Method and system for predicting drilling gas emission quantity in dynamic drilling process

Abstract

The invention discloses a method and system for predicting the drilling gas emission amount in the dynamic drilling process, and relates to the technical field of rock stratum drilling. Constructing a gas migration model, and deploying the gas migration model in a mine supervision platform in an embedded manner; uploading a first drilling parameter to a well drilling supervision platform, calculating a first damage circle range in a gas migration model, performing range detection through data interaction with a geological radar, determining a geological detection signal, performing three-thread parallel prediction and superposition in combination with vibration parameters from a vibration sensor, and determining gas prediction data; and performing dynamic drilling process control. The technical problems that in the prior art, gas emission prediction precision is low, and the coupling effect of coal damage and multi-mechanism gas release under dynamic drilling disturbance cannot be reflected are solved, and the technical effects that the real-time performance and accuracy of gas emission amount prediction in the dynamic drilling process are improved, and effective risk pre-control in the drilling process is achieved are achieved.

Description

Technical Field

[0001] This invention relates to the field of rock drilling technology, specifically to a method and system for predicting borehole gas emission during dynamic drilling. Background Technology

[0002] During underground drilling operations in coal mines, gas migration is influenced by multiple mechanisms, including drill bit disturbance, coal body damage propagation, pore structure evolution, and instantaneous desorption of coal cuttings. This coupled behavior exhibits significant time-varying and nonlinear characteristics. Traditional prediction methods often rely on steady-state or single diffusion models, making it difficult to characterize the impact of dynamic factors such as drilling speed, vibration, and changes in the damage zone on the gas release process. Furthermore, they fail to adequately reflect the short-term, rapid outburst characteristics caused by the evolution of the coal body's dual-pore structure and coal cuttings desorption, resulting in significant discrepancies between predicted and actual outburst volumes. This makes it difficult to meet the needs of rapid early warning and refined safety management underground. Summary of the Invention

[0003] This application provides a method and system for predicting borehole gas emission during dynamic drilling, which addresses the technical problems of low gas emission prediction accuracy and inability to reflect the coupling effect of coal body damage and multi-mechanism gas release under dynamic drilling disturbance in the prior art.

[0004] In view of the above problems, this application provides a method and system for predicting borehole gas emission during dynamic drilling process.

[0005] A first aspect of this application provides a method for predicting borehole gas emission during a dynamic drilling process, the method comprising:

[0006] Vibration sensors and ground-penetrating radar are installed on the front-end execution components of the drilling rig. A gas migration model is constructed based on the initial calculation of the damage zone range and parallel predictions based on dual-hole structure evolution, coal dust analysis, and vibration impact pulse terms. This model is embedded in the mine monitoring platform. First drilling parameters are uploaded to the drilling monitoring platform. The first damage zone range is calculated in the gas migration model. Range detection is performed through data interaction with the ground-penetrating radar to determine the geological detection signal. Three-thread parallel prediction and superposition are combined with vibration parameters from the vibration sensor to determine the gas prediction data. Dynamic drilling process control is then implemented based on the gas prediction data.

[0007] A second aspect of this application provides a dynamic drilling process borehole gas emission prediction system, the system comprising:

[0008] The assembly module is used to assemble vibration sensors and ground-penetrating radar at the front end of the drilling rig; the model building module is used to construct a gas migration model based on the first calculation of the damage zone range and parallel prediction based on the evolution of the dual-hole structure, coal cuttings analysis, and vibration impact pulse terms, and is embedded in the mine supervision platform; the calculation module is used to upload the first drilling parameters to the drilling supervision platform, calculate the first damage zone range in the gas migration model, perform range detection through data interaction with the ground-penetrating radar, determine the geological detection signal, and combine the vibration parameters from the vibration sensor to perform three-thread parallel prediction and superposition to determine the gas prediction data; the control module is used to perform dynamic drilling process control based on the gas prediction data.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application equips the front-end execution components of the drilling rig with vibration sensors and ground-penetrating radar; it constructs a gas migration model based on the first calculation of the damage zone range and parallel prediction based on the evolution of the dual-hole structure, coal cuttings analysis, and vibration impact pulse terms, and embeds it in the mine supervision platform; it uploads the first drilling parameters to the drilling supervision platform, calculates the first damage zone range in the gas migration model, performs range detection through data interaction with the ground-penetrating radar to determine the geological detection signal, and combines the vibration parameters from the vibration sensor to perform three-thread parallel prediction and superposition to determine the gas prediction data; based on the gas prediction data, it performs dynamic drilling process control. This invention solves the technical problems of low gas emission prediction accuracy and inability to reflect the coupling effect of coal body damage and multi-mechanism gas release under dynamic drilling disturbance in the prior art. By constructing a multi-threaded gas migration model that integrates damage zone calculation, dual-hole structure evolution, coal cuttings desorption, and vibration pulse effects, it improves the real-time performance and accuracy of gas emission prediction during dynamic drilling, and achieves the technical effect of effective risk pre-control during the drilling process. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic flowchart of a method for predicting borehole gas emission during dynamic drilling provided in this application embodiment;

[0013] Figure 2 This is a schematic diagram of a dynamic drilling process borehole gas emission prediction system provided in an embodiment of this application.

[0014] Figure labeling: Assembly module 11, Model building module 12, Calculation module 13, Control module 14. Detailed Implementation

[0015] This application provides a method and system for predicting borehole gas emission during dynamic drilling. It addresses the technical problems of low accuracy in gas emission prediction and the inability to reflect the coupling effect of coal body damage and multi-mechanism gas release under dynamic drilling disturbances in existing technologies. By constructing a multi-threaded gas migration model that integrates damage circle calculation, dual-hole structure evolution, coal dust desorption, and the influence of vibration pulses, the method improves the real-time performance and accuracy of gas emission prediction during dynamic drilling, achieving effective risk pre-control during the drilling process.

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0017] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0018] Example 1, as Figure 1 As shown, this application provides a method for predicting borehole gas emission during dynamic drilling, the method comprising:

[0019] Step S100: Install vibration sensors and ground-penetrating radar on the front-end execution components of the drilling rig.

[0020] In this embodiment of the application, a vibration sensor and a ground-penetrating radar are mounted on the front-end execution component of the drilling rig. By fixing the vibration sensor to the force-bearing end face of the execution component, the axial vibration signal and the rotary impact signal are collected in real time during the drilling process. At the same time, the ground-penetrating radar is embedded in the detection position of the execution component, so that it can continuously transmit and receive electromagnetic waves when the drill bit advances, in order to obtain the geological detection signal of the coal body ahead.

[0021] Step S200: Based on the first calculation of the damage zone range, and the parallel prediction based on the evolution of the dual-pore structure, coal dust analysis and vibration impact pulse term, a gas migration model is constructed and embedded in the mine supervision platform.

[0022] In this embodiment, the damage circle formed by drilling disturbance is first calculated to obtain the damage circle range based on drilling speed, drilling pressure, and coal compressive strength. Then, a univariate prediction thread based on the evolution of the dual-pore structure is introduced to analyze the relationship between macropore porosity changes and micropore connectivity evolution regarding gas migration. Subsequently, a binary prediction thread based on coal dust analysis is combined to measure the unsteady diffusion of coal dust through instantaneous desorption rate and dynamic surface area changes.

[0023] Simultaneously, a ternary prediction thread based on vibration and shock pulse terms is used to take the pulsed gas release caused by axial vibration and rotational impact as transient disturbance inputs.

[0024] By running and synchronizing the above three types of prediction threads in parallel, a gas migration model is formed.

[0025] Finally, the gas migration model was deployed in an embedded manner on the mine monitoring platform, enabling the gas migration model to access drilling parameters and vibration parameters in real time during the drilling process, thereby achieving continuous prediction and dynamic response of gas outburst.

[0026] Furthermore, the method provided in the application embodiments, in constructing the gas migration model, further includes:

[0027] A first calculation node is deployed within the drilling damage zone; a second prediction node is deployed for multi-threaded gas outburst prediction, wherein the second prediction node includes a univariate prediction thread based on the evolution of a dual-pore structure, a binary prediction thread based on coal dust desorption, and a ternary prediction thread based on pulsed vibration and impact; the gas migration model is deployed based on the first calculation node and the second prediction node.

[0028] In this embodiment, the first calculation node is deployed based on the drilling damage zone range. That is, based on the damage expansion characteristics formed by the stress disturbance of the coal body during the drill bit advance, the spatial scale of the damage zone is quantified by a damage zone radius calculation method that includes parameters such as drilling speed, drilling pressure and coal compressive strength, so as to form the basic boundary conditions for solving the gas migration problem.

[0029] Next, a second prediction node is deployed using a multi-threaded gas outburst prediction method. In this process, three prediction threads based on different mechanisms are set up within the second prediction node to represent the gas release process in separate fields. Specifically, a univariate prediction thread based on the evolution of the dual-pore structure characterizes the evolution of the dual-pore system through the coupling relationship between macropore permeability and micropore effective diffusion coefficient, reflecting the adjustment of gas transport capacity caused by changes in macropore porosity and micropore connectivity with damage degree and fracture expansion. A binary prediction thread based on coal dust desorption analyzes the particle size distribution and dynamic surface area changes of coal dust using an unsteady diffusion model, quantitatively expressing the contribution of instantaneous desorption rate and cumulative mass to coal dust desorption, characterizing the rapid gas release process caused by coal dust precipitation after the drill bit breaks the coal body. Simultaneously, a ternary prediction thread based on pulsed gas release under vibration and impact introduces axial vibration and rotational impact signals. By constructing the correspondence between vibration energy and fracture aperture fluctuations, the transient gas release caused by vibration and impact is treated as a pulsed disturbance term to describe the pulsed gas release peak.

[0030] After the construction of the first calculation node and the second prediction node is completed, the gas migration model is formed by jointly configuring the boundary conditions of the damage circle generated by the first calculation node and the gas migration solution results of the three types of prediction threads in the second prediction node.

[0031] Furthermore, the method provided in the application embodiments also includes:

[0032] Define a dynamic drilling damage zone range, wherein the damage zone range is calculated based on the damage zone radius, and the damage zone radius is calculated as follows: ,in, The first damage propagation coefficient, For drilling speed, The second damage propagation coefficient, For drilling pressure, It refers to the compressive strength of the coal body.

[0033] In this embodiment of the application, the drilling speed v(t) is first calculated by utilizing the relationship between displacement and time changes formed during the drilling process, wherein the drilling speed is obtained by the displacement difference and the corresponding time difference between adjacent moments.

[0034] Subsequently, the drilling pressure P(t) generated during the drilling process is obtained. This drilling pressure originates from the force exerted on the drill bit in the axial direction, and a drilling pressure curve that varies with the drilling process is formed by recording the time series.

[0035] After obtaining the drilling speed v(t) and drilling pressure P(t), the compressive strength f(c) of the coal seam is used as a parameter characterizing the compressive strength of the coal body, and is obtained experimentally. Based on the above three parameters, the drilling speed term and the drilling pressure term are combined in a linear superposition manner using the method for calculating the damage zone radius. The first damage propagation coefficient, This is the second damage propagation coefficient, which is preset by technical experts.

[0036] Furthermore, in the method provided in the application embodiments, the second prediction node includes a unary prediction thread based on the evolution of a dual-pore structure, and further includes:

[0037] A first coupling relationship is defined, wherein the first coupling relationship is defined by the gas migration relationship under the coupling of macropore permeability and micropore effective diffusion coefficient based on the evolution of the dual-pore structure. The evolution law of the dual-pore structure is that macropore porosity is positively correlated with the degree of damage, and micropore connectivity is positively correlated with the degree of fracture propagation. By mining the first coupling relationship, the Zhang's permeability evolution and magnitude prediction under dual-pore gas migration are used as the underlying logic to construct the unary prediction thread, wherein fracture propagation includes drilled fractures and natural fractures.

[0038] In this embodiment, when defining the first coupling relationship, the evolution of the dual-pore structure is used as the overall basis, and the changes of the macropore structure and micropore structure during drilling disturbance are quantitatively characterized according to a unified logic. Specifically, to determine the changes in the macropore structure, the geological exploration signal is first processed to correspond to the undisturbed window and the disturbed window. Through normalization of reflection amplitude and energy attenuation, background subtraction, and attenuation region extraction, the open range of the dominant channel is identified, and this range is determined as the macropore porosity. Simultaneously, by thresholding the boundary continuity, morphological characteristics, and spatial extension range of the fracture zone, the extensional changes of the fracture zone form a quantitative expression of the damage degree, thereby establishing a dual-pore structure evolution law in which macropore porosity is positively correlated with the damage degree.

[0039] In determining the changes in micropore structure, high-frequency energy components are extracted from geological exploration signals and subjected to time-frequency transformation to identify stable intervals of high-frequency attenuation and phase shift. The amplitude of these changes is defined as micropore connectivity, which describes the diffusion conditions of the internal pore structure. Simultaneously, by performing boundary tracing and geometric quantification of the extrapolated range at the fracture zone boundary, the expansion of drilled and natural fractures is combined and expressed as fracture propagation, thus establishing a positive correlation between micropore connectivity and fracture propagation.

[0040] After obtaining macropore porosity, damage degree, micropore connectivity, and fracture propagation, a joint transport expression of macropore permeability and micropore effective diffusion coefficient is constructed using a flux superposition method. Macropore permeability, determined by macropore porosity, represents the seepage conditions of the dominant channels; micropore effective diffusion coefficient, determined by micropore connectivity, represents the diffusion conditions of the pore network. These two are combined within the same spatiotemporal framework, allowing gas transport to be described by the combined mechanism of macropore permeability and micropore effective diffusion coefficient. This forms the first coupling relationship of gas transport based on the macropore permeability-micropore effective diffusion coefficient coupling in the evolution of a dual-pore structure.

[0041] After establishing the first coupling relationship, Zhang's permeability evolution method is introduced. The degree of damage is used as the control quantity for the change of macropore permeability over time, and the micropore connectivity is used as the control quantity for the change of micropore effective diffusion coefficient over time. This allows the two types of seepage conditions and diffusion conditions to advance over time during the drilling disturbance process, thereby forming Zhang's permeability evolution and magnitude prediction under dual-pore gas migration, providing a continuous magnitude change basis for dynamic gas migration calculation.

[0042] After the above structure is determined, the first coupling relationship is used as the core calculation logic. The degree of damage and the degree of fracture propagation are used as inputs. The joint change of macropore permeability and micropore effective diffusion coefficient is used as the control factor of gas migration law. Combined with the magnitude expression obtained by Zhang's permeability evolution, the gas migration trend and its magnitude are generated, forming a univariate prediction thread. The fracture propagation includes drilled fractures and natural fractures, so that the thread can fully reflect the gas migration process under the participation of the dual-pore structure.

[0043] Furthermore, in the method provided in the application embodiments, the second prediction node includes a binary prediction thread based on coal dust desorption, and further includes:

[0044] Determine the unsteady-state diffusion relationship, which includes a first diffusion relationship and a second diffusion relationship. The first diffusion relationship characterizes the instantaneous desorption rate of coal dust with different particle sizes, and the second diffusion relationship characterizes the time window resolution rate based on the cumulative mass of coal dust, time-varying particle size distribution, and dynamic surface area. The mean value obtained from the solutions of the first diffusion relationship and the second diffusion relationship is used as the coal dust resolution metric, and the binary prediction thread is trained under supervision.

[0045] In this embodiment, to determine the unsteady-state diffusion relationship, the drilled coal cuttings are first divided into different particle size ranges, such as 0.5~1mm, 1~2mm, and 2~4mm, using a particle size classification and sieving method. This allows each particle size to form a data set that can participate in diffusion calculations independently. Simultaneously, pressure-time and mass-time records are acquired under constant temperature and pressure conditions, and a time window with a fixed duration is set for segmented analysis of desorption behavior in subsequent diffusion calculations. This yields the particle size-classified coal cuttings set, pressure-time records, mass-time records, and the defined time window.

[0046] To determine the first diffusion relationship, the pressure decay analysis method and the mass difference conversion algorithm were used to process the desorption data of coal dust with different particle sizes. Taking a certain particle size (e.g., 1~2 mm) as an example, equally spaced time points were extracted from the pressure-time and mass-time records. Difference was performed on adjacent time points to calculate the amount of gas released per unit time, and this release amount was used as the instantaneous desorption rate at that moment. This step was repeated for the same particle size to obtain an instantaneous desorption rate sequence. The same processing procedure was performed sequentially for all particle sizes to obtain a set of instantaneous desorption rates indexed by particle size, i.e., the first diffusion relationship.

[0047] To obtain the second diffusion relationship, a time window aggregation method and a dynamic surface area calculation method were employed. Within each time window, the cumulative mass of coal dust was first incrementally statistically analyzed. Then, the time-varying particle size distribution for that time window was updated based on particle size analysis results. Subsequently, the dynamic surface area of ​​that time window was calculated using the equivalent particle size and shape factor for each particle size. Once all three quantities were obtained within the same time window, the cumulative mass of coal dust, the time-varying particle size distribution, and the dynamic surface area were jointly calculated. This involved dividing the cumulative mass of coal dust by the dynamic surface area, and then dividing by the corresponding time window length. This ratio reflected the diffusion rate per unit surface area within that time window, thus yielding the time window resolution rate. Processing all time windows sequentially yielded a complete sequence composed of time window resolution rates, i.e., the second diffusion relationship.

[0048] To generate quantitative metrics for supervised training, after obtaining the first and second diffusion relationships, a window mean method is applied to both. Within the same time window, the instantaneous desorption rate and the time window resolution rate are aligned, and their arithmetic mean is calculated. This mean is defined as the coal dust resolution metric within that time window. By performing the same calculation on all time windows, a sequence of coal dust resolution metrics arranged by time window is obtained.

[0049] Finally, the binary prediction thread was trained under supervision. In this process, regression training combined with cross-validation was employed, using the analytical measure of coal cuttings as the supervisory variable. Observable inputs related to the drilling process were introduced, such as changes in cuttings discharge, vibration parameters, and drilling pressure. Training and validation sets were constructed, and parameter learning was performed, gradually enabling the binary prediction thread to form a mapping structure from observable drilling inputs to the magnitude of coal cuttings diffusion. After training, the binary prediction thread was obtained.

[0050] Furthermore, in the method provided in the application embodiments, the second prediction node includes a ternary prediction thread based on pulse-type vibration and shock, and further includes:

[0051] Define the vibration axial relationship, which characterizes the vector transmission of vibration energy, crack opening fluctuation, and transient outburst; use axial vibration and rotational impact as input data, gas transient outburst based on the vibration axial relationship as the prediction target, and pulsed gas release peak as the output to supervise the training of a ternary prediction thread.

[0052] In this embodiment, to define the axial relationship of vibration, the vibration response of the drill bit in the drilling direction is first continuously collected by a vibration sensor installed on the front-end execution component of the drilling rig, resulting in a vibration sequence along the drilling direction as the axial vibration type. The pulsating impacts during the rotation of the drill bit are collected to obtain a vibration sequence with impact characteristics as the rotational impact type. Simultaneously, a ground-penetrating radar installed on the front-end execution component continuously probes the coal seam in front of the drill bit, acquiring geological detection signals formed by reflected echoes. Furthermore, a wellhead gas monitoring device records the change in gas emissions over time, forming a transient outburst sequence characterizing the disturbance release behavior.

[0053] After obtaining the above data, DC removal, bandpass filtering, and envelope extraction were performed on axial vibration and rotational impact signals, respectively. The envelope signal was then integrated within a fixed time window to form a quantitatively comparable energy index for the vibration signal, yielding vibration energy for disturbance analysis. Subsequently, the geological survey signal underwent fracture zone boundary extraction. The boundary evolution sequence over time was obtained using a connectivity tracking method, and joint differencing was performed on boundary position changes and reflection amplitude changes to form a measurable time series of instantaneous fracture opening and closing, resulting in fracture aperture fluctuations to describe the structural response. Noise reduction and baseline correction were performed on the transient outburst sequence. Locally rapid peak segments were identified using derivative thresholding and peak detection methods, and the peak amplitude and occurrence time were extracted, making these peaks constitute pulsed gas release peaks. A continuous sequence was retained as the quantitative basis for transient outbursts.

[0054] After generating three sequences—vibration energy, crack aperture fluctuation, and transient outburst—the temporal correlation and trend of these three sequences were calculated within a unified time window. To determine the response relationship between vibration energy and crack aperture fluctuation, Pearson correlation coefficients were calculated for the time window change sequences of both to assess their linear dependence. Then, cross-correlation functions were calculated, and the peak position was used to determine the lead time lag of vibration energy relative to crack aperture fluctuation. Similarly, the same correlation coefficient calculation and cross-correlation function analysis were performed on crack aperture fluctuation and transient outburst to determine the lead time lag of crack response relative to transient outburst. This time lag matching establishes a causal chain in time sequence, from vibration input to structural response and then to gas release.

[0055] Subsequently, by constructing incremental vectors for three types of quantities within the same time window, the change in vibration energy is used as the input vector, the change in crack aperture fluctuation is used as the intermediate vector, and the change in transient outburst is used as the output vector. Given a temporal relationship, the transmission of these three vectors in terms of directional consistency and magnitude response forms a vector transmission of vibration energy – crack aperture fluctuation – transient outburst, which is defined as the vibration axial relationship.

[0056] After establishing the vibration axial relationship, axial vibration and rotational impact are used as inputs to substitute into the relationship. The transient gas outburst based on the vibration axial relationship is obtained through the transfer calculation within the time window, which serves as the prediction target for supervised training. The pulsed gas release peak is used as the training output to establish a learnable correspondence between the disturbance input, structural response and peak release.

[0057] Finally, by constructing axial vibration, rotational impact, transient gas outburst based on vibration axial relationship, and pulsed gas release peak as training samples, the supervised training of the ternary prediction thread is completed by regression training combined with cross-validation, so that the thread can infer the intensity change of transient gas outburst based on vibration disturbance conditions and output the corresponding pulsed gas release peak.

[0058] Step S300: Upload the first drilling parameters to the drilling monitoring platform, calculate the range of the first damage zone in the gas migration model, conduct range detection through data interaction with the ground-penetrating radar, determine the geological detection signal, and combine the vibration parameters from the vibration sensor to perform three-thread parallel prediction and superposition to determine the gas prediction data.

[0059] In this embodiment, firstly, first drilling parameters are uploaded to the drilling monitoring platform. These parameters are collected and transmitted in a fixed format, including drilling pressure, torque, drilling speed, cuttings discharge volume, and rotational speed, and are used as inputs to the gas migration model as drilling conditions. Subsequently, disturbance calculations are performed in the gas migration model based on the first drilling parameters to obtain the range of the first damage zone.

[0060] After obtaining the range of the first damage zone, it is used to define the analysis area of ​​the ground-penetrating radar and is interactively processed with the echo data of the ground-penetrating radar. By identifying and comparing the reflection amplitude sequence, phase difference sequence and interface change sequence within the defined area, structural change information is extracted and a geological exploration signal characterizing the disturbance structure state of the coal body is formed.

[0061] After the geological exploration signal is generated, it is input into the parallel prediction structure along with the vibration parameters collected by the vibration sensor. The disturbance magnitude is obtained by filtering and envelope processing of the vibration parameters and then input into three prediction threads. The three threads independently calculate the corresponding gas magnitude based on the seepage mechanism, diffusion mechanism and transient response mechanism in the model, respectively, to form seepage prediction results, diffusion prediction results and transient release prediction results.

[0062] Finally, the output results of the three threads are superimposed within the same time window. By merging the seepage flow rate, diffusion rate, and transient release rate, gas prediction data characterizing the intensity of gas release is formed.

[0063] Furthermore, in the method provided in the application embodiments, before calculating the range of the first damage circle in the gas migration model, the method further includes:

[0064] The drilling rig's first drilling parameters are obtained, wherein the first drilling parameters are pre-control parameters for the real-time drilling stage; the drilling rig is driven according to the first drilling parameters, and vibration parameters are collected synchronously based on the vibration sensor mounted at the front end of the drilling rig, wherein the vibration parameters include axial vibration signals and rotary impact signals; the first drilling parameters and the vibration parameters are transmitted to the gas migration model.

[0065] In this embodiment, the first drilling parameters of the drilling rig are first obtained. These first drilling parameters are pre-control parameters for the real-time drilling stage, obtained by real-time reading of drilling control quantities such as drilling pressure and drilling speed.

[0066] Subsequently, the drilling rig is driven according to the first drilling parameters, causing the drill bit to enter the coal seam under predetermined control and initiate a continuous drilling process. Simultaneously, vibration sensors located on the front-end actuator of the drilling rig collect drilling disturbance information. Under conditions of axial loading and rotational impact of the drill bit, the vibration sensors record the dynamic vibration response along the drilling direction to form an axial vibration signal, and record the pulsating impact during the rotation of the drill bit to form a rotational impact signal, thereby obtaining vibration parameters that reflect the characteristics of drilling disturbance.

[0067] Finally, the first drilling parameters and the vibration parameters are organized in a unified format and transmitted to the gas migration model, so that the gas migration model can receive the drilling control quantity and the disturbance response quantity in the same time sequence.

[0068] Furthermore, in the method provided in the application embodiments, determining the gas prediction data further includes:

[0069] The first drilling damage zone is calculated based on the pre-control parameters through the first calculation node; the first drilling damage zone is detected by radar through data interaction between the first calculation node and the ground-penetrating radar mounted at the front end of the drilling rig, and the geological detection signal is determined; based on the first drilling damage zone, the geological detection signal and the vibration parameters, a three-dimensional thread parallel prediction based on the second prediction node is executed, and gas prediction data is integrated and output.

[0070] In this embodiment, firstly, through the first calculation node, based on the drilling speed, drilling pressure, and coal compressive strength included in the pre-control parameters, hourly calculations and radial extrapolation calculations are performed to obtain the position of the outer edge that meets the damage criterion. The radius is determined based on this outer edge position, and a spatial boundary is generated around the drill bit axis to obtain the first drilling damage circle.

[0071] Subsequently, data interaction is conducted between the first measurement node and the ground-penetrating radar mounted at the front end of the drilling rig. Within the space defined by the first drilling damage zone, amplitude normalization, background subtraction, and phase difference processing are performed on the radar echoes. Boundary tracking and connectivity analysis are combined to identify the outer edge of the fracture zone and the direction of the fractures. The changes in reflection amplitude and boundary displacement are combined into a sequence on the same time scale to determine the geological exploration signal used to characterize the state of the disturbed structure, and its radial consistency with the boundary position of the first drilling damage zone is checked.

[0072] Finally, based on the first drilling damage zone, geological exploration signals, and vibration parameters collected by vibration sensors and preprocessed, including axial vibration signals and rotary impact signals, a three-dimensional thread parallel prediction is executed within the second prediction node. In this process, seepage is solved using the first drilling damage zone as the boundary and incorporating the fracture connectivity characteristics of the geological exploration signals, outputting the seepage component; unsteady diffusion is solved using the fracture zone scale, granulation degree, and time window geometric changes reflected in the geological exploration signals, outputting the diffusion component; transient release is solved using the axial vibration signal and rotary impact signal as disturbance inputs, and the fracture opening and closing changes in the geological exploration signals as the structural response, extracting the peak value and outputting the transient outburst component. The outputs of the three threads within the same time window are numerically integrated to form gas prediction data.

[0073] Furthermore, the method provided in the application embodiments also includes:

[0074] Perform ternary thread parallel prediction based on the second prediction node to determine the first prediction data, the second prediction data, and the third prediction data; perform time-series superposition processing on the first prediction data and the second prediction data, and perform time-series node superposition as the third prediction data as a pulse term to determine the gas prediction data.

[0075] In this embodiment, within the second prediction node, the first drilling damage zone, geological exploration signals, and vibration parameters are used as inputs under a unified time reference. Three independent prediction threads are invoked for parallel computation. Specifically, the thread based on the seepage mechanism outputs the first prediction data, which characterizes the seepage level controlled by the degree of fracture connectivity; the thread based on the unsteady diffusion mechanism outputs the second prediction data, which characterizes the diffusion level affected by the fracture scale and granulation degree; and the thread based on the transient response mechanism triggered by vibration disturbance outputs the third prediction data, which characterizes the transient outburst level caused by axial vibration signals and rotational impact signals.

[0076] After obtaining the first, second, and third prediction data, the first and second prediction data are aligned in chronological order and superimposed within the same time window to merge the seepage and diffusion levels into a continuous background gas release sequence. Subsequently, the third prediction data is superimposed onto the background sequence at the corresponding time points using pulse characteristic quantities as perturbations, so that the sudden component of transient outbursts is expressed in the overall change.

[0077] After the background and pulse superpositions described above, the gas prediction data that comprehensively describes the gas release trend and disturbance response is finally formed.

[0078] Step S400: Based on the gas prediction data, perform dynamic drilling process control.

[0079] In this embodiment of the application, when controlling the dynamic drilling process based on gas prediction data, the thresholds and time windows for seepage components, diffusion components and transient outburst components are first set on the monitoring interface.

[0080] Then, by comparing the relationship between the three components and the threshold in real time for the current time window, the corresponding parameter adjustment command is triggered. When the seepage component and diffusion component rise for two consecutive time windows and approach the threshold, the drilling pressure and drilling speed are reduced by a fixed amount, and the amount of slag discharged is slightly increased to reduce the expansion of the fracture zone. When the transient outflow component reaches a peak in a single time window and exceeds the threshold, the drilling speed and torque are immediately reduced, the advance is temporarily stopped and the slag discharge is maintained. After the next time window falls back, the drilling pressure and drilling speed are restored in small steps. When the three components are stably below the threshold and the trend is stable, the original control level is restored according to the preset step size.

[0081] For example, if a peak in transient outflow component occurs in a certain time window while the background, i.e., the seepage component and diffusion component, does not rise, then only a short-term response of rapidly reducing drilling speed and torque is executed; if the background continues to rise, then a continuous response of slowly reducing drilling pressure, drilling speed and increasing slag discharge is executed; the above adjustments are re-determined and fine-tuned in the next time window based on the latest gas prediction data, realizing closed-loop control that changes with prediction.

[0082] In summary, the embodiments of this application have at least the following technical effects:

[0083] This application equips the front-end execution components of the drilling rig with vibration sensors and ground-penetrating radar; it constructs a gas migration model based on the first calculation of the damage zone range and parallel prediction based on the evolution of the dual-hole structure, coal cuttings analysis, and vibration impact pulse terms, and embeds it in the mine supervision platform; it uploads the first drilling parameters to the drilling supervision platform, calculates the first damage zone range in the gas migration model, performs range detection through data interaction with the ground-penetrating radar to determine the geological detection signal, and combines the vibration parameters from the vibration sensor to perform three-thread parallel prediction and superposition to determine the gas prediction data; based on the gas prediction data, it performs dynamic drilling process control. This invention solves the technical problems of low gas emission prediction accuracy and inability to reflect the coupling effect of coal body damage and multi-mechanism gas release under dynamic drilling disturbance in the prior art. By constructing a multi-threaded gas migration model that integrates damage zone calculation, dual-hole structure evolution, coal cuttings desorption, and vibration pulse effects, it improves the real-time performance and accuracy of gas emission prediction during dynamic drilling, and achieves the technical effect of effective risk pre-control during the drilling process.

[0084] Example 2, based on the same inventive concept as the method for predicting borehole gas emission during dynamic drilling in the aforementioned examples, such as... Figure 2 As shown, this application provides a dynamic drilling process borehole gas emission prediction system. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0085] Assembly module 11 is used to assemble vibration sensors and ground-penetrating radar on the front-end execution components of the drilling rig; Model building module 12 is used to construct a gas migration model based on the first calculation of the damage zone range and parallel prediction based on the evolution of the dual-hole structure, coal dust analysis and vibration impact pulse terms, and is embedded in the mine supervision platform; Calculation module 13 is used to upload the first drilling parameters to the drilling supervision platform, calculate the first damage zone range in the gas migration model, perform range detection through data interaction with the ground-penetrating radar, determine the geological detection signal, and combine the vibration parameters from the vibration sensor to perform three-thread parallel prediction and superposition to determine the gas prediction data; Control module 14 is used to perform dynamic drilling process control based on the gas prediction data.

[0086] Furthermore, the system is also used to implement the following functions:

[0087] A first calculation node is deployed within the drilling damage zone; a second prediction node is deployed for multi-threaded gas outburst prediction, wherein the second prediction node includes a univariate prediction thread based on the evolution of a dual-pore structure, a binary prediction thread based on coal dust desorption, and a ternary prediction thread based on pulsed vibration and impact; the gas migration model is deployed based on the first calculation node and the second prediction node.

[0088] Furthermore, the system is also used to implement the following functions:

[0089] The damage zone range is calculated based on the damage zone radius. The damage zone radius is calculated as follows: ,in, The first damage propagation coefficient, For drilling speed, The second damage propagation coefficient, For drilling pressure, It refers to the compressive strength of the coal body.

[0090] Furthermore, the system is also used to implement the following functions:

[0091] A first coupling relationship is defined, wherein the first coupling relationship is defined by the gas migration relationship under the coupling of macropore permeability and micropore effective diffusion coefficient based on the evolution of the dual-pore structure. The evolution law of the dual-pore structure is that macropore porosity is positively correlated with the degree of damage, and micropore connectivity is positively correlated with the degree of fracture propagation. By mining the first coupling relationship, the Zhang's permeability evolution and magnitude prediction under dual-pore gas migration are used as the underlying logic to construct the unary prediction thread, wherein fracture propagation includes drilled fractures and natural fractures.

[0092] Furthermore, the system is also used to implement the following functions:

[0093] Determine the unsteady-state diffusion relationship, which includes a first diffusion relationship and a second diffusion relationship. The first diffusion relationship characterizes the instantaneous desorption rate of coal dust with different particle sizes, and the second diffusion relationship characterizes the time window resolution rate based on the cumulative mass of coal dust, time-varying particle size distribution, and dynamic surface area. The mean value obtained from the solutions of the first diffusion relationship and the second diffusion relationship is used as the coal dust resolution metric, and the binary prediction thread is trained under supervision.

[0094] Furthermore, the system is also used to implement the following functions:

[0095] Define the vibration axial relationship, which characterizes the vector transmission of vibration energy, crack opening fluctuation, and transient outburst; use axial vibration and rotational impact as input data, gas transient outburst based on the vibration axial relationship as the prediction target, and pulsed gas release peak as the output to supervise the training of a ternary prediction thread.

[0096] Furthermore, the system is also used to implement the following functions:

[0097] The drilling rig's first drilling parameters are obtained, wherein the first drilling parameters are pre-control parameters for the real-time drilling stage; the drilling rig is driven according to the first drilling parameters, and vibration parameters are collected synchronously based on the vibration sensor mounted at the front end of the drilling rig, wherein the vibration parameters include axial vibration signals and rotary impact signals; the first drilling parameters and the vibration parameters are transmitted to the gas migration model.

[0098] Furthermore, the system is also used to implement the following functions:

[0099] The first drilling damage zone is calculated based on the pre-control parameters through the first calculation node; the first drilling damage zone is detected by radar through data interaction between the first calculation node and the ground-penetrating radar mounted at the front end of the drilling rig, and the geological detection signal is determined; based on the first drilling damage zone, the geological detection signal and the vibration parameters, a three-dimensional thread parallel prediction based on the second prediction node is executed, and gas prediction data is integrated and output.

[0100] Furthermore, the system is also used to implement the following functions:

[0101] Perform ternary thread parallel prediction based on the second prediction node to determine the first prediction data, the second prediction data, and the third prediction data; perform time-series superposition processing on the first prediction data and the second prediction data, and perform time-series node superposition as the third prediction data as a pulse term to determine the gas prediction data.

[0102] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0103] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for predicting borehole gas emission during dynamic drilling, characterized in that, The method includes: Vibration sensors and ground-penetrating radar are installed on the front-end actuators of the drilling rig; Based on the initial calculation of the damage zone range, and the parallel prediction based on the evolution of the dual-pore structure, coal dust analysis, and vibration impact pulse terms, a gas migration model is constructed and embedded in the mine supervision platform. The first drilling parameters are uploaded to the drilling supervision platform, the range of the first damage zone is calculated in the gas migration model, the range detection is carried out through data interaction with the ground-penetrating radar, the geological detection signal is determined, and the gas prediction data is determined by combining the vibration parameters from the vibration sensor with three-thread parallel prediction and superposition. Dynamic drilling process control is implemented based on the gas prediction data.

2. The method for predicting borehole gas emission during dynamic drilling as described in claim 1, characterized in that, Constructing a gas migration model includes: Deploy the first measurement node within the drilling damage zone; Multi-threaded gas outburst prediction is used to deploy a second prediction node, which includes a unary prediction thread based on dual-pore structure evolution, a binary prediction thread based on coal dust desorption, and a ternary prediction thread based on pulsed vibration impact. The gas migration model is deployed based on the first calculation node and the second prediction node.

3. The method for predicting borehole gas emission during dynamic drilling as described in claim 2, characterized in that, Define a dynamic drilling damage zone range, wherein the damage zone range is calculated based on the damage zone radius, and the damage zone radius is calculated as follows: ,in, The first damage propagation coefficient, For drilling speed, The second damage propagation coefficient, For drilling pressure, It refers to the compressive strength of the coal body.

4. The method for predicting borehole gas emission during dynamic drilling as described in claim 3, characterized in that, The second prediction node includes a unary prediction thread based on the evolution of the two-pore structure, comprising: Define the first coupling relationship, wherein the first coupling relationship is defined by the gas migration relationship under the coupling of macropore permeability and micropore effective diffusion coefficient based on the evolution of dual-pore structure. The evolution law of dual-pore structure is that macropore porosity is positively correlated with the degree of damage, and micropore connectivity is positively correlated with the crack propagation degree. By exploring the first coupling relationship, and using the evolution and magnitude prediction of Zhang's permeability under dual-pore gas migration as the underlying logic, the unary prediction thread is constructed, wherein fracture propagation includes drilled fractures and natural fractures.

5. The method for predicting borehole gas emission during dynamic drilling as described in claim 4, characterized in that, The second prediction node includes a binary prediction thread based on coal dust desorption, comprising: Determine the unsteady-state diffusion relationship, wherein the unsteady-state diffusion relationship includes a first diffusion relationship and a second diffusion relationship. The first diffusion relationship characterizes the instantaneous desorption rate of coal dust with different particle sizes, and the second diffusion relationship characterizes the time window resolution rate based on the cumulative mass of coal dust, time-varying particle size distribution and dynamic surface area. The mean value obtained from the solutions of the first and second diffusion relationships is used to perform an analytical metric for coal dust, and a binary prediction thread is trained under supervision.

6. The method for predicting borehole gas emission during dynamic drilling as described in claim 5, characterized in that, The second prediction node includes a ternary prediction thread based on pulse-based vibration and shock, comprising: Define the vibration axial relationship, wherein the vibration axial relationship characterizes the vector transmission of vibration energy-crack opening fluctuation-transient outburst; Using axial vibration and rotational impact as input data, the transient gas outburst based on the vibration axial relationship as the prediction target, and the pulsed gas release peak as the output, a ternary prediction thread is trained under supervision.

7. The method for predicting borehole gas emission during dynamic drilling as described in claim 2, characterized in that, Before calculating the extent of the first damage ring in the gas migration model, the following is included: Obtain the first drilling parameters of the drilling rig, wherein the first drilling parameters are the pre-control parameters of the real-time drilling stage; Based on the first drilling parameters, the drilling rig is driven, and vibration parameters are collected synchronously based on the vibration sensor mounted at the front end of the drilling rig. The vibration parameters include axial vibration signals and rotary impact signals. The first drilling parameters and the vibration parameters are transmitted to the gas migration model.

8. The method for predicting borehole gas emission during dynamic drilling as described in claim 7, characterized in that, Determine gas forecast data, including: The first drilling damage zone is calculated based on the pre-control parameters through the first calculation node; Through data interaction between the first calculation node and the ground-penetrating radar mounted at the front end of the drilling rig, radar detection is performed on the first drilling damage zone to determine the geological detection signal; Based on the first drilling damage zone, geological exploration signals, and vibration parameters, a three-dimensional thread parallel prediction based on the second prediction node is performed, and gas prediction data is integrated and output.

9. The method for predicting borehole gas emission during dynamic drilling as described in claim 8, characterized in that, include: Perform ternary thread parallel prediction based on the second prediction node to determine the first prediction data, the second prediction data, and the third prediction data; The first and second predicted data are superimposed under time-series shift processing, and the third predicted data is superimposed as a pulse term to determine the gas predicted data.

10. A dynamic drilling process borehole gas emission prediction system, characterized in that, The system is used to execute a method for predicting borehole gas emission during a dynamic drilling process as described in any one of claims 1-9, the system comprising: Assembly module, used to assemble vibration sensors and ground-penetrating radar at the front end of the drilling rig; The model building module is used to construct a gas migration model based on the first calculation of the damage zone range and the parallel prediction based on the evolution of the dual-pore structure, coal dust analysis and vibration impact pulse terms. It is embedded in the mine supervision platform. The calculation module is used to upload the first drilling parameters to the drilling supervision platform, calculate the range of the first damage zone in the gas migration model, perform range detection through data interaction with the ground-penetrating radar, determine the geological detection signal, and combine the vibration parameters from the vibration sensor to perform three-thread parallel prediction and superposition to determine the gas prediction data. The control module is used to dynamically control the drilling process based on the gas prediction data.

Images