Fire water supply remote monitoring system based on Internet of Things

Abstract

The invention discloses a fire water supply remote monitoring system based on the Internet of Things, and relates to the technical field of fire safety. The method comprises the following steps: establishing a three-dimensional digital twinborn model according to visual thermal data, mapping steady-state hydraulic data to the three-dimensional digital twinborn model according to intelligent monitoring node position distribution, setting a difference threshold value, and comparing a pressure baseline and a flow baseline in the steady-state hydraulic data with the difference threshold value to obtain effective water use event data. Establishing a standard hydraulic flow model according to the effective water use event data, judging whether to trigger a double-layer wakeup mechanism according to the instantaneous hydraulic data and the visual thermodynamic data, calling the visual thermodynamic data according to a judgment result to update a three-dimensional digital twinborn model, and generating a local hydraulic situation map according to the instantaneous hydraulic data; and comparing the local hydraulic situation map with a standard hydraulic flow model, and generating and executing a pipeline regulation and control decision according to a comparison result.

Description

Technical Field

[0001] This invention relates to the field of fire safety technology, and specifically to a fire water supply remote monitoring system based on the Internet of Things. Background Technology

[0002] In the field of fire protection, ensuring that fire protection pipe networks can provide a stable and effective water supply during a fire is crucial. However, existing fire protection pipe network monitoring and control technologies have many shortcomings.

[0003] On the one hand, there is a lack of comprehensive and accurate collection of hydraulic data for fire protection pipe networks. Previously, it was difficult to collect steady-state and transient hydraulic data of fire protection pipe networks specifically based on different building types (such as high-rise residential buildings, commercial complexes, underground parking garages, etc.) and the characteristics of protected areas, making it impossible to accurately grasp the true hydraulic conditions of fire protection pipe networks under different scenarios. At the same time, the collection of visual thermal data from fire scenes is also inadequate, making it difficult to accurately capture the actual situation at the fire scene.

[0004] On the other hand, existing fire protection network models are not scientifically sound or precise enough. They fail to establish standard hydraulic models that meet actual firefighting needs based on accurate collected data. Furthermore, the data screening, classification, and processing during model building are not detailed enough, making it difficult to effectively distinguish between valid water use event data and background noise data. This results in models that cannot accurately reflect the hydraulic conditions of the fire protection network under ideal firefighting conditions. Moreover, when facing different hazard levels and water supply demands in protected areas, it is difficult to reasonably set pressure distribution and expected values ​​on the model, significantly reducing its practicality and guidance value.

[0005] Furthermore, existing fire control mechanisms are not intelligent or efficient enough during fires. They cannot respond to fire alarm signals in a timely and accurate manner, and the process of confirming and verifying abnormal sources is cumbersome and inaccurate, easily leading to misjudgments or omissions. When the water supply status of the fire pipeline network does not meet the fire-fighting requirements, it cannot quickly and accurately generate and execute effective pipeline control decisions, resulting in water waste or poor fire-fighting effects. To address these issues, an IoT-based remote monitoring system for fire water supply is proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a fire water supply remote monitoring system based on the Internet of Things to address the shortcomings of the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A fire water supply remote monitoring system based on the Internet of Things includes a cloud-edge collaborative platform, which is communicatively connected to a fire hydraulic acquisition module, a standard hydraulic model construction module, and a fire scene adaptive control module. The fire hydraulic data acquisition module is used to deploy intelligent monitoring nodes in the fire pipeline network and set data acquisition events, and then collect steady-state or instantaneous hydraulic data and visual thermal data under different data acquisition events through the intelligent monitoring nodes. The standard hydraulic model construction module is used to build a three-dimensional digital twin model based on visual thermal data, and to map steady-state hydraulic data onto the three-dimensional digital twin model according to the distribution of intelligent monitoring node locations. A difference threshold is set, and the pressure baseline and flow baseline in the steady-state hydraulic data are compared with the difference threshold to obtain effective water use event data. Then, a standard hydraulic flow model is built based on the effective water use event data. The fire-adaptive control module is used to set up a dual-layer wake-up mechanism. It determines whether to trigger the dual-layer wake-up mechanism based on instantaneous hydraulic data and visual thermal data. Based on the judgment result, it retrieves visual thermal data to update the three-dimensional digital twin model and generates a local hydraulic situation map based on instantaneous hydraulic data. It compares the local hydraulic situation map with the standard hydraulic flow model and generates and executes pipeline control decisions based on the comparison results.

[0008] Furthermore, the acquisition process of the steady-state hydraulic data and visual thermal data includes: Several standard fire-fighting scenarios are preset, and the standard fire-fighting scenarios include building type, hazard level and protected area plan; The fire hydraulic data acquisition module is communicatively connected to n intelligent monitoring nodes deployed on the fire pipeline network. Each intelligent monitoring node integrates a multi-parameter sensor, a high-definition infrared camera, and an edge computing unit, where n is a natural number greater than 0. Each intelligent monitoring node is assigned the number N1, N2, N3, ..., N n Several data collection events are set according to standard fire scenarios, and there is a difference between the building type or protected area corresponding to different data collection events; Each data acquisition event is executed m times. During the execution of each data acquisition event, the multi-parameter sensors on the intelligent monitoring node continuously collect steady-state hydraulic data of the fire protection pipeline network. Meanwhile, during fire drills or real-time fire monitoring events, high-definition infrared cameras collect visual thermal data of the fire scene or simulated heat source, where m is a natural number greater than 0.

[0009] Furthermore, the process of acquiring the instantaneous hydraulic data includes: During fire drills or real-time fire monitoring events at various intelligent monitoring nodes, multi-parameter sensors and high-definition infrared cameras located at the intelligent monitoring nodes use a synchronous triggering method to collect transient hydraulic data and real-time visual thermal data. The transient hydraulic data contains the same types of data as the steady-state hydraulic data, and the collection frequency of transient hydraulic data and real-time visual thermal data is once every 10ms.

[0010] Furthermore, the process of establishing the three-dimensional digital twin model includes: After all data acquisition events are executed m times, the fire hydraulic acquisition module will label the data acquisition events for each steady-state hydraulic data and visual thermal data, and then send all steady-state hydraulic data to the standard hydraulic model construction module, and send transient hydraulic data and real-time visual thermal data to the fire adaptive control module, where m is a natural number greater than 20; The standard hydraulic model construction module first establishes a three-dimensional digital twin model of the fire scene based on visual thermal data. Since the installation position of the intelligent monitoring node is fixed, its corresponding collected data has definite spatial coordinates in the three-dimensional digital twin model. Then, the steady-state hydraulic data obtained in each data collection event is mapped to the corresponding spatial position in the three-dimensional digital twin model.

[0011] Furthermore, the process of obtaining effective water use event data includes: Obtain the pressure baseline and flow baseline from each steady-state hydraulic data set, and starting from the data of the first intelligent monitoring node, perform correlation analysis on the pressure baseline and flow baseline of subsequent intelligent monitoring nodes based on the topology of the fire protection network and the direction of hydraulic transmission. Compare the data of adjacent intelligent monitoring nodes at the same time, and divide the fire scene into several fire protection zones. Set a first difference threshold and a second difference threshold, wherein the first difference threshold is less than the second difference threshold, and determine whether the difference in pressure and flow rate between adjacent smart monitoring nodes is greater than or equal to the second difference threshold. If the difference in variation is greater than or equal to the second difference threshold, the data segment is marked as valid water use event data; if the difference in variation is less than the second difference threshold, the data segment is marked as background noise data.

[0012] Furthermore, the process of establishing a standard hydraulic flow model includes: Based on the temporal distribution of effective water use event data in each steady-state hydraulic data under the same data acquisition event, and the spatiotemporal location distribution of other types of data in the three-dimensional digital twin model, the effective water use event data are spliced ​​together to obtain the standard hydraulic flow model of the fire-fighting area under the corresponding data acquisition event. For the standard hydraulic flow model, isobars are generated on the model surface according to its pressure distribution. The standard hydraulic flow model generated for the same data collection event but with different execution counts is statistically analyzed to determine the number of times the expected pressure value is reached in each fire zone, and a frequency threshold is set, where the frequency threshold is less than m. Pressure data points that occur less than the frequency threshold are smoothed and replaced with the average pressure value of the adjacent fire zone; otherwise, no action is taken. After all pressure data points have been processed, the optimized standard hydraulic flow model is obtained, and then the isobars on the model surface are mapped onto the optimized standard hydraulic flow model. Based on the hazard level and water supply demand of the protected area, a three-dimensional pressure point cloud distribution is set at the corresponding location on the optimized standard hydraulic flow model, and a reference pressure height is set. Then, based on the reference pressure height and the isobar distribution, the relative pressure expectation value is labeled for each three-dimensional pressure point cloud.

[0013] Furthermore, the dual-layer wake-up mechanism includes: First-level wake-up: If any intelligent monitoring node triggers an alarm, the edge computing unit is woken up and uses sensors to confirm whether the abnormal source exists; Second-level wake-up: Calls the sensors of other smart monitoring nodes near the smart monitoring node that triggered the alarm to collect data, and cross-validates the data collected by the smart monitoring node that triggered the alarm.

[0014] Furthermore, the process of determining whether the current water supply status meets the firefighting requirements includes: Once the fire adaptive control module is fully activated, the three-dimensional digital twin model is updated by establishing a three-dimensional digital twin model based on visual thermal data. Set a temperature threshold and compare the real-time temperature values ​​of each fire zone on each three-dimensional digital twin model with the temperature threshold. Mark the fire zone with a real-time temperature value greater than or equal to the temperature threshold as a fire source zone; otherwise, do not mark it. Meanwhile, since the location of the fire source and the intensity of the fire are different, the required water supply pressure and flow rate are also different. Therefore, based on the transient hydraulic data generated at each acquisition frequency, corresponding local hydraulic situation maps are generated, and each local hydraulic situation map is superimposed and compared with the optimized standard hydraulic flow model. Set a pressure difference threshold and compare the total pressure difference between the real-time pressure value and the corresponding expected relative pressure value of the local hydraulic situation map generated at each acquisition frequency. If the total pressure difference is less than or equal to the pressure difference threshold, it is determined that the water supply status of the corresponding fire-fighting pipeline network meets the fire-fighting requirements at the current sampling frequency. If the total pressure difference is greater than the pressure difference threshold, it is determined that the water supply status of the corresponding fire-fighting pipeline network at the current sampling frequency does not meet the fire-fighting requirements. At the same time, it is determined whether the local hydraulic situation map generated at each acquisition frequency covers the non-fire source protection area in the optimized standard hydraulic flow model. If so, it is determined that the current water supply status does not meet the fire extinguishing requirements. If it does not exist, then the current water supply status is determined to meet the fire extinguishing requirements.

[0015] Furthermore, the process of establishing pipeline control decisions includes: If it is determined that the water supply status of the fire protection pipeline network meets the fire extinguishing requirements at the current sampling frequency, no operation will be performed. If it is determined that the fire extinguishing requirements are not met, a corresponding pipeline control decision is generated and executed based on the reason for the determination. The pipeline control decision includes pump station speed regulation decision and zone valve regulation decision. If the cause is determined to be that the total pressure difference is greater than the pressure difference threshold, and the real-time pressure value of half of the smart monitoring points is greater than the corresponding relative pressure expectation value, or the real-time pressure value of half of the points is less than the corresponding relative pressure expectation value, then a pump station speed regulation decision or a zone valve adjustment decision is generated to reduce or increase the water supply pressure at the corresponding location in the fire protection pipeline network. If the cause is determined to be that the local hydraulic situation map covers the non-fire source protection area, then a zone valve adjustment decision is generated, including immediately and dynamically closing the branch valves leading to the non-fire source area through remote control. The generated biomimetic control decisions are sent to the corresponding fire protection zones to ensure that the water supply status of the fire protection facilities in the corresponding fire protection zones meets the fire extinguishing requirements. Repeat the above process of generating pipeline control decisions until it is determined that the real-time temperature values ​​of all fire protection zones are less than the temperature threshold.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This invention, through meticulous screening and classification of collected data, effectively distinguishes between valid water use event data and background noise data, and further subdivides the background noise data, enabling the standard hydraulic flow model to accurately reflect the hydraulic conditions of the fire-fighting network under ideal fire-fighting conditions. Simultaneously, pressure data points are smoothed and optimized, and a three-dimensional pressure point cloud distribution and relative pressure expectation value are set according to the hazard level and water supply demand of the protected area, improving the scientific rigor and practicality of the model and providing a reliable reference for the hydraulic control of fire-fighting networks.

[0017] 2. This invention, by comprehensively considering transient hydraulic data and real-time visual thermal data when judging the water supply status of the fire protection pipeline network, quickly generates and executes effective pipeline control decisions, including pump station speed adjustment decisions and zone valve adjustment decisions. This enables the water supply status of the fire protection pipeline network to be adjusted in a timely manner, effectively avoiding the waste of water resources, improving fire extinguishing efficiency, and to a certain extent ensuring that the water supply status of fire protection facilities meets fire extinguishing requirements. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a system framework diagram of a fire water supply remote monitoring system based on the Internet of Things as described in this invention.

[0020] Figure 2 This is a system flowchart of a fire water supply remote monitoring system based on the Internet of Things as described in this invention. Detailed Implementation

[0021] 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 only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 as well as Figure 2 As shown, a fire water supply remote monitoring system based on the Internet of Things includes a cloud-edge collaborative platform, which is communicatively connected to a fire hydraulic acquisition module, a standard hydraulic model construction module, and a fire scene adaptive control module. The fire hydraulic data acquisition module is used to deploy intelligent monitoring nodes in the fire pipeline network and set data acquisition events, and then collect steady-state or instantaneous hydraulic data and visual thermal data under different data acquisition events through the intelligent monitoring nodes. The standard hydraulic model construction module is used to build a three-dimensional digital twin model based on visual thermal data, and to map steady-state hydraulic data onto the three-dimensional digital twin model according to the distribution of intelligent monitoring node locations. A difference threshold is set, and the pressure baseline and flow baseline in the steady-state hydraulic data are compared with the difference threshold to obtain effective water use event data. Then, a standard hydraulic flow model is built based on the effective water use event data. The fire-adaptive control module is used to set up a dual-layer wake-up mechanism. It determines whether to trigger the dual-layer wake-up mechanism based on instantaneous hydraulic data and visual thermal data. Based on the judgment result, it retrieves visual thermal data to update the three-dimensional digital twin model and generates a local hydraulic situation map based on instantaneous hydraulic data. It compares the local hydraulic situation map with the standard hydraulic flow model and generates and executes pipeline control decisions based on the comparison results.

[0023] The working principle of the present invention will be explained below through examples: Staff members preset several standard fire protection scenarios on the cloud-edge collaboration platform. The standard fire protection scenarios include building type (such as high-rise residential buildings, commercial complexes, underground parking garages), hazard level, and protected area plan. The fire hydraulic data acquisition module is communicatively connected to n intelligent monitoring nodes deployed on the fire pipeline network. Each intelligent monitoring node integrates multi-parameter sensors (pressure sensor, flow meter, sound wave sensor, vibration sensor, smoke detector, etc.), high-definition infrared camera, and edge computing unit, where n is a natural number greater than 0. Each intelligent monitoring node is assigned the number N1, N2, N3, ..., N n Several data collection events are set according to standard fire protection scenarios. The building type or protected area corresponding to different data collection events may differ in any way, such as high-rise residential buildings or underground garages. Each data acquisition event is executed m times. During the execution of each data acquisition event, the multi-parameter sensors on the intelligent monitoring node continuously collect steady-state hydraulic data of the fire protection pipeline network. The steady-state hydraulic data includes pressure baseline, flow baseline, ambient sound spectrum, etc. Meanwhile, during fire drills or real-time fire monitoring events, high-definition infrared cameras collect visual thermal data of the fire scene or simulated heat source, where m is a natural number greater than 0. It should be noted that the field of view of the high-definition infrared camera is consistent with the core location of the fire hydrant or protected area it is associated with, so as to ensure that it can effectively capture the heat source information of the fire scene. During fire drills or real-time fire monitoring events at various intelligent monitoring nodes, multi-parameter sensors and high-definition infrared cameras located at the intelligent monitoring nodes use a synchronous triggering method to collect transient hydraulic data and real-time visual thermal data. The transient hydraulic data contains the same types of data as the steady-state hydraulic data, and the collection frequency of transient hydraulic data and real-time visual thermal data is once every 10ms.

[0024] Furthermore, after all data acquisition events are executed m times, the fire hydraulic acquisition module will label the data acquisition events for each steady-state hydraulic data and visual thermal data, and then send all steady-state hydraulic data to the standard hydraulic model construction module, and send transient hydraulic data and real-time visual thermal data to the fire adaptive control module, where m is a natural number greater than 20; The standard hydraulic model construction module first establishes a three-dimensional digital twin model of the fire scene based on visual thermal data. Since the installation position of the intelligent monitoring node is fixed, its corresponding collected data has definite spatial coordinates in the three-dimensional digital twin model. Then, the steady-state hydraulic data obtained in each data collection event is mapped to the corresponding spatial position in the three-dimensional digital twin model. Obtain the pressure baseline and flow baseline from each steady-state hydraulic data set, and starting from the data of the first intelligent monitoring node (N1), perform correlation analysis on the pressure baseline and flow baseline of subsequent intelligent monitoring nodes (N2, N3, ...) based on the topology of the fire protection network and the direction of hydraulic transmission. Compare the data of adjacent intelligent monitoring nodes at the same time, and divide the fire protection scenario into several fire protection zones. Set a first difference threshold and a second difference threshold, wherein the first difference threshold is less than the second difference threshold, and determine whether the difference in pressure and flow rate between adjacent smart monitoring nodes is greater than or equal to the second difference threshold. If the difference in variation is greater than or equal to the second difference threshold, the data segment is marked as valid water use event data; if the difference in variation is less than the second difference threshold, the data segment is marked as background noise data. The effective water use event data and background noise data annotation process were used again to determine the relationship between the fluctuation of the same node data under different collection events and the first difference threshold in the effective water use event data. Then, the background noise data was subdivided into fire pipeline fluctuation noise and equipment inherent vibration noise.

[0025] Furthermore, based on the temporal distribution of effective water use event data in each steady-state hydraulic data under the same data acquisition event, and the spatiotemporal location distribution of other types of data in the three-dimensional digital twin model, the effective water use event data are spliced ​​together to obtain the standard hydraulic flow model of the fire-fighting area under the corresponding data acquisition event. For the standard hydraulic flow model, isobars are generated on the model surface according to its pressure distribution. It should be noted that the standard hydraulic flow model is used to represent the hydraulic conditions that the fire protection pipe network should have under ideal fire extinguishing conditions; The standard hydraulic flow model generated under the same data collection event but with different execution times is statistically analyzed to determine the number of times the expected pressure value (such as the pressure requirement at the most unfavorable point) is reached in each fire zone, and a frequency threshold is set, where the frequency threshold is less than m. Pressure data points that occur less than the frequency threshold are smoothed and replaced with the average pressure value of the adjacent fire zone; otherwise, no action is taken. After all pressure data points have been processed, for pressure data points that are determined to have occurred less than the frequency threshold, the spatial coordinates of the pressure data point in the three-dimensional digital twin model are first located, and the pressure data of its corresponding pipeline branch and adjacent fire protection area are traced. After removing abnormal data that are also low-frequency in adjacent areas, the arithmetic mean of the pressure data in the effective adjacent areas is obtained and used to replace the original pressure data point to avoid local distortion of the model caused by individual abnormal data collection. For pressure data points that meet the frequency threshold requirements, the original values ​​are kept unchanged to ensure the authenticity of the data in the core area of ​​the model. After all pressure data points have been judged and processed, the updated full pressure data is remapped to the corresponding spatial location in the three-dimensional digital twin model. Combining the pipeline topology and hydraulic transmission law, the hydraulic parameters of each area in the three-dimensional digital twin model are calibrated a second time to correct the minor deviations generated during the data mapping process. Finally, the previously generated isobars are accurately superimposed onto the calibrated model surface to visualize the pressure distribution. At the same time, based on the hazard level classification of the protected area (such as high-risk area, medium-risk area, and low-risk area) and the differentiated water supply demand, a three-dimensional pressure point cloud distribution is constructed at the corresponding location of the model. A benchmark pressure height is set (usually based on the lowest point or the most unfavorable point of the pipeline network). Combined with the isobar gradient distribution, the corresponding relative pressure expectation value is labeled for each pressure point cloud. Finally, an optimized standard hydraulic flow model with accuracy, stability, and practicality is formed. The optimized standard hydraulic flow model is obtained, and then the isobars on the model surface are mapped onto the optimized standard hydraulic flow model. During firefighting, the pressure distribution of the fire-fighting pipeline changes with the water supply demand. Therefore, based on the hazard level and water supply demand of the protected area, a three-dimensional pressure point cloud distribution is set at the corresponding location on the optimized standard hydraulic flow model, and a reference pressure height is set. Then, based on the reference pressure height and the isobar distribution, the relative pressure expectation value is labeled for each three-dimensional pressure point cloud.

[0026] Furthermore, when the fire scene adaptive control module receives transient hydraulic data and real-time visual thermal data from the fire hydraulic acquisition module, it triggers a dual-layer wake-up mechanism: First-level wake-up: When any intelligent monitoring node (such as a smoke detector or a sudden pressure drop signal) triggers an alarm, the edge computing unit is woken up and uses sensors to confirm whether the abnormal source exists. Second-level wake-up: Call the sensors of other smart monitoring nodes near the smart monitoring node that triggered the alarm to collect data, and cross-validate the data collected by the smart monitoring node that triggered the alarm. For example, a high-definition infrared camera is called to confirm the presence of a heat source, or an adjacent acoustic sensor is called to confirm the sound of cracking or water spraying. The fire adaptive control module is only fully activated when two or more intelligent monitoring nodes confirm the fire (complete the second touch). Once the fire adaptive control module is fully activated, the three-dimensional digital twin model is updated by establishing a three-dimensional digital twin model based on visual thermal data. Set a temperature threshold and compare the real-time temperature values ​​of each fire zone on each three-dimensional digital twin model with the temperature threshold. Mark the fire zone with a real-time temperature value greater than or equal to the temperature threshold as a fire source zone; otherwise, do not mark it. Meanwhile, since the location of the fire source and the intensity of the fire are different, the required water supply pressure and flow rate are also different. Therefore, based on the transient hydraulic data generated at each acquisition frequency, corresponding local hydraulic situation maps are generated, and each local hydraulic situation map is superimposed and compared with the optimized standard hydraulic flow model. Set a pressure difference threshold and compare the total pressure difference between the real-time pressure value and the corresponding expected relative pressure value of the local hydraulic situation map generated at each acquisition frequency. If the total pressure difference is less than or equal to the pressure difference threshold, it is determined that the water supply status of the corresponding fire-fighting pipeline network meets the fire-fighting requirements at the current sampling frequency. If the total pressure difference is greater than the pressure difference threshold, it is determined that the water supply status of the corresponding fire-fighting pipeline network at the current sampling frequency does not meet the fire-fighting requirements. At the same time, it is determined whether the local hydraulic situation map generated at each acquisition frequency covers the non-fire source protection area in the optimized standard hydraulic flow model (i.e., there is a risk of hydraulic overload or accidental spraying). If so, it is determined that the current water supply status does not meet the fire extinguishing requirements. If it does not exist, then the current water supply status is determined to meet the fire extinguishing requirements.

[0027] Furthermore, if it is determined that the water supply status of the fire protection pipeline network meets the fire extinguishing requirements at the current sampling frequency, no operation will be performed. If it is determined that the fire extinguishing requirements are not met, a corresponding pipeline control decision is generated and executed based on the reason for the determination. The pipeline control decision includes pump station speed regulation decision and zone valve regulation decision. If the cause is determined to be that the total pressure difference is greater than the pressure difference threshold, and the real-time pressure value of half of the smart monitoring points is greater than the corresponding relative pressure expectation value (indicating that the local pressure of the pipeline is too high, and there is a risk of overpressure waste or damage), or the real-time pressure value is less than the corresponding relative pressure expectation value (indicating that the pipeline pressure is insufficient, and the range or flow is insufficient), then a pump station speed regulation decision or a zone valve regulation decision is generated to reduce or increase the water supply pressure at the corresponding location in the fire protection pipeline network. If the cause is determined to be that the local hydraulic situation map covers the non-fire source protection area, then a zone valve adjustment decision is generated, including immediately and dynamically closing the branch valves leading to the non-fire source area through remote control, while maintaining or increasing the opening of the main valves leading to the fire source area, so that the hydraulic resources are accurately wrapped around the fire site and effectively avoid water waste. The generated biomimetic control decisions are sent to the corresponding fire protection zones to ensure that the water supply status of the fire protection facilities in the corresponding fire protection zones meets the fire extinguishing requirements. Repeat the above process of generating pipeline control decisions until it is determined that the real-time temperature values ​​of all fire protection zones are less than the temperature threshold.

[0028] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A fire water supply remote monitoring system based on the Internet of Things, comprising a cloud-edge collaborative platform, characterized in that, The cloud-edge collaborative platform has a communication connection to a fire hydraulic data acquisition module, a standard hydraulic model construction module, and a fire scene adaptive control module. The fire hydraulic data acquisition module is used to deploy intelligent monitoring nodes in the fire pipeline network and set data acquisition events, and then collect steady-state or instantaneous hydraulic data and visual thermal data under different data acquisition events through the intelligent monitoring nodes. The standard hydraulic model construction module is used to build a three-dimensional digital twin model based on visual thermal data, and to map steady-state hydraulic data onto the three-dimensional digital twin model according to the distribution of intelligent monitoring node locations. A difference threshold is set, and the pressure baseline and flow baseline in the steady-state hydraulic data are compared with the difference threshold to obtain effective water use event data. Then, a standard hydraulic flow model is built based on the effective water use event data. The fire-adaptive control module is used to set up a dual-layer wake-up mechanism. It determines whether to trigger the dual-layer wake-up mechanism based on instantaneous hydraulic data and visual thermal data. Based on the judgment result, it retrieves visual thermal data to update the three-dimensional digital twin model and generates a local hydraulic situation map based on instantaneous hydraulic data. It compares the local hydraulic situation map with the standard hydraulic flow model and generates and executes pipeline control decisions based on the comparison results.

2. The fire water supply remote monitoring system based on the Internet of Things according to claim 1, characterized in that, The acquisition process for the steady-state hydraulic data and visual thermal data includes: Several standard fire-fighting scenarios are preset, and the standard fire-fighting scenarios include building type, hazard level and protected area plan; In the n intelligent monitoring nodes deployed on the fire protection pipeline, each intelligent monitoring node integrates a multi-parameter sensor, a high-definition infrared camera and an edge computing unit, where n is a natural number greater than 0; Each intelligent monitoring node is assigned the number N1, N2, N3, ..., N n Several data collection events are set according to standard fire scenarios, and there is a difference between the building type or protected area corresponding to different data collection events; Each data acquisition event is executed m times. During the execution of each data acquisition event, the multi-parameter sensors on the intelligent monitoring node continuously collect steady-state hydraulic data of the fire protection pipeline network. Meanwhile, during fire drills or real-time fire monitoring events, high-definition infrared cameras collect visual thermal data of the fire scene or simulated heat source, where m is a natural number greater than 0.

3. The fire water supply remote monitoring system based on the Internet of Things according to claim 2, characterized in that, The process of acquiring the instantaneous hydraulic data includes: During fire drills or real-time fire monitoring events at various intelligent monitoring nodes, multi-parameter sensors and high-definition infrared cameras located at the intelligent monitoring nodes collect transient hydraulic data and real-time visual thermal data through synchronous triggering.

4. The fire water supply remote monitoring system based on the Internet of Things according to claim 3, characterized in that, The process of establishing the three-dimensional digital twin model includes: After all data acquisition events are executed m times, the fire hydraulic acquisition module will label the data acquisition events for each steady-state hydraulic data and visual thermal data, and then send all steady-state hydraulic data to the standard hydraulic model construction module, and send transient hydraulic data and real-time visual thermal data to the fire adaptive control module, where m is a natural number greater than 20; First, a three-dimensional digital twin model of the fire scene is established based on visual thermal data. Then, the steady-state hydraulic data obtained from each data acquisition event is mapped to the corresponding spatial location in the three-dimensional digital twin model.

5. A fire water supply remote monitoring system based on the Internet of Things according to claim 4, characterized in that, The process of obtaining effective water use event data includes: Obtain the pressure baseline and flow baseline from each steady-state hydraulic data set, and starting from the data of the first intelligent monitoring node, perform correlation analysis on the pressure baseline and flow baseline of subsequent intelligent monitoring nodes based on the topology of the fire protection network and the direction of hydraulic transmission. Compare the data of adjacent intelligent monitoring nodes at the same time, and divide the fire scene into several fire protection zones. Set a first difference threshold and a second difference threshold, wherein the first difference threshold is less than the second difference threshold, and determine whether the difference in pressure and flow rate between adjacent smart monitoring nodes is greater than or equal to the second difference threshold. If the difference in variation is greater than or equal to the second difference threshold, the data segment is marked as valid water use event data; if the difference in variation is less than the second difference threshold, the data segment is marked as background noise data.

6. The fire water supply remote monitoring system based on the Internet of Things according to claim 5, characterized in that, The process of establishing a standard hydraulic flow model includes: Based on the temporal distribution of effective water use event data in each steady-state hydraulic data under the same data acquisition event, and the spatiotemporal location distribution of other types of data in the three-dimensional digital twin model, the effective water use event data are spliced ​​together to obtain the standard hydraulic flow model of the fire-fighting area under the corresponding data acquisition event. For the standard hydraulic flow model, isobars are generated on the model surface according to its pressure distribution. The standard hydraulic flow model generated for the same data collection event but with different execution counts is statistically analyzed to determine the number of times the expected pressure value is reached in each fire zone, and a frequency threshold is set, where the frequency threshold is less than m. Pressure data points that occur less than the frequency threshold are smoothed and replaced with the average pressure value of the adjacent fire zone; otherwise, no action is taken. After all pressure data points have been processed, the optimized standard hydraulic flow model is obtained, and then the isobars on the model surface are mapped onto the optimized standard hydraulic flow model. Based on the hazard level and water supply demand of the protected area, a three-dimensional pressure point cloud distribution is set at the corresponding location on the optimized standard hydraulic flow model, and a reference pressure height is set. Then, based on the reference pressure height and the isobar distribution, the relative pressure expectation value is labeled for each three-dimensional pressure point cloud.

7. A fire water supply remote monitoring system based on the Internet of Things according to claim 6, characterized in that, The dual-layer wake-up mechanism includes: First-level wake-up: If any intelligent monitoring node triggers an alarm, the edge computing unit is woken up and uses sensors to confirm whether the abnormal source exists; Second-level wake-up: Calls the sensors of other smart monitoring nodes near the smart monitoring node that triggered the alarm to collect data, and cross-validates the data collected by the smart monitoring node that triggered the alarm.

8. A fire water supply remote monitoring system based on the Internet of Things according to claim 7, characterized in that, The process of determining whether the current water supply status meets the firefighting requirements includes: Once the fire adaptive control module is fully activated, the three-dimensional digital twin model is updated by establishing a three-dimensional digital twin model based on visual thermal data. Set a temperature threshold and compare the real-time temperature values ​​of each fire zone on each three-dimensional digital twin model with the temperature threshold. Mark the fire zone with a real-time temperature value greater than or equal to the temperature threshold as a fire source zone; otherwise, do not mark it. Based on the transient hydraulic data generated at each acquisition frequency, corresponding local hydraulic situation maps are generated, and each local hydraulic situation map is overlaid and compared with the optimized standard hydraulic flow model. Set a pressure difference threshold, and compare the total pressure difference between the real-time pressure value and the corresponding expected relative pressure value of the local hydraulic situation map generated at each acquisition frequency. By judging the relationship between the total pressure difference and the pressure difference threshold, determine whether the water supply status meets the fire extinguishing requirements.

9. A fire water supply remote monitoring system based on the Internet of Things according to claim 8, characterized in that, The process of establishing pipeline control decisions includes: If it is determined that the water supply status of the fire protection pipeline network meets the fire extinguishing requirements at the current sampling frequency, no operation will be performed. If it is determined that the fire extinguishing requirements are not met, a corresponding pipeline control decision is generated and executed based on the reason for the determination. The pipeline control decision includes pump station speed regulation decision and zone valve regulation decision. If the cause is determined to be that the total pressure difference is greater than the pressure difference threshold, and the real-time pressure value of half of the smart monitoring points is greater than the corresponding relative pressure expectation value, or the real-time pressure value of half of the points is less than the corresponding relative pressure expectation value, then a pump station speed regulation decision or a zone valve adjustment decision is generated to reduce or increase the water supply pressure at the corresponding location in the fire protection pipeline network. If the cause is determined to be that the local hydraulic situation map covers the non-fire source protection area, then a zone valve adjustment decision is generated, including immediately and dynamically closing the branch valves leading to the non-fire source area through remote control. The generated biomimetic control decisions are sent to the corresponding fire protection zones to ensure that the water supply status of the fire protection facilities in the corresponding fire protection zones meets the fire extinguishing requirements.

Images