Visual coal field fire evolutionary process similar simulation test device and method

Abstract

The invention belongs to the technical field of coal field fire evolutionary analysis, and particularly provides a visual coal field fire evolutionary process analog simulation test device and method. The test device comprises a test bench, a physical similarity model, a fiber bragg grating temperature and stress integrated monitoring system, a multi-gas monitoring system, an image and video monitoring system and a computer analysis system. A plurality of opening frames are sequentially fixed to form test box bodies, the physical similarity model is arranged in the test box bodies, the inclination angle of the test box bodies is adjusted through an inclination angle adjusting mechanism, and the gas supply system and the ignition system are connected between the test box bodies. The fiber bragg grating temperature and stress integrated monitoring system, the multi-gas monitoring system and the image and video monitoring system are respectively in signal connection with the computer analysis system. According to the invention, visualization of coal field fire evolution processes under different conditions can be realized, and real-time monitoring of change rules of a temperature field, a seepage field and a fracture field in the coal field fire evolution processes under different conditions can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of coalfield fire evolution analysis, and particularly relates to a visual similar simulation test device and method for the evolution process of coalfield fires. Background Technique

[0002] In northwest China, coal resources are abundant. However, affected by arid climate and geological conditions, the problem of coalfield fires is prominent. Moreover, in these areas, the coal seams are thick, shallowly buried, have widespread outcrops, a short spontaneous combustion period, and the previous random mining and excavation of small coal mines have led to most coal seams being extremely prone to oxidation and spontaneous combustion, making it difficult to completely control. And existing fire-fighting technologies (such as the stripping method, grouting method, etc.) face many limitations in on-site applications and are difficult to effectively meet the requirements for controlling coalfield fires under complex combustion conditions, often lacking systematic and targeted treatment plans. Therefore, in-depth study of the combustion evolution law under the multi-field coupling action in coalfield fire areas, especially establishing a visual similar simulation test device for the evolution process of coalfield fires to visualize its evolution process and proposing targeted treatment plans, is of great significance for studying the evolution mechanism under the multi-field coupling action in coalfield fire areas and guiding the prevention and control of coalfield fires.

[0003] At present, the research on the combustion evolution process in coalfield fire areas mainly includes: 1) In terms of physical simulation, researchers construct a similar model composed of a heat-insulating base, coal seams, and multiple layers of overlying rocks in an experimental device and use an electric heating system to simulate the actual combustion environment; 2) In terms of numerical simulation, computational simulation technology is used to conduct a coupled analysis of the temperature distribution, fracture seepage, and rock layer deformation in the fire area to reveal the combustion propagation law; 3) In terms of on-site monitoring, a distributed temperature sensing network, gas component analyzer, and stress monitoring device are used to achieve fire source location and combustion state assessment; 4) In terms of engineering control, comprehensive measures such as excavation, grouting, and covering are combined to implement fire area control.

[0004] However, there are still the following deficiencies: (1) Most of the existing research speculates on the development process of fractures by collecting stress data during the combustion evolution process in coalfield fire areas, fails to truly and intuitively reflect the fracture development process, establish a corresponding fracture network, and achieve effective monitoring and visualization of the fracture development in coalfield fires.

[0005] (2) Most of the current research on physical similarity simulation in coalfield fire areas monitors parameters such as temperature and gas concentration to speculate on the combustion evolution process in coalfield fire areas, fails to observe and record the evolution process of coalfield fires, and achieve its visualization.

[0006] In summary, although the prior art has accumulated a certain experimental and theoretical basis, it still lacks sufficient originality and comprehensiveness to effectively monitor the crack development during the combustion evolution process of coalfield fire areas and visualize the evolution process of coalfield fires. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a similar simulation test device and method for visualizing the evolution process of coalfield fires in view of the deficiencies of the above-mentioned prior art, so as to solve the problems raised in the above background art.

[0008] To solve the above technical problems, the technical solution adopted by the present invention is as follows: In a first aspect, a similar simulation test device for visualizing the evolution process of coalfield fires includes a test bench, a physical similarity model, an integrated monitoring system for fiber Bragg grating temperature and stress, a multi-gas monitoring system, an image and video monitoring system, and a computer analysis system; Among them, a test box is connected to the test bench. The test box is formed by sequentially assembling a plurality of open frames. The open frames are slidably installed on the test bench, and clamping grooves for clamping heat-insulating glass are provided on both sides of the open frames; The physical similarity model includes a first rock layer, a coal layer, a second rock layer, a third rock layer, a fourth rock layer, and a fifth rock layer from bottom to top. The physical similarity model is arranged in the test box, and an inclination adjustment mechanism for adjusting the inclination angle of the test box is also connected to the test bench; A gas supply system and an ignition system are also provided on the test bench and connected to the test box. Gas is supplied into the test box through the gas supply system, and the physical similarity model is ignited through the ignition system to simulate a coalfield fire for a similar simulation test of the evolution process of coalfield fires; The integrated monitoring system for fiber Bragg grating temperature and stress is used to monitor the temperature and stress of the physical similarity model in the test box during the similar simulation test of the evolution process of coalfield fires in real time. The multi-gas monitoring system is used to monitor the gas concentration value in the test box after cracks are generated in the physical similarity model during the similar simulation test of the evolution process of coalfield fires. The image and video monitoring system is used to monitor the changes in the physical similarity model during the similar simulation test of the evolution process of coalfield fires, and to collect two-dimensional slice images by extracting the open frames after the test. The integrated monitoring system for fiber Bragg grating temperature and stress, the multi-gas monitoring system, and the image and video monitoring system are respectively connected to the computer analysis system through signal connections. The computer analysis system has a dynamic visualization model with multi-physical field coupling built in, and a three-dimensional crack field evolution model is embedded inside the visualization model. Data analysis and visualization processing operations are carried out through the computer analysis system.

[0009] As a further illustration of the present invention, a plurality of screw holes are further provided on the outer side of the opening frame. Two adjacent opening frames can be fixed through the cooperation of bolts and screw holes. A plurality of detection holes are also provided on the opening frame.

[0010] As a further illustration of the present invention, a set of pulleys are provided at the bottom of each opening frame. A plurality of parallel rail grooves are also formed on the base. A set of pulleys are respectively slidably connected in each rail groove. The pulleys are self-locking pulleys. A heat-resistant fiber cotton is also attached to the inner surface of the opening frame.

[0011] As a further illustration of the present invention, the inclination adjustment mechanism includes a lifting hydraulic cylinder and a hydraulic pump. Three lifting hydraulic cylinders are provided. One lifting hydraulic cylinder is provided at the side end of the base, and the other two lifting hydraulic cylinders are provided below the test box body on the base. The telescopic shafts of the lifting hydraulic cylinders are respectively connected to the test box body, and the lifting hydraulic cylinders are powered by the hydraulic pump.

[0012] As a further illustration of the present invention, the gas supply system includes an oxygen cylinder, a nitrogen cylinder, a gas flow meter, and an air compressor. The oxygen cylinder and the nitrogen cylinder are respectively communicated with the air compressor. Gas flow meters are also respectively connected to the gas outlet sections of the oxygen cylinder and the nitrogen cylinder. After the input gas is pressurized by the air compressor to reach the specified pressure, it is input into the test box body.

[0013] As a further illustration of the present invention, the ignition system includes a digital display electronic temperature controller, a heating rod, and a temperature sensor. The heating rod and the temperature sensor are respectively arranged in the physical similarity model. The heating rod and the temperature sensor are connected to the digital display electronic temperature controller. The temperature of the heating rod is set by the digital display electronic temperature controller and monitored by the temperature sensor.

[0014] As a further illustration of the present invention, the fiber Bragg grating temperature and stress integrated monitoring system includes an ultra-high temperature fiber Bragg grating sensor, a fiber Bragg grating network analyzer, and a terminal box; The ultra-high temperature fiber Bragg grating sensor can monitor the temperature and stress of the physical similarity model in the test box body during the similar simulation test of the coalfield fire evolution process. The ultra-high temperature fiber Bragg grating sensor is connected to the fiber Bragg grating network analyzer through the terminal box, and the fiber Bragg grating network analyzer is connected to the computer analysis system.

[0015] As a further description of the present invention, the multi-gas monitoring system consists of a multi-gas monitor and a signal line. The multi-gas monitor is connected to the air outlet at the top of the test chamber. When cracks occur in the physical similarity model in the test chamber during the similar simulation test of the coalfield fire evolution process, the detection head of the multi-gas monitor is inserted into the cracks to monitor and display the concentration values of gases such as CO, C2H4, CH4, CO2, C2H6, C2H2, O2, H2, and H2S, and transmit the data to the computer analysis system through the signal line. The image and video monitoring system consists of a bracket and three dual-spectrum high-definition cameras arranged on the bracket. The process of the similar simulation test of the coalfield fire evolution process in the test chamber is recorded by the dual-spectrum high-definition cameras, and the data is transmitted to the computer analysis system.

[0016] As a further description of the present invention, the computer analysis system records the data obtained by the fiber Bragg grating temperature and stress integrated monitoring system, the multi-gas monitoring system, and the image and video monitoring system.

[0017] In a second aspect, a method for similar simulation test of visualizing the coalfield fire evolution process includes the following steps: S1. Calculate and design the size and conditions of the physical similarity model of the actual experimental coalfield fire according to the similarity criterion. Build a test bench according to the size and conditions of the physical similarity model. Assemble and fix multiple open frames in sequence, and insert heat-insulating glass on the outermost open frame to form a test chamber. Arrange the physical similarity model in the test chamber, and use the inclination adjustment mechanism to adjust the test chamber to the set inclination. S2. Arrange the fiber Bragg grating temperature and stress integrated monitoring system in the physical similarity model for temperature and stress monitoring. S3. After the physical similarity model is left to stand and air-dried and solidified, signal-connect the fiber Bragg grating temperature and stress integrated monitoring system, the multi-gas monitoring system, and the image and video monitoring system to the computer analysis system respectively. S4. Supply gas to the physical similarity model through the gas supply system, and turn on the ignition system to ignite the physical similarity model for the similar simulation test of the coalfield fire evolution process. At this time, start the fiber Bragg grating temperature and stress integrated monitoring system and the image and video monitoring system for data monitoring. When cracks occur due to the collapse of the coalfield fire area in the test chamber, insert the detection head of the multi-gas monitor of the multi-gas monitoring system into the cracks for gas concentration value monitoring. After the combustion of the physical similarity model in S5, let it stand still to allow the test box to cool naturally to room temperature. Insert heat-insulating glass into the card slots on both sides of each open rack. Cut slices of the burned physical similarity model, disassemble the multiple open racks assembled, use the image and video monitoring system to record the two-dimensional slice images after each open rack slides out, obtain the two-dimensional slice image data set, and transmit the two-dimensional slice image data set to the computer analysis system. The computer analysis system uses Matlab to reconstruct the three-dimensional fracture model for the two-dimensional slice image data set, and uses ParaView to render and analyze the spatio-temporal distribution data measured by each system to construct a dynamic visualization model of multi-physical field coupling.

[0018] The present invention has the following advantages compared with the prior art: 1. The present invention can realize the visualization of the evolution process of coalfield fires under different conditions. Through the fiber Bragg grating temperature and stress integrated monitoring system, multi-gas monitoring system, image and video monitoring system, and computer analysis system, it can realize the real-time monitoring of the variation laws of the temperature field, seepage field, and fracture field during the evolution process of coalfield fires under different conditions.

[0019] 2. After the experiment on the test bench in the present invention is completed, each opening frame can be drawn out to collect a two-dimensional slice image dataset. The internal fracture field is reconstructed through Matlab, and a three-dimensional fracture field evolution model is constructed to visualize the fracture distribution and size inside the coalfield fire area evolution process. The three-dimensional fracture model can be imported into Fluent or Comsol software to further provide data support for numerical simulations related to the coalfield fire area evolution. The development of surface fractures and the law of smoke diffusion are recorded through the image and video monitoring system, so that the influence of the fire on the internal structure of the coal body can be intuitively observed, the temperature distribution on the coalfield surface can be captured in real time, and the high-temperature area and the fire spread trend can be displayed. The spatio-temporal distribution data measured by each system is rendered and analyzed through ParaView to construct a dynamic visualization model of multi-physical field coupling, realizing the visualization of the coalfield fire evolution process. Through the superposition display of the high-temperature area and the fracture expansion, the "critical state" can be located, such as the temperature mutation point or the fracture penetration area, providing a scientific basis for early warning. More intuitively revealing the coupling mechanism of heat-seepage-fracture in fire evolution deepens the understanding of the self-sustaining mechanism of coalfield fires. A visual decision-making sand table is provided for active prevention and control. The visualization model can intuitively display the high-temperature area and fracture distribution in the coalfield, helping researchers accurately locate the fire risk area and providing a basis for formulating targeted prevention and control measures. Evaluating the possible disaster degree caused by coalfield fires, including the spread range of the fire and the degree of damage to the surrounding environment, can provide an important reference for formulating fire extinguishing plans. For example, the position and grouting volume of fire extinguishing grouting can be determined according to the fracture distribution, improving the fire extinguishing efficiency and guiding the targeted implementation of engineering measures such as grouting and blocking. And the large amount of accumulated data can provide training samples for machine learning prediction models, promoting the development of fire prevention and control towards the intelligent direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic diagram of the overall structure of the present invention; Figure 2 is a schematic diagram of the opening frame of the present invention; Figure 3 is a schematic diagram after the angle adjustment of the physical similarity model of the present invention; Figure 4 is a schematic diagram of a single opening frame sliding out; Figure 5 is a schematic diagram of the air supply system of the present invention.

[0021] Description of the reference numerals: 1 - Test bench; 11 - Base; 111 - Groove track; 12 - Open frame; 121 - Card slot; 122 - Pulley; 123 - Detection hole; 124 - Screw hole; 13 - Heat - insulating glass; 14 - Hydraulic lifting system; 141 - Hydraulic pump; 142 - Lifting hydraulic cylinder; 151 - Oxygen cylinder; 152 - Nitrogen cylinder; 153 - Gas flowmeter; 154 - Air compressor; 15 - Gas supply system; 16 - Ignition system; 161 - Digital display electronic temperature controller; 162 - Heating rod; 163 - Temperature sensor; 2 - Physical similarity model; 21 - First rock layer; 22 - Coal seam; 23 - Second rock layer; 24 - Third rock layer; 25 - Fourth rock layer; 26 - Fifth rock layer; 3 - Fiber Bragg grating temperature and stress integrated monitoring system; 31 - Ultra - high temperature fiber Bragg grating sensor; 32 - Fiber Bragg grating network analyzer; 33 - Terminal box; 4 - Multi - gas monitoring system; 5 - Image and video monitoring system; 51 - Dual - spectrum high - definition camera; 52 - Bracket; 6 - Computer analysis system. Detailed implementation manners

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] As Figures 1-5 shown, the present invention provides a technical solution: a visual similar simulation test device for the evolution process of coalfield fires, including a test bench 1, a physical similarity model 2, a fiber Bragg grating temperature and stress integrated monitoring system 3, a multi - gas monitoring system 4, an image and video monitoring system 5, and a computer analysis system 6.

[0024] Among them, the test bench 1 includes a base 11, an open frame 12, a heat - insulating glass 13, and an inclination adjustment mechanism 14. A plurality of mutually parallel rail grooves 111 are opened on the base 11. Each rail groove 111 is slidably connected with an open frame 12. A pulley 122 is arranged at the bottom of the open frame 12. The pulley 122 is slidably connected in the rail groove 111, and the pulley 122 is a self - locking pulley.

[0025] The size of the base 11 is 4000mm×2000mm×150mm, the size of the rail groove 111 on the base 11 is 3600mm×100mm×100mm, and the size of the open frame 12 is 1800mm×120mm×1800mm.

[0026] The inner surface of the opening frame 12 is also attached with high-temperature resistant fiber cotton. A plurality of screw holes 124 are also provided on the outer side of the opening frame 12, and the distance between two adjacent screw holes 124 is 180 mm. Adjacent opening frames 12 can be fixed through the cooperation of bolts and the screw holes 124. A plurality of opening frames 12 are sequentially assembled and fixed to form a test box body. The opening frame 12 is a U-shaped frame, and clamping grooves 121 are also provided on both sides of the opening frame 12. A heat preservation glass 13 is clamped in the clamping grooves 121.

[0027] The physical similarity model 2 successively includes a first rock stratum 21, a coal seam 22, a second rock stratum 23, a third rock stratum 24, a fourth rock stratum 25, and a fifth rock stratum 26 from bottom to top. The physical similarity model 2 is arranged inside the test box body.

[0028] An inclination angle adjustment mechanism 14 is also connected to the test bench 1. The inclination angle adjustment mechanism 14 includes a lifting hydraulic cylinder 142 and a hydraulic pump 141. Three lifting hydraulic cylinders 142 are provided. One lifting hydraulic cylinder 142 is arranged at the side end of the base 11, and the other two lifting hydraulic cylinders 142 are arranged below the test box body on the base 11. The telescopic shafts of the lifting hydraulic cylinders 142 are respectively connected to the test box body, and the lifting hydraulic cylinders 142 are powered by the hydraulic pump 141. The inclination angle α of the test box body is adjusted through the inclination angle adjustment mechanism 14, and the adjustment range is 0 - 90 degrees, so as to realize the simulation evolution of coalfield fire areas with different inclination angles.

[0029] An air supply system 15 and an ignition system 16 which are connected to the test box body are also arranged on the test bench 1; The air supply system 15 includes an oxygen cylinder 151, a nitrogen cylinder 152, a gas flowmeter 153, and an air compressor 154. The oxygen cylinder 151 and the nitrogen cylinder 152 are respectively communicated with the air compressor 154. Gas flowmeters 153 are also respectively connected to the gas outlet sections of the oxygen cylinder 151 and the nitrogen cylinder 152. After the input gas is pressurized by the air compressor 154 to reach the specified pressure, it is input into the test box body.

[0030] The ignition system 16 includes a digital display electronic temperature controller 161, a heating rod 162, and a temperature sensor 163. The heating rod 162 and the temperature sensor 163 are respectively arranged inside the physical similarity model 2. The heating rod 162 and the temperature sensor 163 are connected to the digital display electronic temperature controller 161. The temperature of the heating rod 162 is set through the digital display electronic temperature controller 161 and monitored through the temperature sensor 163.

[0031] The fiber Bragg grating temperature and stress integrated monitoring system 3 is used to monitor the temperature and stress of the physical similarity model 2 inside the test box body in the similarity simulation test of the coalfield fire evolution process in real time; The multi-gas monitoring system 4 is used to monitor the gas concentration values in the test chamber after fractures are generated in the physical similarity model 2 during the similar simulation test of coalfield fire evolution; The image and video monitoring system 5 is used to monitor the changes of the physical similarity model 2 during the similar simulation test of coalfield fire evolution, and is used to extract two-dimensional slice images by the extraction opening frame 12 after the test; The fiber Bragg grating temperature and stress integrated monitoring system 3, the multi-gas monitoring system 4, and the image and video monitoring system 5 are respectively connected to the computer analysis system 6 by signals. The computer analysis system 6 has a dynamic visualization model integrating multiple physical fields. A three-dimensional fracture field evolution model is embedded inside the visualization model. Data analysis and visualization processing operations are performed through the computer analysis system 6.

[0032] The fiber Bragg grating temperature and stress integrated monitoring system 3 includes an ultra-high temperature fiber Bragg grating sensor 31, a fiber Bragg grating network analyzer 32, and a terminal box 33; a plurality of detection holes 123 are also provided on the opening frame 12, and the distance between two adjacent detection holes 123 is 180 mm. The ultra-high temperature fiber Bragg grating sensor 31 can monitor the temperature and stress in the test chamber. The ultra-high temperature fiber Bragg grating sensor 31 is connected to the fiber Bragg grating network analyzer 32 through the terminal box 33, and the fiber Bragg grating network analyzer 32 is connected to the computer analysis system 6.

[0033] The multi-gas monitoring system 4 consists of a multi-gas monitor and a signal line. The model of the multi-gas monitor is GREENG7000. The multi-gas monitor is connected to the air outlet at the top of the test chamber. When fractures are generated in the test chamber, the detection head is inserted into the fractures to monitor and display the concentration values of gases such as CO, C2H4, CH4, CO2, C2H6, C2H2, O2, H2, and H2S, and transmit the data to the computer analysis system 6 through the signal line. The image and video monitoring system 5 consists of a bracket 52 and three dual-spectrum high-definition cameras 51 arranged on the bracket 52. The three dual-spectrum high-definition cameras 51 are respectively arranged on the sides and directly above the two heat-insulating glasses of the test chamber. The test process in the test chamber is recorded by the dual-spectrum high-definition cameras 51 and the data is transmitted to the computer analysis system 6. The model of the dual-spectrum high-definition camera 51 can be the intrinsically safe thermal imaging dual-spectrum integrated camera HB-8501R-F9432-T9310. Combining visible light and infrared thermal imaging technologies, it can capture the temperature distribution on the coalfield surface in real time, generate a thermal map, superimpose it with the visible light video, intuitively display the high-temperature area and the fire spread trend, and record the surface fracture development and smoke diffusion rules.

[0034] The computer analysis system 6 is used to record data such as the temperature, stress, fractures, gas concentration values, and images in the spatio-temporal distribution obtained by the fiber Bragg grating temperature and stress integrated monitoring system 3, the multi-gas monitoring system 4, and the image and video monitoring system 5. The computer analysis system 6 has a built-in dynamic visualization model of multi-physical field coupling, and a three-dimensional fracture field evolution model is embedded inside the visualization model.

[0035] The dynamic visualization model of multi-physical field coupling uses ParaView for dynamic rendering of the time series data of multiple physical fields such as stress, temperature, fractures, and gas concentration collected. It ensures the consistency of the time steps and spatial grids of each physical field through data alignment and fusion, and uses the Temporal Interpolator or Resample filter to process non-uniform data; adopts a multi-field coupling visualization strategy, such as using Volume Rendering for the temperature field to highlight high-temperature regions, using Warp by Vector for the fracture field to display displacement deformation, using Stream Tracer for the gas concentration to generate dynamic streamlines, and realizing superimposed display by adjusting the transparency of each field through Opacity.

[0036] Subsequently, use the time axis control to drag the slider or write a Python script to accurately trace back the state at any moment, and record the evolution process from the actual measurement through the Animation View, and export it as an interactive HTML report or a high-resolution animation to realize the dynamic visualization of its multi-physical field coupling.

[0037] The three-dimensional fracture field evolution model uses Matlab to reconstruct the three-dimensional fracture model from the two-dimensional slice image dataset measured by the image and video monitoring system, and reversely deduces the dynamic development process of fractures through finite element or discrete element methods such as PFC and UDEC to generate a time-sequential three-dimensional fracture field evolution model.

[0038] The following is the construction process of the three-dimensional fracture field evolution model: ① Load image data: First, load the two-dimensional slice image dataset into Matlab, and the imread or dicomread function can be used to read the image data; ② Image preprocessing: Perform binary preprocessing on the loaded two-dimensional slice images, with black for coal and rock layers and white for fractures to improve the subsequent reconstruction effect; ③ Create a spatial voxel: According to the size and resolution of the image, create a three-dimensional voxel space in Matlab that matches it to store the reconstructed image data; ④ Realize the reconstruction process: Use the trained generative adversarial network SAGAN model based on the attention mechanism for three-dimensional reconstruction; 5. 3D visualization and analysis: Use 3D visualization tools in Matlab to display and analyze the reconstructed 3D images. Functions such as isosurace or slice can be used to display images from different perspectives and calculate the crack rate. Voxel data can also be used for other related analyses. ⑥ Dynamic crack evolution simulation and reverse modeling: Using the final 3D reconstructed model as the boundary condition, combined with the surface crack expansion law recorded in the video and internal monitoring data, temperature gradient, and stress changes, a correlation model between crack development and internal stress field and temperature field is established. Through finite element or discrete element methods such as PFC and UDEC, the dynamic crack development process is reversely deduced to generate a time-series 3D crack field evolution model, providing data for the dynamic visualization model of multi-physics field coupling.

[0039] The dynamic visualization model of multi-physics field coupling can automatically update the rendering screen according to the real-time updated data and record the entire process. After the experiment is completed, it can be traced back to any time in real time for visualization.

[0040] In this embodiment, the attention-based generative adversarial network SAGAN model inputs Gaussian noise into the generator G to generate a fake three-dimensional image, and inputs the two-dimensional slice image into the discriminator D. During the training process, the generator G and the discriminator D are trained alternately. The generator G attempts to generate increasingly realistic fake three-dimensional images to "cheat" the discriminator D, while the discriminator D attempts to become smarter to better identify fake data. When the current epoch does not exceed the predetermined maximum number of epochs, the SAGAN model is saved, a more realistic three-dimensional model is constructed, and the image is reconstructed. When the current epoch exceeds the predetermined maximum number of epochs, Gaussian noise is again input into the generator G to generate a fake three-dimensional image, and the training is repeated until the previous epoch does not exceed the predetermined maximum number of epochs.

[0041] The generator G consists of four 3D transposed convolutional layers, eight residual blocks, a channel attention module (CTAM), and a 3D convolutional layer. The discriminator D consists of four 2D convolutional layers with ReLUs and a 2D convolutional layer. The residual blocks can mitigate the possibility of degradation. The CTAM helps the model focus on important features of the 2D slice image.

[0042] The Gaussian noise shape is 8×32×4×4×4 (batch size: 8; number of channels: 32; depth, width, and height are all 4).

[0043] The input 2D image shape of the discriminator D is 8×2×64×64 (batch size: 8; number of phases: 2; width and height are both 64).

[0044] The loss function of the SAGAN model based on the attention mechanism is as follows: (1); Wherein, represents the loss function; represents the average of the discriminator outputs of all the fake two-dimensional slice images in the current batch; represents the average of the discriminator outputs of all the real two-dimensional slice images in the current batch; represents the average of the gradient penalty terms corresponding to all the interpolation samples k in the current batch; is the actual data distribution; is the generated data distribution; represents the fake two-dimensional slice images extracted from the generated fake three-dimensional images; represents the real two-dimensional slice images; is the distribution of the interpolation samples; involves the sampling points uniformly distributed on the straight line between the paired sampling points from and ; is the gradient with respect to k; is the weight, with the default setting of 10.

[0045] To test the performance of the learning model, the FID value is used to measure the similarity between the two image distributions, so as to evaluate the quality of the reconstructed images.

[0046] A method for conducting a similar simulation test on the coalfield fire evolution process using the above-mentioned similar simulation test device for the coalfield fire evolution process includes the following steps: S1. Calculate and design the size and conditions of the physical similarity model 2 of the actual experimental coalfield fire according to the similarity criterion, build the test bench 1 according to the size and conditions of the physical similarity model 2, sequentially assemble and fix multiple open frames 12, insert heat-insulating glasses 13 on the two outermost open frames 12 to form a test box body, arrange the physical similarity model 2 in the test box body, fix another heat-insulating glass 13 on the top of the test box body, and use the inclination adjustment mechanism 14 to adjust the test box body to reach the set inclination; ​​S2. Place the integrated fiber Bragg grating temperature and stress monitoring system 3 in the physical similarity model 2 for temperature and stress monitoring; S3. After the physical similarity model 2 is left standing to air-dry and solidify, signal-connect the integrated fiber Bragg grating temperature and stress monitoring system 3, the multi-gas monitoring system 4, and the image and video monitoring system 5 to the computer analysis system 6 respectively; S4. Supply gas to the physical similarity model 2 through the gas supply system 15, and turn on the ignition system 16 to ignite the physical similarity model 2 to conduct a similar simulation experiment on the evolution process of coalfield fires. At this time, start the integrated fiber Bragg grating temperature and stress monitoring system 3 and the image and video monitoring system 5 for data monitoring. When cracks are generated due to the collapse of the coalfield fire area in the test chamber, insert the detection head of the multi-gas monitor of the multi-gas monitoring system 4 into the cracks for gas concentration value monitoring; S5. After the physical similarity model 2 finishes burning, leave it standing to allow the test chamber to cool naturally to room temperature. First, remove the thermal insulation glass 13 on the top of the test chamber. Insert the thermal insulation glass 13 into the card slots on both sides of each opening frame 12 to slice the burned physical similarity model 2. Remove the assembled multiple opening frames 12, use the image and video monitoring system 5 to record the two-dimensional slice images after each opening frame 12 slides out, obtain the two-dimensional slice image dataset, and transmit the two-dimensional slice image dataset to the computer analysis system 6. The computer analysis system 6 uses Matlab to reconstruct the three-dimensional crack model of the two-dimensional slice image dataset, and uses ParaView to render and analyze the spatio-temporal distribution data measured by each system to construct a dynamic visualization model of multi-physical field coupling.

[0047] In summary, the present invention can realize the visualization of the evolution process of coalfield fires under different conditions. Through the integrated fiber Bragg grating temperature and stress monitoring system 3, the multi-gas monitoring system 4, the image and video monitoring system 5, and the computer analysis system 6, it is possible to realize the real-time monitoring of the variation laws of the temperature field, seepage field, and crack field during the evolution process of coalfield fires under different conditions.

[0048] At the same time, after the experiment of the test bench of the present invention, each opening frame 12 can be pulled out. The internal crack field is reconstructed by Matlab, and the finite element or discrete element method, such as PFC and UDEC, is used to reversely deduce the dynamic development process of the cracks to generate a time-sequential three-dimensional crack field evolution model, making the internal crack distribution and size during the combustion evolution process of the coalfield fire area visible. The three-dimensional crack model can be imported into Fluent or Comsol software to further provide data support for the numerical simulation related to the evolution of the coalfield fire area; record the surface crack development and smoke diffusion laws through the image and video monitoring system 5, so that the impact of the fire on the internal structure of the coal body can be intuitively observed, and the temperature distribution on the surface of the coalfield can be captured in real time to display the high-temperature area and the fire spread trend.

[0049] Render and analyze the spatio-temporal distribution data measured by each system through ParaView, construct a dynamic visualization model of multi-physical field coupling, and realize the visualization of the coalfield fire evolution process. By superimposing the display of the high-temperature area and the crack expansion, the "critical state" can be located, such as the temperature mutation point or the crack penetration area, providing a scientific basis for early warning. More intuitively reveal the coupling mechanism of thermal-seepage-crack in fire evolution, and deepen the understanding of the self-sustaining mechanism of coalfield fires. Provide a visual decision-making sand table for active prevention and control. The visualization model can intuitively display the high-temperature areas and crack distributions in the coalfield, helping researchers accurately locate the fire risk areas and providing a basis for formulating targeted prevention and control measures. Evaluate the possible disaster levels caused by coalfield fires, including the spread range of the fire and the degree of damage to the surrounding environment, which can provide important references for formulating fire extinguishing plans. For example, the location and grouting volume of fire extinguishing grouting can be determined according to the crack distribution, improving the fire extinguishing efficiency and guiding the targeted implementation of engineering measures such as grouting and blocking. And the large amount of accumulated data can provide training samples for machine learning prediction models, promoting the development of fire prevention and control towards the intelligent direction.

[0050] Moreover, the present invention can achieve the introduction of oxygen in the concentration range of 0-100% by adjusting the gas flowmeter of the oxygen supply system, so as to simulate the on-site gas environment.

[0051] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0052] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A visualization similarity simulation test device for the evolution process of coalfield fires, characterized in that: It includes a test bench (1), a physical similarity model (2), a fiber Bragg grating temperature and stress integrated monitoring system (3), a multi-gas monitoring system (4), an image and video monitoring system (5), and a computer analysis system (6); Among them, a test box is connected to the test bench (1). The test box is formed by sequentially assembling a plurality of open frames (12). The open frames (12) are slidably installed on the test bench (1). Clamping grooves (121) for clamping heat-insulating glass (13) are also provided on both sides of the open frames (12); The physical similarity model (2) is arranged inside the test box. An inclination angle adjustment mechanism (14) for adjusting the inclination angle of the test box is also connected to the test bench (1); A gas supply system (15) and an ignition system (16) connected to the test box are also arranged on the test bench (1). Gas is supplied into the test box through the gas supply system (15), and the physical similarity model (2) is ignited through the ignition system (16) to simulate a coalfield fire, which is used for a similarity simulation test of the evolution process of a coalfield fire; The fiber Bragg grating temperature and stress integrated monitoring system (3) is used to monitor the temperature and stress of the physical similarity model (2) inside the test box in the similarity simulation test of the evolution process of a coalfield fire in real time. The multi-gas monitoring system (4) is used to monitor the gas concentration value inside the test box after cracks are generated in the physical similarity model (2) in the similarity simulation test of the evolution process of a coalfield fire. The image and video monitoring system (5) is used to monitor the changes of the physical similarity model (2) in the similarity simulation test of the evolution process of a coalfield fire, and is used to extract two-dimensional slice images by pulling out the open frames (12) after the test. The fiber Bragg grating temperature and stress integrated monitoring system (3), the multi-gas monitoring system (4), and the image and video monitoring system (5) are respectively connected to the computer analysis system (6) through signals. The computer analysis system (6) has a dynamic visualization model with multi-physical field coupling built in, and a three-dimensional crack field evolution model is embedded inside the visualization model. Data analysis and visualization processing operations are carried out through the computer analysis system (6).

2. The visualization similar simulation test device for the evolution process of coalfield fires according to claim 1, characterized in that, A plurality of screw holes (124) are also arranged on the outer side of the open frame (12). Adjacent two open frames (12) can be fixed through the cooperation of bolts and the screw holes (124). A plurality of detection holes (123) are also arranged on the open frame (12).

3. A visualization similar simulation test device for the evolution process of coalfield fires according to claim 1, characterized in that, The open frame (12) is movably connected to the base (11). A set of pulleys (122) is arranged at the bottom of each open frame (12). A plurality of mutually parallel rail grooves (111) are also formed in the base (11). A set of pulleys (122) is respectively slidably connected in each rail groove (111). The pulleys (122) are self-locking pulleys. A heat-resistant fiber cotton is also attached to the inner surface of the open frame (12).

4. A visualization similarity simulation test device for the evolution process of coalfield fires according to claim 1, characterized in that, The inclination angle adjustment mechanism (14) includes a lifting hydraulic cylinder (142) and a hydraulic pump (141). There are three lifting hydraulic cylinders (142). One lifting hydraulic cylinder (142) is arranged at the side end of the base (11), and the other two lifting hydraulic cylinders (142) are arranged below the test box body on the base (11). The telescopic shafts of the lifting hydraulic cylinders (142) are respectively connected to the test box body, and the lifting hydraulic cylinders (142) are powered by the hydraulic pump (141).

5. A visualization similar simulation test device for the evolution process of coalfield fires according to claim 1, characterized in that, The gas supply system (15) includes an oxygen cylinder (151), a nitrogen cylinder (152), a gas flow meter (153) and an air compressor (154). The oxygen cylinder (151) and the nitrogen cylinder (152) are respectively communicated with the air compressor (154). Gas flow meters (153) are respectively connected to the gas outlet sections of the oxygen cylinder (151) and the nitrogen cylinder (152). After the input gas is pressurized by the air compressor (154) to reach the specified pressure, it is input into the test box body.

6. The similar simulation test device for visualizing the evolution process of coalfield fires according to claim 1, characterized in that, The ignition system (16) includes a digital display electronic temperature controller (161), a heating rod (162) and a temperature sensor (163). The heating rod (162) and the temperature sensor (163) are respectively arranged in the physical similarity model (2). The heating rod (162) and the temperature sensor (163) are connected to the digital display electronic temperature controller (161). The temperature of the heating rod (162) is set by the digital display electronic temperature controller (161) and monitored by the temperature sensor (163).

7. A visualization similar simulation test device for the evolution process of coalfield fires according to claim 1, characterized in that, The fiber Bragg grating temperature and stress integrated monitoring system (3) includes an ultra-high temperature fiber Bragg grating sensor (31), a fiber Bragg grating network analyzer (32) and a terminal box (33); The ultra-high temperature fiber Bragg grating sensor (31) can monitor the temperature and stress of the physical similarity model (2) in the test box body during the similar simulation test of the coalfield fire evolution process. The ultra-high temperature fiber Bragg grating sensor (31) is connected to the fiber Bragg grating network analyzer (32) through the terminal box (33), and the fiber Bragg grating network analyzer (32) is connected to the computer analysis system (6).

8. A visualization similarity simulation test device for the evolution process of coalfield fires according to claim 1, characterized in that, The multi-gas monitoring system (4) consists of a multi-gas monitor and a signal line. The multi-gas monitor is connected to the air outlet at the top of the test box body. When cracks are generated in the physical similarity model (2) in the test box body during the similar simulation test of the coalfield fire evolution process, the detection head of the multi-gas monitor is inserted into the cracks to monitor and display the concentration value of the gas, and the data is transmitted to the computer analysis system (6) through the signal line. The image and video monitoring system (5) consists of a bracket (52) and three dual-spectrum high-definition cameras (51) arranged on the bracket (52). The process of the similar simulation test of the coalfield fire evolution process in the test box body is recorded by the dual-spectrum high-definition cameras (51), and the data is transmitted to the computer analysis system (6).

9. A visualization similarity simulation test device for the evolution process of coalfield fires according to claim 1, characterized in that, The computer analysis system (6) records the data obtained by the fiber Bragg grating temperature and stress integrated monitoring system (3), the multi-gas monitoring system (4) and the image and video monitoring system (5).

10. A method for similar simulation test of visualizing the evolution process of coalfield fires, using the visual similar simulation test device for the evolution process of coalfield fires described in any one of claims 1-9, characterized in that, It includes the following steps: S1. Calculate and design the size and conditions of the physical similarity model (2) of the actual experimental coalfield fire according to the similarity criterion. Build a test bench (1) according to the size and conditions of the physical similarity model (2). Assemble and fix multiple opening frames (12) in sequence, and insert heat-insulating glass (13) on the outermost opening frame (12) to form a test box body. Arrange the physical similarity model (2) in the test box body, and use the inclination adjustment mechanism (14) to adjust the test box body to reach the set inclination angle; S2. Arrange the fiber Bragg grating temperature and stress integrated monitoring system (3) in the physical similarity model (2) to conduct temperature and stress monitoring; S3. After the physical similarity model (2) is left to stand and air-dried and solidified, signal-connect the fiber Bragg grating temperature and stress integrated monitoring system (3), the multi-gas monitoring system (4), and the image and video monitoring system (5) to the computer analysis system (6) respectively; S4. Supply gas to the physical similarity model (2) through the gas supply system (15), and turn on the ignition system (16) to ignite the physical similarity model (2) to conduct a similarity simulation test on the evolution process of coalfield fire. At this time, start the fiber Bragg grating temperature and stress integrated monitoring system (3) and the image and video monitoring system (5) to conduct data monitoring. When a crack occurs due to the collapse of the coalfield fire area in the test box body, insert the detection head of the multi-gas monitor of the multi-gas monitoring system (4) into the crack to conduct gas concentration value monitoring; S5. After the physical similarity model (2) finishes burning, leave it to stand and let the test box body cool naturally to room temperature. Insert heat-insulating glass (13) into the card slots on both sides of each opening frame (12), slice the burned physical similarity model (2), disassemble the assembled multiple opening frames (12), use the image and video monitoring system (5) to record the two-dimensional slice images after each opening frame (12) slides out, obtain the two-dimensional slice image data set, transmit the two-dimensional slice image data set to the computer analysis system (6), and the computer analysis system (6) uses Matlab to reconstruct the three-dimensional crack model for the two-dimensional slice image data set, and uses ParaView to render and analyze the spatio-temporal distribution data measured by each system to construct a dynamic visualization model of multi-physical field coupling.

Images