Well-ground collaborative geological monitoring method and system for deep coal underground gasification

Abstract

The application provides a kind of well-ground collaborative geological monitoring method and system for deep coal underground gasification, relating to computer technology field, first obtains target deep coal seam geological survey data, delimits near field, middle field and far field monitoring layer range and lays out well-ground integrated three-dimensional monitoring network;Start monitoring equipment to collect data, integrate to form multi-source monitoring data and process in time and space synchronization;Obtain geophysical logging and other various data to construct target coupled geological model, input synchronous multi-source monitoring data to jointly invert output inversion data set;Integrate inversion data set into model to present combustion cavity and surrounding rock state, generate early warning signal and control instruction;Obtain production feedback data, integrate and compare with early warning signal, adjust monitoring network and model parameters, output optimization parameters.The application realizes deep coal underground gasification multi-scale, multi-field coupling, high-precision comprehensive monitoring and safety warning.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and more specifically, to a well-ground collaborative geological monitoring method and system for deep underground coal gasification. Background Technology

[0002] With the increasing depletion of shallow coal resources, developing deeper coal resources has become an inevitable direction for ensuring energy security and meeting energy demands. Underground coal gasification (UCG) technology, at depths exceeding one kilometer, is a highly promising clean development method for deep coal resources, demonstrating broad application prospects. However, the deep underground coal gasification process faces numerous unprecedented technical challenges and difficulties.

[0003] First, the deep environment is extremely complex, characterized by high ground stress, high rock temperature, and high pore / fracture pressure – a "triple threat." These extreme conditions make the rock's mechanical behavior exceptionally complex, making the surrounding rock around the combustion chamber highly susceptible to unsteady fractures, which can trigger dynamic disasters and seriously threaten the safe and stable operation of the gasification process. Second, monitoring signals attenuate significantly during propagation at depth. Traditional surface geophysical methods, such as seismic and electromagnetic methods, experience substantial energy attenuation during their long propagation paths at depth, resulting in extremely low signal-to-noise ratios. This makes it difficult to perform detailed imaging of targets at depths of over 1,000 meters, and to accurately obtain crucial information about the deep underground coal gasification process. Third, drilling deep monitoring wells is costly, and the wellbore operates under harsh conditions of high temperature and pressure. Conventional downhole instruments face severe challenges to their lifespan and reliability under these conditions, making it difficult to meet long-term, stable monitoring requirements.

[0004] Furthermore, during deep underground coal gasification, the thermal field generated by high-temperature combustion strongly couples with the stress field and seepage field formed by the "three-high" geological environment (high temperature, high humidity, and high temperature), simultaneously triggering complex physicochemical reactions and forming a chemical field. These multi-field coupling effects are very pronounced, and traditional single monitoring methods are simply unable to capture their chain reactions, making it difficult to comprehensively and accurately reflect the actual situation of the gasification process. Finally, during deep underground coal gasification, the risk of cavity runaway increases significantly, and gas leakage paths and impacts on deep aquifers become more difficult to predict and effectively manage, resulting in a multiplied increase in safety risks.

[0005] Currently, monitoring technologies for shallow or medium-deep underground coal gasification cannot be directly applied to underground coal gasification processes at depths of over 1,000 meters due to insufficient consideration of the unique and complex nature of the deep environment, and are therefore unable to effectively address the aforementioned extreme challenges at deep depths. Summary of the Invention

[0006] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a well-surface collaborative geological monitoring method for deep underground coal gasification, the method comprising:

[0007] Obtain geological exploration data of the target deep coal seam, delineate the layout range and location of near-field monitoring layer, mid-field monitoring layer and far-field monitoring layer based on the geological exploration data of the target deep coal seam, deploy an integrated well-ground three-dimensional monitoring network including near-field monitoring layer, mid-field monitoring layer and far-field monitoring layer, and output the monitoring network layout results;

[0008] Activate all monitoring equipment in the well-ground integrated three-dimensional monitoring network, collect near-field monitoring data, mid-field monitoring data and far-field monitoring data within the coverage area of ​​the well-ground integrated three-dimensional monitoring network, integrate all monitoring data to form multi-source monitoring data, perform spatiotemporal synchronization processing on the multi-source monitoring data, and output synchronized multi-source monitoring data;

[0009] Acquire geophysical logging data, core test data, and geostress measurement data of the target deep coal seam; integrate the geophysical logging data, core test data, and geostress measurement data of the target deep coal seam; construct a target coupled geological model; input synchronous multi-source monitoring data into the target coupled geological model for joint inversion processing; and output an inversion data set.

[0010] The inversion dataset is integrated into the target coupled geological model. The target coupled geological model after integrating the inversion dataset presents the state of the deep underground coal gasification combustion chamber and the surrounding rock response state. Early warning signals are generated based on the presented state. Control measures are generated in combination with the early warning signals. The generated control measures are decomposed into executable control instructions, and early warning signals and control instructions are output.

[0011] Acquire actual production gas data and regulation effect data, integrate production gas data, regulation effect data with early warning signals, regulation commands and inversion data sets, perform multi-data comparison processing, adjust monitoring network layout parameters and target coupled geological model parameters based on comparison processing results, and output optimized monitoring network parameters and model parameters.

[0012] Furthermore, embodiments of the present invention also provide a well-surface collaborative geological monitoring system for deep underground coal gasification, characterized in that it includes:

[0013] A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described well-to-surface collaborative geological monitoring method for deep underground coal gasification by executing the machine-executable instructions.

[0014] Based on the above, by acquiring geological exploration data of the target deep coal seam, the deployment range and location of near-field, mid-field, and far-field monitoring layers are accurately delineated, and an integrated well-ground three-dimensional monitoring network is constructed. This achieves multi-level and all-round monitoring coverage of the deep underground coal gasification area, effectively overcoming the problems of severe signal attenuation and limited monitoring range of traditional monitoring methods at depth. All equipment in the monitoring network is then activated to collect and integrate multi-source monitoring data, performing spatiotemporal synchronous processing to accurately reflect the actual conditions at different locations and times during the gasification process. A target-coupled geological model is constructed by combining geophysical logging data, core test data, and geostress measurement data. Synchronous multi-source monitoring data is input for joint inversion processing, enabling the model to more accurately simulate the physicochemical processes of deep underground coal gasification and improving the prediction accuracy of combustion chamber state and surrounding rock response state. By integrating the inversion data set into the target-coupled geological model, the model presents the state of the combustion chamber in deep underground coal gasification and the response state of the surrounding rock, and generates early warning signals and control measures. This enables advanced early warning and timely control of safety risks in the gasification process, effectively reducing safety risks such as chamber runaway and gas leakage. Finally, actual production feedback data is obtained and compared with early warning signals, control commands, and the inversion data set. Based on the results, the parameters of the monitoring network deployment and the target-coupled geological model are adjusted, realizing the dynamic optimization of the monitoring method and model. This allows the model to better adapt to the complex changes in the deep underground coal gasification process, further improving the accuracy and reliability of monitoring. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the execution flow of the well-ground collaborative geological monitoring method for deep coal underground gasification provided in an embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of exemplary hardware and software components of a well-to-surface collaborative geological monitoring system for deep coal underground gasification provided in an embodiment of the present invention. Detailed Implementation

[0017] Figure 1 This is a schematic flowchart of a well-ground collaborative geological monitoring method for deep coal underground gasification provided by an embodiment of the present invention, which will be described in detail below.

[0018] Step S110: Obtain geological exploration data of the target deep coal seam, delineate the layout range and location of near-field monitoring layer, mid-field monitoring layer and far-field monitoring layer based on the geological exploration data of the target deep coal seam, deploy an integrated well-ground three-dimensional monitoring network including near-field monitoring layer, mid-field monitoring layer and far-field monitoring layer, and output the monitoring network layout results.

[0019] This embodiment focuses on a deep, high-quality bituminous coal reservoir at a depth of over one kilometer. This area contains small fault structures, and while the coal seam is stable, local fractures are present due to tectonic influences. To effectively monitor the underground coal gasification process in this area, geological exploration data of the target deep coal seam must first be collected. This data, obtained through previous geological exploration projects, covers key information such as the geological structural characteristics, physical and mechanical properties of the rock strata, and the coal seam's occurrence state. Based on this fundamental data, a multi-scale, layered monitoring approach is adopted. The spatial ranges of three monitoring layers—near-field, mid-field, and far-field—are sequentially defined, and corresponding monitoring equipment is deployed accordingly, ultimately integrating them into a well-to-surface integrated three-dimensional monitoring network.

[0020] Step S111: Obtain geological exploration data of the target deep coal seam, the geological exploration data including the burial depth distribution, rock strata structure, tectonic distribution and fracture development status of the target deep coal seam.

[0021] In this embodiment, geological exploration data is acquired through a combination of various geological exploration methods. Depth distribution data is primarily obtained through the interpretation of reflected waves from seismic profiles, corrected using borehole logging data to form a three-dimensional data volume reflecting the elevation changes of the coal seam's top and bottom surfaces. Rock strata structure data is based on core sampling and well logging analysis to determine the thickness, lithological combination characteristics, and spatial distribution patterns of each lithological layer, including the lithological type, thickness variations, and integrity of the coal seam's top and bottom plates. Structural distribution data is obtained through comprehensive analysis of surface geological mapping, seismic profile interpretation, and borehole exposure data, focusing on identifying the occurrence, scale, and spatial distribution characteristics of faults, folds, and other structures, particularly providing detailed characterization of fault fracture zones that may affect the gasification process. Fracture development status data is obtained through core fracture statistics, imaging logging interpretation, and in-situ water pressure tests, including fracture density, occurrence, filling conditions, and permeability characteristics.

[0022] Step S112: Based on the geological survey data, calculate and delineate the area adjacent to the deep underground coal gasification combustion chamber as the deployment range of the near-field monitoring layer, and output the deployment range data of the near-field monitoring layer.

[0023] The deployment range of the near-field monitoring layer is determined with the combustion chamber as the core, comprehensively considering the coal seam occurrence characteristics, structural development, and gasification process parameters. First, based on the coal seam distribution model in the geological exploration data and combined with the gasifier design parameters, the possible expansion range of the combustion chamber is estimated. Then, considering the influence range of the temperature and stress fields during combustion, the area within a certain distance around the combustion chamber is determined as the core area for near-field monitoring. Simultaneously, referring to the structural distribution in the geological exploration data, especially the location of discontinuities such as faults, the initially delineated range is adjusted to ensure that the near-field monitoring layer can cover the key areas that the combustion chamber may affect. During the delineation process, a three-dimensional spatial analysis method is used, combining the planar range and vertical depth to form a three-dimensional spatial region encompassing the combustion chamber and its directly affected zone, which serves as the deployment range of the near-field monitoring layer.

[0024] Step S113: Based on the near-field monitoring layer deployment range data, generate deep monitoring well deployment parameters corresponding to the near-field monitoring layer. The deep monitoring well deployment parameters include well location coordinates, design depth, and relative positional relationship with the target deep coal seam.

[0025] After determining the deployment range of the near-field monitoring layer, deep monitoring wells need to be deployed within this range. The selection of well location coordinates comprehensively considers the representativeness of the monitoring target, the suitability of geological conditions, and the feasibility of construction. By analyzing the rock strata stability, structural development, and relative position to the combustion chamber within the near-field monitoring layer range, a location that can maximize the acquisition of monitoring data and has the lowest construction risk is selected as the well location. The design depth is determined based on the burial depth distribution data of the target deep coal seam, ensuring that the monitoring well can penetrate the target coal seam and extend to a certain depth to monitor the geological response below the coal seam floor. The relative positional relationship with the target deep coal seam is achieved through well trajectory design. Depending on the monitoring requirements, it can be designed as a vertical well or a directional well, so that the monitoring well can maintain a specific spatial relationship with the coal seam in the coal seam section, such as vertically crossing or extending along the strike of the coal seam, to obtain more comprehensive monitoring data.

[0026] Step S114: Based on the deep monitoring well deployment parameters, generate a near-field monitoring layer logical configuration scheme for deploying a distributed optical fiber sensing system at the corresponding monitoring location. The near-field monitoring layer logical configuration scheme includes the deployment logic of distributed temperature sensing optical fibers, distributed acoustic sensing optical fibers, and distributed strain sensing optical fibers, the spatial coverage range, and the logical connection relationship with the well-ground integrated three-dimensional monitoring network.

[0027] Step S1141: Calculate and determine the sensor system layout logic parameters of the distributed optical fiber sensing system based on the deep monitoring well layout parameters. The sensor system layout logic parameters include the depth range of the sensing optical fiber, the spatial arrangement, and the theoretical coupling model with the well wall.

[0028] The determination of the logical parameters for the sensor system deployment needs to be combined with the wellbore structure and monitoring targets of the deep monitoring well. The deployment depth range is determined based on the vertical range of the near-field monitoring layer and the estimated location of the combustion chamber, ensuring that the sensing fiber can cover the entire area potentially affected by combustion, from a certain distance above the coal seam roof to a certain distance below the coal seam floor. Considering the spatial resolution and reliability of the monitoring data, the spatial arrangement typically employs multiple fibers deployed along different orientations around the well to acquire monitoring data from different directions. The theoretical coupling model with the wellbore is established based on wellbore conditions and fiber optic sensing principles, considering the physical properties of the wellbore rock, the cement sheath bonding quality, and the fiber installation method to ensure that the fiber can effectively sense temperature, acoustic waves, and strain changes in the surrounding rock mass.

[0029] Step S1142: In the sensor system model library, match and select a distributed optical fiber sensor system logic model that is compatible with the environmental conditions represented by the deployment parameters of the deep monitoring well. The distributed optical fiber sensor system logic model includes the functional module definitions and performance parameters of distributed temperature sensing optical fiber, distributed acoustic sensing optical fiber, and distributed strain sensing optical fiber.

[0030] In this embodiment, the sensor system model library stores the technical parameters of various distributed fiber optic sensor systems, including operating temperature range, pressure tolerance, spatial resolution, and measurement accuracy. Based on the downhole environmental conditions reflected by the deep monitoring well deployment parameters, such as temperature, pressure, and well depth, the system is screened and matched within the model library. The selected distributed fiber optic sensor system must be able to operate stably in harsh downhole environments and meet the requirements for monitoring accuracy and spatial resolution. The functional module definitions clearly define the functions and working principles of each type of fiber optic (temperature measurement, acoustic wave, strain measurement), while the performance parameters specify the technical indicators of each module in detail, such as temperature measurement range, acoustic wave frequency response range, and strain measurement range.

[0031] Step S1143: Based on the distributed optical fiber sensing system logic model and the deep monitoring well deployment parameters, generate virtual encapsulation processing parameters for the distributed optical fiber sensing system logic model. The virtual encapsulation processing parameters are used to define the thermal insulation and thermal conductivity properties and temperature and high pressure resistance logic properties of the sensing optical fiber.

[0032] The generation of virtual encapsulation processing parameters is to ensure the normal operation of the distributed fiber optic sensing system in the high-temperature and high-pressure environment downhole. Based on the performance requirements of the distributed fiber optic sensing system's logical model and the environmental conditions reflected by the deep monitoring well deployment parameters, the fiber optic encapsulation structure is designed. Thermal insulation and conductivity properties are achieved by selecting appropriate encapsulation materials and structures to reduce the impact of external temperature on the fiber optic measurement accuracy, while ensuring that the fiber optic can accurately sense temperature changes in the rock mass. Temperature and pressure resistance properties are determined based on the downhole temperature and pressure conditions, specifying the temperature resistance rating and pressure resistance of the encapsulation material, as well as the sealing performance of the encapsulation structure, to prevent damage to the fiber optic from high-temperature and high-pressure fluids.

[0033] Step S1144: Based on the sensor system deployment logic parameters and the virtual encapsulation processing parameters, generate the logic flow and spatial coordinates for embedding and fixing the distributed optical fiber sensor system logic model in the virtual deep monitoring well, and output the optical fiber embedding and fixing logic.

[0034] The design of the fiber optic cable installation and fixing logic process needs to consider the safety and feasibility of construction. First, the installation sequence and method of the fiber optic cables are determined based on the sensor system deployment logic parameters, such as using a winch for lowering or tubular delivery. In the virtual deep monitoring well, the position and direction of the fiber optic cable within the well are determined based on the wellbore trajectory and spatial coordinates. The fixing method must ensure good coupling between the fiber optic cable and the wellbore wall, while preventing damage to the fiber optic cable during downhole operations. The spatial coordinates are generated based on the wellbore trajectory data, corresponding the fiber optic cable installation depth to the three-dimensional coordinates of the wellbore for subsequent spatial positioning of monitoring data.

[0035] Step S1145: Based on the deployment range data of the near-field monitoring layer, generate control instructions to adjust the monitoring parameters of the logic model of the distributed optical fiber sensing system. The control instructions are used to set the monitoring angle and spatial sensitivity distribution so that its logical monitoring range covers the deployment range of the near-field monitoring layer, and output the monitoring parameter adjustment logic.

[0036] The generation of monitoring parameter adjustment logic ensures that the monitoring range of the distributed fiber optic sensing system fully covers the near-field monitoring layer. Based on the near-field monitoring layer deployment data, the adjustment targets for monitoring angle and spatial sensitivity distribution are determined. The monitoring angle is adjusted by controlling the directionality of the fiber optic sensor, ensuring the sensor faces the target monitoring area. The spatial sensitivity distribution is adjusted by setting different sensitivity levels based on the importance of different locations within the monitoring area and the expected signal strength, increasing sensitivity in critical areas to obtain more detailed data. The format and content of control commands are determined according to the communication protocol of the distributed fiber optic sensing system, ensuring that the commands can be correctly recognized and executed by the system.

[0037] Step S1146: Generate a virtual connection channel definition for data transmission between the distributed optical fiber sensing system logic model and the ground data center. The virtual connection channel definition includes a communication protocol, a data flow interface, and logical link parameters for simulating high-temperature and high-voltage cables, and outputs the data transmission link logic.

[0038] The generation of virtual connection channel definitions needs to consider the reliability and real-time performance of data transmission. The choice of communication protocol is based on the characteristics of the monitoring data and transmission requirements, such as using TCP / IP or a dedicated industrial bus protocol. The data flow interface defines the data format, transmission rate, and verification method to ensure that data is not lost or corrupted during transmission. Logical link parameters simulate the characteristics of temperature- and high-voltage resistant cables, including signal attenuation, transmission delay, and interference immunity, to evaluate the quality and reliability of data transmission. By defining these parameters, a stable and efficient data transmission link is established between the distributed fiber optic sensing system and the ground data center.

[0039] Step S1147: Based on the fiber optic insertion and fixing logic, monitoring parameter adjustment logic, and data transmission link logic, generate system debugging logic for joint testing of the virtual sensing transmission system of the near-field monitoring layer. The system debugging logic includes sending virtual test signals and receiving and analyzing feedback data to verify the completeness of the data acquisition and transmission logic.

[0040] The generation of system debugging logic is to verify the proper functioning of the near-field monitoring layer virtual sensing transmission system. The debugging process includes sending virtual test signals, which simulate various physical quantity changes that may occur during actual monitoring, such as temperature rise, acoustic signal changes, and strain changes. After receiving these test signals, the system acquires and transmits data, sending the feedback data to the ground data center. By analyzing the consistency between the feedback data and the virtual test signals, the system's data acquisition accuracy, transmission reliability, and response speed are evaluated, and the completeness of the data acquisition and transmission logic is verified. If any problems are found, the system parameters are adjusted based on the debugging results.

[0041] Step S1148: Based on the running results of the system debugging logic, generate logic instructions to fine-tune the parameters of the virtual sensing transmission system to optimize its logic performance and obtain the optimized virtual sensing transmission system.

[0042] The logic instructions for parameter fine-tuning are generated based on the results of the system debugging logic. By analyzing the feedback data during the debugging process, problems with system performance are identified, such as large data acquisition errors, excessive transmission delays, or severe signal interference. For these problems, parameters requiring adjustment are determined, such as sensor gain, sampling frequency, filter parameters, or transmission protocol timeout settings. The generation of logic instructions follows a specific optimization algorithm. Through gradual parameter adjustments and retesting, the system performance is brought to its optimal state, resulting in an optimized virtual sensing and transmission system.

[0043] Step S1149: Generate a process to start the optimized virtual sensing transmission system, collect virtual near-field monitoring data, and perform logical verification with preset monitoring data, and output the system test verification logic.

[0044] The system testing and verification process includes starting the optimized virtual sensing transmission system and bringing it into normal operation. This begins the collection of virtual near-field monitoring data, which simulates the physical changes in the near-field monitoring layer during actual combustion. Simultaneously, preset monitoring data is generated based on expected monitoring results from numerical simulations or empirical data. The collected virtual near-field monitoring data is then logically verified against the preset monitoring data to compare their consistency. Verification includes checking data trends, numerical ranges, and characteristic parameters. Through this verification, the system's monitoring performance in a simulated environment is evaluated, ensuring it accurately reflects the state of the actual monitored object.

[0045] Step S11410: Integrate the sensor system deployment logic parameters, the selected distributed optical fiber sensor system logic model, encapsulation logic parameters, optical fiber insertion and fixing logic, monitoring parameter adjustment logic, data transmission link logic, system debugging logic, system fine-tuning logic, and system testing and verification logic to generate a near-field monitoring layer logic configuration scheme, and mark the virtual monitoring points and logical positions.

[0046] The integration of the near-field monitoring layer logic configuration scheme involves systematically organizing and combining the logic parameters and processes generated in the preceding steps. It organically combines elements such as the sensor system deployment logic parameters, the distributed fiber optic sensor system logic model, and encapsulated logic parameters to form a complete configuration scheme. Simultaneously, it marks the location and logical relationships of virtual monitoring points, clarifying the monitoring parameters and data transmission paths for each monitoring point.

[0047] Step S115: Based on the geological exploration data, calculate and delineate the areas of deep coal underground gasification heat influence and stress disturbance, as well as key structural zones, as the locations for the mid-field monitoring layer, and output the mid-field monitoring layer location data.

[0048] The location of the mid-field monitoring layer is determined primarily by considering the extent of thermal impact and stress disturbance during deep underground coal gasification, as well as the distribution of key structural zones. Based on heat conduction parameters and geostress data from geological exploration, numerical simulation methods are used to calculate the distribution characteristics of the temperature and stress fields during combustion, thus determining the extent of the thermally affected and stress-disturbed regions. Simultaneously, combined with structural distribution data, key structural zones, such as faults and fold axes, are included in the mid-field monitoring layer's deployment area. Through comprehensive analysis of these factors, a three-dimensional spatial region encompassing the thermally affected zone, stress-disturbed zone, and key structural zones is delineated as the location for the mid-field monitoring layer.

[0049] Step S116: Based on the location data of the mid-field monitoring layer, generate mid-field monitoring well deployment parameters. Based on the mid-field monitoring well deployment parameters, generate a mid-field monitoring layer logical configuration scheme for deploying multi-parameter integrated monitoring stations at the corresponding monitoring locations. The mid-field monitoring layer logical configuration scheme determines the monitoring items used by the multi-parameter integrated monitoring stations to collect in-situ three-dimensional stress, pore pressure, temperature gradient, and gas composition, as well as their logical nodes in the well-ground integrated three-dimensional monitoring network.

[0050] The generation process for mid-field monitoring well deployment parameters is similar to that of near-field monitoring wells, but the characteristics and monitoring requirements of the mid-field monitoring layer must be considered. Well locations should be chosen in representative positions within thermally affected and stress-disturbed areas and key structural zones to obtain critical geological parameters for these areas. The design depth is determined based on the vertical range of the mid-field monitoring layer, ensuring that the monitoring well can penetrate the main target monitoring layers. The deployment of multi-parameter integrated monitoring stations is the core of the mid-field monitoring layer, and its monitoring items include in-situ three-dimensional stress, pore pressure, temperature gradient, and gas composition. Logical nodes are determined based on the topology of the integrated well-ground monitoring network, assigning a unique logical address to each monitoring item to achieve unified management and transmission of monitoring data.

[0051] Step S117: Based on the geological exploration data, calculate and delineate the overlying strata and corresponding surface area of ​​the target deep coal seam as the deployment area of ​​the far-field monitoring layer, and output the data of the far-field monitoring layer deployment area.

[0052] The deployment area of ​​the far-field monitoring layer is determined by the overlying strata of the target deep coal seam and the corresponding area on the surface. Based on the rock strata structure and tectonic distribution in the geological exploration data, the stability and possible deformation characteristics of the overlying strata are analyzed to determine the monitoring range of the overlying strata. The corresponding area on the surface is determined by projecting the underground monitoring area onto the surface and considering the potential impact on the surface. The deployment area of ​​the far-field monitoring layer is usually large, needing to cover the entire area from the top of the overlying strata of the coal seam to the surface, as well as a certain area on the surface, in order to comprehensively monitor the impact of underground gasification processes on the surface and overlying strata.

[0053] Step S118: Based on the data of the far-field monitoring layer deployment area, generate a logical configuration scheme for the far-field monitoring layer of the surface monitoring equipment and the airborne monitoring equipment. The surface monitoring equipment includes an ultra-dense array broadband seismograph and a high-sensitivity gas flux monitoring station. The airborne monitoring equipment is configured to perform time-series interferometric synthetic aperture radar monitoring.

[0054] The generation of the logical configuration scheme for the far-field monitoring layer needs to consider the characteristics and monitoring requirements of both surface and airborne monitoring equipment. Among the surface monitoring equipment, ultra-dense array broadband seismometers are used to monitor vibration signals from underground rock masses. Their deployment should be based on the size of the monitoring area and geological conditions, employing either a uniform distribution or a denser deployment in key areas. High-sensitivity gas flux monitoring stations are used to monitor the release of surface gases; their deployment locations should be selected in areas where gas leaks may occur or on critical structural zones. Airborne monitoring equipment uses temporal interferometric synthetic aperture radar technology to acquire surface deformation data via satellite or UAVs. Its monitoring range covers the entire far-field monitoring layer deployment area, and the monitoring cycle is determined based on the required monitoring accuracy. The logical configuration scheme clearly defines the deployment location, quantity, technical parameters, and data transmission methods for each type of equipment.

[0055] Step S119: Logically integrate the near-field monitoring layer logical configuration scheme, the mid-field monitoring layer logical configuration scheme, and the far-field monitoring layer logical configuration scheme to generate a unified monitoring network topology, communication protocol, and data interface specification, and generate the monitoring network deployment result. The monitoring network deployment result includes the virtual structure of the well-ground integrated three-dimensional monitoring network and the logical configuration parameters of each node.

[0056] The logical integration of the monitoring network organically combines the configuration schemes of the three monitoring layers to form a unified, integrated well-to-ground three-dimensional monitoring network. First, the topology of the monitoring network is constructed, clarifying the connections and data flows between each monitoring layer and between each monitoring device. Then, a unified communication protocol is established to ensure that different types of monitoring devices can achieve data interconnection. The data interface specification defines the data format, transmission rate, and interaction method. The monitoring network deployment result includes a virtual structure model and logical configuration parameters for each node, such as device ID, IP address, and monitoring parameter type.

[0057] Step S120: Activate all monitoring equipment in the well-ground integrated three-dimensional monitoring network, collect near-field monitoring data, mid-field monitoring data and far-field monitoring data within the coverage area of ​​the well-ground integrated three-dimensional monitoring network, integrate all monitoring data to form multi-source monitoring data, perform spatiotemporal synchronization processing on the multi-source monitoring data, and output synchronized multi-source monitoring data.

[0058] After the monitoring network is deployed, all monitoring equipment needs to be activated according to a predetermined procedure. The activation process includes sequentially activating the equipment at each monitoring layer to ensure it is in normal working order. Once activated, the equipment begins collecting monitoring data within its respective coverage area. Near-field monitoring data mainly includes temperature, sound waves, and strain; mid-field monitoring data includes in-situ stress, pore pressure, temperature gradient, and gas composition; and far-field monitoring data includes seismic signals, gas flux, and surface deformation. The collected data is transmitted to a ground data center via its respective transmission links for integration and processing, forming multi-source monitoring data. To ensure the validity and comparability of the data, spatiotemporal synchronization processing of the multi-source monitoring data is required, unifying the time reference and spatial coordinates of the data, ultimately outputting synchronized multi-source monitoring data.

[0059] Step S121: Obtain the monitoring network deployment results and output the monitoring equipment positioning results based on all monitoring equipment in the integrated three-dimensional monitoring network.

[0060] The location of monitoring equipment is determined based on the equipment information and network topology in the monitoring network deployment results, identifying the actual spatial position of each monitoring device. By analyzing the equipment coordinate parameters and wellbore trajectory data in the deployment results, each monitoring device is correlated with its specific physical location. For downhole equipment, its accurate underground location is determined by combining the well location coordinates and design depth of the deep monitoring well; for surface equipment, its location is directly determined based on the geographical coordinates in the deployment plan; and for airborne equipment, its coverage area is determined through its monitoring range and trajectory parameters.

[0061] Step S122: Output a start signal to the distributed optical fiber sensing system of the near-field monitoring layer. After receiving the start signal, the distributed optical fiber sensing system enters the working state and outputs near-field monitoring data acquisition signals.

[0062] The start-up signal is generated by the ground control center sending commands to the distributed fiber optic sensing system in the near-field monitoring layer. The command format and content conform to the system's communication protocol and include information such as the device identification code and the start-up command. Upon receiving the start-up signal, the distributed fiber optic sensing system performs internal initialization and self-testing, and enters operational mode after confirming that all modules are functioning correctly. Subsequently, the system begins acquiring data according to preset monitoring parameters and generates a near-field monitoring data acquisition signal, which is fed back to the ground control center, indicating that the system has begun normal data acquisition.

[0063] Step S123: Collect temperature data, acoustic data and strain data of the area covered by the near-field monitoring layer through the distributed optical fiber sensing system of the near-field monitoring layer, and output the near-field monitoring data after summarizing and integrating them.

[0064] The distributed fiber optic sensing system detects physical changes in the area covered by the near-field monitoring layer through its internal sensing fibers. Temperature data is acquired by measuring the intensity changes of Raman scattered light in the fiber, acoustic data is obtained by monitoring the frequency changes of Brillouin scattered light in the fiber, and strain data is calculated based on the photoelastic effect of the fiber by analyzing the polarization state changes of the scattered light. The acquired raw data is amplified, filtered, and digitized by the system's internal signal processing module, and then aggregated and integrated according to a preset data format. During the integration process, the data is time- and spatially labeled to ensure that each data point corresponds to a specific monitoring location and time, ultimately outputting structured near-field monitoring data.

[0065] Step S124: Output a start signal to the multi-parameter integrated monitoring station of the field monitoring layer. After receiving the start signal, the multi-parameter integrated monitoring station enters the working state and outputs the field monitoring data acquisition signal.

[0066] Similar to the near-field monitoring layer, the activation signal for the multi-parameter integrated monitoring station in the mid-field monitoring layer is also transmitted via the ground control center. The activation signal includes the station identification code and activation instructions. Upon receiving the signal, the multi-parameter integrated monitoring station activates the power supply and data acquisition units of each monitoring module. After a brief warm-up and calibration, the station enters operational mode and begins acquiring in-situ three-dimensional stress, pore pressure, temperature gradient, and gas composition monitoring data. Simultaneously, the station outputs a mid-field monitoring data acquisition signal to notify the ground control center that data acquisition has begun.

[0067] Step S125: Collect in-situ three-dimensional stress data, pore pressure data, temperature gradient data and gas composition data of the area covered by the mid-field monitoring layer through the multi-parameter integrated monitoring station of the mid-field monitoring layer, and output the mid-field monitoring data after summarizing and integrating them.

[0068] Each monitoring module of the multi-parameter integrated monitoring station is responsible for acquiring different parameters. In-situ three-dimensional stress data is obtained by measuring the stress state of the rock mass through stress sensors; pore pressure data is obtained by monitoring the pressure changes of pore fluids through pressure sensors; temperature gradient data is obtained by acquiring temperature values ​​at different depths and calculating gradients through distributed temperature sensors; and gas composition data is obtained by analyzing the content of various components in collected gas samples through gas sensors. The data acquired by each module is transmitted to the station's central processing unit via an internal bus for data aggregation, verification, and formatting. The processed field monitoring data includes the measured values ​​of each parameter, acquisition time, monitoring location, and other information, and is output in a unified format.

[0069] Step S126: Output a start signal to the surface monitoring equipment and airborne monitoring equipment of the far-field monitoring layer. After receiving the start signal, the surface monitoring equipment and airborne monitoring equipment enter the working state and output far-field monitoring data acquisition signals.

[0070] The activation signal for the far-field monitoring layer is sent to both surface and airborne monitoring equipment. For surface monitoring equipment, such as ultra-dense broadband seismograph arrays and high-sensitivity gas flux monitoring stations, the activation command is sent directly from the ground control center. For airborne monitoring equipment, such as satellites or UAVs performing time-series interferometric synthetic aperture radar monitoring, the activation is initiated through a pre-set mission plan or remote control command. Upon receiving the activation signal, each device completes system initialization and parameter settings, enters data acquisition mode, and outputs a far-field monitoring data acquisition signal, indicating the start of data acquisition.

[0071] Step S127: Collect surface deformation data through the airborne monitoring equipment of the far-field monitoring layer, summarize and integrate the data collected by the surface monitoring equipment of the far-field monitoring layer and the airborne monitoring equipment of the far-field monitoring layer, and output the far-field monitoring data.

[0072] Step S1271: Obtain the geographic coordinate data of the far-field monitoring layer deployment area, plan the flight path and monitoring points of the far-field monitoring layer airborne monitoring equipment according to it, so that the flight path covers the entire far-field monitoring layer deployment area, and output the flight monitoring plan.

[0073] In this embodiment, the geographic coordinate data of the far-field monitoring layer deployment area adopts the Gauss-Kruger projection coordinate system, including the latitude and longitude coordinates of the area boundary. First, the area is divided into several regular rectangular sub-regions, with the side length of each sub-region determined according to the spatial resolution of the airborne monitoring equipment. Then, flight paths are designed based on the distribution of the sub-regions, employing an "S"-shaped round-trip flight pattern to ensure a certain degree of overlap between adjacent flight paths to meet the requirements of interferometry. Monitoring points are set at the center of each sub-region, while the density of monitoring points is increased near the area boundary and key structural zones. The flight altitude is determined based on the sensor parameters and imaging resolution requirements of the airborne monitoring equipment, while the flight speed comprehensively considers data acquisition efficiency and imaging quality. The final generated flight monitoring scheme includes parameters such as the latitude and longitude coordinate sequence of the flight path, flight altitude, speed, monitoring point distribution, and acquisition time window.

[0074] Step S1272: Start the airborne monitoring equipment of the far-field monitoring layer, carry out flight monitoring according to the flight monitoring plan, and start the time-series interferometric synthetic aperture radar sensor to perform ground imaging and output the airborne monitoring start signal.

[0075] Once the ground control center receives the flight monitoring plan, it sends a start command to the airborne monitoring equipment. The flight control system of the airborne monitoring equipment autonomously navigates according to the flight path parameters in the plan, adjusting its flight attitude and altitude. Simultaneously, the temporal interferometric synthetic aperture radar sensor begins power-on preheating and initializes internal parameters, including radar operating frequency, pulse repetition frequency, and antenna gain. After the equipment enters the designated monitoring area and reaches a stable flight state, the sensor begins transmitting radar signals and receiving echoes to image the ground surface. At this point, the airborne monitoring equipment sends an airborne monitoring start signal to the ground control center. This start signal includes the equipment ID, current position coordinates, imaging parameters, and a start timestamp.

[0076] Step S1273: The ground surface of the far-field monitoring layer deployment area is continuously imaged by the time-series interferometric synthetic aperture radar sensor of the far-field monitoring layer airborne monitoring equipment, and the ground surface image data of multiple time periods are collected, the imaging time and location are recorded, and the ground surface image data is output.

[0077] The temporal interferometric synthetic aperture radar (TIAR) sensor continuously images the far-field monitoring layer area at fixed time intervals according to set imaging parameters. During each imaging session, the sensor emits radar waves that illuminate the ground surface, and the echo signal is processed to form a single radar image. During the imaging process, the device's built-in GPS records the precise time and location coordinates of each image in real time and associates them with the corresponding image data for storage. The time intervals for multi-period surface image data are determined according to monitoring requirements, typically ranging from several days to several weeks, to capture the dynamic process of surface deformation. The acquired surface image data includes radar echo intensity information, phase information, and corresponding time and location tags, and is stored in a standard radar data format.

[0078] Step S1274: Denoise the surface image data by using an image denoising algorithm to eliminate interference signals and noise points, and output the denoised surface image data after enhancing the clarity.

[0079] Surface image data is subject to various noise interferences during acquisition, such as thermal noise and speckle noise. This embodiment employs an image denoising algorithm based on wavelet transform. First, the surface image data is decomposed into multi-scale wavelet components to obtain wavelet coefficients for different frequency components. Then, based on the noise distribution characteristics in the wavelet domain, high-frequency coefficients are thresholded, with coefficients below the threshold considered noise and set to zero, while larger coefficients are retained as signal components. Finally, the processed coefficients are reconstructed into denoised image data through inverse wavelet transform. During the denoising process, an adaptive threshold selection method is used, dynamically adjusting according to the noise level in different areas to maximize the preservation of image detail while removing noise. The signal-to-noise ratio of the denoised surface image data is significantly improved.

[0080] Step S1275: Extract surface deformation features from the denoised surface image data, compare data from different time periods to identify deformation areas and states, record the geographic coordinates and morphology of the deformation areas, and output surface deformation feature data.

[0081] First, the denoised multi-time-period surface image data are registered to precisely align images acquired at different times to the same geographic coordinate system. Then, differential interferometric synthetic aperture radar (DISAR) technology is used to perform interferometric processing on image pairs from adjacent time periods, generating interferometric patterns. By unwrapping the phase of the interferometric patterns, the phase change information of the surface is obtained and then converted into deformation. Based on the magnitude and distribution of the deformation, deformation regions on the surface are identified. For each deformation region, its geographic coordinates of the boundary, shape (e.g., circular, elliptical, strip-shaped), and the direction and approximate range of deformation are recorded. Simultaneously, the correlation between the distribution of deformation regions and geological structures and underground gasification activities is analyzed to preliminarily determine the causes of deformation. The surface deformation feature data is stored in the form of vector polygons, containing attribute information for each deformation region.

[0082] Step S1276: Perform time-series processing on the surface deformation characteristic data, track the dynamic changes in deformation, record the amount and rate of deformation changes at different time periods, and output time-series surface deformation data.

[0083] The extracted surface deformation characteristic data are arranged chronologically to construct a deformation time series. For each deformation region, the difference in deformation between adjacent time periods is calculated to obtain the deformation change. The deformation rate is obtained by dividing the deformation change by the time interval. During time series processing, linear or polynomial fitting methods are used to perform trend analysis on the deformation time series to identify acceleration, deceleration, or stable states of deformation. Simultaneously, abnormal deformation data points are detected and removed to ensure the reliability of the time series data. The time series surface deformation data includes the deformation, deformation rate, and trend analysis results of each deformation region at different time points, and is stored in the form of a time series database.

[0084] Step S1277: Associate and bind the time series surface deformation data with the geographic coordinate data of the far-field monitoring layer deployment area, mark the specific location and status of each deformation area, and output the surface deformation data.

[0085] Each deformation region in the time-series surface deformation data is precisely correlated with the geographic coordinate system of the far-field monitoring layer deployment area to ensure the accuracy of the deformation region's location information. Through coordinate transformation, the vector polygon coordinates of the deformation region are converted to a format consistent with the geographic coordinate data. Then, a status label, such as "expanding," "stable," or "shrinking," is added to each deformation region. These statuses are determined based on the deformation rate and trend analysis results obtained from time-series processing. Simultaneously, the deformation amount and deformation rate data of the deformation region are stored as attribute information correlated with the geographic coordinates. The final output surface deformation data is a geographic information dataset containing spatial location, time-series deformation information, and status labels.

[0086] Step S1278: Acquire seismic data and gas flux data collected by the surface monitoring equipment in the far-field monitoring layer, classify and organize them, remove invalid data, and output valid surface monitoring data.

[0087] The far-field monitoring layer surface monitoring equipment includes ultra-dense array broadband seismographs and high-sensitivity gas flux monitoring stations. Seismic data is continuous waveform data, containing information such as arrival time, amplitude, and frequency of seismic waves; gas flux data includes the concentration, flux value, and acquisition time of different gas components. First, the data is classified and stored separately for seismic and gas flux data. Then, data quality checks are performed. For seismic data, records with instrument malfunctions, strong interference, or waveform distortion are removed; for gas flux data, records exceeding the instrument's measurement range, with abnormal sampling, or missing data are removed. Simultaneously, valid data undergoes format standardization, unifying the timestamp format and units. Valid surface monitoring data is stored in a structured manner for easy integration with other data later.

[0088] Step S1279: Associate and store the surface deformation data with the effective surface monitoring data according to the monitoring time, spatial location and physical quantity type to generate initial far-field monitoring data.

[0089] A unified database model is established to store surface deformation data and effective surface monitoring data. Each record in the database includes the monitoring time, spatial location (latitude and longitude coordinates), physical quantity type (such as deformation, seismic amplitude, gas concentration, etc.), and corresponding measurement value. Different types of data are linked by monitoring time to ensure that monitoring data from the same moment can be queried. Simultaneously, a spatial index is established based on spatial location to facilitate data retrieval by region. The physical quantity type serves as an attribute field to distinguish different monitoring parameters.

[0090] Step S12710: Perform integrity processing on the initial far-field monitoring data, supplement missing data and correct erroneous data to fully and accurately reflect the actual state of the far-field monitoring layer, and output the far-field monitoring data.

[0091] Interpolation methods were used to supplement missing values ​​in the initial far-field monitoring data. For missing time-series data, linear or spline interpolation was employed; for spatially missing data, spatial interpolation was performed based on data from surrounding monitoring points. Data with obvious errors, such as outliers exceeding reasonable limits, were corrected by comparing with historical data or combining with other relevant data; data that could not be corrected were discarded. Simultaneously, data consistency was checked to ensure temporal and spatial coordination of data collected by different monitoring devices. After integrity processing, the far-field monitoring data became more comprehensive and accurate, truly reflecting the actual state of the far-field monitoring layer.

[0092] Step S128: Classify, store, and label the near-field monitoring data, mid-field monitoring data, and far-field monitoring data according to their respective monitoring types, physical quantities, and units to generate structured multi-source monitoring data.

[0093] The classification, storage, and annotation of multi-source monitoring data are crucial aspects of data management. First, the data is initially classified according to monitoring type (near-field, mid-field, far-field). Then, within each monitoring type, further subdivisions are made based on physical quantities (such as temperature, stress, seismic waves, etc.) and units. During storage, a database management system is used to organize the data in a structured manner, facilitating retrieval and analysis. Annotation establishes connections between data from different sources, such as linking data from different monitoring points at the same time using timestamps, or linking monitoring data from different locations to geological models using spatial coordinates, thus forming a multi-source monitoring dataset.

[0094] Step S129: Use a high-precision time-stamping system to perform spatiotemporal synchronization processing on multi-source monitoring data, so that the monitoring data collected by different monitoring equipment and different monitoring areas have a unified time and space reference, and output synchronized multi-source monitoring data.

[0095] Spatiotemporal synchronization is crucial for ensuring the validity of multi-source monitoring data. A high-precision time-stamping system provides a unified time reference, ensuring clock consistency across all monitoring devices through satellite time synchronization or atomic clock synchronization. Based on time synchronization, the spatial coordinates of different monitoring areas are uniformly transformed, mapping the data to the same spatial reference frame. For time-series data, interpolation or resampling methods are used to align data from different sampling frequencies along the time axis. These processes ensure the consistency of multi-source monitoring data in both time and space, facilitating comprehensive analysis and comparison.

[0096] Step S1210: Transmit the synchronous multi-source monitoring data to the ground data center via an optical fiber composite cable. During the transmission process, the data is verified and retransmitted, and the synchronous multi-source monitoring data after transmission is completed is output.

[0097] The transmission of synchronous multi-source monitoring data uses a fiber optic composite cable as the transmission medium. This cable simultaneously performs power transmission and data communication functions, making it suitable for complex underground environments. Data transmission employs packet switching technology, dividing the data into multiple data packets. Each data packet contains data content, a checksum, and address information. During transmission, the ground data center verifies the received data packets, checking their integrity and correctness. If a data packet is lost or erroneous, a feedback mechanism requests retransmission from the sending end. After verification and retransmission, all synchronous multi-source monitoring data is ensured to be accurately transmitted to the ground data center, outputting the completed synchronous multi-source monitoring data.

[0098] Step S130: Obtain geophysical logging data, core test data, and geostress measurement data of the target deep coal seam; integrate the geophysical logging data, core test data, and geostress measurement data of the target deep coal seam; construct a target coupled geological model; input synchronous multi-source monitoring data into the target coupled geological model for joint inversion processing; and output an inversion data set.

[0099] To gain a deeper understanding of the deep underground coal gasification process, it is necessary to construct a target-coupled geological model and perform joint inversion processing. First, geophysical logging data from the target deep coal seam is collected, reflecting the physical properties of the coal seam and surrounding rock. Core testing data provides the mechanical and physical parameters of the rock, while geostress measurement data reveals the stress state of the underground rock mass. These data are integrated, outliers are removed, and missing data is supplemented to form the foundational data for model construction. Based on this data, a target-coupled geological model incorporating geological structure, rock mass physical and mechanical properties, and initial stress field is constructed using numerical simulation methods. Then, synchronous multi-source monitoring data is input into the model, and a joint inversion algorithm is used to invert key information such as the morphology of the combustion chamber, thermal field distribution, and stress field evolution, ultimately outputting an inversion dataset.

[0100] Step S131: Collect resistivity data, sonic transit time data and density data of the target deep coal seam, organize and archive them, and output the geophysical logging data of the target deep coal seam.

[0101] Geophysical logging data acquisition is accomplished during the drilling process using logging instruments. Resistivity data is measured using a resistivity logging tool, reflecting the electrical conductivity of the rock formation; sonic transit time data is acquired using a sonic logging tool, used to calculate the porosity and elastic parameters of the rock formation; density data is measured using a density logging tool, providing density information of the rock formation. The acquired raw data undergoes preprocessing, including environmental correction, depth correction, and curve smoothing, to remove the influence of interfering factors. Then, it is organized and archived according to logging projects and depths, forming standardized geophysical logging data, stored in a specific format for subsequent model building.

[0102] Step S132: Drill core samples from different areas of the target deep coal seam, perform physical and mechanical property tests and thermophysical parameter tests on them, record the data and output the core test data of the target deep coal seam.

[0103] Core samples were drilled from different areas within the target deep coal seam to ensure representativeness. The drilled core samples underwent physical and mechanical property testing in the laboratory, including determination of parameters such as uniaxial compressive strength, elastic modulus, and Poisson's ratio; simultaneously, thermophysical parameters such as thermal conductivity, specific heat capacity, and coefficient of thermal expansion were tested. The testing process was strictly conducted in accordance with relevant standards to ensure the accuracy and reliability of the data. Test data were recorded in standardized forms, including sample number, sampling location, test items, and test results, and the core test data was output after processing.

[0104] Step S133: Use the in-situ geostress measurement method to measure the geostress of the target deep coal seam and surrounding rock strata, collect three-dimensional geostress data including the maximum principal stress data, minimum principal stress data and intermediate principal stress data, and output the geostress measurement data of the target deep coal seam.

[0105] In-situ stress measurement employs methods suitable for deep environments, such as stress relief or hydraulic fracturing. During the measurement process, measuring instruments are installed in the borehole to acquire stress values ​​in the rock mass in different directions. The measurement data includes the magnitude and direction of the maximum principal stress, minimum principal stress, and intermediate principal stress. The measurement data is analyzed and processed to remove measurement errors and correct the calculated stress results. The final output of the stress measurement data is expressed in tensor form, containing the magnitude and direction angle of the three principal stresses.

[0106] Step S134: Integrate geophysical logging data, core test data, and geostress measurement data of the target deep coal seam, remove abnormal data, and uniformly organize the valid data to output basic geological data.

[0107] The integration of basic geological data is a process of comprehensively processing geophysical well logging data, core testing data, and geostress measurement data. First, a quality check is performed on all types of data to identify and remove outliers, such as those caused by instrument malfunction or operational errors. Then, valid data undergoes standardized format conversion and unit conversion to ensure consistency. For data from different sources, spatial coordinates and depth information are used to establish correspondences between the data. The integrated data is then organized according to the needs of the geological model, forming basic geological data that includes information on rock strata physical and mechanical parameters, geophysical properties, and geostress states.

[0108] Step S135: Construct a target coupled geological model based on basic geological data, which includes the initial geological conditions, rock mass structure parameters, thermophysical parameters and mechanical parameters of the target deep coal seam. After the construction is completed, output the target coupled geological model.

[0109] The target-coupled geological model is constructed based on fundamental geological data using 3D geological modeling software. First, based on the rock strata structure and tectonic distribution in the geological exploration data, the geometric framework of the model is established, including the spatial morphology of geological bodies such as coal seams, roof and floor strata, and faults. Then, the rock mass physical and mechanical parameters, thermophysical parameters, and initial geostress data from the fundamental geological data are assigned to the corresponding geological bodies in the model. Rock mass structural parameters are determined based on fracture development data, such as fracture density, occurrence, and permeability. After model construction, mesh generation is performed, discretizing the continuous geological body into a finite number of elements for numerical calculations. The final output target-coupled geological model contains the model's geometric information, physical parameters, and initial conditions, and can be used for subsequent joint inversion analysis.

[0110] Step S136: Extract near-field monitoring data from the synchronous multi-source monitoring data, filter out the acoustic data and strain data and convert them into formats to meet the requirements of joint inversion processing, and output the near-field monitoring data to be inverted.

[0111] Near-field monitoring data is extracted from synchronous multi-source monitoring data. Then, based on the requirements of joint inversion processing, acoustic and strain data are selected. Acoustic data includes information such as sound wave propagation time, amplitude, and frequency; strain data includes the magnitude and direction of strain in the rock mass. The selected data undergoes format conversion to a format recognizable by numerical simulation software, such as text or binary files. Simultaneously, data preprocessing is performed, including noise removal, smoothing filtering, and data normalization, to improve the accuracy of the inversion results. The processed near-field monitoring data to be inverted includes information such as the data's time series, spatial location, and physical quantity values.

[0112] Step S137: Input the near-field monitoring data to be inverted into the target coupled geological model, and use microseismic tomography and moment tensor inversion to invert it, locate the deep underground coal gasification combustion front and the surrounding rock fracture area, and output the near-field inversion data.

[0113] After the inverted near-field monitoring data is input into the target coupled geological model, microseismic tomography is used to process the acoustic data. By analyzing the propagation path and arrival time information of microseismic events, the wave velocity distribution of the rock mass is inverted, thereby determining the location and morphology of the combustion front. Simultaneously, the focal mechanism of the microseismic events is analyzed using the moment tensor inversion method to obtain information such as the type, scale, and direction of surrounding rock fracturing, thereby locating the surrounding rock fracturing area. During the inversion process, the model parameters are continuously adjusted to minimize the residuals between the inversion results and the monitoring data. The final output near-field inversion data includes the three-dimensional coordinates of the combustion front, the distribution range of the surrounding rock fracturing area, and fracturing mechanism parameters.

[0114] Step S138: Extract temperature data from near-field monitoring data and temperature gradient data from mid-field monitoring data from synchronous multi-source monitoring data, use them as independent input parameter sets to input into the target coupled geological model, combine with the heat conduction-convection model for inversion processing, and output thermal field inversion data.

[0115] Step S1381: Acquire synchronous multi-source monitoring data, extract near-field monitoring data and filter out near-field temperature data, which is collected by distributed temperature measurement fiber of the distributed fiber optic sensing system of the near-field monitoring layer.

[0116] Synchronous multi-source monitoring data is stored in the database of the ground data center, and near-field monitoring data is retrieved using a structured query language. Near-field monitoring data includes various types such as temperature, sound waves, and strain. Temperature data is filtered based on data tags. This temperature data is acquired by distributed temperature-sensing optical fibers, possessing high spatial and temporal resolution. The data format is a three-dimensional array containing the coordinates of the monitoring point location, a timestamp, and the temperature value. The filtered near-field temperature data undergoes a preliminary check to remove obvious noise points and outliers, ensuring data reliability.

[0117] Step S1382: Extract the mid-field monitoring data from the synchronous multi-source monitoring data and filter out the temperature gradient data. The temperature gradient data is collected by the multi-parameter integrated monitoring station of the mid-field monitoring layer and outputs the mid-field temperature data.

[0118] Similarly, midfield monitoring data is extracted through database queries, and temperature gradient data is filtered based on data type tags. Midfield temperature gradient data is collected by a distributed temperature sensor array in a multi-parameter integrated monitoring station, and the temperature gradient is calculated by measuring temperature values ​​at different depths. The data format includes station number, depth, timestamp, and temperature gradient value. Quality control is performed on the midfield temperature data to check its continuity and rationality, and invalid data caused by sensor malfunctions or environmental interference is removed.

[0119] Step S1383: Collect and label the near-field temperature data and the mid-field temperature gradient data according to their monitoring points, time, depth and physical quantity type, remove abnormal data, and output the collected temperature field related data.

[0120] Create a unified data table to categorize and arrange near-field temperature data and mid-field temperature gradient data according to the geographical coordinates of the monitoring points, the acquisition time, the depth, and the type of physical quantity (temperature or temperature gradient). Add annotation information to each data record, such as the data source (near-field or mid-field) and sensor number. Then, use statistical methods to identify outliers, such as calculating the mean and standard deviation of the data, and considering data exceeding the mean plus or minus three times the standard deviation as outliers and removing them. The aggregated temperature field data forms a complete dataset containing temperature and temperature gradient information at different locations, depths, and times.

[0121] Step S1384: Convert the format of the integrated temperature data and output the converted temperature data.

[0122] The target-coupled geological model requires input data in a specific grid data format; therefore, the aggregated temperature field data needs to be converted. First, based on the model's grid division scheme, discrete monitoring point data are interpolated onto the model's computational grid nodes. For temperature data, distance-weighted interpolation is used; for temperature gradient data, it is converted into corresponding grid node gradient values ​​based on its physical meaning. Then, the interpolated data is organized according to the model's required format, including grid node numbers, coordinates, temperature values, or temperature gradient values. After format conversion, the temperature data can be directly read and used by the target-coupled geological model.

[0123] Step S1385: Obtain the thermophysical parameter data from the basic geological data, which includes the thermal conductivity, specific heat capacity and density data of the rock strata, and input them into the target coupled geological model to construct a heat conduction-convection model.

[0124] Thermophysical parameters in the basic geological data were obtained through previous core testing and laboratory analysis and stored in a geological database. Thermal conductivity, specific heat capacity, and density data for each rock layer were extracted from the database; these parameters are key inputs for constructing the heat conduction-convection model. These parameters were assigned values ​​according to the spatial distribution of the rock layers in the target coupled geological model, specifying corresponding thermophysical parameter values ​​for each grid cell. The heat conduction-convection model is based on the law of conservation of energy, considering both heat conduction and fluid convection as heat transfer mechanisms. Its governing equations include time derivative terms, heat conduction terms, and convection terms.

[0125] Step S1386: Input the converted temperature data into the heat conduction-convection model, set its initial boundary conditions and inversion parameters, start the inversion process, present the thermal field distribution state corresponding to the temperature data, and output the initial thermal field inversion data.

[0126] After format conversion, the temperature data is used as the observational data input for the heat conduction-convection model. Initial boundary conditions include the initial temperature distribution of the model's computational domain (set based on geological background and previous temperature measurements) and the heat flux density or temperature value at the boundaries. Inversion parameters include the spatial distribution of thermal conductivity, fluid velocity, and thermal convection intensity. A gradient descent-based optimization algorithm is used for inversion processing, adjusting model parameters to minimize the residual between the temperature field calculated by the model and the observed data. During the inversion process, the thermal field distribution corresponding to the temperature data is presented in real-time, displayed in the form of contour maps or 3D cloud maps. The initial thermal field inversion data includes the temperature and heat flux density values ​​of each grid node calculated by the model.

[0127] Step S1387: Present the thermal field expansion range of the deep underground coal gasification combustion chamber using a heat conduction-convection model, delineate the thermal field boundary location and expansion direction, record relevant data, and output thermal field expansion data.

[0128] The extent of thermal field expansion is determined by analyzing the temperature field distribution calculated using a heat conduction-convection model. A temperature threshold (e.g., above a specific background temperature) is set, and the region with a temperature above this threshold is defined as the heat-affected zone, i.e., the extent of thermal field expansion. The dynamic changes in the boundary position of the heat-affected zone are determined by tracking the boundary at different times. The direction of thermal field expansion is determined by calculating the trajectory of the center of gravity of the heat-affected zone or the directional distribution of the boundary expansion rate. The area, volume, boundary coordinates, and expansion direction angle of the thermal field expansion are recorded, and the output is the thermal field expansion data.

[0129] Step S1388: Present the spatiotemporal distribution of thermal power using a heat conduction-convection model, record the thermal power changes in different monitoring areas and at different times, and output thermal power distribution data.

[0130] The calculation of thermal power is based on the energy equation in the heat conduction-convection model. The total thermal power is obtained by integrating the heat flux density within the calculation region of the model. The spatiotemporal distribution is obtained by allocating the total thermal power to each grid cell and combining it with time series analysis. For different monitoring areas, the thermal power values ​​are calculated separately, and the spatial distribution differences of thermal power are analyzed. Simultaneously, the thermal power values ​​at different time points are recorded, and curves of thermal power variation over time are plotted to analyze its trend. The thermal power distribution data includes the thermal power values ​​of each grid cell, the statistical values ​​of thermal power in different regions, and the time series data of thermal power.

[0131] Step S1389: Integrate the initial thermal field inversion data, thermal field extension data, and thermal power distribution data; compare the correspondence between the initial thermal field inversion data, thermal field extension data, and thermal power distribution data; adjust the inversion deviation; and output the adjusted thermal field inversion data.

[0132] The initial thermal field inversion data, thermal field expansion data, and thermal power distribution data are spatially and temporally matched to establish a correspondence among them. For example, changes in the thermal field expansion range should correspond to changes in thermal power. Through comparative analysis, inconsistencies between the data are identified, such as situations where thermal power increases but the thermal field expansion is slow, which may indicate a deviation in the inversion parameters. Based on the comparison results, the inversion parameters of the heat conduction-convection model are adjusted, such as correcting the thermal conductivity of the rock strata or the fluid convection velocity, and the inversion calculation is re-performed. After multiple iterative adjustments, the correspondence between the various data points is made more reasonable, and the adjusted thermal field inversion data is output.

[0133] Step S13810: Organize and archive the adjusted thermal field inversion data, mark the corresponding monitoring time period and area, and generate complete thermal field inversion data.

[0134] The adjusted thermal field inversion data was categorized and organized according to monitoring time period and monitoring area. Detailed metadata was added to each data file, including data acquisition time, coverage area, inversion parameters, and data quality assessment. Standardized data formats, such as HDF5, were used for storage to support efficient data querying and access. Simultaneously, a data summary report was generated, summarizing the main results of the thermal field inversion, such as thermal field distribution characteristics, expansion patterns, and trends in thermal power variation.

[0135] Step S139: Extract the field monitoring data from the synchronous multi-source monitoring data, and input the in-situ three-dimensional stress data, pore pressure data and gas composition data into the target coupled geological model as mechanical, seepage and chemical field constraints, respectively. Perform joint calibration on the near-field inversion data and thermal field inversion data, and output the calibrated inversion data.

[0136] The in-situ three-dimensional stress data, pore pressure data, and gas composition data from the field monitoring data reflect the mechanical state, fluid seepage characteristics, and chemical environment of the rock mass, respectively. These data are input as constraints into the target-coupled geological model to jointly calibrate the near-field and thermal field inversion data. Mechanically, the in-situ three-dimensional stress data is used to adjust the stress field distribution in the model, ensuring consistency between the inversion results and the actual stress state; the pore pressure data serves as the boundary condition for the seepage field, constraining fluid flow and pressure distribution; and the gas composition data reflects the progress of chemical reactions and is used to calibrate the chemical field model parameters. Through multi-field coupling analysis, comprehensively considering the interactions between various fields, the near-field and thermal field inversion data are corrected and optimized, outputting calibrated inversion data and improving the reliability of the inversion results.

[0137] Step S1310: Extract far-field monitoring data from the synchronous multi-source monitoring data, input the seismic data and surface deformation data into the target coupled geological model, perform cross-validation on the calibrated inversion data, integrate the near-field inversion data, thermal field inversion data and calibrated inversion data, and output the inversion data set.

[0138] The seismic and surface deformation data from the far-field monitoring data reflect the impact of underground gasification processes on large-scale rock masses. These data are input into the target-coupled geological model and compared with the calibrated inversion data to verify the rationality and accuracy of the inversion results. Seismic data can be used to determine whether the inverted fracture areas match actual microseismic activity; surface deformation data can verify the accuracy of the model's predictions of rock mass movement and deformation. Through cross-validation, the inversion data is further adjusted and improved. Finally, the near-field inversion data, thermal field inversion data, and calibrated inversion data are integrated to form an inversion data set containing information on combustion chamber status, thermal field distribution, stress field evolution, and fluid migration.

[0139] Step S140: Integrate the inversion data set into the target coupled geological model. The target coupled geological model after integrating the inversion data set presents the state of the deep underground coal gasification combustion chamber and the surrounding rock response state. Generate early warning signals based on the presented state, generate control measures in combination with the early warning signals, decompose the generated control measures into executable control instructions, and output early warning signals and control instructions.

[0140] After the inversion dataset is integrated into the target coupled geological model, the model can dynamically reflect the state of the deep underground coal gasification combustion chamber and the surrounding rock response. Through 3D visualization technology, the morphology, thermal field distribution, stress concentration areas, and fluid migration paths of the combustion chamber can be visually displayed. Based on this state information, early warning indicators and thresholds are set. When the monitoring data or inversion results exceed the threshold, a corresponding early warning signal is generated. Early warning signals are divided into different levels according to risk level, such as prompt, warning, and alarm. Based on the type and severity of the early warning signal, targeted control measures are constructed, such as adjusting gas injection parameters, controlling the combustion rate, or taking reinforcement measures. These control measures are decomposed into specific executable control instructions, specifying the operation object, parameter adjustment values, and execution time, and outputting early warning signals and control instructions to guide on-site production operations.

[0141] Step S141: Obtain the inversion data set and the target coupled geological model. Integrate all the data in the inversion data set into the corresponding modules of the target coupled geological model according to their types. After completing the data integration, output the integrated geological model.

[0142] The data integration process involves associating and fusing various data types from the inversion dataset with the corresponding modules of the target coupled geological model. First, based on the data type (e.g., combustion cavity morphology data, thermal field data, stress field data, etc.), the corresponding module in the model is determined. Then, the inversion data is input into the model through a data interface, replacing or supplementing the original data in the model. During the integration process, it is necessary to ensure that the data format and units are consistent with the model requirements. For spatial data, coordinate transformation is also required to match it with the model's spatial reference system. After data integration is completed, the model is validated to ensure the correctness and completeness of the data, and the integrated geological model is output.

[0143] Step S142: Present the three-dimensional morphology, volume, spatial distribution, boundary range and internal structure of the deep underground coal gasification combustion chamber through the integrated geological model, and output the combustion chamber morphology display results.

[0144] The integrated geological model possesses 3D visualization capabilities, clearly presenting the three-dimensional morphology of the combustion chamber. In the model, the combustion chamber is displayed as a 3D solid, its surface colored according to temperature or other physical parameters, intuitively reflecting the chamber's characteristics. The volume is calculated from the model; based on the chamber's 3D morphology and mesh division, the number of units contained within the chamber is counted, and the total volume is obtained by multiplying by the unit volume. The spatial distribution is determined using a coordinate system, clearly defining the chamber's specific location underground. The boundary is represented by the contour lines of the chamber's surface, while the internal structure can be displayed through cross-sectional cutting or transparent displays, such as the internal temperature distribution and fracture development. The output combustion chamber morphology display can be in the form of 3D images, animations, or data tables.

[0145] Step S143: The integrated geological model presents the thermal field distribution, expansion range, thermal power distribution and dynamic change process of the deep underground coal gasification combustion chamber, and outputs the thermal field analysis results of the combustion chamber.

[0146] The thermal field distribution is presented by plotting isotherms or temperature cloud maps in the integrated geological model. Different colors represent different temperature values, visually displaying the temperature gradient and high-temperature region distribution around the combustion chamber. The expansion range is determined by analyzing the thermal field distribution at different time points, identifying the expansion of areas where temperatures exceed a certain threshold, and plotting dynamic curves of thermal field expansion. The heat power distribution is calculated based on the thermal field data and the thermophysical parameters of the rock mass, reflecting the energy release and transfer during combustion. The dynamic change process is demonstrated through animation, showcasing the evolution of the thermal field over time, including temperature rise, propagation, and decay. The output combustion chamber thermal field analysis results include a thermal field distribution map, expansion curves, heat power data, and dynamic change animations.

[0147] Step S144: The stress field evolution, stress anomaly region and stability state of the surrounding rock of the deep underground coal gasification combustion chamber are presented through the integrated geological model, and the stress evolution analysis results of the surrounding rock are output.

[0148] The stress field evolution is presented based on stress data calculated by the model. Principal stress traces or stress contour maps are used to display the distribution and evolution trend of stress in different directions. Stress anomaly regions are identified by setting stress thresholds, marking areas exceeding these thresholds; these areas are typically prone to rock failure. Stability is assessed by calculating the safety factor or failure probability of the surrounding rock; areas with a safety factor less than 1 indicate a risk of instability. The output of the surrounding rock stress evolution analysis includes a stress field distribution map, stress anomaly region markings, a stability assessment report, and stress-time curves.

[0149] Step S145: Present the fluid migration path, migration state, dominant fluid channels, and fluid leakage state during the deep underground coal gasification process using the integrated geological model, and output the fluid migration analysis results.

[0150] The fluid migration path is presented using the seepage field calculation results in the model. Flow path diagrams or particle tracking animations are used to demonstrate the flow direction and path of the fluid within the rock mass. Migration status includes parameters such as fluid velocity, flow rate, and pressure distribution, displayed through contour maps or data tables. Dominant fluid channels refer to highly permeable regions where fluid flows preferentially, determined by analyzing permeability distribution and streamline density. Fluid leakage status is assessed by monitoring fluid flow rates at model boundaries or specific locations; a leakage risk is considered present when the flow rate exceeds a certain threshold. The output fluid migration analysis results include streamline diagrams, velocity distribution maps, dominant channel identification maps, and leakage status assessment reports.

[0151] Step S146: Based on the independent state information represented by the combustion chamber morphology display results, combustion chamber thermal field analysis results, surrounding rock stress evolution analysis results, and fluid migration analysis results, classify and identify the safety risks in the deep underground coal gasification process, and output the comprehensive risk judgment results.

[0152] The classification and identification of safety risks is based on a comprehensive analysis of various status information. First, based on changes in the combustion chamber morphology, it is determined whether there is a risk of excessive expansion or deformation of the chamber; based on thermal field analysis results, the risks of high-temperature damage to the surrounding rock and thermal shock are identified; based on stress evolution analysis, the risks of rockbursts or collapses that may be triggered by stress concentration areas are determined; based on fluid migration analysis, the risks of fluid leakage and gas explosions are assessed. These risks are then classified, such as chamber instability risk, thermal damage risk, stress disaster risk, and fluid leakage risk. Then, combining the severity and probability of occurrence of each risk, a comprehensive assessment is conducted, outputting a comprehensive risk judgment result, including risk type, level, scope of impact, and possible consequences.

[0153] Step S147: Generate multi-level early warning signals based on the comprehensive risk assessment results, and mark the triggering conditions and corresponding monitoring areas of the multi-level early warning signals.

[0154] The generation of multi-level early warning signals is determined based on the risk level in the comprehensive risk assessment results. Early warning signals are typically divided into multiple levels, such as Level 1 (alert), Level 2 (warning), and Level 3 (alarm). Each level corresponds to a different degree of risk severity and corresponding countermeasures. Triggering conditions are set based on the thresholds of various monitoring indicators. When the monitoring data or inversion results reach or exceed the threshold, the corresponding level of early warning signal is triggered. Simultaneously, the monitoring area corresponding to the early warning signal is marked, clearly indicating the specific location of the risk. The generation of early warning signals can be automated through a software system. When the triggering conditions are met, the system automatically issues the corresponding early warning information, including the early warning level, triggering conditions, monitoring area, and recommended measures.

[0155] Step S148: Based on the safety risk type corresponding to the multi-level early warning signal and combined with the parameters of the integrated geological model, generate targeted control measures, refine and decompose the generated control measures, and decompose each control measure into control instructions with clear operational objects.

[0156] For example, step S1481: Extract the safety risk type corresponding to the multi-level early warning signal from the comprehensive risk assessment result, which includes combustion chamber instability risk, surrounding rock fracture risk and fluid leakage risk.

[0157] The comprehensive risk assessment results are presented in a structured report, including information such as risk type, level, and scope of impact. By analyzing the report content, the types of safety risks directly associated with multi-level warning signals are extracted. Combustion chamber instability risk manifests as abnormal changes in chamber morphology, excessively rapid volume expansion, or reduced structural integrity; surrounding rock fracturing risk is mainly reflected in increased microseismic activity in stress concentration areas or rock deformation exceeding thresholds; fluid leakage risk is reflected by indicators such as abnormal gas composition, pressure changes, or excessive surface gas flux. Each risk type is clearly defined to ensure the targeted nature of subsequent control measures.

[0158] Step S1482: Extract the scope and status of the safety risks from the comprehensive risk assessment results, define the safety risk levels and determine the response order for different levels, and output the safety risk levels.

[0159] Based on the impact range parameters (such as area, volume, and number of monitored areas involved) and status descriptions (such as "expanding," "stable," or "weakening") in the comprehensive risk assessment results, safety risks are classified into levels. Typically, risk levels are divided into four categories: minor, moderate, severe, and urgent. The classification criteria are based on the size of the impact range, the speed of development, and the severity of the potential consequences. Simultaneously, based on the risk level and the urgency of the risk type, the response order is determined, prioritizing higher-level and rapidly developing risks. The output of the safety risk level includes information such as risk type, level, impact range, status, and response priority.

[0160] Step S1483: Obtain the parameters of the integrated geological model, including combustion chamber parameters, surrounding rock parameters, and thermal field parameters. Compare the safety risk type with the model parameters, locate the monitoring data corresponding to the core factors that cause safety risks, and output the core risk factor data.

[0161] The parameters of the integrated geological model are stored in the model database. Combustion chamber parameters (such as chamber morphology, volume, and temperature distribution), surrounding rock parameters (such as rock mechanical properties, fracture distribution, and stress state), and thermal field parameters (such as thermal power, temperature gradient, and thermal conductivity) are extracted through the model interface. Safety risk types are correlated with these parameters. For example, combustion chamber instability risk may be related to excessively high internal pressure, insufficient surrounding rock strength, or excessive temperature gradient; surrounding rock fracturing risk may be related to stress concentration exceeding rock strength or excessively rapid fracture propagation rate; and fluid leakage risk may be related to abnormal permeability or excessively large fluid pressure gradient. Through comparative analysis, the core factors causing each safety risk are identified, and corresponding monitoring data, such as pressure monitoring data, stress monitoring data, and permeability data, are extracted, outputting the core risk factor data.

[0162] Step S1484: In response to the risk of combustion chamber instability, based on the core risk factor data and the combustion chamber parameters in the integrated geological model, generate control measures to adjust the injection pressure and injection components, and output the chamber stability control measures.

[0163] Based on the core risk factor data, the specific causes of combustion chamber instability are identified, such as excessive gas injection pressure leading to excessive chamber expansion, or unreasonable gas composition resulting in overly vigorous combustion. Combining the combustion chamber parameters from the integrated geological model, such as the current gas injection pressure, injection rate, internal temperature and pressure distribution, reasonable gas injection pressure adjustment values ​​and gas composition ratios are calculated. For example, if the chamber pressure exceeds a safety threshold, the gas injection pressure can be reduced to a set value; if combustion is too vigorous, the ratio of oxygen and inert gas can be adjusted to slow the combustion reaction rate. The chamber stabilization control measures specify in detail the adjusted gas injection pressure values, the specific proportions of gas composition, the adjustment time window, and the implementation steps.

[0164] Step S1485: In response to the risk of surrounding rock fracturing, combine the core risk factor data and the surrounding rock parameters in the integrated geological model to generate control measures for injecting corrosion inhibitors and reinforcing the surrounding rock, and output the surrounding rock reinforcement control measures.

[0165] The core risk factor data indicates the causes of surrounding rock fracturing, such as stress concentration, low rock mass strength, or chemical corrosion. The surrounding rock parameters in the integrated geological model provide information on the mechanical properties, fracture distribution, and chemical environment of the rock mass. Based on this information, control measures for injecting corrosion inhibitors and reinforcing the surrounding rock are generated. The corrosion inhibitor injection measures include the type, concentration, injection volume, and injection location of the inhibitor to mitigate the damage caused by chemical corrosion to the surrounding rock. Surrounding rock reinforcement measures can employ grouting or anchor bolt support, specifically including parameters such as the selection of grouting materials, grouting pressure, grouting range, and the arrangement, length, and density of anchor bolts. These surrounding rock reinforcement and control measures aim to improve the stability of the surrounding rock and prevent further fracture propagation.

[0166] Step S1486: In response to the risk of fluid leakage, based on the core risk factor data and the fluid migration parameters in the integrated geological model, generate control measures to block the dominant fluid channels and strengthen fluid monitoring, and output fluid prevention and control measures.

[0167] Fluid migration parameters include permeability distribution, fluid pressure gradient, and the location of dominant channels. Based on core risk factor data, the path and cause of fluid leakage are determined, such as the presence of dominant channels or abnormally high permeability. Control measures to seal dominant fluid channels include selecting appropriate sealing materials (such as cement grout, gel, etc.), determining the sealing location and extent, and controlling injection pressure and volume. Measures to enhance fluid monitoring include increasing monitoring point density, increasing monitoring frequency, and expanding the range of monitoring parameters (such as adding monitoring items for gas composition, pressure, flow rate, etc.). The goal of fluid control and prevention measures is to stop fluid leakage and promptly detect signs of leakage.

[0168] Step S1487: Based on the safety risk level and the scope of impact, generate control measures to adjust the monitoring frequency, increase the monitoring frequency in the safety risk area, and output the monitoring adjustment control measures.

[0169] Based on the level of safety risk and the size of the impact area, a monitoring frequency adjustment plan is determined. For high-risk areas or areas with a large impact area, the sampling frequency of monitoring equipment is increased and the data collection interval is shortened to more timely grasp the changes in risk; for low-risk areas, the original monitoring frequency can be maintained or appropriately reduced. Simultaneously, for newly added risk areas or areas where the risk is aggravated, temporary monitoring points are added or backup monitoring equipment is activated. The monitoring adjustment and control measures clearly define the new monitoring frequency, sampling interval, and the location and parameters of newly added monitoring points for each monitoring area.

[0170] Step S1488: Based on the actual production status of deep underground coal gasification, conduct feasibility verification of cavity stability control measures, surrounding rock reinforcement control measures, fluid control control measures, and monitoring and adjustment control measures, and output the feasibility verification results.

[0171] The actual production status includes current gasification process parameters, equipment operating status, resource supply, and on-site construction conditions. Various control measures are compared and analyzed against these actual conditions to assess their operability and implementation difficulty. For example, adjusting the gas injection pressure requires considering the capacity of the gas injection equipment and the current production plan; injecting corrosion inhibitors requires confirming the supply of corrosion inhibitors and the availability of the injection equipment; reinforcing the surrounding rock requires assessing on-site construction space and time constraints. Feasibility verification results include the feasibility level of each measure (e.g., "feasible," "requires adjustment," "infeasible") and adjustment suggestions, such as modifying parameters, changing materials, or adjusting the implementation time.

[0172] Step S1489: Adjust the implementation details of each control measure based on the feasibility verification results, optimize the implementation process, and output the optimized control measures.

[0173] For control measures whose feasibility verification results indicate "adjustment is required," the implementation details should be modified according to specific adjustment suggestions. This could include adjusting the specific value of the gas injection pressure, replacing the sealing material with a more readily available one, or optimizing the grouting process parameters. The implementation process should be optimized to reduce unnecessary steps and improve the efficiency and safety of the measures. For example, the implementation times of multiple control measures should be coordinated to avoid conflicts; the transmission and analysis processes of monitoring data should be optimized to ensure timely feedback on the control effects. The optimized control measures will be more in line with actual production conditions and have higher operability.

[0174] Step S14810: Integrate and optimize the control measures, classify and organize them according to the type of safety risk and the order of response, mark the safety risk level and core risk factor data corresponding to each control measure, and output the control measures.

[0175] The optimized control measures are categorized according to safety risk type (combustion chamber instability, surrounding rock fracturing, fluid leakage), and the control measures under each risk type are arranged in the order of response. Each control measure is tagged with its corresponding safety risk level and core risk factor data, such as "combustion chamber instability - severe - excessive injection pressure". Simultaneously, the implementation time, target objects, key parameters, and expected effects of each measure are summarized to form a complete control measure document.

[0176] Step S149: Integrate early warning signals and control instructions, classify and organize them according to monitoring areas and risk types, and output early warning signals and control instructions.

[0177] Integrating early warning signals and control instructions involves systematically organizing the previously generated early warning information and control measures. Early warning signals and corresponding control instructions are grouped according to monitoring areas, making the risks and corresponding countermeasures for each area readily apparent. Simultaneously, they are categorized by risk type, facilitating managers to adopt appropriate management strategies based on different risk types. The integrated early warning signals and control instructions are output in report or table format, including information such as early warning level, risk type, monitoring area, triggering conditions, control measures, and execution instructions.

[0178] Step S150: Obtain the actual production gas data and control effect data from production feedback, integrate the production gas data, control effect data, early warning signals, control instructions, and inversion data sets, perform multi-data comparison processing, adjust the monitoring network layout parameters and target coupled geological model parameters based on the comparison processing results, and output the optimized monitoring network parameters and model parameters.

[0179] Actual production feedback data is crucial for optimizing monitoring networks and model parameters. Acquiring produced gas data, such as gas composition, flow rate, and pressure, reflects the efficiency of the gasification process and product characteristics. Acquiring control effect data, i.e., changes in various monitoring indicators after implementing control measures, allows for the evaluation of the effectiveness of these measures. Integrating this data with early warning signals, control commands, and inversion data sets, and through multi-data comparative analysis, identifies discrepancies between monitoring data and actual conditions, and deviations between control measures and expected results. Based on the comparison results, adjusting the deployment parameters of the monitoring network, such as the location of monitoring equipment, sampling frequency, and monitoring range, improves the accuracy and relevance of monitoring. Simultaneously, adjusting the parameters of the target-coupled geological model, such as rock mechanics parameters, thermophysical parameters, and permeability coefficient, enables the model to better reflect actual geological conditions and the gasification process, outputting optimized monitoring network and model parameters.

[0180] Step S151: Obtain the produced gas data during the deep underground coal gasification process, wherein the produced gas data includes the composition, flow rate and pressure data of the produced gas.

[0181] Produced gas data is collected using gas analysis instruments and flow and pressure sensors installed at the wellhead of the production well. Gas composition analysis employs gas chromatography or infrared spectroscopy to monitor the content of various gas components in the produced gas in real time, such as hydrogen, carbon monoxide, methane, and carbon dioxide. Flow data is measured using flow meters, and pressure data is acquired using pressure sensors. The collected data undergoes preprocessing, such as noise removal and zero-point and range calibration, to ensure accuracy. Produced gas data is recorded in time series, including information such as sampling time, gas component concentration, flow rate, and pressure, and is stored in the form of a database or data file.

[0182] Step S152: Obtain the control effect data after the control command is executed, which includes the combustion chamber state change data, surrounding rock stress change data, and fluid migration change data after the control measures are executed.

[0183] Obtaining control effect data requires comparative analysis of relevant monitoring data before and after the execution of control commands. Combustion chamber state change data is obtained through the inversion results of the integrated geological model, comparing changes in the morphology, volume, and temperature distribution of the combustion chamber before and after control. Surrounding rock stress change data is collected by stress sensors in the mid-field monitoring layer, analyzing the trend of stress value changes. Fluid migration change data is acquired through fluid monitoring equipment, such as pore pressure sensors and gas composition sensors, monitoring changes in fluid pressure and composition. Control effect data includes information such as the execution time of control measures, the amount and rate of change of various monitoring indicators, used to evaluate the actual effectiveness of the control measures.

[0184] Step S153: Obtain the early warning signal, control command and inversion data set, correlate and compare the produced gas data and control effect data with the corresponding physical quantities in the early warning signal, control command and inversion data set, and generate multi-data comparison analysis results.

[0185] Multi-data comparison analysis involves correlating data from different sources to compare their consistency and differences. First, based on timestamps, produced gas data, control effect data, and corresponding physical quantities in early warning signals, control commands, and inversion datasets are matched. Then, the values ​​of the same physical quantity in different datasets are compared and analyzed. For example, the gas composition data obtained from inversion is compared with the composition data of actual produced gas to evaluate the accuracy of the inversion model; the control effect data is compared with the expected effects in the control commands to determine whether the control measures have achieved the expected goals. By calculating statistical indicators such as data deviation and correlation coefficients, the multi-data comparison analysis results are generated.

[0186] Step S154: Verify the accuracy of the warning signal based on the data association results, compare the correspondence between the warning signal and the actual safety risk, adjust the trigger threshold of the warning signal, and output the warning signal deviation processing result.

[0187] The accuracy of early warning signals is verified by comparing the triggering of the early warning signals with the actual safety risks that occur. Frequent false alarms or missed alarms indicate that the trigger threshold for the early warning signals is set unreasonably. Based on data correlation results, the relationship between the monitoring data at the time the early warning signal is triggered and the data at the time the actual safety risk occurs is analyzed to re-determine a reasonable trigger threshold. For example, if the monitoring values ​​when a certain type of risk actually occurs are generally higher than the original trigger threshold, the threshold is appropriately increased; if there are many false alarms, the threshold is appropriately decreased or auxiliary judgment conditions are added. The adjusted trigger threshold is used to update the early warning system, outputting the early warning signal deviation processing results, including a comparison of the thresholds before and after adjustment and the basis for the adjustment.

[0188] Step S155: Verify the validity of the control instructions based on the data association results, compare the correspondence between control measures and control effects, adjust the execution details of control measures, and output the control instruction effect processing results.

[0189] The effectiveness of control commands is verified by analyzing the correlation between the execution of control measures and their effects. If the control effect does not meet expectations, it is necessary to check whether there are problems with the execution details of the control measures, such as whether the parameter adjustment values ​​are reasonable or the execution time is appropriate. Based on the data correlation results, key factors affecting the control effect are identified, and the execution details of the control measures are adjusted. For example, if increasing the injection pressure fails to effectively control cavity expansion, it may be necessary to adjust the injection components or injection rate simultaneously. The adjusted execution details of the control measures are used to optimize the control commands and output the control command effect processing results, including the adjusted content of the control measures and the expected effect.

[0190] Step S156: Based on the results of multi-data comparison analysis, compare the various inversion results in the inversion dataset with the corresponding actual monitoring or production data, analyze the differences, adjust the inversion algorithm or parameters of the target coupled geological model, and output the inversion data difference processing results.

[0191] Data discrepancy processing is a crucial step in improving model predictive capabilities. This involves comparing various inversion results from the dataset, such as combustion chamber morphology, thermal field distribution, and stress field, with corresponding actual monitoring data (e.g., near-field temperature monitoring data, stress monitoring data) or production data (e.g., produced gas data). The error between the inversion results and the actual data is calculated, and the causes of the error are analyzed, potentially including inaccurate model parameters or defects in the inversion algorithm. Based on the analysis results, the inversion algorithm or parameters of the target coupled geological model are adjusted, such as correcting the thermal conductivity coefficient of the rock mass or adjusting the initial value of in-situ stress. Through iterative adjustments, the discrepancy between the inversion results and the actual data is minimized, and the results of the data discrepancy processing are output, including adjusted model parameters and suggestions for improving the inversion algorithm.

[0192] Step S157: Adjust the trigger threshold of the warning signal according to the result of the warning signal deviation processing, optimize the classification standard of the warning signal, and output the optimized warning standard.

[0193] Based on the results of the early warning signal deviation processing, the trigger thresholds for the early warning signals are adjusted to better reflect actual safety risk conditions. Simultaneously, the grading standards for early warning signals are optimized, clearly defining the risk levels and corresponding response measures for different levels of warning signals. For example, the subdivision of warning levels is increased, allowing the warning signals to more accurately reflect the severity of risks. The optimized early warning standards include the triggering conditions, judgment criteria, and response measures for each level of warning signal, and are output as a standardized early warning standard document to guide the operation and management of the early warning system.

[0194] Step S158: Correct the parameters of the target coupled geological model based on the inversion data difference processing results. These parameters include rock mechanics parameters, thermophysical parameters, and fluid parameters. Optimize the model inversion algorithm and output the corrected model parameters.

[0195] Based on the results of the inversion data difference processing, various parameters in the target coupled geological model are corrected. Rock mechanics parameters such as elastic modulus, Poisson's ratio, and internal friction angle are adjusted based on actual stress monitoring data; thermophysical parameters such as thermal conductivity and specific heat capacity are corrected based on temperature monitoring data; and fluid parameters such as permeability and porosity are optimized based on fluid pressure and flow rate data. Simultaneously, the model's inversion algorithm is improved, for example, by adopting a more advanced optimization algorithm or considering more influencing factors. The corrected model parameters and optimized inversion algorithm are used to update the target coupled geological model, improving the model's prediction accuracy and reliability, and outputting the corrected model parameters.

[0196] Step S159: Based on the processing results of the control command effect and the actual production needs, adjust the monitoring network deployment parameters, including the location of the monitoring equipment, the monitoring frequency and the monitoring range, and output the optimized monitoring network parameters.

[0197] If monitoring data in certain areas is insufficient to accurately assess the control effects or if safety risks exist, it is necessary to adjust the location of monitoring equipment and increase the density of monitoring points; adjust the monitoring frequency according to the rate and importance of data changes, and conduct high-frequency monitoring of key parameters; adjust the monitoring range according to the expansion of the combustion chamber and changes in risk areas to ensure that the monitoring network can fully cover key areas. The optimized monitoring network parameters include the coordinates of the equipment deployment locations, sampling frequency settings, and monitoring range boundaries, and the output is a monitoring network optimization plan to guide the upgrading and transformation of the monitoring system.

[0198] Step S1510: Integrate the optimized early warning criteria, the corrected model parameters, and the optimized monitoring network parameters, and output the optimized monitoring network parameters and model parameters.

[0199] The optimized early warning criteria, corrected model parameters, and optimized monitoring network parameters are integrated to form a complete optimization scheme, which includes the deployment parameters of the monitoring network, the parameters of the target-coupled geological model, and the standards of the early warning system. During the integration process, the coordination and consistency among the various parameters are ensured. For example, the optimization of the monitoring network should match the correction of the model parameters, and the adjustment of the early warning criteria should consider changes in the model's predictive capability. The final output is the optimized monitoring network parameters and model parameters.

[0200] Based on the same inventive concept, please refer to Figure 2 The diagram shows a schematic block diagram of a well-ground collaborative geological monitoring system 100 for deep underground coal gasification provided in an embodiment of this application. The well-ground collaborative geological monitoring system 100 for deep underground coal gasification may include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0201] In this embodiment, the machine-readable storage medium 120 can also be integrated into the processor 130 and can communicate and interact with external systems through the communication unit 110. The machine-readable storage medium 120 stores machine-executable instructions for executing the scheme of this application, and the processor 130 executes the machine-executable instructions stored in the machine-readable storage medium 120 to implement the well-to-surface collaborative geological monitoring method for deep coal underground gasification provided in the aforementioned method embodiments.

[0202] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A well-ground collaborative geological monitoring method for deep underground coal gasification, characterized in that, The method includes: Obtain geological exploration data of the target deep coal seam, delineate the layout range and location of near-field monitoring layer, mid-field monitoring layer and far-field monitoring layer based on the geological exploration data of the target deep coal seam, deploy an integrated well-ground three-dimensional monitoring network including near-field monitoring layer, mid-field monitoring layer and far-field monitoring layer, and output the monitoring network layout results; Activate all monitoring equipment in the well-ground integrated three-dimensional monitoring network, collect near-field monitoring data, mid-field monitoring data and far-field monitoring data within the coverage area of ​​the well-ground integrated three-dimensional monitoring network, integrate all monitoring data to form multi-source monitoring data, perform spatiotemporal synchronization processing on the multi-source monitoring data, and output synchronized multi-source monitoring data; Acquire geophysical logging data, core test data, and geostress measurement data of the target deep coal seam; integrate the geophysical logging data, core test data, and geostress measurement data of the target deep coal seam; construct a target coupled geological model; input synchronous multi-source monitoring data into the target coupled geological model for joint inversion processing; and output an inversion data set. The inversion dataset is integrated into the target coupled geological model. The target coupled geological model after integrating the inversion dataset presents the state of the deep underground coal gasification combustion chamber and the surrounding rock response state. Early warning signals are generated based on the presented state. Control measures are generated in combination with the early warning signals. The generated control measures are decomposed into executable control instructions, and early warning signals and control instructions are output. Acquire actual production gas data and regulation effect data, integrate production gas data, regulation effect data with early warning signals, regulation commands and inversion data sets, perform multi-data comparison processing, adjust monitoring network layout parameters and target coupled geological model parameters based on comparison processing results, and output optimized monitoring network parameters and model parameters.

2. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 1, characterized in that, The process involves acquiring geological exploration data of the target deep coal seam, delineating the deployment range and location of near-field, mid-field, and far-field monitoring layers based on this data, deploying an integrated well-to-surface three-dimensional monitoring network encompassing these layers, and outputting the monitoring network deployment results, including: Obtain geological exploration data of the target deep coal seam, including the burial depth distribution, rock strata structure, structural distribution and fracture development status of the target deep coal seam; Based on the geological survey data, the area adjacent to the deep underground coal gasification combustion chamber is calculated and delineated as the deployment range of the near-field monitoring layer, and the deployment range data of the near-field monitoring layer is output. Based on the near-field monitoring layer deployment range data, deep monitoring well deployment parameters corresponding to the near-field monitoring layer are generated. The deep monitoring well deployment parameters include well location coordinates, design depth, and relative positional relationship with the target deep coal seam. Based on the deep monitoring well deployment parameters, a near-field monitoring layer logical configuration scheme is generated for deploying a distributed optical fiber sensing system at the corresponding monitoring location. The near-field monitoring layer logical configuration scheme includes the deployment logic of distributed temperature sensing optical fiber, distributed acoustic sensing optical fiber, and distributed strain sensing optical fiber, the spatial coverage range, and the logical connection relationship with the well-ground integrated three-dimensional monitoring network. Based on the geological exploration data, the areas affected by underground coal gasification heat and stress disturbance, as well as key structural zones, are calculated and delineated as the locations for the mid-field monitoring layer, and the mid-field monitoring layer location data is output. Based on the location data of the mid-field monitoring layer, mid-field monitoring well deployment parameters are generated. Based on the mid-field monitoring well deployment parameters, a mid-field monitoring layer logical configuration scheme for deploying multi-parameter integrated monitoring stations at the corresponding monitoring locations is generated. The mid-field monitoring layer logical configuration scheme determines the monitoring items used by the multi-parameter integrated monitoring stations to collect in-situ three-dimensional stress, pore pressure, temperature gradient, and gas composition, as well as their logical nodes in the well-ground integrated three-dimensional monitoring network. Based on the geological exploration data, calculate and delineate the overlying strata and corresponding surface area of ​​the target deep coal seam as the deployment area of ​​the far-field monitoring layer, and output the data of the far-field monitoring layer deployment area. Based on the data of the far-field monitoring layer deployment area, a logical configuration scheme for the far-field monitoring layer of the surface monitoring equipment and the airborne monitoring equipment is generated. The surface monitoring equipment includes an ultra-dense array broadband seismograph and a high-sensitivity gas flux monitoring station. The airborne monitoring equipment is configured to perform time-series interferometric synthetic aperture radar monitoring. The logical configuration schemes for the near-field monitoring layer, the mid-field monitoring layer, and the far-field monitoring layer are logically integrated to generate a unified monitoring network topology, communication protocol, and data interface specification. This generates the monitoring network deployment result, which includes the virtual structure of the well-ground integrated three-dimensional monitoring network and the logical configuration parameters of each node.

3. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 1, characterized in that, The aforementioned activation of all monitoring equipment in the integrated well-ground monitoring network collects near-field, mid-field, and far-field monitoring data within the network's coverage area. All monitoring data is integrated to form multi-source monitoring data. This multi-source monitoring data undergoes spatiotemporal synchronization processing to output synchronized multi-source monitoring data, including: Obtain the monitoring network deployment results, and output the monitoring equipment positioning results based on all monitoring equipment in the integrated well-ground three-dimensional monitoring network. A start signal is output to the distributed optical fiber sensing system of the near-field monitoring layer. After receiving the start signal, the distributed optical fiber sensing system enters the working state and outputs near-field monitoring data acquisition signals. Temperature data, acoustic data, and strain data of the area covered by the near-field monitoring layer are collected by the distributed optical fiber sensing system of the near-field monitoring layer, and the near-field monitoring data is output after being summarized and integrated. A start signal is output to the multi-parameter integrated monitoring station at the midfield monitoring layer. After receiving the start signal, the multi-parameter integrated monitoring station enters the working state and outputs the midfield monitoring data acquisition signal. In-situ three-dimensional stress data, pore pressure data, temperature gradient data, and gas composition data of the area covered by the mid-field monitoring layer are collected by multi-parameter integrated monitoring stations in the mid-field monitoring layer, and the mid-field monitoring data are output after being summarized and integrated. A start signal is output to the surface monitoring equipment and airborne monitoring equipment of the far-field monitoring layer. After receiving the start signal, the surface monitoring equipment and airborne monitoring equipment enter the working state and output far-field monitoring data acquisition signals. Surface deformation data is collected by the airborne monitoring equipment of the far-field monitoring layer. The data collected by the surface monitoring equipment of the far-field monitoring layer and the airborne monitoring equipment of the far-field monitoring layer are summarized and integrated to output far-field monitoring data. Near-field monitoring data, mid-field monitoring data, and far-field monitoring data are classified, stored, and labeled according to their respective monitoring types, physical quantities, and units to generate structured multi-source monitoring data; A high-precision time-stamping system is used to perform spatiotemporal synchronization processing on multi-source monitoring data, so that monitoring data collected by different monitoring equipment and different monitoring areas have a unified time and space reference, and output synchronized multi-source monitoring data; Synchronous multi-source monitoring data is transmitted to the ground data center via an optical fiber composite cable. During the transmission process, the data is verified and retransmitted, and the completed synchronous multi-source monitoring data is output.

4. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 1, characterized in that, The process involves acquiring geophysical logging data, core test data, and geostress measurement data of the target deep coal seam; integrating these data to construct a target coupled geological model; inputting synchronous multi-source monitoring data into the target coupled geological model for joint inversion processing; and outputting an inversion data set, including: Collect resistivity data, sonic transit time data and density data of the target deep coal seam, and output geophysical well logging data of the target deep coal seam after sorting and archiving. Core samples were drilled from different areas of the target deep coal seam, and their physical and mechanical properties and thermophysical parameters were tested. After the data was recorded, the core test data of the target deep coal seam was output. In-situ geostress measurement method is used to measure the geostress of the target deep coal seam and surrounding rock strata, and three-dimensional geostress data including the maximum principal stress data, minimum principal stress data and intermediate principal stress data are collected, and geostress measurement data of the target deep coal seam are output. Integrate geophysical logging data, core test data, and geostress measurement data of the target deep coal seam, remove abnormal data, and uniformly organize the effective data to output basic geological data; Based on basic geological data, a target coupled geological model is constructed, which includes the initial geological conditions, rock mass structure parameters, thermophysical parameters, and mechanical parameters of the target deep coal seam. After the construction is completed, the target coupled geological model is output. Near-field monitoring data is extracted from synchronous multi-source monitoring data. Acoustic and strain data are selected and converted to adapt to the joint inversion processing requirements. The near-field monitoring data to be inverted is then output. The near-field monitoring data to be inverted is input into the target coupled geological model, and microseismic tomography and moment tensor inversion are used to invert the data to locate the front edge of deep underground coal gasification and combustion and the surrounding rock fracture area, and output the near-field inversion data. Temperature data from near-field monitoring data and temperature gradient data from mid-field monitoring data are extracted from synchronous multi-source monitoring data. These are used as independent input parameter sets and input into the target coupled geological model. The model is then combined with a heat conduction-convection model for inversion processing to output thermal field inversion data. Extract the field monitoring data from the synchronous multi-source monitoring data, and input the in-situ three-dimensional stress data, pore pressure data and gas composition data into the target coupled geological model as mechanical, seepage and chemical field constraints, respectively. Perform joint calibration on the near-field inversion data and thermal field inversion data, and output the calibrated inversion data. Far-field monitoring data is extracted from synchronous multi-source monitoring data. Seismic data and surface deformation data are input into the target coupled geological model. Cross-validation is performed on the calibrated inversion data. Near-field inversion data, thermal field inversion data and calibrated inversion data are integrated to output an inversion data set.

5. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 1, characterized in that, The process involves integrating the inverted data set into a target coupled geological model. The integrated model then presents the state of the deep underground coal gasification combustion chamber and the surrounding rock response. Based on this state, an early warning signal is generated. Combined with the early warning signal, control measures are generated, which are then decomposed into executable control commands. The early warning signal and control commands are then output, including: Obtain the inversion dataset and the target coupled geological model. Integrate all the data in the inversion dataset into the corresponding modules of the target coupled geological model according to their types. After completing the data integration, output the integrated geological model. The integrated geological model presents the three-dimensional morphology, volume, spatial distribution, boundary range and internal structure of the deep underground coal gasification combustion chamber, and outputs the combustion chamber morphology display results. The integrated geological model presents the thermal field distribution, expansion range, thermal power distribution and dynamic change process of the deep underground coal gasification combustion chamber, and outputs the thermal field analysis results of the combustion chamber. The integrated geological model presents the stress field evolution, stress anomaly regions, and stability of the surrounding rock of the deep underground coal gasification combustion chamber, and outputs the stress evolution analysis results of the surrounding rock. The integrated geological model presents the fluid migration path, migration state, dominant fluid channels, and fluid leakage state during deep underground coal gasification, and outputs fluid migration analysis results. Based on the independent state information represented by the combustion chamber morphology display results, combustion chamber thermal field analysis results, surrounding rock stress evolution analysis results, and fluid migration analysis results, the safety risks in the deep underground coal gasification process are classified and identified, and a comprehensive risk assessment result is output. Based on the comprehensive risk assessment results, multi-level early warning signals are generated, and the triggering conditions and corresponding monitoring areas of the multi-level early warning signals are marked. Based on the safety risk type corresponding to the multi-level early warning signals and combined with the parameters of the integrated geological model, targeted control measures are generated. The generated control measures are then further refined and decomposed into control instructions with clear operational targets. Integrate early warning signals and control instructions, classify and organize them according to monitoring areas and risk types, and output early warning signals and control instructions.

6. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 1, characterized in that, The process involves acquiring actual production feedback data on gas output and control effect data, integrating the gas output data, control effect data, early warning signals, control commands, and inversion data sets, performing multi-data comparison processing, adjusting the monitoring network deployment parameters and target-coupled geological model parameters based on the comparison processing results, and outputting optimized monitoring network parameters and model parameters, including: Acquire produced gas data during the deep underground coal gasification process, wherein the produced gas data includes the composition, flow rate and pressure data of the produced gas; Acquire data on the control effect after the execution of control commands, including data on changes in the state of the combustion chamber, changes in surrounding rock stress, and changes in fluid migration after the execution of control measures; Acquire early warning signals, control commands, and inversion data sets; correlate and compare the produced gas data and control effect data with the corresponding physical quantities in the early warning signals, control commands, and inversion data sets to generate multi-data comparison analysis results. The accuracy of the warning signal is verified based on the data correlation results. The correspondence between the warning signal and the actual safety risk is compared. The trigger threshold of the warning signal is adjusted and the result of the warning signal deviation processing is output. The validity of the control instructions is verified based on the data correlation results. The correspondence between the control measures and the control effects is compared, the execution details of the control measures are adjusted, and the processing results of the control instructions are output. Based on the results of multi-data comparison analysis, the various inversion results in the inversion dataset are compared with the corresponding actual monitoring or production data. The differences are analyzed and the inversion algorithm or parameters of the target coupled geological model are adjusted. The results of inversion data difference processing are then output. Adjust the trigger threshold of the warning signal based on the result of the warning signal deviation processing, optimize the classification standard of the warning signal, and output the optimized warning standard. Based on the results of the difference processing of the inversion data, the parameters of the target coupled geological model are corrected, including rock mechanics parameters, thermophysical parameters and fluid parameters. After optimizing the model inversion algorithm, the corrected model parameters are output. Based on the processing results of the control instructions and actual production needs, adjust the monitoring network deployment parameters, including the location of monitoring equipment, monitoring frequency and monitoring range, and output the optimized monitoring network parameters. The optimized early warning criteria, corrected model parameters, and optimized monitoring network parameters are integrated and output as optimized monitoring network parameters and model parameters.

7. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 2, characterized in that, The step of generating a near-field monitoring layer logical configuration scheme for deploying a distributed fiber optic sensing system at the corresponding monitoring location based on the deep monitoring well deployment parameters includes: Based on the deep monitoring well deployment parameters, the sensor system deployment logic parameters of the distributed optical fiber sensing system are calculated and determined. The sensor system deployment logic parameters include the deployment depth range of the sensing optical fiber, the spatial arrangement method, and the theoretical coupling model with the well wall. In the sensor system model library, a distributed optical fiber sensor system logic model that is compatible with the environmental conditions represented by the deployment parameters of the deep monitoring well is matched and selected. The distributed optical fiber sensor system logic model includes the functional module definitions and performance parameters of distributed temperature sensing fiber, distributed acoustic sensing fiber and distributed strain sensing fiber. Based on the distributed optical fiber sensing system logic model and the deep monitoring well deployment parameters, virtual encapsulation processing parameters are generated for the distributed optical fiber sensing system logic model. The virtual encapsulation processing parameters are used to define the thermal insulation and thermal conductivity properties and temperature and high pressure resistance logical properties of the sensing optical fiber. Based on the sensor system deployment logic parameters and the virtual encapsulation processing parameters, a logic flow and spatial coordinates are generated to insert and fix the distributed optical fiber sensor system logic model in the virtual deep monitoring well, and the optical fiber insertion and fixing logic is output. Based on the deployment range data of the near-field monitoring layer, control instructions are generated to adjust the monitoring parameters of the logic model of the distributed optical fiber sensing system. The control instructions are used to set the monitoring angle and spatial sensitivity distribution so that its logical monitoring range covers the deployment range of the near-field monitoring layer, and output the monitoring parameter adjustment logic. A virtual connection channel definition is generated between the logical model of the distributed optical fiber sensing system and the ground data center for data transmission. The virtual connection channel definition includes communication protocol, data flow interface and logical link parameters for simulating temperature and high voltage resistant cables, and outputs data transmission link logic. Based on the fiber optic insertion and fixing logic, monitoring parameter adjustment logic, and data transmission link logic, a system debugging logic is generated to perform joint debugging tests on the virtual sensing transmission system of the near-field monitoring layer. The system debugging logic includes sending virtual test signals and receiving and analyzing feedback data to verify the completeness of the data acquisition and transmission logic. Based on the running results of the system debugging logic, logic instructions are generated to fine-tune the parameters of the virtual sensing and transmission system in order to optimize its logic performance and obtain the optimized virtual sensing and transmission system. Generate a process to start the optimized virtual sensing and transmission system, collect virtual near-field monitoring data, and perform logical verification with preset monitoring data, and output the system test and verification logic; By integrating the deployment logic parameters of the aforementioned sensing system, the selected distributed optical fiber sensing system logic model, encapsulation logic parameters, optical fiber insertion and fixing logic, monitoring parameter adjustment logic, data transmission link logic, system debugging logic, system fine-tuning logic, and system testing and verification logic, a near-field monitoring layer logic configuration scheme is generated, and virtual monitoring points and logical locations are marked.

8. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 3, characterized in that, The data collected by the space-based monitoring equipment in the far-field monitoring layer includes surface deformation data. The data collected by the surface monitoring equipment and the space-based monitoring equipment in the far-field monitoring layer are then aggregated and integrated to output far-field monitoring data, including: Obtain the geographic coordinate data of the far-field monitoring layer deployment area, and plan the flight path and monitoring points of the far-field monitoring layer airborne monitoring equipment according to it, so that the flight path covers the entire far-field monitoring layer deployment area, and output the flight monitoring plan. The airborne monitoring equipment in the far-field monitoring layer is activated, and flight monitoring is carried out according to the flight monitoring plan. The time-series interferometric synthetic aperture radar sensor on board is activated and performs surface imaging, outputting an airborne monitoring activation signal. The time-series interferometric synthetic aperture radar sensor of the airborne monitoring equipment in the far-field monitoring layer continuously images the surface of the area where the far-field monitoring layer is deployed, collects surface image data for multiple time periods, records the imaging time and location, and outputs surface image data. The surface image data is denoised by using image denoising algorithms to eliminate interference signals and noise points, and the denoised surface image data is output after enhancing the clarity. Extract surface deformation features from denoised surface image data, compare data from different time periods to identify deformation areas and states, record the geographic coordinates and morphology of deformation areas, and output surface deformation feature data. Time-series processing of surface deformation characteristic data is performed to track the dynamic changes in deformation, record the amount and rate of deformation changes at different time periods, and output time-series surface deformation data. The time-series surface deformation data is associated and bound with the geographic coordinate data of the far-field monitoring layer deployment area, the specific location and status of each deformation area are marked, and the surface deformation data is output. Acquire seismic and gas flux data collected by surface monitoring equipment in the far-field monitoring layer, classify and organize them, remove invalid data, and output valid surface monitoring data; The surface deformation data and the effective surface monitoring data are associated and stored according to the monitoring time, spatial location and physical quantity type to generate the initial far-field monitoring data; The initial far-field monitoring data is processed to ensure its completeness, by supplementing missing data and correcting erroneous data, so that it fully and accurately reflects the actual state of the far-field monitoring layer and outputs the far-field monitoring data.

9. The well-surface collaborative geological monitoring method for deep underground coal gasification according to claim 4, characterized in that, The temperature data from near-field monitoring data and the temperature gradient data from mid-field monitoring data are extracted from synchronous multi-source monitoring data. These are used as independent input parameter sets into the target coupled geological model, and inverted using a heat conduction-convection model to output thermal field inversion data, including: Acquire synchronous multi-source monitoring data, extract near-field monitoring data and filter out near-field temperature data, which is collected by distributed temperature measurement fiber of the distributed fiber optic sensing system of the near-field monitoring layer. Midfield monitoring data is extracted from synchronous multi-source monitoring data and temperature gradient data is filtered out. This temperature gradient data is collected by multi-parameter integrated monitoring stations in the midfield monitoring layer, and midfield temperature data is output. Near-field temperature data and mid-field temperature gradient data are collected and labeled according to their monitoring points, time, depth and physical quantity type, abnormal data are removed, and the collected temperature field related data are output. Convert the format of the integrated temperature data and output the converted temperature data. Acquire thermophysical parameters from basic geological data, including thermal conductivity, specific heat capacity, and density data of rock strata, and input them into the target coupled geological model to construct a heat conduction-convection model; Input the converted temperature data into the heat conduction-convection model, set its initial boundary conditions and inversion parameters, and start the inversion process to present the thermal field distribution state corresponding to the temperature data and output the initial thermal field inversion data. The thermal field expansion range of the deep underground coal gasification combustion chamber is presented by a heat conduction-convection model. The location of the thermal field boundary and the direction of expansion are defined and relevant data are recorded. The thermal field expansion data is then output. The heat power distribution in time and space is presented by a heat conduction-convection model, and the heat power change status in different monitoring areas and at different times is recorded, and heat power distribution data is output. Integrate the initial thermal field inversion data, thermal field extension data, and thermal power distribution data; compare the correspondence between the initial thermal field inversion data, thermal field extension data, and thermal power distribution data; adjust the inversion deviation; and output the adjusted thermal field inversion data. The adjusted thermal field inversion data is organized and archived, and the corresponding monitoring time period and area are marked to generate complete thermal field inversion data.

10. A well-surface collaborative geological monitoring system for deep underground coal gasification, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the well-to-surface collaborative geological monitoring method for deep underground coal gasification as described in any one of claims 1 to 9 by executing the machine-executable instructions.

Images