based on the Internet of Things according to claim 6 is characterized in that: The fire detection area associated with the smoke source tracing vector is recorded as the fire abnormality area, and when a scene object has a real-time temperature value greater than or equal to the corresponding temperature alarm threshold, the fire detection area associated with the scene object is recorded as the fire abnormality area.

8. The multi-parameter comprehensive fire monitoring system based on the Internet of Things according to claim 7 is characterized in that: The process of establishing equipment triggering decisions based on fire anomaly areas includes: According to the location of the fire abnormality area in the multi-parameter visualization scene model, the fire-fighting equipment closest to the fire abnormality area is linked, and the corresponding equipment triggering decision is generated according to the abnormality type corresponding to the fire abnormality area; When the abnormal type corresponding to the fire abnormality area is abnormal smoke concentration, the equipment triggers the decision to call the smoke exhaust system and fire extinguishing device to eliminate the fire abnormality in the fire abnormality area; If the abnormality type corresponding to the fire abnormality area is abnormal device temperature, the device trigger decision is to turn off the power of the corresponding scene object and call the fire extinguishing device to eliminate the fire abnormality in the fire abnormality area; When the device triggers the decision execution, the fire anomaly detection module re-judges whether there is a fire anomaly area in the multi-parameter visualization scene model at the next fire supervision time point; If there is a fire decision-making planning module that generates a device trigger decision again and executes it, if it is judged that there is a fire abnormality area at the same location for three consecutive fire supervision time points, a fire abnormality alarm will be generated to the fire scene management personnel until there is no fire abnormality area in the multi-parameter visualization scene model.

Images