A coal mine fluidized mining multi-field coupling physical simulation test method and system

Abstract

This invention provides a method and system for multi-field coupled physical simulation experiments in fluidized bed mining of coal mines, relating to the field of mine safety science and technology. The method includes simultaneously monitoring stress, pore pressure, temperature, microseismic activity, and fracture imaging on a three-dimensional sample to construct raw data for a multi-field monitoring matrix; extracting stress increments and fracture connectivity changes through denoising and gridded registration to establish a stress-fracture-seepage coupling model; delineating the source zone, propagation path zone, and protection zone based on the coupling identification results, and executing programmable loading and medium injection tests; calculating the source intensity, propagation attenuation, and energy absorption displacement determination quantities using process response data, performing comparisons and dynamic corrections; and backfilling the comparison results into the model to complete parameter calibration and generate a field mapping calibration data package, achieving multi-field dynamic linkage and closed-loop control of the physical experiment.

Description

Technical Field

[0001] This invention relates to the field of mine safety science and technology, and in particular to a multi-field coupled physical simulation test method and system for fluidized coal mining. Background Technology

[0002] With the accelerated pace of deep and intelligent coal mining, traditional mechanical mining and passive anti-scour methods such as regional pressure relief are no longer adequate for the prevention and control of dynamic disasters in deep, high-stress, and high-permeability coupled environments. Fluidized bed mining technology, proposed in recent years, directly transforms solid coal and rock into gaseous, liquid, or gas-solid-liquid mixed products underground, achieving an integrated resource development approach encompassing in-situ mining, beneficiation, metallurgy, charging, electrification, and gasification. While altering the coal and rock structure and energy field distribution, this approach also introduces more complex stress redistribution, fracture propagation, and seepage coupling behaviors, making the incubation and release of dynamic disasters more nonlinear and sudden. Existing experimental studies primarily focus on single mechanical loading or seepage analysis, making it difficult to physically reproduce the co-evolution process of multiple fields such as stress, fractures, and seepage.

[0003] Current research on the mechanisms of dynamic disasters in coal mines is gradually shifting from macroscopic mechanical responses to multi-field coupling and spatiotemporal evolution, urgently requiring the verification of dynamic coupling and feedback mechanisms of stress fields, fracture fields, and seepage fields through experimental platforms. Some studies have attempted to achieve coupled experiments using triaxial loading devices combined with fluid injection systems, but these generally suffer from insufficient monitoring dimensions, ambiguous parameter coupling relationships, and data that cannot be correlated with numerical models. Furthermore, most existing simulation experiments only focus on the interaction between stress and seepage, failing to form a closed-loop model that can be used for integrated source-path-protection analysis, leading to significant discrepancies between experimental results and field responses.

[0004] Existing coupled experimental methods for coal and rock fluidization generally lack a unified multi-field monitoring system and programmable loading control mechanism, making it impossible to simultaneously acquire stress, fracture, and seepage data within the same time-space framework, resulting in insufficient calibration accuracy of the coupled models. Furthermore, the lack of a full-process parameter mapping and calibration mechanism from energy incubation in the source region to energy absorption response in the protection zone makes it difficult to provide reliable references for intelligent on-site rockfall prevention. Therefore, there is an urgent need for a multi-field coupled physical simulation experimental method for coal mine fluidized mining that can achieve multi-field collaborative loading, real-time monitoring, and data-driven calibration under three-dimensional sample conditions. This method would establish a physical verification system for the entire process of "stress-fracture-seepage" coupling, providing experimental support for the prevention and control of dynamic disasters in deep fluidized mining. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide a multi-field coupled physical simulation test method and system for fluidized coal mining. Through multi-field collaborative monitoring and coupled modeling, dynamic linkage control of stress field, fracture field and seepage field is realized, and a calibrable and mappable physical test closed-loop system is constructed, thereby significantly improving the accuracy and reliability of dynamic disaster simulation and prevention in fluidized coal mining.

[0006] To achieve the above objectives, the present invention provides the following solution: A multi-field coupled physical simulation test method for fluidized coal mining includes: S1. Unify the spatial coordinates and time reference on the three-dimensional sample, set up stress, pore pressure, temperature, microseismic and crack imaging channels, continuously collect data and synchronize the timing to obtain the original data of the multi-field monitoring matrix. S2. Denoise and grid registration are performed on the original data of the multi-field monitoring matrix, stress increment and crack connectivity change are extracted, and spatiotemporal alignment is completed to construct a stress-crack-seepage coupling model and a set of coupling parameters. S3. Identify the coupling based on the stress-crack-seepage coupling model to obtain the coupling identification result; S4. Based on the coupling identification results, the source area, propagation path area and protection area are delineated in the three-dimensional sample, control command data for partitioned loading and injection are generated, and programmable loading and medium injection are executed. Micro-vibration, stress, displacement and pore pressure response are recorded online to obtain process response data. S5. Calculate the three types of judgment quantities—source intensity, propagation attenuation, and energy absorption displacement—from the process response data, and compare them with the control index set to obtain the comparison results; S6. After adjusting the control command data according to the comparison result, repeat S4 to output the judgment result when the comparison result meets the standard. S7. Backfill the process response data and the judgment result into the stress-crack-seepage coupling model, complete the parameter calibration, and generate a test calibration data package for field mapping.

[0007] Preferably, step S1 includes: Stress loading surfaces, pore pressure interfaces, temperature control zones, and micro-vibration sensing zones are set on the outer wall and key interfaces of the three-dimensional specimen, and crack imaging windows are arranged in the transparent or transmissive areas of the three-dimensional specimen. In a multi-field monitoring system, a unified spatial coordinate reference and time reference are set, the output signals of each sensor channel are formatted in a unified manner, and a mapping relationship between monitoring points and coordinate grids is established. By synchronously acquiring the controller, the stress, pore pressure, temperature, microseismic and crack imaging channels are triggered in sequence to acquire stress values, pore pressure, temperature field changes, microseismic event waveforms and crack image frames at each monitoring point in real time. After synchronizing and aligning the real-time acquired data in time and space, the synchronized multi-type data are arranged in the order of the coordinate grid and integrated into the original data of the multi-field monitoring matrix.

[0008] Preferably, step S2 includes: In the original data of the multi-field monitoring matrix, data cleaning is performed on each type of signal based on the monitoring channel identifier and time reference to remove noise waveforms and outliers, forming a filtered and smoothed basic dataset. According to the spatial grid division rules of the three-dimensional specimen, the stress values, pore pressure, temperature, microseismic events and crack image results in the basic dataset are mapped to the corresponding grid cells, and registration is performed with the grid center point as a reference to obtain the registered multi-field monitoring matrix. The stress difference of each grid cell in adjacent time periods is calculated to form the stress increment distribution. At the same time, the fracture skeleton is extracted from the fracture image sequence and the rate of change of the connectivity path is calculated to form the fracture connectivity change distribution. Within a unified time window, the stress increment distribution is matched with the fracture connectivity change distribution, the correlation coefficient between the stress increment distribution and the fracture connectivity change distribution is calculated, and the coupling parameters characterizing their mutual influence are determined. A stress-fracture-seepage coupling model and a set of coupling parameters are then established.

[0009] Preferably, the process of determining the spatial grid division rules includes: Based on the geometry of the three-dimensional sample and the distribution density of monitoring points, the volume of the three-dimensional sample is divided into several spatial units; each spatial unit has a unique coordinate index; the size of the spatial unit is determined according to the monitoring resolution and the sample size, so that the side length of the spatial unit is no more than twice the minimum monitoring interval, in order to ensure the spatial uniformity of multi-field data within the unit. Non-uniform layered grids are set according to the geostress gradient and pore pressure gradient in the depth direction, so that the grid density in the high gradient region is higher than the average density, thereby improving the accuracy of local stress-fracture response capture. Increasing the grid sampling density along the main direction of fracture development in the horizontal direction ensures that the grid division is consistent with the main axis of fracture connectivity, thereby maintaining the spatial continuity of stress-fracture-seepage coupling characteristics.

[0010] Preferably, step S4 includes: Within the three-dimensional specimen, the energy conversion rate of the coupling control region and the distribution of fracture connectivity of the coupling identification results are spatially superimposed. The source region is defined based on the region with energy conversion rate higher than a set threshold, the propagation path region is defined based on the region with large fracture connectivity and continuous energy transmission, and the protection region is defined based on the region with significant energy attenuation and structural integrity. Within each zone, loading and injection boundary conditions are set. A loading axis system is established based on the principal stress direction of the source zone and the dominant seepage direction of the propagation path zone. Control parameters for load rate, medium injection pressure and flow rate are determined to form control command data for zone loading and injection. The control command data is input into the programmable loading system and the injection control system, and the partitioned loading and media injection process is executed through a unified time base to achieve multi-field coordinated response under simulated conditions; During loading and injection, microseismic sensors, stress gauges, displacement gauges, and pore pressure gauges deployed continuously collect response signals. Based on a time synchronization mechanism, they record the microseismic energy, stress changes, displacement response, and pore pressure fluctuations throughout the entire process, forming process response data.

[0011] Preferably, the process of determining the loading and injection boundary conditions includes: Based on the spatial distribution of the source region, propagation path region, and protection region, the initial values ​​of the stress boundary, pore pressure boundary, and constraint boundary of each region are calculated. A linear loading procedure is applied in the principal stress direction according to the stress boundary, and the injection rate and flow state of the pore pressure boundary are adjusted synchronously so that the loading stress and seepage pressure change synergistically under a unified time reference. Displacement constraints are established around the protected area, and the boundary constraint parameters are dynamically corrected by real-time monitoring of the sample boundary displacement and reaction force signals to ensure that the stress and seepage transmission during the loading and injection process are in a closed state.

[0012] Preferably, step S5 includes: The process response data is time-synchronized and spatially registered, and the microseismic energy, stress change rate, displacement response and pore pressure change are mapped to three-dimensional mesh cells according to the sampling time to obtain a spatiotemporally corresponding dataset. The source intensity determination quantity is calculated based on the dataset corresponding to the spatiotemporal time; the source intensity determination quantity is determined by the coupling relationship between the microseismic energy release rate and the stress reduction amplitude, and is used to reflect the degree of energy accumulation and release in the source area; The propagation attenuation determination quantity is calculated based on the change in stress wave amplitude at measuring points before and after the energy-reducing barrier; the propagation attenuation determination quantity is used to characterize the energy dissipation capability of the stress wave in the propagation path region. The energy-absorbing displacement judgment quantity is calculated by using the difference between the peak displacement and the recovered displacement of the displacement monitoring points in the protected area; the energy-absorbing displacement judgment quantity is used to characterize the energy absorption and deformation capacity of the support system against external dynamic loads; The source intensity determination quantity, propagation attenuation determination quantity, and energy absorption displacement determination quantity are combined into three sets of determination quantities, and then compared with the corresponding source intensity upper limit, path attenuation target, and energy absorption displacement target in the control index set to obtain the comparison results.

[0013] Preferably, step S6 includes: Based on the type of judgment quantity that does not meet the standard in the comparison results, determine the region to which the control command data to be corrected belongs, and extract the parameter set of the corresponding source region, propagation path region or protection region; When the non-compliant item is the source strength determination quantity, adjust the loading rate, injection pressure and medium phase state of the source area to keep the energy release rate and stress reduction within the control index constraints. When the non-compliance item is the propagation attenuation judgment quantity, the arrangement parameters of the energy-cutting barrier and the seepage channel in the propagation path area are modified, including their thickness, location and wave impedance difference, in order to enhance the energy dissipation effect. When the non-compliant item is the energy absorption displacement judgment quantity, adjust the number of components, spacing and pre-tightening state of the energy absorption support system in the protected area to improve the deformation energy absorption capacity of the support system; After the adjustment is completed, the corrected control command data is regenerated, and the partition loading and medium injection process defined in step S4 is executed. New micro-vibrations, stresses, displacements and pore pressure responses are recorded in real time, and the process response data is updated until all judgment quantities in the comparison results reach the set threshold and the judgment result is output.

[0014] Preferably, step S7 includes: The process response data and the judgment result are aligned by time series and spatial coordinates and then imported into the stress-fracture-seepage coupling model. The stress concentration parameters, fracture guiding parameters and seepage connectivity parameters in the stress-fracture-seepage coupling model are matched. The calibration coefficient is calculated based on the deviation between the predicted values ​​and the measured response data of the stress-fracture-seepage coupling model, and the model parameter set is updated using the calibration coefficient to ensure that the stress distribution, fracture connectivity and pore pressure change output by the stress-fracture-seepage coupling model are consistent with the measured results. After calibration, the calibrated coupled model is subjected to steady-state verification. The stability and response accuracy of the model output are verified through multiple rounds of virtual loading and medium injection simulation, and the final parameter set is determined. The calibrated and verified model parameters, process response characteristics, and judgment results are encapsulated together into a field-oriented test calibration data package; the test calibration data package is used to guide the initial setting and parameter tuning of the field stress-crack-seepage dynamic monitoring and linkage control system.

[0015] A multi-field coupled physical simulation test system for fluidized coal mining includes: The multi-field monitoring unit is used to unify the spatial coordinates and time reference on the three-dimensional sample, set up stress, pore pressure, temperature, microseismic and crack imaging channels, continuously collect data and synchronize the timing to obtain the original data of the multi-field monitoring matrix; The data registration and coupling modeling unit is used to denoise and mesh the original data of the multi-field monitoring matrix, extract stress increment and crack connectivity change, and complete spatiotemporal alignment to construct a stress-crack-seepage coupling model and a set of coupling parameters. The coupling identification unit is used to identify the coupling based on the stress-crack-seepage coupling model and obtain the coupling identification result. The partitioned loading and injection control unit is used to delineate the source zone, propagation path zone and protection zone in the three-dimensional specimen according to the coupling identification result, generate control command data for partitioned loading and injection, execute programmable loading and medium injection, record micro-vibration, stress, displacement and pore pressure response online, and obtain process response data. The judgment quantity calculation unit is used to calculate three types of judgment quantities—source intensity, propagation attenuation, and energy absorption displacement—from the process response data, and compare them with the control index set to obtain the comparison results. The control command correction unit is used to adjust the control command data according to the comparison result and repeat the implementation steps of the partition loading and injection control unit, so as to output the judgment result when the comparison result meets the standard. The model calibration and output unit is used to backfill the process response data and the judgment result into the stress-crack-seepage coupling model, complete the parameter calibration, and generate a test calibration data package for field mapping.

[0016] The present invention discloses the following technical effects: (1) This invention constructs stress, pore pressure, temperature, microseismic and fracture imaging channels on a three-dimensional sample to form a multi-field synchronous acquisition system, realizing real-time coordinated monitoring of stress field, fracture field and seepage field. This design can obtain continuous dynamic data under a unified time and space reference, and truly reproduce the physical evolution process of multi-field interaction in fluidized mining, providing a reliable basis for subsequent coupled modeling and quantitative analysis.

[0017] (2) This invention constructs a stress-fracture-seepage coupling model based on multi-field monitoring data, and achieves differentiated control of the source area, propagation path area and protection area through programmable loading and directional injection, enabling the test process to have dynamic feedback adjustment capability. Through the stress concentration parameters and fracture guidance parameters output by the coupling model, fine control of the modification direction, energy distribution and seepage channel formation can be achieved, thereby improving the physical controllability and repeatability of the simulation test.

[0018] (3) This invention constructs a data closed-loop judgment and parameter correction mechanism by collecting process response data in real time and calculating three types of judgment quantities: source intensity, propagation attenuation, and energy absorption displacement, and comparing them with the control index set. This mechanism can automatically adjust the loading and injection parameters according to the test feedback, ensuring that the coupled model gradually approaches the real mining conditions, realizing the self-calibration and dynamic optimization of model parameters, and improving the mapping accuracy between test results and field conditions.

[0019] (4) After the experiment is completed, the process response data and judgment results are backfilled into the coupled model to generate a test calibration data package for field mapping, which can be directly used to guide the anti-scouring design and parameter selection of deep coal mine fluidized mining. Through this calibration system, the experimental results can quantitatively guide the field stress control and fracture guidance layout, realize the technology transformation and risk pre-control from the laboratory to the actual mine, and significantly improve the safety and intelligence level of deep fluidized mining. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart of the method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the system structure provided in an embodiment of the present invention. Detailed Implementation

[0022] 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.

[0023] The purpose of this invention is to provide a multi-field coupled physical simulation test method and system for fluidized coal mining. By constructing a multi-field coupled physical simulation system, the synergistic evolution and dynamic feedback control of the three fields of stress, fracture and seepage are realized. A quantifiable and calibrable test verification mechanism is established, which effectively improves the scientificity and engineering applicability of dynamic disaster simulation and prevention in fluidized coal mining.

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] Figure 1 The method flowchart provided in the embodiments of the present invention is as follows: Figure 1 As shown, this invention provides a multi-field coupled physical simulation test method for fluidized coal mining, comprising: S1. Unify the spatial coordinates and time reference on the three-dimensional sample, set up stress, pore pressure, temperature, microseismic and crack imaging channels, continuously collect data and synchronize the timing to obtain the original data of the multi-field monitoring matrix. S2. Denoise and grid registration are performed on the original data of the multi-field monitoring matrix, stress increment and crack connectivity change are extracted, and spatiotemporal alignment is completed to construct a stress-crack-seepage coupling model and coupling parameter set. S3. Identify the coupling based on the stress-crack-seepage coupling model and obtain the coupling identification results; S4. Based on the coupling identification results, the source area, propagation path area and protection area are delineated in the three-dimensional sample. Control command data for partitioned loading and injection are generated, and programmable loading and medium injection are executed. Microseismic, stress, displacement and pore pressure response are recorded online to obtain process response data. S5. Calculate the three types of judgment quantities—source intensity, propagation attenuation, and energy absorption displacement—from the process response data, and compare them with the control index set to obtain the comparison results; S6. After adjusting the control command data according to the comparison results, repeat S4 to output the judgment result when the comparison results meet the standards. S7. Backfill the process response data and judgment results into the stress-crack-seepage coupling model, complete the parameter calibration, and generate a test calibration data package for field mapping.

[0026] Specifically, in step S1 of this embodiment, a stress loading surface, a pore pressure interface, a temperature control zone, and a microseismic sensing zone are sequentially arranged on the outer wall and key interfaces of the three-dimensional specimen. A crack imaging window is also arranged in the transparent or transmissive area of ​​the three-dimensional specimen. The crack imaging window is used to acquire visual images of crack evolution, and its position covers the expected dominant crack direction. The window thickness and area satisfy both imaging clarity and structural strength (e.g., thickness not less than 10 mm, effective imaging area of ​​a single window not less than 100 cm²). The stress loading surface and the pore pressure interface are arranged in pairs around the window to reduce boundary effects; the microseismic sensing zone is arranged in an equidistant ring around the shell to improve the accuracy of event localization.

[0027] This embodiment establishes a unified spatial coordinate reference and a unified time reference in the multi-field monitoring system: the spatial coordinate reference uses the geometric center of the sample as the origin and the loading principal axis as the coordinate axis; the time reference is uniformly timed by the master clock and distributed to each sensing channel. Before acquisition, the output signals of each channel (stress, pore pressure, temperature, micro-vibration waveform, crack image frame) are uniformly formatted (uniform dimensions, uniform timestamp accuracy, uniform channel number), and a mapping relationship between monitoring points and a coordinate grid is established; the coordinate grid is a three-dimensional regular grid, and the side length of the grid cell is determined according to the sensor deployment density and target resolution to ensure approximately uniform measurement within the cell (e.g., side length 20–50 mm). Through this mapping, data from any monitoring point can be uniquely located to the corresponding grid cell.

[0028] This embodiment employs a synchronous acquisition controller to sequentially trigger stress, pore pressure, temperature, microseismic, and fracture imaging channels according to a unified time reference, enabling real-time acquisition of stress values, pore pressure, temperature field changes, microseismic event waveforms, and fracture image frames at each monitoring point. The synchronous acquisition controller refers to a device that integrates multi-channel triggering, timing, and buffering into a single control unit to ensure cross-channel timing consistency (the synchronization error is preferably no greater than 0.01s). After acquisition, the data is synchronized in time and aligned spatially: the timestamps are aligned with the master clock, the spatial indexes are aligned with the coordinate grid, and the data is sorted and merged according to grid cells and time series, integrating it into the original data of the multi-field monitoring matrix. The original data of the multi-field monitoring matrix refers to the original data set formed by aggregating stress, pore pressure, temperature, microseismic characteristics, and fracture image indices according to grid cells under the same time-space reference, which serves as the sole input for denoising and grid registration in step S2.

[0029] Specifically, in step S2 of this embodiment, the original data of the multi-field monitoring matrix is ​​cleaned according to the monitoring channel identifier and time reference: drift correction and abrupt change point removal are performed on the stress and pore pressure signals, bounded bandwidth denoising and saturation segment removal are performed on the microseismic waveform, steady-state segment identification and mismatch segment removal are performed on the temperature sequence, and dark field and artifacts are removed from the crack image; then the units and timestamp accuracy are unified, and a quality identifier is generated for each record to form a filtered and smoothed basic dataset for subsequent gridded registration.

[0030] This embodiment maps stress values, pore pressure, temperature, microseismic events, and fracture image results from the basic dataset to corresponding grid cells based on the spatial grid division rules. Spatial registration and temporal alignment are completed using the grid center point as a reference, resulting in a registered multi-field monitoring matrix. Specifically, the fracture image sequence is segmented and a "fracture skeleton" is extracted. The fracture skeleton refers to the centerline network formed by connecting the main fracture channels in the image, used to calculate the number of connected paths and intersections within the grid. Microseismic events are mapped to the grid using their source location coordinates and aligned with stress and pore pressure records, achieving synchronous representation of multiple data types within the same grid and time window.

[0031] In this embodiment, within a unified time window, the stress difference between adjacent time periods of each grid cell is calculated to generate a stress increment distribution. Simultaneously, within the same time window, the fracture skeleton of the fracture image sequence is temporally compared to calculate the rate of change of connectivity paths, generating a fracture connectivity variation distribution. Subsequently, the stress increment distribution and the fracture connectivity variation distribution are matched cell-by-cell to obtain the correlation intensity reflecting their mutual influence. Combined with the modulation effect of pore pressure on their relationship, coupling parameters characterizing the mutual influence are determined (including the driving sensitivity of stress on fractures, the feedback sensitivity of fractures to corresponding forces, and the seepage modulation coefficient). Based on this, a stress-fracture-seepage coupling model and a set of coupling parameters are established. The above coupling model and set of coupling parameters serve as inputs for the identification and analysis in step S3.

[0032] Further, in step S2 of this embodiment, the sample volume is first globally discretized based on the geometry of the three-dimensional sample and the distribution density of monitoring points: the shape and arrangement of spatial units are determined, a unique coordinate index is assigned to each spatial unit, and the origin and principal axis direction are established for subsequent data mapping. The size of the spatial unit is determined jointly according to the monitoring resolution and the sample size, ensuring that the side length of the spatial unit is no more than twice the minimum monitoring interval, so as to ensure that various monitoring values ​​have spatial uniformity within the unit; the consistency of stress, pore pressure, temperature, and microseismic positioning accuracy is checked by using the variance within the unit and the difference between adjacent units. If the uniformity within the unit does not meet the threshold, the unit size and boundary position are finely adjusted without changing the coordinate index system until the uniformity requirement is met. After completing the above steps, a spatial grid skeleton and a monitoring point-spatial unit mapping table are formed for subsequent data positioning, resulting in a data spatial framework for registration.

[0033] This embodiment employs a non-uniform layered mesh (a layering method that adaptively changes the element thickness along the thickness direction based on the gradient magnitude) based on the geostress and pore pressure gradients in the depth direction. Mesh density is increased in high-gradient sections to enhance the accuracy of capturing local stress-fracture responses. In the horizontal direction, anisotropic meshing is performed according to the main fracture development direction, ensuring that the mesh division aligns with the main fracture connectivity axis, thus maintaining the spatial continuity of stress-fracture-seepage coupling characteristics. During the meshing and layering process, the aspect ratio of the spatial elements is controlled within a preset range to avoid element distortion affecting subsequent registration and calculation. Transition zones are set at the boundaries of different density zones to ensure topological continuity of the spatial elements. After completing depth layering and horizontal meshing, the spatial meshing rules and the final spatial mesh are output, serving as the spatial mapping basis for mesh registration in step S2 and subsequent coupling identification.

[0034] Optionally, step S3 in this embodiment includes: Using the stress-fracture-seepage coupling model established in step S2 as input, three types of identification quantities are calculated for each grid cell within a unified time window and spatial grid: the first is the coupling strength index, used to quantify the temporal synchronicity and spatial consistency of stress increment and fracture connectivity changes; the second is the energy conversion rate, used to characterize the degree of conversion of elastic energy release into microseismic energy and surface energy of newly formed fractures within the cell; and the third is the seepage modulation index, used to measure the modulation intensity of pore pressure changes on the aforementioned relationship. Simultaneously, this embodiment extracts the energy transfer direction vector from the spatial gradient field of coupling strength and energy conversion rate within the grid domain, and extracts the seepage guidance vector based on the fracture skeleton and pore pressure gradient field. The above four types of quantities are output as multi-layer grids and corresponding direction fields within the same grid system, serving as the basic data for zonal identification.

[0035] A combined threshold and connectivity criterion is used to perform region growing and clustering on the aforementioned grid and orientation field: connected clusters with coupling strength and energy conversion rates both above the threshold and with good orientation field consistency are marked as source candidate zones; connected clusters with continuous energy transfer direction vectors, continuous seepage guidance vectors, and high-level seepage modulation indices are marked as propagation path candidate zones; and connected clusters with low energy conversion rates, low-to-medium coupling strength, and continuous and complete structures are marked as protection candidate zones. Morphological smoothing and minimum gap checks are performed at the boundaries of the candidate zones, and a comprehensive dataset including grid partition masks, dominant directions for each zone, statistical thresholds within each zone, and a zone-level parameter table is output as the coupling identification result. This dataset is then directly used in subsequent step S4 to delineate the source zone, propagation path zone, and protection zone in the three-dimensional specimen, as well as to generate the control command data for partition loading and injection.

[0036] Further, in step S4 of this embodiment, the coupling identification result is read, and the energy conversion rate distribution and fracture connectivity distribution of the coupling master control region are spatially superimposed in the three-dimensional sample coordinate system. A source region mask is generated based on connected clusters with energy conversion rates higher than a set threshold; a propagation path region mask is generated based on connected clusters with high fracture connectivity and continuous energy transfer; and a protection region mask is generated based on connected clusters with significant energy attenuation and complete structure. Boundary smoothing and minimum area verification are performed on the three types of masks, and an index table of "mask-mesh unit-monitoring point" is established as the spatial constraint input for partition loading and injection.

[0037] In this embodiment, loading and injection boundary conditions are determined in the source region, propagation path region, and protection region, respectively, and loading and injection axis systems are established. The loading axis system is based on the principal stress direction in the source region, used to constrain the application direction and rate of triaxial loading; the injection axis system is based on the dominant seepage direction in the propagation path region, used to constrain the timing of medium injection pressure and flow rate. Based on the mask index table and axis system directions, the load rate, target stress level, medium injection pressure, and flow rate of each zone are determined and encoded into control command data for zone loading and injection; the control command data is a structured dataset of time step-spatial zone-parameter group, containing execution order, triggering conditions, and safety limits, used to drive subsequent equipment execution.

[0038] In this embodiment, the control command data is sent to the programmable loading system and the injection control system respectively. Execution is triggered by a unified time base, ensuring that partitioned loading and media injection are carried out collaboratively within the same timing framework. The loading system performs linear or piecewise linear loading and holding according to the loading axis, while the injection control system performs steady-state or pulsed injection according to the injection axis and maintains back pressure conditions. When any execution parameter approaches the safety limit, the amplitude is automatically reduced or paused according to the protection logic in the control command data. During execution, the command execution status and equipment feedback are recorded, forming an execution log associated with the spatial mask for subsequent result interpretation.

[0039] In this embodiment, microseismic sensors, stress gauges, displacement gauges, and pore pressure gauges deployed within the partition boundaries and key grid cells continuously acquire response signals under a unified time reference, and the partitions and grid cells to which the data belong are marked according to the mask index table. After acquisition, the microseismic energy, stress changes, displacement response, and pore pressure fluctuations are synchronized in time and aligned spatially, and integrated together with the execution log into process response data. The process response data includes timestamps, partition labels, grid indexes, and synchronization sequences of four types of physical quantities, serving as the sole input for step S5 to calculate the three judgment quantities: source intensity, propagation attenuation, and energy absorption displacement.

[0040] Optionally, in step S5 of this embodiment, the process response data is first aligned according to a unified time reference, and coordinate mapping is completed according to the spatial grid division rules. The microseismic energy, stress change rate, displacement response, and pore pressure change are labeled to the corresponding grid cell and sampling time, forming a spatiotemporally corresponding dataset. The spatiotemporally corresponding dataset is used to achieve the comparability and aggregation of multi-source data within the same time window and the same spatial cell, providing a unique input for the calculation of the decision quantity.

[0041] This embodiment performs sliding time window statistics on the spatiotemporally corresponding dataset within the source region mask, extracting two quantities within the window: microseismic energy release rate and synchronous stress reduction amplitude. The source intensity determination quantity is obtained by coupling these two quantities within the same window. When the microseismic energy release rate increases and is accompanied by a significant stress reduction amplitude, a high value is taken for the determination quantity; otherwise, a low value is taken. To avoid the influence of sporadic pulses, a consistency constraint between adjacent windows and an extreme value suppression strategy are adopted, retaining only high-value regions that stably appear within continuous windows, and outputting the distribution and representative values ​​of the source intensity determination quantity at the regional level.

[0042] In this embodiment, paired data from measurement points before and after the energy-reducing barrier are selected within the propagation path region. Frequency band selection and distance normalization are performed based on the dominant frequency band and path length. The amplitude reduction ratio between the incident and transmitted sides is calculated to obtain the propagation attenuation determination quantity. Within the protection zone, the peak displacement under loading and the recovered displacement after unloading are extracted from displacement monitoring points. The difference between these two values ​​is used as the single-point energy absorption displacement, and a weighted summary is performed by region to obtain the energy absorption displacement determination quantity. Both types of determination quantities, along with their spatial labels, are written into the determination quantity set for easy comparison with the control index set item by item.

[0043] This embodiment uses three sets of judgment quantities—source intensity judgment quantity, propagation attenuation judgment quantity, and energy absorption displacement judgment quantity—to form a set of judgment quantities. These are then compared one by one with the corresponding upper limit of source intensity, path attenuation target, and energy absorption displacement target in the control index set. If the source intensity judgment quantity does not exceed the upper limit of source intensity, the propagation attenuation judgment quantity is not lower than the path attenuation target, and the energy absorption displacement judgment quantity is within the energy absorption displacement target range, the standard is deemed met; otherwise, it is deemed unmet. The final output comparison result includes the district-level representative values ​​of the three judgment quantities, the deviation from the control index, and the compliance indicator. This comparison result serves as the sole basis for adjusting the control command data in step S6 and repeating step S4.

[0044] Further, in step S6 of this embodiment, the comparison results output in step S5 are first read, and the non-compliant judgment quantities are located to the source region, propagation path region, or protection region, respectively. The corresponding parameter sets are then extracted from the control command data. In the source region, the parameter sets include loading rate, injection pressure, and medium phase; in the propagation path region, they include the thickness, location, and wave impedance difference of the energy-absorbing barrier, as well as the connectivity and direction of the seepage channel; and in the protection region, they include the number, spacing, and pre-tightening status of energy-absorbing support components. Through the mapping relationship of "non-compliant type - region - parameter set," a list of items to be corrected is formed, serving as input for subsequent adjustments.

[0045] When the non-compliant item is the source strength determination quantity, this embodiment adjusts the control command data in the source region according to a gradient strategy: firstly, the loading rate is slightly reduced and the injection pressure is reduced in conjunction; secondly, the medium phase state is adjusted to reduce the driving force for fracture propagation; and if necessary, the injection orientation and on / off ratio are finely adjusted to suppress instantaneous energy concentration. After the adjustment is completed, the corrected control command data is generated, and over-limit protection thresholds and backoff values ​​are set to ensure that the energy release rate and stress reduction are controlled and meet the constraints of the control index set.

[0046] When the non-compliant item is the propagation attenuation determination quantity, this embodiment corrects the parameters of the energy-absorbing barrier and the seepage channel in the propagation path area: first, it increases the barrier thickness or extends the continuous length, then it optimizes the barrier position to increase the wave impedance difference relative to the surrounding rock, and at the same time, it adjusts the connectivity and direction of the seepage channel to avoid energy short circuits without reducing seepage conductivity. When the non-compliant item is the energy-absorbing displacement determination quantity, this embodiment increases the number of energy-absorbing support components or reduces the component spacing in the protection zone, and appropriately increases the pre-tightening state or lowers the trigger threshold according to the component type to improve the deformation energy absorption capacity and ensure that the displacement response reaches the target range of energy-absorbing displacement.

[0047] In this embodiment, the corrected control command data is sent to the programmable loading system and the injection control system, and the partitioned loading and medium injection process in step S4 is repeated. Under a unified time reference, the new micro-vibrations, stresses, displacements, and pore pressure responses are recorded in real time, the process response data is updated, and the three types of judgment quantities are calculated according to step S5 and compared with the control index set again. If any judgment quantity still fails to meet the standard, the iterative correction is continued according to this step. When the source intensity judgment quantity, propagation attenuation judgment quantity, and energy absorption displacement judgment quantity all meet the corresponding thresholds, the judgment result is output, and the control command data and process response data before and after this round of correction are archived as the basis for subsequent model calibration and field mapping.

[0048] Furthermore, in step S7 of this embodiment, the process response data and the judgment result are aligned with the spatial grid line by line according to a unified time reference and then imported into the stress-fracture-seepage coupling model. An "observation-prediction" pairing table is established at the grid cell level of the model: the observation side includes time-series curves and representative values ​​at the regional level for three types of quantities: stress distribution, fracture connectivity, and pore pressure change; the prediction side calls the current parameter set of the model to output results of the same dimension. To facilitate matching, this embodiment marks stress concentration parameters (used to characterize the principal stress amplitude and gradient peak position), fracture guidance parameters (used to characterize the direction and intensity of the fracture connectivity axis), and seepage connectivity parameters (used to characterize equivalent connectivity and pressure transmission efficiency) in the model, and binds the three types of parameters to their corresponding observations one by one, completing the parameter-observation mapping table.

[0049] This embodiment calculates the deviation between the model's predicted values ​​and the measured responses at both the grid and zone levels based on a pairing table, generating a deviation distribution map and a zone-level deviation summary table. Calibration coefficients are then calculated, and updates are implemented step-by-step according to parameter groups: stress concentration parameters are prioritized for correcting the amplitude and gradient of stress distribution; fracture guidance parameters are used to correct the direction and intensity of the main axis of connectivity; and seepage connectivity parameters are used to correct the pore pressure propagation delay and amplitude. The update employs a limiting and rollback mechanism, with the amount of parameter correction in a single cycle limited to a preset proportional range (e.g., no more than 0.05–0.15 of the original value per round) to avoid oscillations. When all three types of deviations are below a preset threshold in two consecutive calibration rounds (e.g., the zone-level relative deviation is no greater than 0.10), calibration is considered converged, and the current model parameter set is frozen.

[0050] This embodiment performs steady-state verification of the coupled model after calibration: Multiple rounds of virtual loading and medium injection simulation scenarios are set, including three typical operating conditions: constant-rate loading and steady-state injection, segmented loading and pulse injection, and load holding and backpressure holding. The stability (no non-physical drift), consistency (consistent results when running under the same conditions repeatedly), and accuracy (error compared to the process response data remains within the threshold) of the model output within a long time window are checked. If any operating condition fails to meet the criteria, the system returns to the previous parameter set for fine-tuning and verification. After all operating conditions are met, the final parameter set is determined, and version information and the effective operating condition range are recorded.

[0051] This embodiment encapsulates the calibrated and verified model parameters, the feature quantities extracted from the process response data, the judgment results and their compliance thresholds, the partition mask and mesh configuration, and the effective and recommended intervals of the control command data into a field-oriented test calibration data package. The test calibration data package includes a data structure description, time reference, spatial mesh specifications, parameter value range, initial setting suggestions for partition loading and injection, and a correspondence table of monitoring quantity, judgment quantity, and adjustment quantity. It can be directly used to guide the initial setting and parameter tuning of the field stress-crack-seepage dynamic monitoring and linkage control system, and serve as the basis for alignment and verification when the field data is reinjected into the model.

[0052] Corresponding to the above methods, such as Figure 2 As shown, this embodiment also provides a multi-field coupled physical simulation test system for fluidized coal mining, including: The multi-field monitoring unit is used to unify the spatial coordinates and time reference on the three-dimensional sample, set up stress, pore pressure, temperature, microseismic and crack imaging channels, continuously collect data and synchronize the timing to obtain the original data of the multi-field monitoring matrix; The data registration and coupling modeling unit is used to denoise and mesh the original data of the multi-field monitoring matrix, extract stress increment and crack connectivity change, and complete spatiotemporal alignment to construct a stress-crack-seepage coupling model and a set of coupling parameters. The coupling identification unit is used to identify the coupling based on the stress-crack-seepage coupling model and obtain the coupling identification result. The partitioned loading and injection control unit is used to delineate the source zone, propagation path zone and protection zone in the three-dimensional specimen according to the coupling identification result, generate control command data for partitioned loading and injection, execute programmable loading and medium injection, record micro-vibration, stress, displacement and pore pressure response online, and obtain process response data. The judgment quantity calculation unit is used to calculate three types of judgment quantities—source intensity, propagation attenuation, and energy absorption displacement—from the process response data, and compare them with the control index set to obtain the comparison results. The control command correction unit is used to adjust the control command data according to the comparison result and repeat the implementation steps of the partition loading and injection control unit, so as to output the judgment result when the comparison result meets the standard. The model calibration and output unit is used to backfill the process response data and the judgment result into the stress-crack-seepage coupling model, complete the parameter calibration, and generate a test calibration data package for field mapping.

[0053] The beneficial effects of this invention are as follows: (1) This invention constructs a multi-field dynamic test system with real geological environment constraints by simultaneously deploying and uniformly acquiring stress, pore pressure, temperature, microseismic and fracture imaging on a three-dimensional sample. This system can reproduce the entire process of stress evolution, fracture propagation and seepage transmission in deep coal and rock during fluidized mining under indoor conditions, significantly improving the experiment's ability to reproduce complex multi-physics field linkage behavior in the field, and providing a controllable and repeatable experimental platform for studying the mechanism of deep in-situ fluidization.

[0054] (2) This invention utilizes synchronously acquired multi-field monitoring matrix data to establish a stress-fracture-seepage coupling model through gridded registration and feature extraction, and outputs a set of coupling parameters. By identifying the spatiotemporal correlation between stress increment, fracture connectivity change and pore pressure fluctuation, the model reveals the driving and feedback mechanism between the three fields, and can accurately quantify the energy accumulation characteristics and seepage guidance effect in the stress concentration area, thereby achieving quantitative analysis of the micro-fracture evolution and energy transfer law during fluidized bed transformation.

[0055] (3) This invention introduces a three-zone active control logic of "source-path-field" for the first time in physical simulation experiments. Based on the coupling identification results, the source zone, propagation path zone and protection zone are automatically delineated, and control command data for zone loading and injection are generated. Through the programmable loading system and injection control system, the stress loading, medium injection and seepage response are coordinated and controlled, so that energy release, wave propagation and structural protection can be linked in a closed loop under the same time base. This breaks through the limitations of single-field loading and passive observation in traditional experiments and forms an active and intelligent dynamic disaster simulation system.

[0056] (4) This invention calculates three types of judgment quantities—source intensity, propagation attenuation, and energy-absorbing displacement—using process response data and compares them with a set of control indicators. When any indicator fails to meet the standard, the system automatically locates the corresponding partition and executes parameter correction and control command updates. This cyclical mechanism of "real-time judgment—parameter correction—process reproduction" enables the experiment to have online self-learning and self-calibration capabilities, dynamically track energy evolution and structural response changes, and provide real-time adjustable technical basis for on-site dynamic disaster prevention and control.

[0057] (5) This invention backfills the process response data and judgment results of the entire process into the coupled model, and forms a highly reliable parameter set through deviation calibration and steady-state verification, and encapsulates it into a test calibration data package for field mapping. This data package can not only be used for the initial parameter tuning of the field monitoring system, but also provide model boundaries and control benchmarks for the establishment of digital twin mines, realizing the connection of the entire chain from physical test to field control, and from data acquisition to intelligent decision-making, and significantly improving the systematicness and intelligence level of dynamic disaster prevention and control in deep fluidized mining.

[0058] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0059] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A multi-field coupled physical simulation test method for fluidized coal mining, characterized in that, include: S1. Unify the spatial coordinates and time reference on the three-dimensional sample, set up stress, pore pressure, temperature, microseismic and crack imaging channels, continuously collect data and synchronize the timing to obtain the original data of the multi-field monitoring matrix. S2. Denoise and grid registration are performed on the original data of the multi-field monitoring matrix, stress increment and crack connectivity change are extracted, and spatiotemporal alignment is completed to construct a stress-crack-seepage coupling model and a set of coupling parameters. S3. Identify the coupling based on the stress-crack-seepage coupling model to obtain the coupling identification result; S4. Based on the coupling identification results, the source area, propagation path area and protection area are delineated in the three-dimensional sample, control command data for partitioned loading and injection are generated, and programmable loading and medium injection are executed. Micro-vibration, stress, displacement and pore pressure response are recorded online to obtain process response data. S5. Calculate the three types of judgment quantities—source intensity, propagation attenuation, and energy absorption displacement—from the process response data, and compare them with the control index set to obtain the comparison results; S6. After adjusting the control command data according to the comparison result, repeat S4 to output the judgment result when the comparison result meets the standard. S7. Backfill the process response data and the judgment result into the stress-crack-seepage coupling model, complete the parameter calibration, and generate a test calibration data package for field mapping.

2. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 1, characterized in that, Step S1 includes: Stress loading surfaces, pore pressure interfaces, temperature control zones, and micro-vibration sensing zones are set on the outer wall and key interfaces of the three-dimensional specimen, and crack imaging windows are arranged in the transparent or transmissive areas of the three-dimensional specimen. In a multi-field monitoring system, a unified spatial coordinate reference and time reference are set, the output signals of each sensor channel are formatted in a unified manner, and a mapping relationship between monitoring points and coordinate grids is established. By synchronously acquiring the controller, the stress, pore pressure, temperature, microseismic and crack imaging channels are triggered in sequence to acquire stress values, pore pressure, temperature field changes, microseismic event waveforms and crack image frames at each monitoring point in real time. After synchronizing and aligning the real-time acquired data in time and space, the synchronized multi-type data are arranged in the order of the coordinate grid and integrated into the original data of the multi-field monitoring matrix.

3. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 1, characterized in that, Step S2 includes: In the original data of the multi-field monitoring matrix, data cleaning is performed on each type of signal based on the monitoring channel identifier and time reference to remove noise waveforms and outliers, forming a filtered and smoothed basic dataset. According to the spatial grid division rules of the three-dimensional specimen, the stress values, pore pressure, temperature, microseismic events and crack image results in the basic dataset are mapped to the corresponding grid cells, and registration is performed with the grid center point as a reference to obtain the registered multi-field monitoring matrix. The stress difference of each grid cell in adjacent time periods is calculated to form the stress increment distribution. At the same time, the fracture skeleton is extracted from the fracture image sequence and the rate of change of the connectivity path is calculated to form the fracture connectivity change distribution. Within a unified time window, the stress increment distribution is matched with the fracture connectivity change distribution, the correlation coefficient between the stress increment distribution and the fracture connectivity change distribution is calculated, and the coupling parameters characterizing their mutual influence are determined. A stress-fracture-seepage coupling model and a set of coupling parameters are then established.

4. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 3, characterized in that, The process of determining the spatial grid division rules includes: Based on the geometry of the three-dimensional sample and the distribution density of monitoring points, the volume of the three-dimensional sample is divided into several spatial units; each spatial unit has a unique coordinate index; the size of the spatial unit is determined according to the monitoring resolution and the sample size, so that the side length of the spatial unit is no more than twice the minimum monitoring interval, in order to ensure the spatial uniformity of multi-field data within the unit. Non-uniform layered grids are set according to the geostress gradient and pore pressure gradient in the depth direction, so that the grid density in the high gradient region is higher than the average density, thereby improving the accuracy of local stress-fracture response capture. Increasing the grid sampling density along the main direction of fracture development in the horizontal direction ensures that the grid division is consistent with the main axis of fracture connectivity, thereby maintaining the spatial continuity of stress-fracture-seepage coupling characteristics.

5. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 1, characterized in that, Step S4 includes: Within the three-dimensional specimen, the energy conversion rate of the coupling control region and the distribution of fracture connectivity of the coupling identification results are spatially superimposed. The source region is defined based on the region with energy conversion rate higher than a set threshold, the propagation path region is defined based on the region with large fracture connectivity and continuous energy transmission, and the protection region is defined based on the region with significant energy attenuation and structural integrity. Within each zone, loading and injection boundary conditions are set. A loading axis system is established based on the principal stress direction of the source zone and the dominant seepage direction of the propagation path zone. Control parameters for load rate, medium injection pressure and flow rate are determined to form control command data for zone loading and injection. The control command data is input into the programmable loading system and the injection control system, and the partitioned loading and media injection process is executed through a unified time base to achieve multi-field coordinated response under simulated conditions; During loading and injection, microseismic sensors, stress gauges, displacement gauges, and pore pressure gauges deployed continuously collect response signals. Based on a time synchronization mechanism, they record the microseismic energy, stress changes, displacement response, and pore pressure fluctuations throughout the entire process, forming process response data.

6. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 5, characterized in that, The process of determining the loading and injection boundary conditions includes: Based on the spatial distribution of the source region, propagation path region, and protection region, the initial values ​​of the stress boundary, pore pressure boundary, and constraint boundary of each region are calculated. A linear loading procedure is applied in the principal stress direction according to the stress boundary, and the injection rate and flow state of the pore pressure boundary are adjusted synchronously so that the loading stress and seepage pressure change synergistically under a unified time reference. Displacement constraints are established around the protected area, and the boundary constraint parameters are dynamically corrected by real-time monitoring of the sample boundary displacement and reaction force signals to ensure that the stress and seepage transmission during the loading and injection process are in a closed state.

7. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 1, characterized in that, Step S5 includes: The process response data is time-synchronized and spatially registered, and the microseismic energy, stress change rate, displacement response and pore pressure change are mapped to three-dimensional mesh cells according to the sampling time to obtain a spatiotemporally corresponding dataset. The source intensity determination quantity is calculated based on the dataset corresponding to the spatiotemporal time; the source intensity determination quantity is determined by the coupling relationship between the microseismic energy release rate and the stress reduction amplitude, and is used to reflect the degree of energy accumulation and release in the source area; The propagation attenuation determination quantity is calculated based on the change in stress wave amplitude at measuring points before and after the energy-reducing barrier; the propagation attenuation determination quantity is used to characterize the energy dissipation capability of the stress wave in the propagation path region. The energy-absorbing displacement judgment quantity is calculated by using the difference between the peak displacement and the recovered displacement of the displacement monitoring points in the protected area; the energy-absorbing displacement judgment quantity is used to characterize the energy absorption and deformation capacity of the support system against external dynamic loads; The source intensity determination quantity, propagation attenuation determination quantity, and energy absorption displacement determination quantity are combined into three sets of determination quantities, and then compared with the corresponding source intensity upper limit, path attenuation target, and energy absorption displacement target in the control index set to obtain the comparison results.

8. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 1, characterized in that, Step S6 includes: Based on the type of judgment quantity that does not meet the standard in the comparison results, determine the region to which the control command data to be corrected belongs, and extract the parameter set of the corresponding source region, propagation path region or protection region; When the non-compliant item is the source strength determination quantity, adjust the loading rate, injection pressure and medium phase state of the source area to keep the energy release rate and stress reduction within the control index constraints. When the non-compliance item is the propagation attenuation judgment quantity, the arrangement parameters of the energy-cutting barrier and the seepage channel in the propagation path area are modified, including their thickness, location and wave impedance difference, in order to enhance the energy dissipation effect. When the non-compliant item is the energy absorption displacement judgment quantity, adjust the number of components, spacing and pre-tightening state of the energy absorption support system in the protected area to improve the deformation energy absorption capacity of the support system; After the adjustment is completed, the corrected control command data is regenerated, and the partition loading and medium injection process defined in step S4 is executed. New micro-vibrations, stresses, displacements and pore pressure responses are recorded in real time, and the process response data is updated until all judgment quantities in the comparison results reach the set threshold and the judgment result is output.

9. The multi-field coupled physical simulation test method for fluidized coal mining according to claim 1, characterized in that, Step S7 includes: The process response data and the judgment result are aligned by time series and spatial coordinates and then imported into the stress-fracture-seepage coupling model. The stress concentration parameters, fracture guiding parameters and seepage connectivity parameters in the stress-fracture-seepage coupling model are matched. The calibration coefficient is calculated based on the deviation between the predicted values ​​and the measured response data of the stress-fracture-seepage coupling model, and the model parameter set is updated using the calibration coefficient to ensure that the stress distribution, fracture connectivity and pore pressure change output by the stress-fracture-seepage coupling model are consistent with the measured results. After calibration, the calibrated coupled model is subjected to steady-state verification. The stability and response accuracy of the model output are verified through multiple rounds of virtual loading and medium injection simulation, and the final parameter set is determined. The calibrated and verified model parameters, process response characteristics, and judgment results are encapsulated together into a field-oriented test calibration data package; the test calibration data package is used to guide the initial setting and parameter tuning of the field stress-crack-seepage dynamic monitoring and linkage control system.

10. A multi-field coupled physical simulation test system for fluidized coal mining, characterized in that, include: The multi-field monitoring unit is used to unify the spatial coordinates and time reference on the three-dimensional sample, set up stress, pore pressure, temperature, microseismic and crack imaging channels, continuously collect data and synchronize the timing to obtain the original data of the multi-field monitoring matrix; The data registration and coupling modeling unit is used to denoise and mesh the original data of the multi-field monitoring matrix, extract stress increment and crack connectivity change, and complete spatiotemporal alignment to construct a stress-crack-seepage coupling model and a set of coupling parameters. The coupling identification unit is used to identify the coupling based on the stress-crack-seepage coupling model and obtain the coupling identification result. The partitioned loading and injection control unit is used to delineate the source zone, propagation path zone and protection zone in the three-dimensional specimen according to the coupling identification result, generate control command data for partitioned loading and injection, execute programmable loading and medium injection, record micro-vibration, stress, displacement and pore pressure response online, and obtain process response data. The judgment quantity calculation unit is used to calculate three types of judgment quantities—source intensity, propagation attenuation, and energy absorption displacement—from the process response data, and compare them with the control index set to obtain the comparison results. The control command correction unit is used to adjust the control command data according to the comparison result and repeat the implementation steps of the partition loading and injection control unit, so as to output the judgment result when the comparison result meets the standard. The model calibration and output unit is used to backfill the process response data and the judgment result into the stress-crack-seepage coupling model, complete the parameter calibration, and generate a test calibration data package for field mapping.

Images