Fire-fighting information management method based on Internet of Things

Abstract

The invention relates to the technical field of fire-fighting remote monitoring systems, and provides a fire-fighting information management method based on the Internet of Things, and the method comprises the steps: collecting environment data and facility state data in real time through an Internet of Things sensing terminal, and transmitting the environment data and facility state data to an edge computing gateway; the edge computing gateway preprocesses the data and identifies fire features of video images; a processing result is transmitted to the cloud management platform for multi-source information fusion, and a fire-fighting situation map is constructed; the cloud management platform calculates a fire probability value by using an analysis model and judges a risk level; when the fire alarm level is reached, fire-fighting equipment linkage control is automatically triggered, a rescue route is generated in combination with a GIS, and the rescue route is pushed to a terminal, through edge calculation and cloud cooperation, early intelligent recognition and quick response of a fire disaster are achieved, the problems that traditional fire-fighting management lags behind, the false alarm rate is high, and data are isolated are effectively solved, and the fire-fighting management efficiency is improved. And the efficiency and reliability of fire control management are improved.

Description

Technical Field

[0001] This invention relates to the field of fire remote monitoring system technology, and in particular to a fire information management method based on the Internet of Things. Background Technology

[0002] With the acceleration of urbanization, the number of high-rise buildings, large complexes, and industrial plants has increased dramatically, making fire safety management increasingly difficult.

[0003] As disclosed in application number CN201420010967.3, a fire information management system based on the Internet of Things includes a static monitoring component comprising RFID electronic tags attached to fire information documents; a dynamic monitoring component comprising RFID electronic tags attached to buildings and RFID electronic tags attached to fire data acquisition modules; a front-end card reader capable of reading and / or modifying the information stored in the RFID electronic tags of the static and dynamic monitoring components in a contactless manner; a data exchange center located in the building's property management center; the front-end card reader comprising a communication module for interacting with the data exchange center; and the data exchange center exchanging data with the data center headquarters via a public communication network, with a firewall established between the public communication network and the data center headquarters.

[0004] However, existing fire management methods lack real-time monitoring of facility data such as fire water pressure, fire extinguisher pressure, and fire door opening status. Equipment malfunctions are often only discovered when a fire occurs. Furthermore, fire data, building information, and geographical location information are independent of each other, making it difficult to conduct comprehensive analysis and assist in decision-making. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a fire information management method based on the Internet of Things (IoT).

[0006] To achieve the above objectives, the present invention provides the following technical solution: A fire information management method based on the Internet of Things includes the following steps: S1: Real-time collection of environmental data and facility status data is achieved through IoT sensing terminals deployed in the fire monitoring area. The data collection includes periodic collection and event-triggered collection. S2: The IoT sensing terminal transmits the collected data to the edge computing gateway. The edge computing gateway preprocesses the received data and performs fire feature recognition and analysis on the image stream of the video surveillance equipment. S3: The edge computing gateway transmits the pre-processed data and identification results to the cloud management platform. The cloud management platform performs multi-source information fusion on the data to construct a fire situation map. S4: The cloud management platform uses analytical models to comprehensively analyze the fused data, calculate the probability value of fire occurrence, and determine the fire risk level based on the threshold range of the probability value. S5: When the determined fire risk level reaches the preset fire alarm level, the system automatically triggers the linkage control of fire protection facilities and generates rescue routes and related information through the geographic information system and pushes them to the rescue terminal.

[0007] Preferably, in step S1, the IoT sensing terminal includes a smoke sensor, a temperature sensor, a combustible gas sensor, a fire water pressure sensor, a fire liquid level sensor, and a video monitoring device; the event-triggered acquisition specifically means that when the collected environmental data exceeds a preset threshold or the data change rate exceeds a preset limit, the IoT sensing terminal immediately generates an alarm data packet and uploads it.

[0008] Preferably, in step S2, the preprocessing of the edge computing gateway includes data cleaning, noise reduction, and format unification; the fire feature identification and analysis specifically involves the edge computing gateway using a lightweight deep learning algorithm to analyze the real-time image stream of the video surveillance equipment, identify flame color features, smoke dynamic features, or temperature thermal imaging features, and generate an early warning signal if a feature is identified.

[0009] Preferably, in step S4, the analysis model calculates the probability value P of a fire occurrence based on the sensor value change rate, duration, and multi-sensor correlation; the graded alarm logic includes: if P < P1, it is determined to be a normal state, and only data is stored; if P1 ≤ P < P2, it is determined to be a hidden danger state, a hidden danger work order is generated and pushed to the maintenance terminal; if P ≥ P2, it is determined to be a fire state, and a level one fire alarm response is triggered.

[0010] Preferably, in step S5, the fire protection facility linkage control includes: sending instructions to the controller of the fire area to cut off non-fire protection power, start emergency lighting and evacuation indication system, and start fire broadcast system to play evacuation recordings.

[0011] Preferably, in step S5, the cloud management platform pre-stores the three-dimensional structural information of the building and the distribution information of fire protection facilities; when a fire alarm response is triggered, the cloud management platform combines the geographic information system to locate the fire point, automatically plans the optimal rescue route from the rescue starting point to the fire point, and pushes the information package containing fire alarm information, three-dimensional structural diagram of the building, distribution diagram of fire protection facilities and the optimal rescue route to the fire command center screen and mobile rescue terminal.

[0012] Preferably, the method further includes step S6: recording the entire process data from alarm to rescue end to form an electronic archive, and using machine learning algorithms to analyze historical rescue data to dynamically adjust the threshold parameters and linkage control strategies in the fire risk level model.

[0013] An IoT-based fire information management system includes a perception layer, a network layer, a platform layer, and an application layer. The perception layer consists of IoT sensing terminals composed of various sensors and cameras deployed in the fire monitoring area, used to collect environmental data and facility status data. The network layer includes an edge computing gateway, a wireless communication module, and an internet transmission network, used for data transmission, preprocessing, and edge-side analysis. The platform layer includes a data storage module, a data processing engine, an AI analysis module, and a GIS map service module, used for data fusion, intelligent judgment, and command issuance. The application layer includes a large screen in the fire command center, a mobile APP, and a fire maintenance management terminal, used to display monitoring information, receive alarm push notifications, and execute management operations.

[0014] Preferably, the edge computing gateway is configured with a ZigBee, LoRa, NB-IoT or 5G communication module to be compatible with different types of IoT sensing terminals; the edge computing gateway has a fire identification algorithm library deployed locally, which can independently perform local fire early warning and control in the event of an interruption of the connection with the cloud network.

[0015] The advantages of this invention are as follows: By combining edge computing video AI recognition with cloud-based sensor multi-dimensional data fusion models, the accuracy of fire identification is significantly improved compared to single-sensor alarms; by performing data cleaning and preliminary fire identification at the edge, the local gateway can still work independently and issue warnings even if the network is interrupted, ensuring the robustness of the system; an automated closed loop from perception to linkage is achieved, and by combining GIS and 3D building information, precise navigation and operational maps are provided for rescue forces, significantly improving rescue efficiency; through machine learning analysis of historical data, the system can self-optimize threshold parameters, becoming increasingly intelligent with use. Attached Figure Description

[0016] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart of the process of this invention; Figure 2 This is a structural diagram of the IoT sensing terminal of the present invention; Figure 3 This is a probability value distribution diagram of the present invention. Detailed Implementation

[0017] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0018] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0019] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0020] Example 1, combined with Figures 1-3 Explanation: A fire information management method based on the Internet of Things includes the following steps: S1: Real-time collection of environmental data and facility status data is achieved through IoT sensing terminals deployed in the fire monitoring area. The data collection includes periodic collection and event-triggered collection. S2: The IoT sensing terminal transmits the collected data to the edge computing gateway. The edge computing gateway preprocesses the received data and performs fire feature recognition and analysis on the image stream of the video surveillance equipment. S3: The edge computing gateway transmits the pre-processed data and identification results to the cloud management platform. The cloud management platform performs multi-source information fusion on the data to construct a fire situation map. S4: The cloud management platform uses analytical models to comprehensively analyze the fused data, calculate the probability value of fire occurrence, and determine the fire risk level based on the threshold range of the probability value. S5: When the determined fire risk level reaches the preset fire alarm level, the system automatically triggers the linkage control of fire protection facilities and generates rescue routes and related information through the geographic information system and pushes them to the rescue terminal.

[0021] In step S1, the IoT sensing terminal includes a smoke sensor, a temperature sensor, a combustible gas sensor, a fire water pressure sensor, a fire liquid level sensor, and a video monitoring device; the event-triggered acquisition specifically means that when the collected environmental data exceeds a preset threshold or the data change rate exceeds a preset limit, the IoT sensing terminal immediately generates an alarm data packet and uploads it.

[0022] Smoke sensors detect the concentration of smoke particles in the air; temperature sensors monitor the rate of increase or absolute temperature of the ambient environment; combustible gas sensors detect leaks of flammable and explosive gases such as methane, carbon monoxide, and liquefied petroleum gas; fire water pressure sensors and fire water level sensors monitor the status of the fire water supply network; water pressure sensors measure pipeline pressure, and water level sensors measure the water level in water tanks / pools; video surveillance equipment collects real-time images or video streams from the site. Periodic data collection refers to the equipment uploading data at preset times, such as every hour or 30 minutes, when there is no fire risk. This informs the system that the current environment is normal, thus saving power and bandwidth. Event-triggered data collection, on the other hand, changes its logic immediately upon the occurrence of an anomaly. Event-triggered data collection has two triggering conditions. First, it exceeds a preset threshold; If the threshold for smoke concentration is set to 0.5 mg / m³ and the threshold for temperature is set to 60℃, an alarm will be triggered immediately if the value detected by the sensor is greater than or equal to this value, regardless of whether the value increases slowly or suddenly, as long as it exceeds the limit.

[0023] Second, the rate of change of data exceeds the preset limit; For example, if the current room temperature is 25℃ and the set absolute threshold is 60℃, even though the temperature has not yet reached 60℃, the system detects that the temperature has soared from 25℃ to 35℃ in 1 second. The set change rate limit is 10℃ / minute. If the change rate exceeds this limit, even if the absolute temperature is not high, the device will determine it as abnormal and immediately alarm. This can effectively shorten the fire response time and avoid waiting until the temperature is already very high before alarming.

[0024] In step S2, the preprocessing of the edge computing gateway includes data cleaning, noise reduction, and format unification; the fire feature recognition and analysis specifically involves the edge computing gateway using a lightweight deep learning algorithm to analyze the real-time image stream of the video surveillance equipment, identify flame color features, smoke dynamic features, or temperature thermal imaging features, and generate an early warning signal if a feature is identified.

[0025] Data cleaning involves discarding obviously erroneous or meaningless data, noise reduction eliminates random fluctuations or interference signals in the data, and format unification converts sensor data from different brands and types into a standard language that the cloud platform can recognize.

[0026] Based on the color characteristics of flames, the algorithm searches for pixel areas in the image that match the color characteristics of flames and determines whether their shape resembles a burning flame, thereby eliminating interference from static objects such as red clothes and red lights. Based on the dynamic characteristics of smoke, the algorithm analyzes several consecutive frames and distinguishes it from white walls, fog, or dust by recognizing the dynamic pattern of continuously spreading, clump-shaped, grayish-white gas. Based on the characteristics of temperature-based thermal imaging, the algorithm needs to identify clusters of abnormally high-temperature areas in thermal imaging videos. This will improve the overall accuracy and speed of fire alarms.

[0027] In step S4, the analysis model calculates the probability value P of a fire occurrence based on the sensor value change rate, duration, and multi-sensor correlation. The graded alarm logic includes: if P < P1, it is determined to be a normal state, and only data is stored; if P1 ≤ P < P2, it is determined to be a hidden danger state, a hidden danger work order is generated and pushed to the maintenance terminal; if P ≥ P2, it is determined to be a fire state, and a level one fire alarm response is triggered.

[0028] Wherein, P1 and P2 are the probability thresholds preset by the system, with P1 being the upper limit threshold and P2 being the lower limit threshold. It can be assumed that P1 is 0.3 (30%) and P2 is 0.8 (80%).

[0029] The probability value p can be calculated using the following methods.

[0030] Definitions: S1 = Rate of change risk; S2 = Duration risk; S3 = Multi-sensor correlation risk; S1 = Current rate of change / Preset hazardous rate of change limit; For example: if the temperature rises by 1°C per second, and the preset danger limit is 5°C per second, then S1 = 1 / 5 = 0.2; The temperature rises by 5°C per second, the preset danger limit is 5°C per second, S1=5 / 5=1.0; S2 = (Current anomaly duration - Minimum confirmation time) / (Maximum allowable delay time - Minimum confirmation time); For example, the rule is set to consider an abnormality if it exceeds 10 seconds, and to confirm a fire if it exceeds 60 seconds. If it only lasts for 5 seconds, S2=0; If it lasts for 35 seconds, S2 = 0.5; If it lasts for more than 60 seconds, S2=1; S3 is a composite coefficient that depends on how many sensors are alarming simultaneously and whether their results are consistent. If only one sensor alarms, S3 = 0.3; If two sensors trigger an alarm, S3 = 0.7; If three or more sensors trigger an alarm, S3 = 1.0; In summary, the probability value P = (W1 × S1) + (W2 × S2) + (W3 × S3). Where W is the weight, W1+W2+W3=1; W1 is the rate of change weight, which is set to 0.3 because fires are usually very intense, and the rate of change is important. W2 is the duration weight, which is set to 0.2 to resist interference and prevent instantaneous spikes; W3 is the correlation weight, which is set to 0.5 because the correlation weight is the most important, and the data is most reliable through multi-source cross-validation.

[0031] If an item catches fire, the rate of change S1 = 1.0 because the temperature rises extremely quickly, reaching the preset limit. The duration S2, assuming it burns for 45 seconds, is then S2 = 0.7; S3 is calculated based on smoke alarm, heat alarm, and fire detected by video, in which case S3 = 1.0. Substituting into the formula P = (0.3 × 1.0) + (0.2 × 0.7) + (0.5 × 1.0) = 0.94; Since P = 0.94 ≥ P2 (0.8), the system determines that the system is in a fire state.

[0032] In step S5, the fire protection facility linkage control includes: sending instructions to the controller in the fire area to cut off non-fire protection power, activate emergency lighting and evacuation guidance systems, and activate the fire broadcast system to play evacuation recordings. This setup automatically cuts off non-fire protection power and simultaneously activates lighting, broadcasting, and evacuation systems, creating a safe escape environment and guiding people to evacuate quickly and orderly, thereby minimizing casualties and property damage.

[0033] In step S5, the cloud management platform pre-stores the building's three-dimensional structural information and fire protection facility distribution information. When a fire alarm is triggered, the cloud management platform uses a geographic information system to locate the fire location, automatically plans the optimal rescue route from the rescue starting point to the fire location, and pushes an information package containing fire alarm information, a three-dimensional building structural diagram, a fire protection facility distribution map, and the optimal rescue route to the fire command center's large screen and mobile rescue terminals. This setup, by integrating three-dimensional building information with real-time GIS data, automatically plans and pushes accurate operational maps and optimal routes to rescue forces, thereby significantly improving on-site command and dispatch efficiency and shortening rescue response time.

[0034] The system also includes step S6: recording data from the alarm to the end of the rescue process to form an electronic archive, and using machine learning algorithms to analyze historical rescue data to dynamically adjust threshold parameters and linkage control strategies in the fire risk level model. With this setup, by recording data throughout the entire process and using machine learning for self-optimization, the system can continuously correct alarm thresholds and linkage strategies, thereby continuously reducing false alarm rates and improving the accuracy and adaptability of fire management.

[0035] An IoT-based fire information management system comprises a perception layer, a network layer, a platform layer, and an application layer. The perception layer consists of IoT sensing terminals comprised of various sensors and cameras deployed in the fire monitoring area, used to collect environmental and facility status data. The network layer includes an edge computing gateway, a wireless communication module, and an internet transmission network for data transmission, preprocessing, and edge-side analysis. The platform layer includes a data storage module, a data processing engine, an AI analysis module, and a GIS map service module for data fusion, intelligent analysis, and command issuance. The application layer includes a large screen in the fire command center, a mobile app, and a fire maintenance management terminal for displaying monitoring information, receiving alarm notifications, and executing management operations. This configuration, by constructing a four-layer integrated architecture from bottom-level perception to top-level application, achieves closed-loop collaboration between data collection and transmission, edge computing, cloud fusion, and business applications, thereby comprehensively improving the system's data processing capabilities, response speed, and intelligent management level.

[0036] The edge computing gateway is configured with ZigBee, LoRa, NB-IoT, or 5G communication modules to ensure compatibility with different types of IoT sensing terminals. The edge computing gateway also has a locally deployed fire detection algorithm library, enabling independent local fire warning and control even when the connection to the cloud network is interrupted. This configuration, by integrating multi-mode communication modules with local fire detection algorithms, achieves broad compatibility with terminal devices and independent local warning and control when the cloud network is down, thus significantly ensuring the system's communication compatibility, operational reliability, and real-time response capabilities in complex environments.

[0037] The working principle of this invention is as follows: Environmental and facility status data are collected in real time via IoT sensing terminals and transmitted to an edge computing gateway. The edge computing gateway preprocesses the data and identifies fire features in video images. The processing results are transmitted to a cloud management platform for multi-source information fusion to construct a fire situation map. The cloud management platform uses an analysis model to calculate the fire probability value and determine the risk level. When a fire alarm level is reached, the fire-fighting facility linkage control is automatically triggered, and a rescue route is generated and pushed to the terminal using GIS. This invention, through edge computing and cloud collaboration, achieves early intelligent identification and rapid response to fires, effectively solving the problems of lagging traditional fire management, high false alarm rates, and data silos, thus improving the efficiency and reliability of fire management.

[0038] For those skilled in the art, the present invention is not limited to the details of the exemplary embodiments described above, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention; therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0039] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any minor modifications, equivalent substitutions, and improvements made to the above embodiments based on the technical essence of the present invention should be included within the protection scope of the present invention.

Claims

1. A fire information management method based on the Internet of Things, characterized in that, Includes the following steps: S1: Real-time collection of environmental data and facility status data is achieved through IoT sensing terminals deployed in the fire monitoring area. The data collection includes periodic collection and event-triggered collection. S2: The IoT sensing terminal transmits the collected data to the edge computing gateway. The edge computing gateway preprocesses the received data and performs fire feature recognition and analysis on the image stream of the video surveillance equipment. S3: The edge computing gateway transmits the pre-processed data and identification results to the cloud management platform. The cloud management platform performs multi-source information fusion on the data to construct a fire situation map. S4: The cloud management platform uses analytical models to comprehensively analyze the fused data, calculate the probability value of fire occurrence, and determine the fire risk level based on the threshold range of the probability value. S5: When the determined fire risk level reaches the preset fire alarm level, the system automatically triggers the linkage control of fire protection facilities and generates rescue routes and related information through the geographic information system and pushes them to the rescue terminal.

2. The fire information management method based on the Internet of Things according to claim 1, characterized in that, In step S1, the IoT sensing terminal includes a smoke sensor, a temperature sensor, a combustible gas sensor, a fire water pressure sensor, a fire liquid level sensor, and a video monitoring device; the event-triggered acquisition specifically means that when the collected environmental data exceeds a preset threshold or the data change rate exceeds a preset limit, the IoT sensing terminal immediately generates an alarm data packet and uploads it.

3. The fire information management method based on the Internet of Things according to claim 1, characterized in that, In step S2, the preprocessing of the edge computing gateway includes data cleaning, noise reduction, and format unification; the fire feature recognition and analysis specifically involves the edge computing gateway using a lightweight deep learning algorithm to analyze the real-time image stream of the video surveillance equipment, identify flame color features, smoke dynamic features, or temperature thermal imaging features, and generate an early warning signal if a feature is identified.

4. The fire information management method based on the Internet of Things according to claim 1, characterized in that, In step S4, the analysis model calculates the probability value P of a fire occurrence based on the sensor value change rate, duration, and multi-sensor correlation. The graded alarm logic includes: if P < P1, it is determined to be a normal state, and only data is stored; if P1 ≤ P < P2, it is determined to be a hidden danger state, a hidden danger work order is generated and pushed to the maintenance terminal; if P ≥ P2, it is determined to be a fire state, and a level one fire alarm response is triggered.

5. The fire information management method based on the Internet of Things according to claim 1, characterized in that, In step S5, the fire protection facility linkage control includes: sending instructions to the controller of the fire area to cut off non-fire protection power, start emergency lighting and evacuation guidance system, and start fire broadcast system to play evacuation recordings.

6. The fire information management method based on the Internet of Things according to claim 1, characterized in that, In step S5, the cloud management platform pre-stores the three-dimensional structural information of the building and the distribution information of fire protection facilities. When a fire alarm is triggered, the cloud management platform uses a geographic information system to locate the fire location, automatically plans the optimal rescue route from the rescue starting point to the fire location, and pushes an information package containing fire alarm information, a three-dimensional structural diagram of the building, a distribution diagram of fire protection facilities, and the optimal rescue route to the fire command center screen and mobile rescue terminal.

7. A fire information management method based on the Internet of Things according to claim 1, characterized in that, It also includes step S6: recording the entire process data from alarm to rescue end to form an electronic archive, and using machine learning algorithms to analyze historical rescue data to dynamically adjust the threshold parameters and linkage control strategies in the fire risk level model.

8. A fire information management system based on the Internet of Things (IoT), used to implement the fire information management method based on the Internet of Things as described in any one of claims 1-7, characterized in that, include: The system comprises a perception layer, a network layer, a platform layer, and an application layer. The perception layer consists of IoT sensing terminals composed of various sensors and cameras deployed in the fire monitoring area, used to collect environmental data and facility status data. The network layer includes an edge computing gateway, a wireless communication module, and an internet transmission network, used for data transmission, preprocessing, and edge-side analysis. The platform layer includes a data storage module, a data processing engine, an AI analysis module, and a GIS map service module, used for data fusion, intelligent judgment, and command issuance. The application layer includes a large screen in the fire command center, a mobile APP, and a fire maintenance management terminal, used to display monitoring information, receive alarm push notifications, and execute management operations.

9. A fire information management system based on the Internet of Things according to claim 8, characterized in that, The edge computing gateway is equipped with ZigBee, LoRa, NB-IoT or 5G communication modules to be compatible with different types of IoT sensing terminals; the edge computing gateway has a fire identification algorithm library deployed locally, which can independently perform local fire early warning and control in the event of an interruption of the connection with the cloud network.

Images