Skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device and method

Abstract

This invention discloses a skid-mounted AI-controlled active air supply and self-priming three-dimensional fire plume reconstruction device and method, involving the interdisciplinary field of fire protection engineering and intelligent control technology. It includes a three-dimensional fire plume reconstruction combustion system, a modular integrated system, a multimodal monitoring system, a multidimensional wind field simulation system, and an AI intelligent iterative control system. The three-dimensional fire plume reconstruction combustion system can reproduce the three-dimensional flame gradient morphology. The modular integrated system consists of multiple skid-mounted units, integrating gas supply, power distribution control, and explosion-proof devices, supporting all-terrain mobile deployment. The multimodal monitoring system collects data through panoramic cameras, thermocouples, and ultrasonic anemometers, and accurately simulates complex wind fields in conjunction with the multidimensional wind field simulation system. The AI ​​intelligent iterative control system integrates flame morphology, temperature field, and wind field data in real time, dynamically optimizing the gas valve opening and fan tilt angle. This achieves high-fidelity reproduction of three-dimensional flame gradient morphology fire scenes under the coupling effect of real-world wind fields.

Description

Technical Field

[0001] This invention relates to the field of interdisciplinary technology of fire protection engineering and intelligent control technology, and more specifically to a skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device and method. Background Technology

[0002] High-precision reproduction of fire scenarios is a core challenge in fire science research and fire-resistant engineering design. Real fires possess highly complex dynamic characteristics: flames in a fire often exhibit a pattern of high temperature in the middle and a natural decreasing temperature gradient at both ends, and are constantly swaying due to environmental winds, forming a dynamically changing, non-uniform temperature field throughout the fire space. Furthermore, structures such as buildings, bridges, and tunnels are typically subjected to intermittent direct exposure to flames in real fires. Current mainstream fire testing equipment, including enclosed-space homogenizing electric furnaces and natural gas fire testing furnaces, has significant limitations: on the one hand, it cannot simulate the dynamic characteristics of real fires; on the other hand, it is difficult to reproduce the direct exposure of structures by flames. This inadequacy of fire testing equipment capabilities prevents the conduct of high-precision damage mechanism research under real fire scenarios, becoming a key bottleneck restricting the development of fire science and the improvement of engineering protection design; specifically as follows:

[0003] Firstly, traditional uniform temperature electric furnace technology is mostly limited to the size of the combustion space: the size of the specimen that can be accommodated in the furnace cavity is no more than 1m, and the entire space is a uniform temperature environment. No open flame can be generated throughout the process, and the coupling effect of the environmental wind field cannot be considered.

[0004] Secondly, existing natural gas fire test furnaces generally suffer from poor scene adaptability: the most commonly used fixed combustion furnaces can only be installed in fixed laboratory settings, requiring a large space and making them difficult to deploy quickly to different sites; moreover, the fire furnaces have limited functions, only able to simulate specific standard temperature rise curves, and although they have natural gas flares, they do not have flame control or regulation, and are only used for heating the furnace space, so the temperature inside the furnace is still close to a uniform temperature.

[0005] Third, existing fire experiment equipment completely fails to consider the three-dimensional combustion characteristics of fire plumes under real fire conditions. It cannot adjust the dynamic development process of flames, reproduce the non-uniform temperature gradient distribution and the significant deviation of environmental wind field coupling effects from real fires. It is difficult to adapt to the fire reproduction needs of special scenarios such as bridges and tunnels, resulting in a disconnect between fire scene reproduction and real disaster scenarios. At the same time, the lack of high-precision real non-uniform fire experiment combustion equipment and technology also leads to low reliability of material fire resistance performance testing and structural damage assessment data, which cannot provide accurate and effective support for complex fire-resistant design.

[0006] Furthermore, the contradiction between safety and testing efficiency further restricts technological development; traditional devices often adopt low-power combustion mode to reduce risks, but their energy release intensity is far lower than that of a real fire, resulting in limited practical value of the test results; while high-power equipment has safety hazards such as gas leakage and explosion due to lagging monitoring systems and slow emergency response, and the test process has poor controllability.

[0007] To address the aforementioned issues, there is an urgent need to develop a new type of combustion device that can overcome scene limitations, accurately reproduce the three-dimensional dynamic changes of flames and non-uniform fire environments under the influence of ambient wind, while simultaneously ensuring both safety and efficiency. Summary of the Invention

[0008] In view of this, the present invention provides a skid-mounted AI intelligent control active air supply self-absorption three-dimensional fire plume reconstruction device and method, which aims to solve the above-mentioned technical problems.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] This invention provides a three-dimensional flame plume reconstruction combustion system, including a frame-type combustion vehicle. A multi-directional flame injection unit is formed on the top of the frame-type combustion vehicle. Each flame injection unit includes several parallel gas collecting pipes, each with several flame pipes evenly distributed within it. The middle of each gas collecting pipe is connected to the gas input interface of the frame-type combustion vehicle via a multi-branching pipe. The upper part of the multi-branching pipe extends into the gas collecting pipe, forming an inverted V-shaped cut. A flow equalization component is installed inside the frame-type combustion vehicle. Air enters the frame-type combustion vehicle from its self-priming air supply interface and is supplied to the multi-directional flame injection unit through the flow equalization component. The inverted V-shaped cut enables a uniform, decreasing diffusion and diversion of gas from the middle of the gas collecting pipe to both ends. Combined with the flow equalization component, the system synchronously ignites each gas collecting pipe, reproducing the natural gradient morphology of a real three-dimensional flame and presenting a non-uniform temperature field in a real fire space.

[0011] Preferably, the flow equalization assembly includes an air flow equalization plate arranged along the airflow direction and two arc-shaped flow guide baffles. The two arc-shaped flow guide baffles divide the air inlet of the frame-type combustion vehicle into three paths: left, middle, and right, to perform initial air diversion. The air inlet is connected to a self-priming air supply interface. The air flow equalization plate is located at the lower part of the multi-directional flame jet unit and cooperates with the arc-shaped flow guide baffles to perform secondary air flow equalization.

[0012] Preferably, two arc-shaped baffles are symmetrically arranged inside the frame-type combustion vehicle, dividing the frame-type combustion vehicle into three air channels leading to the air distribution plate. The air distribution plate is a grid-type air distribution plate, and square through holes are evenly distributed on the grid-type air distribution plate.

[0013] Preferably, the gas collecting pipes are arranged at equal intervals, and each gas collecting pipe end is equipped with a high-energy pulse electric spark igniter. The gas inlet end of each multi-way branch pipe is connected to the gas input interface after the gas flows through the gas inlet pipe.

[0014] Preferably, the gas input interface and the self-priming air supply interface are located on the side of the frame-type combustion vehicle, and the self-priming air supply interface forms a rectangular gradually expanding flow channel into the interior of the frame-type combustion vehicle.

[0015] This invention also provides a skid-mounted AI-controlled active air supply and self-priming three-dimensional fire plume reconstruction device, including the aforementioned three-dimensional fire plume reconstruction combustion system, as well as a modular movable skid-mounted integrated system, a multimodal fire scene environment monitoring system, a multi-dimensional wind field simulation system, and an AI intelligent iterative control system. The three-dimensional fire plume reconstruction combustion system is used to simulate the three-dimensional flame gradient morphology. The modular movable skid-mounted integrated system is used to provide the three-dimensional fire plume reconstruction combustion system with adjustable gas flow. The multi-dimensional wind field simulation system corresponds to the three-dimensional fire plume reconstruction combustion system and is used to simulate complex wind field conditions. The multimodal fire scene environment monitoring system is used to monitor flame morphology and gradient, non-uniform temperature field, and wind field data. The AI ​​intelligent iterative control system uses an LSTM model to fuse flame morphology, non-uniform temperature field, and wind field data in real time, dynamically optimizes gas flow and wind field conditions, reproduces the real three-dimensional natural gradient morphology of flames in real fire scenarios, and forms a non-uniform spatial temperature field.

[0016] Preferably, the modular movable skid-mounted integrated system includes a first skid-mounted unit and a second skid-mounted unit; the first skid-mounted unit integrates a gas cylinder or fuel tank, a vaporizer, an explosion-proof fire extinguishing device, a first explosion-proof lighting device, and a gas pipeline for connecting to the gas input interface; the second skid-mounted unit integrates a power distribution cabinet, an electrical control cabinet, a frequency converter, and a second explosion-proof lighting device.

[0017] Preferably, the explosion-proof fire extinguishing device includes a linkage fire extinguishing system consisting of an explosion-proof control cabinet, an explosion-proof fan, a combustible gas detector, and several suspended ultra-fine dry powder fire extinguishers; the power distribution cabinet is used to provide power supply; the electrical control cabinet has a built-in PLC controller and communication module for coordinated control of various systems; the gas pipeline includes a main pipeline with a multi-channel manifold interface, one end of the main pipeline is connected to a gas cylinder or fuel tank through a high-pressure hose and an explosion-proof quick connector, and the other end of the main pipeline away from the gas cylinder or fuel tank is connected in series with a vaporizer, a main gas shut-off valve, an intelligent pressure reducing valve with pressure feedback, and an intelligent valve positioner with an integrated LCD screen, the input end of the intelligent valve positioner is connected to the pressure stabilizing branch of the air compressor through a filter, and the output end of the intelligent valve positioner is connected to the gas input interface through a corrugated flame arrester.

[0018] Preferably, the multi-dimensional wind field simulation system includes several high-dynamic axial flow fans, which are distributed in a matrix topology. Each high-dynamic axial flow fan is equipped with an independent frequency converter. The multi-modal fire field environment monitoring system includes a fire field panoramic video acquisition unit, a fire field spatial temperature field monitoring unit, and a wind speed monitoring unit. The fire field panoramic video acquisition unit captures visible light and thermal radiation images of the three-dimensional flame gradient morphology fire field in real time through a fire field panoramic camera, and monitors the fire field panoramic video. The fire field spatial temperature field monitoring unit consists of distributed thermocouples and infrared thermometers, used to collect fire field spatial temperature distribution data in real time and generate a three-dimensional temperature field model. The wind speed monitoring unit consists of ultrasonic anemometers and pressure sensors, used to monitor the wind speed, wind direction, and wind pressure data of the multi-dimensional wind field simulation system in real time. The data monitored by the multi-modal fire field environment monitoring system is transmitted in real time to the AI ​​intelligent iterative control system on the computer through a communication network.

[0019] This invention also provides a method for implementing a skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device, comprising the following steps:

[0020] S1: Transport the first skid-mounted unit, the second skid-mounted unit, the frame-type combustion vehicle, the multimodal fire environment monitoring system, and the multidimensional wind field simulation system to the target site, and deploy the first skid-mounted unit, the second skid-mounted unit, and the frame-type combustion vehicle to the preset positions.

[0021] S2: Deploy a multi-dimensional wind field simulation system according to the wind direction of the fire scene, and install and deploy a multi-modal fire scene environmental monitoring system;

[0022] S3: Activate the explosion-proof fire extinguishing device in the first skid-mounted unit;

[0023] S4: Connect the gas pipeline of the first skid-mounted unit to the gas input interface of the frame-type combustion vehicle through a quick interface, connect the output end of the distribution cabinet of the second skid-mounted unit to the power supply port of the high dynamic axial flow fan, and connect the communication interface of the electrical control cabinet to the PLC controller of the frame-type combustion vehicle and the high dynamic axial flow fan through a shielded twisted pair cable.

[0024] S5: Input the flame height and gradient, temperature gradient and wind speed vector of the target fire scene into the AI ​​intelligent iterative control system to generate initial control commands;

[0025] S6: Start the gas supply flow control valve and air compressor, adjust the gas flow through the intelligent valve positioner, and adjust the high dynamic axial flow fan speed through the frequency converter;

[0026] S7: Activate the multimodal fire environment monitoring system;

[0027] S8: Start the test, trigger the high-energy pulse electric spark igniter to ignite, and the multi-modal fire scene environment monitoring system collects and transmits data in real time to the AI ​​intelligent iterative control system;

[0028] S9: The AI ​​intelligent iterative control system, based on the deviation between real-time collected data and preset target parameters, uses an LSTM model to iteratively optimize the opening degree of the intelligent valve positioner and the wind speed command of the high dynamic axial flow fan to reproduce the real three-dimensional natural gradient and tilt angle of the flame under the influence of ambient wind in a real fire scenario, and present the non-uniform temperature field distribution in a real fire space.

[0029] S10: Test terminated, gas supply stopped, high dynamic axial flow fan shut down;

[0030] S11: The computer terminal automatically stores the data packets for monitoring and controlling the entire test process.

[0031] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device and method, which has the following beneficial effects:

[0032] 1. High-fidelity fire scene reproduction: Through a three-dimensional fire plume reconstruction combustion system and a multi-dimensional wind field simulation system, the system can highly reproduce the real three-dimensional flame natural gradient morphology, non-uniform temperature distribution, and complex wind field conditions in real fire scenes. By setting the inverted V-shaped cut of the multi-way branch pipe, the gas is diffused and diverted uniformly from the middle of the gas collection pipe to both ends. With the uniform air formed by the flow equalization component, each gas collection pipe is ignited synchronously, so that the multi-directional flame injection unit presents a three-dimensional flame gradient morphology. Then, by controlling the gas flow rate and the wind field conditions of the multi-dimensional wind field simulation system, the system finally realizes the physical reality of the three-dimensional combustion characteristics and flame morphology of the three-dimensional fire plume under the control of the AI ​​intelligent iterative control system. This provides a test platform that is closer to the real situation for fire science research and engineering protection design.

[0033] 2. Intelligent and Dynamic Optimization Control: The AI ​​intelligent iterative control system embeds an LSTM model, which can integrate flame morphology, temperature field and wind field data in real time, and dynamically adjust the opening of the intelligent valve positioner and the tilt angle of the high dynamic axial flow fan; it can realize real-time control of the natural gradient morphology of the real three-dimensional flame and high-fidelity presentation of the non-uniform spatial temperature field, improve the automation and control accuracy of the test process, and ensure the precise control of test conditions.

[0034] 3. Comprehensive monitoring and data visualization: The multimodal fire scene environment monitoring system combines panoramic cameras, distributed thermocouples, ultrasonic anemometers and other equipment to collect visible light images, thermal radiation images, temperature distribution data and wind field parameters of the fire scene in real time, and transmits them to the AI ​​control system through a high-speed communication network, supporting data visualization and facilitating real-time monitoring and analysis by test personnel.

[0035] 4. Modularity and Mobility: The modular mobile skid-mounted integrated system consists of two independent skid-mounted units, supporting all-terrain mobile deployment; it improves the flexibility and applicability of the device, making it easy to transport and deploy to different sites quickly, and meeting the fire simulation needs of various scenarios.

[0036] 5. Enhanced Safety: The integration of combustible gas detectors, suspended ultra-fine dry powder fire extinguishers, and explosion-proof fans forms a complete linkage fire extinguishing system. These devices can monitor the gas concentration in real time, and once an abnormality is detected, ventilation and fire extinguishing measures will be activated immediately to effectively prevent explosions and fires caused by gas leaks.

[0037] 6. Optimized gas and air mixing: The design of the inverted V-shaped cut and the grid-type flow distribution plate optimizes the gas flow and air distribution, ensuring that the gas and air are fully mixed and combusted, improving combustion efficiency and flame quality, and enhancing the stability and safety of the combustion process.

[0038] 7. Standardized operating procedures: It provides complete operating methods, clarifying the specific steps for each stage from equipment transportation, installation, commissioning to test operation and data storage; the standardized operating procedures make it easy for test personnel to quickly master and operate, improving test efficiency and data reliability.

[0039] 8. Data Integrity and Analysis Support: Stores full-cycle test data packages, including dynamic sequences of flame morphology, temperature field distribution, and matrix wind field control logs, providing comprehensive and accurate data support for fire scene reproduction analysis. This helps to conduct in-depth research on fire development patterns and structural damage mechanisms, and provides a scientific basis for fire-resistant design and fire prevention.

[0040] 9. Enhanced overall benefits: This device breaks through the limitations of traditional fire simulation equipment, integrating a variety of advanced technologies and functional modules. It not only improves the accuracy and reliability of fire scene reproduction, but also enhances the safety, efficiency and flexibility of the experiment, and has broad application prospects and important scientific research value. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0042] Figure 1 A schematic diagram of the skid-mounted AI intelligent control active air replenishment and self-absorption three-dimensional fire plume reconstruction device provided by the present invention;

[0043] Figure 2 A schematic diagram of the frame-type combustion vehicle provided by the present invention;

[0044] Figure 3 for Figure 2 Enlarged view of point A in the middle;

[0045] Figure 4 This is a schematic diagram of the structure of the multi-branch pipe provided by the present invention;

[0046] Figure 5 A schematic diagram of the structure of the first skid-mounted unit provided by the present invention;

[0047] Figure 6 This is a schematic diagram of the structure of the second skid-mounted unit provided by the present invention;

[0048] Figure 7 This is a schematic diagram of the structure of a gas pipeline provided by the present invention.

[0049] in:

[0050] 1. Multi-dimensional wind field simulation system; 2. Frame-type combustion vehicle; 3. First skid-mounted unit; 4. Second skid-mounted unit; 5. Gas pipeline; 101. High-dynamic axial flow fan; 102. Frequency converter; 103. Universal locking roller; 201. High-temperature resistant alloy steel pipe; 202. Short-section stainless steel flame tube; 203. High-energy pulse electric spark igniter; 204. Gas input interface; 205. Self-priming air supply interface; 206. Flange; 207. Rectangular gradually expanding flow channel; 208. Multi-branch distribution pipe; 209. Inverted V-shaped cut; 210. Arc-shaped flow guide baffle; 211. Grille-type flow distribution plate; 301. Gas cylinder or Fuel tank; 302, Gasification furnace; 303, Explosion-proof control cabinet; 304, Explosion-proof fan; 305, Combustible gas detector; 306, Suspended ultra-fine dry powder fire extinguisher; 307, Container; 308, Explosion-proof louvers; 309, Fire extinguisher ring hoisting bracket; 310, Touch panel; 311, LED explosion-proof light assembly; 312, Explosion-proof junction box; 401, Distribution cabinet; 402, Electrical control cabinet; 403, Second explosion-proof lighting device; 501, Main pipeline; 502, Gas main shut-off valve; 503, Intelligent pressure reducing valve; 504, Intelligent valve positioner; 505, Filter; 506, Corrugated flame arrester. Detailed Implementation

[0051] 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, and 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.

[0052] See appendix Figure 1 This invention discloses a skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device, comprising a three-dimensional fire plume reconstruction combustion system, an AI intelligent iterative control system, a modular movable skid-mounted integrated system, a multimodal fire scene environment monitoring system, and a multi-dimensional wind field simulation system 1. The three-dimensional fire plume reconstruction combustion system can simulate the three-dimensional flame gradient morphology. The modular movable skid-mounted integrated system can provide the three-dimensional fire plume reconstruction combustion system with adjustable gas flow. The multi-dimensional wind field simulation system 1 corresponds to the three-dimensional fire plume reconstruction combustion system and is used to simulate complex wind field conditions. The multimodal fire scene environment monitoring system is used to monitor flame morphology, temperature field, and wind field data. The AI ​​intelligent iterative control system fuses flame morphology, temperature field, and wind field data in real time through an LSTM model, dynamically optimizes gas flow and complex wind field conditions, and reproduces the three-dimensional flame gradient morphology in real fire scenarios.

[0053] See appendix Figure 2-4 The main body of the three-dimensional flame plume reconstruction combustion system is a frame-type combustion vehicle 2. Several sets of parallel, equidistantly distributed high-temperature alloy steel pipes 201 are integrated in the upper part of the inner cavity of the frame-type combustion vehicle 2. The high-temperature alloy steel pipes 201 serve as gas collection pipes. Each set of high-temperature alloy steel pipes 201 is equipped with several short-section stainless steel flame tubes 202 arranged axially at preset intervals. The high-temperature alloy steel pipes 201 and the short-section stainless steel flame tubes 202 are connected, forming a multi-directional flame injection unit. The short-section stainless steel flame tubes 202 are arranged in a vertical array. At the intersection of the outer edges of each set of high-temperature alloy steel pipes 201, a high-energy pulse electric spark igniter 203 corresponding to the short-section stainless steel flame tube 202 is provided. The high-energy pulse electric spark igniter 203 is connected to… The AI ​​intelligent iterative control system synchronizes the control circuit to achieve synchronous flame ignition. A gas input interface 204 and a self-priming air supply interface 205 are located at the bottom of one side wall of the frame-type combustion vehicle 2. The gas input interface 204 is sealed to the gas supply system via a flange 206. A multi-branch pipe 208 is connected to the middle of each set of high-temperature alloy steel pipes 201, and the multi-branch pipe 208 is connected to the gas input interface 204. The self-priming air supply interface 205 forms a rectangular, gradually expanding flow channel 207 into the interior of the frame-type combustion vehicle 2, achieving natural air supply through the negative pressure self-priming principle. Universal locking casters 103 are installed at the bottom of the frame-type combustion vehicle 2 to enable free and convenient movement of the equipment.

[0054] In this embodiment, the gas input interface 204 adopts a high-pressure resistant gas input interface that is well known to those skilled in the art.

[0055] Each multi-way branch pipe 208 has its air inlet end connected to the gas input interface 204 after the air flow is collected through the air inlet pipe. Each multi-way branch pipe 208 corresponds to a circular high-temperature alloy steel pipe 201. The connection end of the multi-way branch pipe 208 to the high-temperature alloy steel pipe 201 extends into the high-temperature alloy steel pipe 201 and forms an inverted V-shaped cut 209. The inverted V-shaped cut 209 reduces flow resistance and evenly distributes the gas to both sides of the high-temperature alloy steel pipe 201. The air inlet corresponding to the self-priming air supply interface 205 is a rectangular cross-section pipe. The frame-type combustion vehicle 2 is equipped with a flow equalization component, which includes two arc-shaped flow guide baffles 210 arranged along the airflow direction and an air flow equalization plate. The two arc-shaped flow guide baffles 210 divide the air inlet of the frame-type combustion vehicle 2 into three paths: left, middle, and right, to perform initial air diversion and ensure that the air is evenly distributed throughout the entire frame-type combustion vehicle 2. The air inlet is connected to the self-priming air supply interface 205. The air flow equalization plate is located at the lower part of the multi-directional flame injection unit and is used in conjunction with the arc-shaped flow guide baffles 210 to perform secondary air flow equalization.

[0056] Two arc-shaped baffles 210 are symmetrically arranged inside the frame-type combustion vehicle 2, dividing the frame-type combustion vehicle 2 into three air channels leading to the air equalization plate. The air equalization plate adopts a grid-type equalization plate 211, on which square through holes are evenly distributed to achieve secondary air equalization, which can realize the full mixing and combustion of gas and air. Each multi-way equalization branch pipe 208 passes through the grid-type equalization plate 211 and is connected to the corresponding high-temperature alloy steel pipe 201, and does not interfere with each other. The high-energy pulse electric spark igniter 203 is located at the end of each group of high-temperature alloy steel pipes 201 and corresponds to the short-section stainless steel flame tube 202 and is connected to the AI ​​intelligent iterative control system. The ignition is triggered by the control circuit of the AI ​​intelligent iterative control system to realize the synchronous ignition of the flame.

[0057] The shape and material of each component in this embodiment are preferred embodiments of this embodiment. Other changes to the shape and material of components that can achieve the effects of this embodiment are also within the scope of protection of this application.

[0058] See appendix Figure 5-6 The modular movable skid-mounted integrated system includes two independent skid-mounted units, namely the first skid-mounted unit 3 and the second skid-mounted unit 4.

[0059] See appendix Figure 5 The first skid-mounted unit 3 integrates a gas cylinder or fuel tank 301, a gasification furnace 302, an explosion-proof fire extinguishing device, a first explosion-proof lighting device, and a gas pipeline 5.

[0060] The explosion-proof fire extinguishing device can ensure safe ventilation within the first skid-mounted unit 3, prevent industrial explosions, and extinguish and retard fires in a timely manner.

[0061] The explosion-proof fire extinguishing device in the first skid-mounted unit 3 includes an explosion-proof control cabinet 303, an explosion-proof fan 304, combustible gas detectors 305, and several suspended ultra-fine dry powder fire extinguishers 306. The explosion-proof fan 304 is symmetrically arranged on both sides of the bottom of the inner wall of the container 307 of the first skid-mounted unit 3 and is electrically connected to the explosion-proof control cabinet 303. Explosion-proof louvers 308 are installed at the exhaust vent of the first skid-mounted unit 3. In this embodiment, three combustible gas detectors 305 are used. One combustible gas detector 305 is vertically suspended 50-100cm above the fitting flange joint of the gas pipeline 5. The other two combustible gas detectors 305 are respectively arranged in the middle of the top beam of the container above the gas cylinder or fuel tank 301. All three combustible gas detectors 305 are electrically connected to the explosion-proof control cabinet 303. The suspended ultra-fine dry powder fire extinguishers 306 are suspended and fixed to the container roof in a ring-shaped hoisting system. On bracket 309, the bottom of the fire extinguisher ring hoisting bracket 309 is 0.7 times the height of the container bottom. The suspended ultra-fine dry powder fire extinguisher 306 is driven by a pneumatic solenoid valve and connected to the explosion-proof control cabinet 303. The explosion-proof fan 304, combustible gas detector 305, and suspended ultra-fine dry powder fire extinguisher 306 are all electrically connected to the explosion-proof control cabinet 303. The explosion-proof control cabinet 303 is connected to a touch panel 310 suspended on the outer wall of container 307. The touch panel 310 displays the gas concentration, fan status, and fire extinguisher pressure data inside container 307 in real time. When any combustible gas detector 305 detects that the gas concentration exceeds the standard, the explosion-proof control cabinet 303 starts the explosion-proof fan 304 for forced ventilation, and at the same time triggers the pneumatic solenoid valve switch to activate the corresponding area of ​​suspended ultra-fine dry powder fire extinguisher 306 to release dry powder, realizing the fully automated handling of explosion prevention, fire extinguishing, and hazard disposal.

[0062] The first explosion-proof lighting device includes LED explosion-proof light groups 311 with an explosion-proof rating greater than or equal to ExdIIBT4, which are symmetrically installed on the inner ceiling of container 307. Each explosion-proof light of the LED explosion-proof light group 311 is connected in parallel to the explosion-proof control cabinet 303 through an explosion-proof junction box 312. The lighting switch control module is integrated into the touch panel 310. When the combustible gas detector 305 alarms or the suspended ultra-fine dry powder fire extinguisher 306 is activated, the touch panel 310 automatically switches the lighting mode to emergency high-brightness state.

[0063] The explosion-proof control cabinet 303, explosion-proof fan 304, combustible gas detector 305, suspended ultra-fine dry powder fire extinguisher 306, container 307, explosion-proof louver 308, fire extinguisher ring hoisting bracket 309, container top beam, LED explosion-proof light group 311, and other structures not described in detail in this embodiment all adopt existing products or structures well known to those skilled in the art, and the connection or control methods between them also adopt existing connection or control methods well known to those skilled in the art.

[0064] See appendix Figure 7 A gas cylinder or fuel tank 301 storing combustible gas is connected to a vaporizer 302 to convert liquid fuel into gaseous fuel. The gas pipeline 5 includes a high-pressure pipeline, valves, and safety devices to ensure stable and safe fuel delivery. The gas pipeline 5 includes a main pipeline 501 with a multi-channel manifold interface. One end of the main pipeline 501 is connected to the gas cylinder or fuel tank 301 via a high-pressure hose and an explosion-proof quick-connect fitting. The other end of the main pipeline 501, away from the gas cylinder or fuel tank 301, is sequentially connected to a vaporizer 302, a main gas shut-off valve 502, an intelligent pressure reducing valve 503 with pressure feedback, and an intelligent valve positioner 504 with an integrated LCD screen. The input end of the intelligent valve positioner 504 is filtered by a filter 5. 05 Connects to the pressure stabilizing branch of an existing air compressor known to those skilled in the art. The output end of the intelligent valve positioner 504 is connected to the gas input interface 204 of the frame-type combustion vehicle 2 through the corrugated flame arrester 506. The intelligent valve positioner 504 is connected to the AI ​​intelligent iterative control system and provides real-time feedback of valve opening information to the AI ​​intelligent iterative control system. The AI ​​intelligent iterative control system generates a valve opening adjustment command for the next period based on the preset target temperature parameter and sends it back to the intelligent valve positioner 504 to perform the opening adjustment, forming a closed-loop control link that dynamically matches the gas flow and temperature field. The AI ​​intelligent iterative control system can also control the intelligent pressure reducing valve 503 to cut off the gas delivery of the gas pipeline 5.

[0065] See appendix Figure 6 The second skid-mounted unit 4 integrates a power distribution cabinet 401, an electrical control cabinet 402, a frequency converter 102, and a second explosion-proof lighting device 403. The power distribution cabinet 401 provides power to the entire unit, and the electrical control cabinet 402 has a built-in PLC controller and communication module to achieve coordinated control of various systems. The two skid-mounted units adopt a standardized interface design, which can be quickly connected and realize seamless transmission of fuel, electricity, wind power and data. They are also equipped with universal locking rollers 103 for easy transportation and fixing.

[0066] The multi-dimensional wind field simulation system 1 includes several high-dynamic axial flow fans 101, which are distributed in a matrix topology. Each high-dynamic axial flow fan 101 is equipped with an independent frequency converter 102, which is integrated in the second skid-mounted unit 4. This enables continuous adjustment of the wind speed of the high-dynamic axial flow fan 101 and precise control of the fan tilt angle. The fan array adopts the PLC intelligent dynamic control strategy in the AI ​​intelligent iterative control system to simulate wind field conditions under complex fire scenarios, including uniform wind fields and gradient wind fields. The bottom of the multi-dimensional wind field simulation system 1 integrates universal locking rollers 103, which support all-terrain movement in various environments. The roller locking mechanism ensures stability during the wind field simulation process.

[0067] The multimodal fire scene environment monitoring system includes a panoramic video acquisition unit, a fire scene spatial temperature field monitoring unit, and a wind speed monitoring unit. The panoramic video acquisition unit uses a high-resolution camera and an infrared imaging device to capture visible light and thermal radiation images of the fire scene in real time and monitor the panoramic video of the fire scene. The temperature field monitoring unit consists of distributed thermocouples and an infrared thermometer, used to collect real-time temperature distribution data of the fire scene space and generate a three-dimensional temperature field model. The wind speed monitoring unit consists of an ultrasonic anemometer and a pressure sensor, used to monitor wind speed, wind direction, and wind pressure data in real time. The monitoring data is transmitted in real time to an AI intelligent iterative control system on a computer via a high-speed communication network to precisely control the experiment, record data throughout the process, and support data visualization.

[0068] In this embodiment, the high-resolution fire scene camera and infrared imaging device, distributed thermocouples and infrared thermometers, ultrasonic anemometers and pressure sensors are arranged in a manner well known to those skilled in the art.

[0069] The multi-dimensional wind field simulation system 1 is connected to the ultrasonic wind speed and direction sensor and pressure sensor in the multi-modal fire environment monitoring system to monitor multi-dimensional parameters of the wind field in real time and feed the monitoring data back to the AI ​​intelligent iterative control system. The system performs multi-scale error analysis and adjustment on the wind field monitoring data and preset target parameters to ensure accurate reproduction of wind field conditions.

[0070] The high-resolution camera and infrared imaging device, distributed thermocouples and infrared thermometers, ultrasonic anemometers and pressure sensors, high-dynamic axial flow fan 101, frequency converter 102, power distribution cabinet 401, electrical control cabinet 402, intelligent pressure reducing valve 503, intelligent valve positioner 504, and touch panel 310 in this embodiment adopt existing products or structures well known to those skilled in the art, and their connection or control method with the AI ​​intelligent iterative control system also adopts existing connection or control methods well known to those skilled in the art.

[0071] Other components not disclosed in detail in this embodiment adopt existing products or structures known to those skilled in the art, and their connection with other structures in this embodiment or with each other also adopt existing connection methods known to those skilled in the art.

[0072] The AI ​​intelligent iterative control system embeds a deep spatiotemporal sequence prediction model based on a long short-term memory (LSTM) architecture, well-known to those skilled in the art. It receives real-time measured data from a multimodal fire environment monitoring system, including fire scene images, spatiotemporal temperature distribution, and multidimensional wind field parameters, such as wind speed vectors, fan tilt angles, and fan turbulence intensity. The system performs spatiotemporal feature fusion and nonlinear modeling on the data using the LSTM model, outputting optimized dynamic opening values ​​for the gas control valve and target wind speed and angle control commands for the high-dynamic axial flow fan 101 in the multidimensional wind field simulation system 1. The predicted values ​​are transmitted in real-time to the PLC intelligent dynamic control system of the AI ​​intelligent iterative control system. Based on the real-time monitoring data, the PLC intelligent dynamic control system dynamically adjusts the opening of the gas control valve and the wind speed and tilt angle of each high-dynamic axial flow fan 101 through the electrical control cabinet 402, ensuring accurate control and reproduction of real fire scene test conditions. The PLC intelligent dynamic control system is interconnected with the main control computer via the OPC UA protocol, providing real-time visualization of the fire scene holographic heat map, gas valve opening deviation curve, fan operating status matrix, and control parameter iteration process.

[0073] In this embodiment, the electrical control cabinet 402, the intelligent valve positioner 504 with integrated LCD screen, the intelligent pressure reducing valve 503, and the AI ​​intelligent iterative control system all adopt existing products or structures known to those skilled in the art, and the connection or communication methods between them also adopt existing connection or communication methods known to those skilled in the art.

[0074] Other necessary structures not disclosed in this example also adopt existing products or structures known to those skilled in the art, and their connection or control methods with other structures and with each other also adopt existing connection or control methods known to those skilled in the art.

[0075] This embodiment also provides an implementation method for the above-mentioned skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device, which specifically includes the following steps:

[0076] Step 1: Transport the first skid-mounted unit 3, the second skid-mounted unit 4, the frame-type combustion vehicle 2, the multimodal fire scene environment monitoring system, and the multidimensional wind field simulation system to the target site, and arrange the first skid-mounted unit 3, the second skid-mounted unit 4, and the frame-type combustion vehicle 2 to the preset positions.

[0077] Step 2: Place the multi-dimensional wind field simulation system 1 according to the wind direction of the fire scene, install and deploy the multi-modal fire scene environmental monitoring system, and arrange the test specimens and frame-type combustion vehicle 2 in accordance with the methods known to those skilled in the art.

[0078] Step 3: Activate the explosion-proof fire extinguishing device in the first skid-mounted unit 3. The explosion-proof control cabinet 303, combustible gas detector 305, and suspended ultra-fine dry powder fire extinguisher 306 form a linked fire extinguishing system, which monitors the gas concentration in the skid-mounted unit in real time and maintains continuous operation. If at any time the combustible gas detector 305 detects that the gas concentration in the first skid-mounted unit 3 is ≥25%LEL, where LEL represents the lower explosive limit, the suspended ultra-fine dry powder fire extinguisher 306 is immediately activated and the power supply is cut off, and the test is immediately stopped.

[0079] Step 4: Connect the gas pipeline 5 of the first skid-mounted unit 3 to the gas input interface 204 of the frame-type combustion vehicle 2 through a quick interface known to those skilled in the art. In this embodiment, a flange 206 is used for connection. The output end of the power distribution cabinet 401 of the second skid-mounted unit 4 is connected to the power interface of the frame-type combustion vehicle 2 and the power supply port of the high dynamic axial flow fan 101 through waterproof cables. The communication interface of the electrical control cabinet 402 is connected to the PLC controller of the frame-type combustion vehicle 2 and the high dynamic axial flow fan 101 through shielded twisted pair cables.

[0080] Step 5: Input the flame height and gradient, temperature gradient and wind speed vector of the target fire scene into the AI ​​intelligent iterative control system to generate initial control commands;

[0081] Step 6: Start the gas supply valve and air compressor, and dynamically adjust the gas flow through the intelligent valve positioner 504 to synchronously drive the high dynamic axial flow fan 101 to generate the preset air field;

[0082] Step 7: Activate the multimodal fire environment monitoring system;

[0083] Step 8: Start the test, trigger the high-energy pulse electric spark igniter 203 to ignite the flame, and the multi-modal fire scene environment monitoring system will provide real-time feedback on flame shape, non-uniform temperature field and wind field data to the AI ​​intelligent iterative control system.

[0084] Step Nine: Use the fluid dynamics software FDS, which is well known to those skilled in the art, to simulate and extract parameters such as the height, gradient, and tilt angle of the flames under the influence of ambient wind in the real fire scene as preset target parameters for the experiment; the AI ​​intelligent iterative control system optimizes the instructions of the intelligent valve positioner 504 opening degree and the frequency converter frequency through the LSTM model, which is well known to those skilled in the art, based on the deviation between the real-time data monitored by the multi-modal fire scene environment monitoring system and the preset target parameters;

[0085] The system control commands for the frame-type combustion vehicle are generated according to the following formula (1);

[0086] (1)

[0087] In the formula: u gas(t) represents the system control command for the frame-type combustion vehicle; K p Let K be the proportional gain coefficient, e(t) be the error signal for analysis step t, and K be the proportional gain coefficient. i For the integral gain coefficient, K d α is the differential gain coefficient; α is the coupling coefficient, determined by the LSTM model embedded in the AI ​​intelligent iterative control system, which can be dynamically adjusted by those skilled in the art according to real-time operating conditions; LSTM gas (t)(T i ,F i ,W) is the correction term for the coupling of temperature and flame morphology of the frame burner output by the LSTM model. The input is the real-time monitoring data of temperature T, flame morphology F, and fan speed W of the frame burner in the t analysis step.

[0088] The system control command for the frequency converter is generated according to the following formula (2):

[0089] (2)

[0090] In the formula f fan (t) represents the system control command for the inverter frequency, and β is the dynamic weighting coefficient for the inverter frequency; LSTM fan (t)(W,θ) is the correction term for the frequency converter frequency in the t-step analysis output of the LSTM model, and the input is the fan speed W and the fan tilt angle θ;

[0091] By adding an intelligent valve positioner to the existing PID control method, the valve opening u is increased. gas Inverter frequency f fan It enables the conversion between target curves and control commands, reproduces the natural gradient morphology of real three-dimensional flames, and presents the non-uniform temperature field of real fire space.

[0092] Step 10: When the test is terminated, the AI ​​intelligent iterative control system closes the intelligent pressure reducing valve 503 and stops the operation of the high dynamic axial flow fan 101;

[0093] Step 11: Store the full-cycle test data package, including the dynamic sequence of flame morphology, temperature field distribution, and matrix wind field control logs, for fire scene reproduction and analysis.

[0094] This embodiment achieves uniform, decreasing diffusion and diversion of gas from the middle to both ends of the gas collecting pipe through the inverted V-shaped cut of the multi-way branch pipe 208. Combined with the uniform airflow formed by the flow equalization component, each gas collecting pipe is ignited synchronously, so that the multi-directional flame jet unit presents a three-dimensional flame gradient morphology, that is, the flame presents a high-temperature core in the middle and turbulent diffusion on the periphery. Then, through the optimization and control of the gas flow rate and the wind field conditions of the multi-dimensional wind field simulation system 1, the physical reality of the three-dimensional combustion characteristics of the flame morphology of the three-dimensional fire plume is realized under the control of the AI ​​intelligent iterative control system. This provides a test platform that is closer to the real situation for fire science research and engineering protection design.

[0095] In this specification, the apparatus disclosed in the embodiments corresponds to the methods disclosed in the embodiments, and the relevant parts can be referred to in the method section.

[0096] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A three-dimensional fire plume reconstruction combustion system, characterized in that, The system includes a frame-type combustion vehicle. A multi-directional flame injection unit is formed on the top of the frame-type combustion vehicle. Each multi-directional flame injection unit includes several parallel gas collecting pipes, each with several flame injectors evenly distributed throughout. The middle of each gas collecting pipe is connected to the gas input interface of the frame-type combustion vehicle via a multi-way branch pipe. The upper part of the multi-way branch pipe extends into the gas collecting pipe, forming an inverted V-shaped cut. A flow equalization component is installed inside the frame-type combustion vehicle. Air enters the frame-type combustion vehicle from its self-priming air supply interface and is supplied to the multi-directional flame injection unit through the flow equalization component. The inverted V-shaped cut enables a uniform, decreasing diffusion and diversion of gas from the middle of the gas collecting pipe to both ends, working in conjunction with the flow equalization component. The uniform airflow is generated, and each gas collection pipe is synchronously ignited to reproduce the natural gradient shape of a real three-dimensional flame, presenting a non-uniform temperature field in a real fire space. The uniform airflow assembly includes an airflow equalization plate set along the airflow direction and two arc-shaped flow guide baffles. The two arc-shaped flow guide baffles divide the air inlet of the frame-type combustion vehicle into three paths: left, middle, and right, to perform initial airflow equalization. The air inlet is connected to the self-priming air supply interface. The airflow equalization plate is set below the multi-directional flame injection unit and cooperates with the arc-shaped flow guide baffles to perform secondary airflow equalization. The gas input interface and the self-priming air supply interface are set on the side of the frame-type combustion vehicle. The self-priming air supply interface forms a rectangular gradually expanding flow channel into the interior of the frame-type combustion vehicle.

2. The three-dimensional fire plume reconstruction combustion system according to claim 1, characterized in that, Two arc-shaped baffles are symmetrically arranged inside the frame-type combustion vehicle, dividing the frame-type combustion vehicle into three air channels leading to the air distribution plate. The air distribution plate is a grid-type air distribution plate, and square through holes are evenly distributed on the grid-type air distribution plate.

3. The three-dimensional fire plume reconstruction combustion system according to claim 1, characterized in that, The gas collecting pipes are evenly spaced, and each gas collecting pipe end is equipped with a high-energy pulse electric spark igniter. The gas inlet end of each multi-way branch pipe is connected to the gas input interface after the gas flows through the gas inlet pipe.

4. A skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device, characterized in that, The system includes the three-dimensional fire plume reconstruction combustion system as described in any one of claims 1-3, and further includes a modular movable skid-mounted integrated system, a multimodal fire scene environment monitoring system, a multi-dimensional wind field simulation system, and an AI intelligent iterative control system. The three-dimensional fire plume reconstruction combustion system is used to simulate the three-dimensional flame gradient morphology. The modular movable skid-mounted integrated system is used to provide the three-dimensional fire plume reconstruction combustion system with adjustable gas flow. The multi-dimensional wind field simulation system corresponds to the three-dimensional fire plume reconstruction combustion system and is used to simulate complex wind field conditions. The multimodal fire scene environment monitoring system is used to monitor flame morphology and gradient, non-uniform temperature field, and wind field data. The AI ​​intelligent iterative control system uses an LSTM model to fuse flame morphology, non-uniform temperature field, and wind field data in real time, dynamically optimizes gas flow and wind field conditions, reproduces the real three-dimensional natural gradient morphology of flames in real fire scenarios, and forms a non-uniform spatial temperature field.

5. The skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device according to claim 4, characterized in that, The modular movable skid-mounted integrated system includes a first skid-mounted unit and a second skid-mounted unit; the first skid-mounted unit integrates a gas cylinder or fuel tank, a vaporizer, an explosion-proof fire extinguishing device, a first explosion-proof lighting device, and a gas pipeline for connecting to the gas input interface; the second skid-mounted unit integrates a power distribution cabinet, an electrical control cabinet, a frequency converter, and a second explosion-proof lighting device.

6. The skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device according to claim 5, characterized in that, The explosion-proof fire extinguishing device includes a linkage fire extinguishing system consisting of an explosion-proof control cabinet, an explosion-proof fan, a combustible gas detector, and several suspended ultra-fine dry powder fire extinguishers; the power distribution cabinet provides power supply; the electrical control cabinet has a built-in PLC controller and communication module for coordinated control of various systems; the gas pipeline includes a main pipeline with a multi-channel manifold interface, one end of which is connected to a gas cylinder or fuel tank via a high-pressure hose and an explosion-proof quick connector, and the other end of the main pipeline away from the gas cylinder or fuel tank is connected in series with a vaporizer, a main gas shut-off valve, an intelligent pressure reducing valve with pressure feedback, and an intelligent valve positioner with an integrated LCD screen; the input end of the intelligent valve positioner is connected to the pressure stabilizing branch of the air compressor via a filter, and the output end of the intelligent valve positioner is connected to the gas input interface via a corrugated flame arrester.

7. A skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device according to claim 4, characterized in that, The multi-dimensional wind field simulation system includes several high-dynamic axial flow fans, which are distributed in a matrix topology. Each high-dynamic axial flow fan is equipped with an independent frequency converter. The multi-modal fire field environment monitoring system includes a fire field panoramic video acquisition unit, a fire field spatial temperature field monitoring unit, and a wind speed monitoring unit. The fire field panoramic video acquisition unit captures visible light and thermal radiation images of the three-dimensional flame gradient morphology fire field in real time through a fire field panoramic camera, and monitors the fire field panoramic video. The fire field temperature field monitoring unit consists of distributed thermocouples and infrared thermometers, used to collect real-time fire field temperature distribution data and generate a three-dimensional temperature field model; the wind speed monitoring unit consists of ultrasonic anemometers and pressure sensors, used to monitor the wind speed, wind direction and wind pressure data of the multi-dimensional wind field simulation system in real time; the data monitored by the multi-modal fire field environment monitoring system is transmitted in real time to the AI ​​intelligent iterative control system on the computer via a communication network.

8. A method for implementing the skid-mounted AI intelligent control active air supply and self-priming three-dimensional fire plume reconstruction device as described in any one of claims 4-7, characterized in that, Includes the following steps: S1: Transport the first skid-mounted unit, the second skid-mounted unit, the frame-type combustion vehicle, the multimodal fire environment monitoring system, and the multidimensional wind field simulation system to the target site, and deploy the first skid-mounted unit, the second skid-mounted unit, and the frame-type combustion vehicle to the preset positions. S2: Deploy a multi-dimensional wind field simulation system according to the wind direction of the fire scene, and install and deploy a multi-modal fire scene environmental monitoring system; S3: Activate the explosion-proof fire extinguishing device in the first skid-mounted unit; S4: Connect the gas pipeline of the first skid-mounted unit to the gas input interface of the frame-type combustion vehicle through a quick interface, connect the output end of the distribution cabinet of the second skid-mounted unit to the power supply port of the high dynamic axial flow fan, and connect the communication interface of the electrical control cabinet to the PLC controller of the frame-type combustion vehicle and the high dynamic axial flow fan through a shielded twisted pair cable. S5: Input the flame height and gradient, temperature gradient and wind speed vector of the target fire scene into the AI ​​intelligent iterative control system to generate initial control commands; S6: Start the gas supply flow control valve and air compressor, adjust the gas flow through the intelligent valve positioner, and adjust the high dynamic axial flow fan speed through the frequency converter; S7: Activate the multimodal fire environment monitoring system; S8: Start the test, trigger the high-energy pulse electric spark igniter to ignite, and the multi-modal fire scene environment monitoring system collects and transmits data in real time to the AI ​​intelligent iterative control system; S9: The AI ​​intelligent iterative control system, based on the deviation between real-time collected data and preset target parameters, uses an LSTM model to iteratively optimize the opening degree of the intelligent valve positioner and the wind speed command of the high dynamic axial flow fan to reproduce the real three-dimensional natural gradient and tilt angle of the flame under the influence of ambient wind in a real fire scenario, and present the non-uniform temperature field distribution in a real fire space. S10: Test terminated, gas supply stopped, high dynamic axial flow fan shut down; S11: The computer terminal automatically stores the data packets for monitoring and controlling the entire test process.

Images