In-situ conversion of coal seams by radio frequency heating downhole

Abstract

The application discloses an in-situ conversion method of coal seams by radio frequency heating underground, and particularly relates to the technical field of coal seam pyrolysis conversion, and comprises the following steps: injecting a preset fluid to form a fracture network after drilling, and then injecting a preset first material to construct a filling structure with electromagnetic response characteristics; expanding the borehole and filling a preset second material to form a structure area; arranging a radio frequency heating device and a temperature monitoring device to obtain temperature change data; periodically and dynamically calculating a temperature change gradient coefficient and a temperature stable offset coefficient, generating a control parameter data combination based on the two through reasoning, and dynamically regulating and controlling a heating period and a power output in an intermittent heating mode; the application constructs a multidirectional permeation channel fracture network and a differential electromagnetic response structure, optimizes a radio frequency energy coupling path, improves the heating uniformity and conversion efficiency of coal seams, introduces a dynamic coefficient reasoning mechanism based on temperature data, realizes intelligent regulation and control of the intermittent heating period and power, and enhances the safety and adaptability of the system.

Description

Technical Field

[0001] This invention relates to the field of multi-well site well network configuration ranging technology, and more specifically, to a method for in-situ conversion of coal seams via downhole radio frequency heating. Background Technology

[0002] As deep coal mining evolves towards lower disturbance, higher efficiency, and greener methods, in-situ coal seam conversion technology is gradually becoming an important alternative to traditional mining methods. This technology, by constructing a fracture network within the coal seam and applying external energy to trigger a pyrolysis reaction, enables the direct underground conversion of coal into gaseous or liquid products, showing great promise in unconventional energy development. Among these technologies, radio frequency heating, due to its non-contact, remote control, and coal seam penetration capabilities, has become one of the important energy forms for in-situ coal seam heating.

[0003] In practical applications, the propagation of radio frequency electromagnetic waves within coal seams is often affected by microstructural inhomogeneities, resulting in non-uniform coupling effects. This leads to abnormal energy accumulation in specific areas, forming high-temperature hotspots. Such phenomena frequently occur in areas of coal seams containing natural interlayers, localized mineral aggregates, or micro-regions with high dielectric constants. This "micro-regional anomalous absorption" causes local temperatures to far exceed those of the surrounding areas, leading to premature coking of the coal, fracture closure, and even sudden coal and rock bursts due to thermal stress concentration. This severely damages the established fracture network and blocks product flow paths. In more severe cases, the high temperature may also conduct along fractures to the near-wellbore area, damaging radio frequency heating devices and related equipment, causing system shutdowns or wellbore structural failures, resulting in irreversible engineering losses. Furthermore, traditional heating systems often employ constant power or periodic start-stop strategies, lacking the ability to dynamically sense and control the internal temperature response of the coal seam. Conventional temperature monitoring devices struggle to detect these "low-probability, high-hazard" local anomalies in a timely manner, and the overall system lacks robustness and predictive capability for the complex thermal evolution of coal seams. Therefore, this invention proposes an in-situ conversion method for coal seams via underground radio frequency heating in order to solve the above-mentioned problems. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A method for in-situ conversion of coal seams via downhole radio frequency heating includes the following steps:

[0006] After drilling in the target coal seam area, a pre-set fluid is injected into the wellbore to form a fracture network with multi-directional permeability channels inside the coal seam.

[0007] A predetermined first material is injected into the crack network to form a filling structure with predetermined electromagnetic response characteristics.

[0008] The aperture is enlarged in the near-wellbore area of ​​the wellbore and filled with a pre-set second material to form a structural region with different electromagnetic response characteristics;

[0009] The radio frequency heating device and the temperature monitoring device are arranged along the shaft to the deep part of the coal body, so that the radio frequency heating device is positioned in the area where the fracture network intersects, and the temperature monitoring device is set in multiple spatial locations to collect temperature change data at different points.

[0010] During the heating process, the time series data collected by the temperature monitoring device is periodically analyzed. Based on the analysis results, the temperature change gradient coefficient and the temperature stability offset coefficient are dynamically calculated. Using these two coefficients as input, a combination of control parameter data is generated through reasoning. This combination of control parameter data is used to dynamically adjust the heating cycle and power output in the intermittent heating mode of the radio frequency heating device until the coal seam temperature rises to the preset pyrolysis temperature range.

[0011] During the aforementioned heating step, the fluid material generated inside the coal mass is caused to migrate along the fracture network toward the shaft and then guided through the shaft to the surface treatment area.

[0012] In a preferred embodiment, the step of injecting a preset fluid into the wellbore includes the following process:

[0013] After the wellbore is formed, water is injected into the coal body according to the first injection pressure to induce initial fracturing of the coal body in the natural weak surface or bedding direction, forming a fracturing channel.

[0014] The preset fluid is injected at the second injection pressure, and the crack is driven to expand at a uniform preset discharge flow rate. The second injection pressure is greater than the first injection pressure.

[0015] By continuously injecting a preset fluid, the fractures eventually form a multi-directional distribution structure in the coal body and connect to the wellbore, thus constructing a fracture network with continuity and permeability.

[0016] In a preferred embodiment, completing wellbore formation means: after setting the drilling path on the ground, drilling through the target coal seam area using vertical or horizontal drilling methods, and finally completing wellbore formation.

[0017] In a preferred embodiment, injecting the preset first material includes the following process:

[0018] Select a preset first material combination formulation that matches the radio frequency, namely a high dielectric loss additive. This formulation has stable dielectric loss characteristics. The preset first material is continuously injected into the crack network according to a preset injection rate, so that the material penetrates along the crack path to a predetermined depth. The crack space is filled by a preset injection duration and injection volume. Finally, after filling is completed, a preset pressure difference is maintained to promote the preset first material to form a tightly fitted filling structure with the crack wall.

[0019] In a preferred embodiment, when filling with a pre-set second material, a mechanical reaming tool is used to enlarge the original wellbore diameter to a pre-set diameter in a defined coal seam wellbore area, and loose debris generated during the reaming process is removed to ensure the compactness of the subsequent filling process. The pre-set second material with low dielectric loss characteristics is continuously injected into the annular space formed by the reaming process, maintaining a stable injection pressure during the filling process. After filling is completed, a pre-set curing time is allowed. After the pre-set second material has cured, a structural region with different electromagnetic response characteristics is formed, so that the pre-set second material forms a stable capping layer around the wellbore and constructs a structural region with significantly different electromagnetic response characteristics from the fracture network region, thereby realizing the redistribution and control of energy coupling paths in the near-wellbore area.

[0020] In a preferred embodiment, the temperature monitoring device is set up by the following process:

[0021] With the radio frequency heating device as the center, multiple temperature monitoring devices are set at intervals along the vertical direction inside the wellbore;

[0022] Temperature monitoring devices are simultaneously installed in the radially adjacent coal seam area of ​​the radio frequency heating device;

[0023] A local spatial temperature monitoring network is constructed using the above vertical and radial arrangement to obtain real-time temperature change information of the coal body at different points, thereby supporting the dynamic control of intermittent heating.

[0024] In a preferred embodiment, the calculation of the temperature change gradient coefficient includes the following steps:

[0025] Based on multiple vertical and radial temperature monitoring devices arranged around the radio frequency heating device, temperature data of each monitoring point at multiple consecutive time nodes are collected to form a three-dimensional temperature dataset with spatial orientation and time series correlation.

[0026] The monitoring points are divided into multiple spatial path groups according to the radial and vertical paths. Within each path group, the temperature values ​​corresponding to three adjacent time nodes are extracted. By calculating the difference between two consecutive temperature increments, the temperature change rate of the path within that time period is obtained.

[0027] The temperature change rate of all paths is normalized and weighted according to the spatial direction of the path. Then, the weighted average change rate of all paths is calculated. This weighted average change rate is the temperature change gradient coefficient, which is used to characterize the non-uniformity of temperature response in the current heating area.

[0028] In a preferred embodiment, the calculation of the temperature stability offset coefficient includes the following steps:

[0029] Multiple vertical and radial monitoring devices around the radio frequency heating device collect temperature data over a continuous period of time, and a sliding window is constructed with a fixed time length;

[0030] At each monitoring point, the difference between the maximum and minimum temperatures within the sliding window is calculated as the local temperature fluctuation value. At the same time, the average value of all temperatures within the sliding window is calculated, and the absolute difference between the average value and the median value of the preset pyrolysis temperature range is taken as the temperature offset value of that point.

[0031] Spatial weights are set according to the physical distance of each monitoring device from the center of the radio frequency heating device. The temperature fluctuation values ​​and temperature offset values ​​of all monitoring points are weighted and summed to construct a temperature stability offset coefficient, which is used to represent the degree of uniformity and stability trend of the current temperature distribution in the heating area.

[0032] In a preferred embodiment, the control parameter data combination is generated as follows:

[0033] A fuzzy inference model is constructed. This model takes the temperature change gradient coefficient and the temperature stability offset coefficient as input variables and uses a fuzzy inference mechanism based on the membership function to establish the mapping relationship between the input and output, which is used to output the combination of heating cycle parameters and power output parameters.

[0034] The temperature change gradient coefficient and the temperature stability offset coefficient are defined as input dimensions, and are divided into three level intervals (low, medium, and high) by a three-segment membership function. Each input variable corresponds to three fuzzy linguistic terms, and the input space forms nine combinations of input state rules. The fuzzy inference model adopts an inference engine based on a fuzzy rule table to map each combination state to the interval level to which the heating cycle parameter belongs. At the same time, based on the position of the fuzzy aggregation center point of the output layer and the statistical regularity of the temperature response data in the historical heating samples, the corresponding power output parameters are generated, forming a combination of control parameter data.

[0035] The technical effects and advantages of this invention are as follows:

[0036] This invention involves injecting a pre-defined fluid into the wellbore after drilling into the target coal seam area. This induces the formation of a fracture network with multi-directional permeable channels within the coal seam. This structure not only achieves structural disturbance and pre-construction of flow channels within the coal seam but also enhances the coal seam's response to thermal energy and fluids through the three-dimensional expansion of the fracture network. The multi-directional permeable channels can cover heterogeneous areas of the coal seam, enabling uniform diffusion of subsequent radio frequency energy transmission over a wider range. This significantly reduces the "hot spot effect" caused by local energy enrichment, providing a structural foundation for the smooth migration of pyrolysis products and the safe operation of equipment. Simultaneously, it improves the spatial continuity of coal seam conversion and the consistency of kinetic energy response.

[0037] This invention involves injecting a pre-set first material into the fracture network and then filling it with a pre-set second material after enlarging the pore size in the near-wellbore area. This creates composite structures with pre-set electromagnetic response characteristics and different electromagnetic response characteristics in the fracture region and around the wellbore, respectively. This construction strategy effectively establishes a gradient absorption zone and a shielding zone for high-frequency energy, guiding radio frequency energy to focus at the intersection of the fracture network along the energy propagation path. This improves the spatial utilization efficiency of the radio frequency heating device's output energy and avoids localized material carbonization or equipment heat loss in the near-wellbore area due to high energy density. Simultaneously, the two-layer structure with different electromagnetic response characteristics exhibits synergistic control under radio frequency action, contributing to the construction of a controlled and stable heating environment and significantly enhancing the reliability of coal seam thermal conversion.

[0038] This invention periodically analyzes time-series data collected by a temperature monitoring device during radio frequency (RF) heating, dynamically calculates the temperature change gradient coefficient and temperature stability offset coefficient, and uses these two coefficients as inputs to generate a combination of control parameter data through inference, thereby dynamically regulating the heating cycle and power output in the intermittent heating mode. This method introduces a control mechanism based on real-time data feedback, enabling closed-loop regulation of the coal seam heating process. It adapts to the non-uniform temperature rise characteristics under different geological structures, dynamically corrects the heating strategy, and prevents fracture structure degradation or thermal stress damage caused by continuous heating. Compared to traditional fixed-parameter heating methods, this technology improves the response sensitivity and energy efficiency of the RF heating device, ensuring that the heating process is both efficient and safe, and contributing to long-term stable pyrolysis conversion operations in deep coal seams. Attached Figure Description

[0039] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0040] Figure 1 This is a schematic diagram of an in-situ conversion method for underground radio frequency heating coal seams according to the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0042] Reference Figure 1 The following examples were obtained:

[0043] Example 1: A method for in-situ conversion of coal seams using downhole radio frequency heating, comprising the following steps: After drilling in the target coal seam area, a preset fluid is injected into the wellbore to form a fracture network with multi-directional permeability channels inside the coal seam; This step aims to form a fracture network by injecting fluid, thereby creating interconnected fractures in multiple directions in the originally dense coal seam structure, constructing multi-directional permeability channels with fluid conductivity, and providing a basic path for subsequent material injection and pyrolysis product flow; This fracture network relies on controlling the injection pressure and drainage parameters to dynamically induce the fracturing of natural weak surfaces and the propagation of fractures, thereby establishing structural continuity and permeability connectivity inside the coal seam.

[0044] A predetermined first material is injected into the crack network to form a filling structure with predetermined electromagnetic response characteristics. This step involves injecting the predetermined first material into the constructed crack network to fill the cracks and form a stable structure. This structure has electromagnetic response characteristics that match the radio frequency field, which can enhance the energy absorption capacity of a specific area during heating. This helps to improve the coupling efficiency of radio frequency energy and the uniformity of the thermal field distribution, and avoid overheating or energy waste in non-target areas.

[0045] In the near-wellbore region of the wellbore, the aperture is enlarged and filled with a pre-set second material to form a structural region with different electromagnetic response characteristics. This step involves enlarging the aperture in the near-wellbore region by mechanical enlargement and injecting a pre-set second material with low dielectric loss characteristics into the enlarged region to form a structural region with significantly different electromagnetic characteristics from the fracture network region. This structure is used to regulate the propagation mode of radio frequency energy in the near-wellbore region, realize the spatial directional transmission of energy, reduce the risk of thermal stress concentration around the wellbore, and improve the energy utilization efficiency in the far-end fracture network.

[0046] The radio frequency (RF) heating device and temperature monitoring device are deployed along the shaft to the depth of the coal seam. The RF heating device is positioned at the intersection of the fracture network, and the temperature monitoring device is set up in multiple spatial locations to collect temperature change data at different points. This step ensures that the energy is concentrated in the most interconnected part of the coal structure by precisely deploying the RF heating device to the intersection of the fracture network, thereby improving heating efficiency. At the same time, by deploying temperature monitoring devices in multiple vertical and radial spatial locations, a temperature acquisition network covering key areas is constructed, providing multi-dimensional, all-time thermal field data support for subsequent dynamic regulation, and realizing precise monitoring and closed-loop control.

[0047] During the heating process, the time-series data collected by the temperature monitoring device is periodically analyzed. Based on the analysis results, the temperature change gradient coefficient and the temperature stability deviation coefficient are dynamically calculated. Using these two coefficients as inputs, a combination of control parameter data is generated through inference. This combination of control parameter data is used to dynamically adjust the heating cycle and power output in the intermittent heating mode of the radio frequency heating device until the coal seam temperature rises to the preset pyrolysis temperature range. This step continuously analyzes the time-series data obtained by the temperature monitoring device to dynamically evaluate the uniformity and fluctuation of the temperature distribution during the coal heating process, and calculates the temperature change gradient coefficient and the temperature stability deviation coefficient respectively, forming a dual-indicator input reflecting heat conduction efficiency and stability. Subsequently, the control parameter data combination is output through an inference mechanism, thereby precisely adjusting the heating cycle and power output to achieve closed-loop control in the intermittent heating mode, so that the coal temperature gradually and stably rises to the target pyrolysis temperature range, effectively suppressing local overheating or energy runaway problems.

[0048] During the aforementioned heating step, the fluid substances generated within the coal mass migrate along the fracture network towards the wellbore and are then guided to the surface processing area through the wellbore. This step utilizes the previously constructed fracture network and wellbore connection structure. After the heating induces a pyrolysis reaction in the coal mass and generates oil and gas-like fluid substances, these substances are guided in an orderly manner into the wellbore through thermal pressure difference and permeation channels. Finally, they are transported to the surface through the wellbore, achieving efficient output of the product stream and separation and processing at the surface, ensuring the integrity and controllability of the conversion process.

[0049] In a specific implementation process, the step of injecting a preset fluid into the wellbore includes the following steps: After the wellbore is formed, a wellbore structure penetrating the target coal body area is obtained according to the preset drilling path and wellbore diameter specifications. Then, clean water is injected into the coal body as the initial fluid according to the first injection pressure. Under high pressure, the water body preferentially propagates along the natural weak surfaces or bedding planes inside the coal body, inducing micro-fractures in the coal body in these structurally weak areas, thereby forming a fracture initiation channel in the area near the wellbore. This process generally maintains the injection pressure in the range of 15 MPa to 20 MPa and the duration is controlled within 3 to 5 minutes to ensure that the fractures can be started in a controlled manner without causing severe damage.

[0050] After the fracturing channel is formed, the process immediately switches to injecting a preset fluid at a second injection pressure. The injection pressure is increased to over 25 MPa and maintained above the aforementioned first injection pressure to enhance the driving force for further fracture propagation. A uniform preset discharge flow rate is used for stable propagation, controlled within the range of 100 to 150 liters per minute, ensuring uniform expansion during fracture extension. Taking a deep, low-permeability coal seam (1000m depth, 5m thickness, 0.1mD permeability) as an example, the specific implementation of this invention is illustrated: Hydraulic fracturing parameters: a preset fluid, i.e., water-based fracturing fluid, is used; the construction pressure is 30 MPa; after fracturing, four main vertical fractures are formed (each fracture has a length Lf = 50m, a width Hf = 0.1m, and a horizontal distance Wf = 10m from the wellbore). During the expansion process, the injection state is maintained continuously, allowing the water-based fracturing fluid to further scour the unevenly distributed joint structures and weak interface areas within the coal body along the fracturing initiation path. This induces the fracture structure to extend in different spatial directions, gradually forming a multi-directional distribution structure. The injection time at this stage is generally controlled between five and fifteen minutes, but can be extended according to actual conditions, specifically adjusted based on the coal body's structure and acoustic monitoring results. The pre-set fluid is then injected into the coal body to promote the gradual expansion of the multi-directional fracture structure to the extent of communication with the wellbore structure. Throughout the process, the mechanical continuity and spatial connectivity of the fractures are maintained, ultimately constructing a fracture network covering the target coal body area with stable fluid permeability. This provides an effective structural channel and heat conduction path for subsequent injection of functional materials and radio frequency heating processes.

[0051] In the implementation of this invention, wellbore formation refers to: after setting the drilling path on the ground, penetrating the target coal seam area using vertical or horizontal drilling methods, and finally completing wellbore formation. Specifically, the operators first pre-set the wellbore trajectory and incident angle on the ground based on the coal seam's burial depth, thickness variation, structural stability, and geological profile data, combined with the layout requirements of the radio frequency heating device, and determine a suitable drilling method. When the geological structure is relatively simple or the coal seam is shallow, vertical drilling is preferred to quickly penetrate the thick coal seam. In application scenarios where the coal seam dip angle is small, the ductility is strong, or the heating area needs to be expanded, horizontal drilling is used to extend along the coal seam strata, so that the wellbore maintains long-distance contact with the coal body, improving subsequent heating efficiency. During the drilling process, existing auxiliary technologies such as mud wall protection, sonic logging, and trajectory navigation are combined to ensure that the drill bit accurately penetrates the target coal seam area and maintains the wellbore diameter to meet the engineering conditions for subsequent fluid injection and device layout, ultimately completing a continuous, through-hole wellbore formation operation with structural integrity.

[0052] In the specific implementation of this invention, a predetermined first material is injected into the crack network to form a filling structure with predetermined electromagnetic response characteristics. The injection of the predetermined first material includes the following process: First, based on the target operating frequency of the radio frequency heating device, a predetermined first material combination formula matching the radio frequency, namely a high dielectric loss additive, is selected. By detecting its dielectric loss tangent, dielectric constant stability, and thermal stability, it is confirmed that the formula has stable dielectric loss characteristics. In this embodiment, carbon nanotube composite resin (dielectric loss tangent of 0.8, matching a 20 MHz radio frequency) is selected as the predetermined first material, enabling it to generate significant electromagnetic energy absorption capability under radio frequency action.

[0053] After material selection, the preset first material is continuously injected into the fracture network at a preset injection rate. The injection rate is stabilized at 0.5 cubic meters per hour by the wellbore flow control system, so that the material can continuously penetrate to the predetermined depth along the fracture path under the action of the pressure gradient. At the same time, real-time pressure monitoring confirms that no reverse blockage or backflow occurs during the penetration process. In this embodiment, the total injection volume is controlled at 10 cubic meters to ensure that the injection volume is proportional to the fracture volume.

[0054] During the injection process, the crack space is filled by pre-setting the injection duration and injection volume. The uniformity of material distribution inside the crack can be confirmed by existing acoustic imaging technology, so that the filling rate reaches or exceeds 90%, thereby ensuring the formation of a stable electromagnetic response channel structure inside the crack. This step not only ensures that the pre-set first material forms a continuous coverage in the crack wall area, but also lays the foundation for the uniform distribution of radio frequency energy in the crack area.

[0055] After filling is completed, a preset pressure difference is maintained to cause the preset first material to form a tightly fitted filling structure with the crack wall. This pressure difference is usually maintained in the range of one to three MPa, so that the material is shaped inside the crack and the later shrinkage rate is reduced, while maintaining the structural adhesion between it and the wall. This results in the crack area eventually forming a stable filling structure with preset electromagnetic response characteristics, providing efficient energy coupling conditions for the subsequent radio frequency heating stage.

[0056] In the specific implementation of this invention, when filling with the pre-set second material, a mechanical reaming tool is used in the determined coal shaft area to enlarge the original shaft diameter to the pre-set diameter, and loose debris generated during the reaming process is removed to ensure the compactness of the subsequent filling process. First, after drilling is completed, a coal shaft of conventional diameter is obtained, which is usually less than 24.1 cm. Then, a mechanical reaming device is used to enlarge the shaft of the target layer to a diameter of 40 cm, forming a ring-shaped expansion area near the well. During the reaming process, a rotating reamer and a high-pressure flushing device are used simultaneously to ensure the integrity of the coal wall and the absence of collapse risk. At the same time, physical scrubbing and repeated flushing with high-flow-rate clean water are used to remove loose debris generated during the reaming process, providing a clean contact interface for material injection.

[0057] After the borehole is enlarged and cleaned, a pre-designed second material with low dielectric loss characteristics is continuously injected into the annular space formed by the borehole enlargement. A composite material system is formed by mixing phenolic resin and quartz sand in a 1:1 mass ratio. This system has fluidity and a controllable reaction rate before curing, and can fill the near-well annular area within one meter. The injection process adopts a continuous pumping method, and the injection rate is maintained between 50 and 80 liters per minute. A pressure sensing system is used to ensure that no cavity residue occurs during the filling process.

[0058] During the injection process, a stable injection pressure is maintained, controlled between 0.6 MPa and 1 MPa, to ensure that the pre-set second material fully penetrates and is evenly distributed in the structure surrounding the wellbore, and to prevent material backflow or leakage caused by excessive pressure differential. After injection, a pre-set curing time is allowed, ranging from three to eight hours, dynamically adjusted according to the temperature inside the wellbore and the material's reaction characteristics, to ensure that the material system completes full cross-linking and structural consolidation. Finally, after the pre-set second material has cured, a structural region with different electromagnetic response characteristics is formed, i.e., a low dielectric loss capping layer is constructed around the wellbore, thereby achieving a significant difference in electromagnetic response characteristics between the region and the fracture network region in space. The existence of this structure enables the redistribution and regulation of energy coupling paths in the near-wellbore area, effectively reducing redundant dissipation of electromagnetic energy in the near-wellbore area, guiding more radio frequency energy to accumulate in the deeper coal body, improving heating efficiency and reducing the risk of thermal damage to the wellbore.

[0059] In the implementation of this invention, the temperature monitoring device is set up in the following manner. First, in order to accurately sense the temperature changes of coal at different depths within the working area of ​​the radio frequency heating device, so as to provide a basis for dynamic control of radio frequency heating, multiple temperature monitoring devices are set up at intervals along the vertical direction inside the shaft, with the radio frequency heating device as the center. These devices can be thermocouple arrays or fiber optic temperature sensor arrays, preferably with a vertical spacing of 0.5 meters to 1 meter. Each group has four to six monitoring nodes, and the installation depth covers the upper and lower ranges of the heating area, so that the temperature distribution differences in each vertical section can be continuously collected and recorded. To compensate for the problem that relying solely on monitoring inside the shaft may ignore the uneven heating of the coal in the lateral direction, temperature monitoring devices are simultaneously set up in the radially adjacent area of ​​the radio frequency heating device to the coal. These devices use encapsulated high-temperature sensing elements that can be inserted into the coal. Multiple temperature points are arranged radially on both sides of the radio frequency heating device, preferably at a distance of 0.3 meters to 0.8 meters from the center of the heating device, forming a typical cross-sectional monitoring layer to obtain information on the lateral temperature rise, heat diffusion rate, and potential hot spot areas of the coal, thereby improving the accuracy of temperature field spatial modeling.

[0060] By employing the aforementioned coordinated vertical and radial arrangement, a local spatial temperature monitoring network is constructed to achieve dynamic data acquisition of the temperature field in three-dimensional space. In practice, all temperature monitoring devices are connected to a ground-based data processing system, with an acquisition cycle set to once every five to ten seconds. Data is uploaded in real-time to the control platform for subsequent temperature gradient calculations, stability analysis, and input for inference algorithms. Finally, this local spatial temperature monitoring network can operate in conjunction with the radio frequency heating control system during intermittent heating to acquire real-time temperature change information of the coal at different points, supporting dynamic control of intermittent heating. The system automatically identifies the temperature rise response and thermal decay characteristics of multiple temperature monitoring points at different stages, thereby determining whether there is hot spot aggregation, uneven temperature rise, or localized cooling lag during the heating cycle. This provides continuous feedback for updating the control parameter data combination, ensuring that the heating process remains within a controllable range.

[0061] In another embodiment, the arrangement of temperature monitoring devices synchronously installed radially adjacent to the coal seam area of ​​the radio frequency heating device adopts a curved controllable tubular column, driving the sensor head to be directionally inserted into the crack network in a radially opening manner to achieve directional expansion deployment; alternatively, multiple temperature sensing elements can be pre-embedded in perforated sleeves, and the network can be completed before radio frequency activation. This method is suitable for areas with irregular coal seam cracks or complex original stress fields, ensuring that a high-resolution three-dimensional temperature sensing scene can still be constructed in complex coal seam structures.

[0062] In the implementation of this invention, the calculation of the temperature change gradient coefficient includes the following steps: To accurately reflect the spatial differences in the temperature rise pattern of the coal body during radio frequency heating, it is necessary to fully utilize the temperature data of each measuring point in the monitoring network to construct a multi-dimensional data foundation. Therefore, based on multiple vertical and radial temperature monitoring devices arranged around the radio frequency heating device, temperature data of each monitoring point at multiple consecutive time nodes are collected to form a three-dimensional temperature dataset with spatial direction and time series correlation. In specific implementation, the temperature acquisition cycle can be set to five seconds, and temperature values ​​can be continuously acquired in the initial, middle and steady-state stages of radio frequency heating, so that the three-dimensional temperature dataset includes not only the vertical temperature change trajectory but also the radial temperature diffusion process. For example, six vertical monitoring points are arranged within a one-meter range, and four monitoring points are arranged at radial positions of 0.5 meters and 0.8 meters from the center of the radio frequency heating device, so that the three-dimensional temperature dataset meets the requirements for dynamic analysis in both spatial scale and temporal resolution.

[0063] After obtaining the three-dimensional temperature dataset, it is necessary to path-code the temperature changes in different spatial directions so that the temperature change trend is reflected not only in single-point data but also in the gradient along the spatial path. To this end, the monitoring points are divided into multiple spatial path groups according to radial and vertical paths. Within each path group, the temperature values ​​corresponding to three adjacent time nodes are extracted. By calculating the difference between two consecutive temperature increments, the rate of temperature change along that path within that time period is obtained. In specific implementation, vertical monitoring points can be combined at fixed intervals to form multiple vertical path groups, and radial monitoring points can be distributed according to their varying distances from the center of the radio frequency heating device to form multiple radial path groups. For example, the length of each path group can be set to three measurement points, forming a three-segment temperature sequence corresponding to three time nodes. Extracting the temperature values ​​of three adjacent time nodes, for example, the temperature values ​​obtained at 10 seconds, 15 seconds, and 20 seconds are 100 degrees Celsius, 120 degrees Celsius, and 140 degrees Celsius respectively, results in a difference of 20 degrees Celsius between two consecutive temperature increments. This difference further reflects the acceleration or stagnation of the temperature change trend. This not only reflects the absolute temperature increase along the path but also the changing characteristics of the temperature rise rate.

[0064] After calculating the temperature change rate of different paths over various time periods, it is necessary to perform a standardized correlation processing on the data of each path so that data from different directions, regions, and paths can be compared under the same evaluation system during the overall analysis. Therefore, the temperature change rates of all paths are normalized and weighted according to the spatial direction of the path. During normalization, the ratio of the temperature change rate of each path to the highest temperature change rate among all paths can be used to ensure the result is within the range of zero to one, allowing for quantitative analysis of the data at different heating stages. The weights are set based on the path's contribution to spatial heat conduction; for example, a weight of 0.4 can be used for vertical paths, and a weight of 0.6 can be used for radial paths to reflect the main role of radial temperature diffusion during radio frequency heating. At this point, the weighted temperature change rate of each path can be used to characterize the influence of different directions on temperature uniformity.

[0065] Finally, the normalized and weighted temperature change rate data are integrated, and the weighted average change rate of all paths is calculated. This weighted average change rate is the temperature change gradient coefficient, used to characterize the non-uniformity of the temperature response within the current heating area. In practical implementation, if the temperature change gradient coefficient is close to or below 0.2, it indicates that the temperature change is relatively uniform; if it is greater than 0.5, it indicates that the temperature change rate in a local area deviates significantly from the overall structure, and hot spot trends may appear. In such cases, dynamic control of the heating cycle and power output of the radio frequency heating device is required through a combination of control parameter data to avoid excessively rapid local temperature rise leading to crack structure damage or material thermal instability. This invention achieves precise quantification of the spatial temperature distribution state through this method, making the radio frequency heating process perceptible and controllable, and providing a reliable basis for closed-loop dynamic adjustment of intermittent heating modes.

[0066] In the implementation of this invention, to achieve a quantitative evaluation of the stability and uniformity of temperature distribution during coal heating, the calculation of the temperature stability deviation coefficient includes the following steps: To construct a data analysis foundation with temporal continuity, representative original temperature sequences need to be obtained from the temperature monitoring network. Specifically, multiple vertical and radial monitoring devices around the radio frequency heating device collect temperature data over a continuous time period, and a sliding window is constructed with a fixed time length. In a specific implementation, temperature monitoring devices can be arranged above, below, and on both sides of the radio frequency heating device as the center, for example, eight vertical points and six radial points are arranged to form a spatial temperature acquisition network; the temperature data acquisition cycle is preferably once every five seconds, and the sliding window time length is set to thirty seconds, forming a window sequence consisting of six consecutive data points for each monitoring point. The sliding window slides according to the time sequence, with one set of sampled data sliding each time, thereby forming a time-space data set covering the entire heating stage, which has dynamic update capability and supports subsequent time-sensitive calculation operations.

[0067] After constructing the sliding window, it is necessary to quantify and extract the local temperature fluctuations and absolute deviations within the window. Therefore, at each monitoring point, the difference between the maximum and minimum temperatures within the sliding window is calculated as the local temperature fluctuation value. Simultaneously, the average value of all temperatures within the sliding window is calculated, and the absolute difference between this average and the median of the preset pyrolysis temperature range is taken as the temperature offset value for that point. For example, if the six temperature values ​​of a radial monitoring point within the current sliding window are 320°C, 330°C, 332°C, 335°C, 331°C, and 329°C, then the maximum value is 335°C, the minimum value is 320°C, and the difference is 15°C, which is recorded as the local temperature fluctuation value for that point. Meanwhile, the average value of the six temperature values ​​is 329.5°C. If the preset pyrolysis temperature range set for this operation is 300 to 600°C, then the median is 450°C. The temperature offset value for that point is the absolute difference of 120.5°C. These indicators comprehensively reflect the intensity of transient temperature fluctuations at the monitoring point and their proximity to the target pyrolysis conditions.

[0068] After calculating the local temperature fluctuation and temperature offset values ​​at all monitoring points, it is necessary to further integrate spatial layout factors to construct a global evaluation factor with physical constraints. Specifically, spatial weights are assigned based on the physical distance of each monitoring device from the center of the radio frequency heating device, and the temperature fluctuation and temperature offset values ​​of all monitoring points are weighted and summed. In the specific weight allocation, a method of using the reciprocal of the distance plus a piecewise function of the distance normalization can be used: for example, monitoring points within 0.5 meters of the center are assigned a weight of 0.45, those between 0.5 and 1 meter are assigned a weight of 0.35, and those greater than 1 meter are assigned a weight of 0.2. The temperature fluctuation and temperature offset values ​​of each monitoring point are multiplied by their corresponding spatial weights and then linearly summed. This weighting strategy reflects a greater sensitivity to near-field (near the heating source) temperature fluctuations and also reflects the importance hierarchy of different locations in temperature control, enhancing the responsiveness and indicative nature of the coefficient calculation.

[0069] After completing all weighting operations, a temperature stability offset coefficient is constructed to represent the uniformity and stability trend of the temperature distribution in the current heating area. This temperature stability offset coefficient is essentially a weighted comprehensive thermal stability index; the smaller the value, the lower the temperature fluctuation within the heating area, the smaller the deviation from the target temperature, and the more stable the system's thermal distribution. If the coefficient shows a continuous upward trend, it indicates that the system is in an unstable state, requiring correction by adjusting the power output or heating cycle settings in the control parameter data combination. In an actual heating experiment, when the coefficient remained above 20 degrees Celsius for more than 120 seconds, the system automatically switched to a low-power intermittent heating mode and extended the cooling cycle, significantly alleviating the local overheating problem. Therefore, this calculation method in this invention not only possesses accurate modeling capabilities but can also form a closed-loop thermal field regulation loop with the control system, exhibiting high engineering application value and inference response characteristics.

[0070] In the implementation of this invention, to achieve adaptive control of energy release behavior during radio frequency heating and ensure that the temperature rise rate and distribution state during coal seam heating are within a controllable range, the generation method for the control parameter data combination is as follows: a fuzzy inference model is constructed. This model receives the temperature change gradient coefficient and temperature stability offset coefficient as input variables, and uses a fuzzy inference mechanism based on membership functions to establish a mapping relationship between input and output, which is used to output the combination of heating cycle parameters and power output parameters. After the coal seam fracture network is formed, to avoid local overheating or decreased energy utilization efficiency, a multi-factor perception-based inference control module needs to be introduced. The temperature change gradient coefficient and temperature stability offset coefficient collected and calculated from the temperature monitoring device are used as input variables to establish a fuzzy logic rule model to output control parameters. In implementation, this model uses fuzzy language to describe nonlinear change trends, abstracting continuous real variables into fuzzy intervals, and is compatible with the fluctuations and dynamics present in the temperature response. The inference model is deployed in the above-ground intelligent control unit and can communicate with the underground temperature monitoring device via wired or wireless means to achieve closed-loop control.

[0071] To ensure the inference model has clear and controllable logical boundaries under complex thermal environments, the temperature change gradient coefficient and temperature stability offset coefficient are defined as input dimensions, and divided into three level intervals (low, medium, and high) using a three-segment membership function. Each input variable corresponds to three fuzzy linguistic terms, forming nine combinations of input state rules in the input space. In specific implementation, the low interval of the temperature change gradient coefficient can be defined as 0 to 0.3, the medium interval as 0.3 to 0.6, and the high interval as above 0.6; the three level intervals of the temperature stability offset coefficient are set in a similar manner. Each interval is modeled using a triangular or trapezoidal membership function, and the fuzzy linguistic terms are defined as "low stability fluctuation," "medium change trend," "high intensity deviation," or similar terms, thereby constructing control semantics in the inference system using natural language logic. The nine combined states constructed in this way, including "low temperature difference - low offset," "medium temperature difference - high offset," and "high temperature difference - medium offset," can reflect typical thermal response states during the radio frequency heating process.

[0072] After constructing the input language rules, to ensure the operability of the control commands, the fuzzy inference model employs an inference engine based on a fuzzy rule table. This engine maps each combined state to the interval level to which the heating cycle parameter belongs. Simultaneously, based on the statistical regularity of the output layer's fuzzy aggregation center point location and the temperature response data in historical heating samples, corresponding power output parameters are generated. The inference engine uses either the maximum membership principle or a weighted average aggregation mechanism. In practical implementation, the heating cycle parameters can be categorized into three types: high-frequency short cycle, standard cycle, and low-frequency long cycle, used for rapid heating start-up, stable operation, and overheat correction phases, respectively. For example, when the temperature change gradient coefficient is medium and the temperature stability offset coefficient is high, the heating cycle parameter output by the inference model is a low-frequency long cycle, and the power output parameter should automatically decrease by 20% to prevent further thermal instability. This power output parameter is mapped based on the temperature rise rate corresponding to similar input states in historical heating samples, through statistical aggregation center point distances. For example, a target limit is set at a local heating rate not exceeding five degrees Celsius per minute to ensure that the output power does not induce the formation of local hot spots.

[0073] In the fuzzy inference model, after the input temperature change gradient coefficient and temperature stability offset coefficient are fuzzified and matched according to rules, a fuzzy membership distribution is obtained in the output layer (i.e., some heating power levels are partially activated). To derive a specific power output value from these fuzzy results, the system calculates the "aggregation center point" of all membership distributions—that is, the numerical result of the weighted average of the output values ​​of multiple fuzzy levels. Next, the system compares this center point result with a large amount of sample data accumulated during historical heating processes, finds the temperature response closest to this center point in these samples, and selects a statistically optimal or safe and stable power value as the power output parameter for the current heating operation based on the temperature rise feedback of the coal body to the power output in these samples. This process is equivalent to "landing" the fuzzy results obtained in real time into a specific heating power, ensuring that this power is a value verified in historical experience that can bring a good thermal response, thereby achieving intelligent, safe, and steady-state heating control.

[0074] The output heating cycle parameters and power output parameters together constitute the control parameter data combination, which drives the radio frequency heating device into the corresponding intermittent heating working mode. Within each control cycle, the coal body thermal response is fed back to the temperature monitoring device, recalculating the updated temperature change gradient coefficient and temperature stability offset coefficient, forming a new inference process and achieving continuous dynamic control of the heating behavior. This method significantly improves coal body heating efficiency, avoids the energy waste and fracture structure degradation problems of traditional continuous heating, and can adaptively identify local abnormal temperature rise states, preventing them in advance through power reduction and cycle adjustment. This provides a strong guarantee for the engineering adaptability and safety of this invention in complex coal seam geological environments.

[0075] It should be noted that during intermittent radio frequency heating, the coal seam needs to be gradually heated to the pyrolysis temperature range (typically 300 to 600 degrees Celsius). Therefore, this invention employs a double-layer structure design for the wellbore and fracture network before heating. The pre-selected first material filling the fracture network is a thermally stable filler with high dielectric loss characteristics, such as carbon nanotube-reinforced composite resin. This material can maintain dielectric response performance for a long time at high temperatures and possesses good adhesion and thermal expansion compatibility, effectively adhering to the fracture wall and preventing melting or structural collapse within the pyrolysis temperature range. This ensures the deep coupling efficiency of electromagnetic energy and maintains the continuity of seepage through the fracture channels. Furthermore, the first material is injected using a volume-controlled + differential pressure retention mode, forming a dense and stable filling structure with thermal stress buffering capacity, significantly reducing the risk of coking closure.

[0076] To prevent excessive energy concentration in the near-wellbore area during heating, which could lead to overheating, this invention proposes injecting a pre-defined second material with low dielectric loss characteristics into the near-wellbore area. A mixture of phenolic resin and quartz particles is preferred, as it exhibits thermal stability exceeding 600 degrees Celsius and does not significantly absorb radio frequency energy, forming an equivalent "energy shielding band." This structure forms a dense covering layer in the borehole expansion area, effectively guiding radio frequency energy distribution to the deeper coal seam and preventing overheating in the near-wellbore area. It also maintains structural integrity under high-temperature conditions, does not participate in the coal seam pyrolysis reaction, and ensures a stable operating environment for the radio frequency device. Therefore, although the heating zone eventually reaches the pyrolysis temperature range, the structure formed by the first and second materials possesses a clear design logic for thermal stability and structural reliability in its material selection and arrangement strategy, enabling it to collaboratively support the safe and efficient operation of the radio frequency heating system throughout the entire process.

[0077] The above algorithms or formulas are all dimensionless and numerical calculations, and the results are obtained by software simulation based on a large amount of collected data to obtain the most recent real-world results. The preset parameters are set by those skilled in the art according to the actual situation.

[0078] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0079] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0080] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0081] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for in-situ conversion of coal seams via underground radio frequency heating, characterized in that, Includes the following steps: After drilling in the target coal seam area, a pre-set fluid is injected into the wellbore to form a fracture network with multi-directional permeability channels inside the coal seam. A predetermined first material is injected into the fracture network to form a filling structure with predetermined electromagnetic response characteristics; the aperture is enlarged and a predetermined second material is filled in the near-wellbore area of ​​the wellbore to form a structural region with different electromagnetic response characteristics. The radio frequency heating device and the temperature monitoring device are arranged along the shaft to the deep part of the coal body, so that the radio frequency heating device is positioned in the area where the fracture network intersects, and the temperature monitoring device is set in multiple spatial locations to collect temperature change data at different points. During the heating process, the time series data collected by the temperature monitoring device is periodically analyzed. Based on the analysis results, the temperature change gradient coefficient and the temperature stability offset coefficient are dynamically calculated. Using these two coefficients as input, a combination of control parameter data is generated through reasoning. This combination of control parameter data is used to dynamically adjust the heating cycle and power output in the intermittent heating mode of the radio frequency heating device. The calculation of the temperature gradient coefficient includes the following steps: Based on multiple vertical and radial temperature monitoring devices arranged around the radio frequency heating device, temperature data of each monitoring point at multiple consecutive time nodes are collected to form a three-dimensional temperature dataset with spatial orientation and time series correlation. The monitoring points are divided into multiple spatial path groups according to the radial and vertical paths. Within each path group, the temperature values ​​corresponding to three adjacent time nodes are extracted. By calculating the difference between two consecutive temperature increments, the temperature change rate of the path within that time period is obtained. The temperature change rate of all paths is normalized and weighted according to the spatial direction of the path. Then, the weighted average change rate of all paths is calculated, which is the temperature change gradient coefficient. The calculation of the temperature stability offset coefficient includes the following steps: Multiple vertical and radial monitoring devices around the radio frequency heating device collect temperature data over a continuous period of time, and a sliding window is constructed with a fixed time length; At each monitoring point, the difference between the maximum and minimum temperatures within the sliding window is calculated as the local temperature fluctuation value. At the same time, the average value of all temperatures within the sliding window is calculated, and the absolute difference between the average value and the median value of the preset pyrolysis temperature range is taken as the temperature offset value of that point. Spatial weights are set according to the physical distance of each monitoring device from the center of the radio frequency heating device. The temperature fluctuation values ​​and temperature offset values ​​of all monitoring points are weighted and added together to construct a temperature stability offset coefficient.

2. The method for in-situ conversion of coal seams via downhole radio frequency heating according to claim 1, characterized in that, The steps for injecting a pre-set fluid into the wellbore include the following: After the wellbore is formed, water is injected into the coal body according to the first injection pressure to induce initial fracturing of the coal body in the natural weak surface or bedding direction, forming a fracturing channel. The preset fluid is injected at the second injection pressure, and the crack is propelled to expand with a uniform preset discharge flow rate. The second injection pressure is greater than the first injection pressure.

3. The method for in-situ conversion of coal seams via downhole radio frequency heating according to claim 2, characterized in that, The completion of wellbore formation refers to: after setting the drilling path on the ground, drilling through the target coal seam area using vertical or horizontal drilling methods.

4. The method for in-situ conversion of coal seams via downhole radio frequency heating according to claim 3, characterized in that, The injection of the pre-set first material includes the following process: The preset first material is continuously injected into the crack network at a preset injection rate, allowing the material to penetrate along the crack path to a predetermined depth. The crack space is filled by the preset injection duration and injection volume. Finally, the preset pressure difference is maintained after the filling is completed.

5. The method for in-situ conversion of coal seams via downhole radio frequency heating according to claim 4, characterized in that, When filling with the pre-set second material, the original wellbore diameter is enlarged to the pre-set diameter in the determined coal body wellbore area. The pre-set second material is then filled into the annular space formed by the enlarged hole through continuous injection. During the filling process, a stable injection pressure is maintained. After the filling is completed, a pre-set curing time is allowed. After the pre-set second material has cured, a structural region with different electromagnetic response characteristics is formed.

6. The method for in-situ conversion of coal seams via downhole radio frequency heating according to claim 5, characterized in that, The setup process for a temperature monitoring device includes the following steps: With the radio frequency heating device as the center, multiple temperature monitoring devices are set at intervals along the vertical direction inside the wellbore; Temperature monitoring devices are simultaneously installed in the radially adjacent coal seam area of ​​the radio frequency heating device; A local spatial temperature monitoring network is constructed using the above vertical and radial arrangement to obtain real-time information on temperature changes in the coal body at different locations.

7. The method for in-situ conversion of coal seams via downhole radio frequency heating according to claim 6, characterized in that, The method for generating the control parameter data combination is as follows: A fuzzy inference model is constructed. This model takes the temperature change gradient coefficient and the temperature stability offset coefficient as input variables and uses a fuzzy inference mechanism based on the membership function to establish the mapping relationship between the input and output, which is used to output the combination of heating cycle parameters and power output parameters. The temperature change gradient coefficient and the temperature stability offset coefficient are defined as input dimensions, and are divided into three level intervals (low, medium, and high) by a three-segment membership function. Each input variable corresponds to three fuzzy linguistic terms, and the input space forms nine combinations of input state rules. The fuzzy inference model adopts an inference engine based on a fuzzy rule table to map each combination state to the interval level to which the heating cycle parameter belongs. At the same time, based on the position of the fuzzy aggregation center point of the output layer and the statistical regularity of the temperature response data in the historical heating samples, the corresponding power output parameters are generated, forming a combination of control parameter data.

Images