Drilling System and Method for Risk Self-Perception and Decision-Making in Directional Drilling for Gas Drainage

By collecting and analyzing borehole parameters in real time, and utilizing an expansion state observer and integral terminal sliding mode control, the problem of singularity and closed-loop control in risk perception and decision-making in directional drilling for gas drainage is solved. This enables the classification and identification of stuck drill and blowout risks and safety interlocking, and is applicable to directional drilling construction for gas drainage in underground coal mines.

CN123109541APending Publication Date: 2026-11-17CHINA UNIV OF MINING & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611422335.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-14
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing directional drilling technology for gas extraction suffers from limitations such as limited risk perception methods, difficulty in distinguishing between stuck drill and blowout risks, reliance on manual adjustment of drilling parameters, lack of autonomous decision-making capabilities, fixed risk assessment thresholds, insufficient adaptability to operating conditions, and a lack of a complete closed loop from risk perception to execution control.

Method used

By collecting real-time information on propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, orifice gas concentration, and borehole attitude, and using the expansion state observer to estimate the lumped disturbance of the loop, the risk of stuck drill and blowout is classified and graded, and differentiated drilling parameters are generated. Closed-loop control is then performed in conjunction with integral terminal sliding mode control.

Benefits of technology

It enables real-time identification and classification of risks of stuck drill and blowout, generates differentiated drilling parameters, enhances the system's adaptability to different coal seam conditions, and realizes safety interlocking and closed-loop control, making it suitable for directional drilling construction for underground gas extraction in coal mines.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123109541A_ABST
    Figure CN123109541A_ABST
Patent Text Reader

Abstract

This invention discloses a drilling system and method for risk self-sensing and decision-making in directional drilling for gas drainage. The system includes a data acquisition unit, a borehole risk assessment unit, a drilling parameter decision-making unit, and a parameter tracking control unit. The data acquisition unit is used to collect data on propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, borehole gas emission concentration, and borehole azimuth and inclination during the drilling process. The borehole risk assessment unit is used to determine the dominant risk type and dominant risk level. The drilling parameter decision-making unit is used to generate target propulsion hydraulic pressure, target slewing speed, and safety control signals. The parameter tracking control unit is used for closed-loop control. This invention can achieve real-time identification of stuck drill and blowout risks, generate differentiated drilling parameters, and complete safety interlocks without relying on a large-scale labeled training set, using existing drilling rig signals combined with gas concentration and borehole attitude information. It is particularly suitable for the construction of directional drilling for gas drainage in underground coal mines.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to a drilling system and method, specifically a drilling system and method for risk self-sensing and decision-making in directional drilling for gas drainage in underground coal mines. It is particularly suitable for real-time sensing, classification, autonomous decision-making, and closed-loop control of stuck drill and blowout risks during the construction of directional drilling for gas drainage in underground coal mines, and belongs to the field of coal mine safety drilling technology. Background Technology

[0002] Directional drilling technology for gas drainage is one of the core methods for coal mine gas control. By precisely controlling the trajectory of underground boreholes, it guides them along a designed path to penetrate specific areas of the coal seam or rock strata for efficient, long-distance gas drainage. Its core advantages lie in "precision" and "long distance," effectively solving problems such as uncontrollable borehole trajectories, small drainage range, and low efficiency in traditional gas drainage drilling.

[0003] During directional drilling for gas drainage, as the drilling depth increases, the drill bit needs to rotate and advance over long distances within the coal seam for extended periods. The drilling process faces complex and variable geological conditions and gas occurrence. Factors such as changes in coal seam hardness, poor drill cuttings removal, borehole instability, and localized increases in gas pressure can cause the drilling rig to experience abnormal conditions such as increased rotational resistance, decreased rotational speed, stuck drill bit, rapid increase in gas concentration at the borehole opening, and blowouts. Continuing to drill at a fixed advance pressure and rotational speed may further exacerbate stuck drill bit or blowouts, and could even lead to serious safety accidents.

[0004] Currently, existing directional drilling technology for gas extraction has the following main shortcomings:

[0005] 1. Risk perception methods are limited, making it difficult to distinguish between stuck drill and blowout risks. Existing hydraulic drilling rigs typically rely on operators manually adjusting feed and rotation parameters based on feed pressure, rotation pressure, or field experience, lacking a unified correlation between monitored parameters. Some solutions only limit pressure or retract the drill bit based on a single circuit pressure, failing to integrate gas concentration, borehole trajectory deviation, and hydraulic circuit disturbances for comprehensive risk assessment. Stuck drill risk is primarily caused by increased rotational resistance, while blowout risk is primarily caused by abnormally high gas pressure; their physical mechanisms are fundamentally different, requiring differentiated perception and handling.

[0006] 2. Drilling parameters rely on manual adjustment, lacking autonomous decision-making capability. In existing technologies, the adjustment of drilling parameters (push pressure, rotation speed, etc.) mainly depends on the operator's field experience and subjective judgment, resulting in slow response speed and low accuracy, making it difficult to take effective measures in the early stages of risk occurrence. When stuck drill or blowout risks occur, manual adjustments often lag behind the risk evolution process.

[0007] 3. Fixed risk assessment thresholds lead to insufficient adaptability to various working conditions. Existing technologies mostly use fixed thresholds for risk assessment. However, different coal seam conditions, different borehole sections, and drill bit wear can cause the baseline parameters for normal drilling conditions to drift. Fixed thresholds cannot adapt to these changes, easily leading to false alarms or missed alarms. Some solutions rely on large-scale labeled training sets for machine learning modeling, but the working conditions of directional drilling in coal mines are complex and variable, making it difficult to obtain sufficient high-quality labeled samples, thus limiting the model's generalization ability.

[0008] 4. Lack of a complete closed loop from risk perception to execution control. In existing technologies, risk perception, parameter decision-making, and execution control are often independent of each other, failing to form a complete closed-loop control chain. Even if risks can be perceived, it is difficult to automatically generate optimal drilling parameter adjustment strategies and execute them accurately. Summary of the Invention

[0009] To address the problems existing in the prior art, this invention provides a drilling system and method for risk self-sensing and decision-making in directional drilling for gas drainage. It can identify the risks of stuck drill and blowout in real time, generate differentiated drilling parameters, and complete safety interlocks without relying on a large-scale labeled training set, by utilizing the existing hydraulic and rotation speed signals of the drilling rig and combining gas concentration and borehole posture information. It is particularly suitable for the construction of directional drilling for gas drainage in underground coal mines.

[0010] To achieve the above objectives, the drilling system for risk self-sensing and decision-making in directional drilling for gas extraction includes a drilling rig frame, a propulsion guide rail, a propulsion hydraulic cylinder, a hydraulic rotary motor, a drill rod, a drill bit, a borehole opening device, a hydraulic pump station, an electro-hydraulic proportional valve for the propulsion circuit, and an electro-hydraulic proportional valve for the rotary circuit. The cylinder body of the propulsion hydraulic cylinder is fixedly connected to the drilling rig frame, and the piston end is fixedly connected to the power head slide. The hydraulic rotary motor is fixedly mounted on the power head slide. The power output end of the hydraulic rotary motor is installed and connected to the rear end of the drill rod, and the front end of the drill rod is installed and connected to the drill bit. The hydraulic pump station is connected to the propulsion hydraulic cylinder through the propulsion circuit electro-hydraulic proportional valve, and the hydraulic pump station is connected to the hydraulic rotary motor through the rotary circuit electro-hydraulic proportional valve.

[0011] It also includes a propulsion hydraulic pressure sensor, a slewing hydraulic pressure sensor, a slewing speed sensor, an orifice gas concentration sensor, a borehole attitude measurement component, and a control device;

[0012] The propulsion hydraulic pressure sensor is installed on the pressure line between the propulsion loop electro-hydraulic proportional valve and the propulsion hydraulic cylinder; the slewing hydraulic pressure sensor is installed on the pressure line between the slewing loop electro-hydraulic proportional valve and the hydraulic slewing motor; the slewing speed sensor is installed at the power output end of the hydraulic slewing motor; the orifice gas concentration sensor is installed in the gas discharge channel of the borehole orifice device; the borehole attitude measurement assembly is installed in the measurement-while-drilling (MWD) sub, which is located in the non-magnetic drill string section. The borehole attitude measurement assembly includes a MWD device, an azimuth measurement element, and an inclination measurement element.

[0013] The control unit includes a processor and, electrically connected to the processor, a memory, a data acquisition unit, a borehole risk assessment unit, a drilling parameter decision unit, and a parameter tracking control unit. The data acquisition unit is electrically connected via data acquisition ports to a propulsion hydraulic pressure sensor, a slewing hydraulic pressure sensor, a slewing speed sensor, a borehole gas concentration sensor, and a borehole attitude measurement assembly. The borehole risk assessment unit includes a slewing loop expansion state observer, a propulsion loop expansion state observer, a stuck drill risk channel, and a blowout risk channel, used to determine the dominant risk type and dominant risk level. The drilling parameter decision unit generates a target propulsion hydraulic pressure based on the dominant risk type and dominant risk level. Target rotational speed and safety control signals; the parameter tracking control unit is electrically connected via its control output port to the electro-hydraulic proportional valve in the propulsion circuit, the electro-hydraulic proportional valve in the rotary circuit, the pressure relief actuator in the hydraulic pump station, and the power-off interlocking actuator in the drilling rig power supply circuit, respectively, to adjust the propulsion hydraulic pressure according to the target hydraulic pressure. Target rotational speed Closed-loop control is performed based on the actual feedback signal.

[0014] A drilling method for gas drainage directional drilling risk self-perception and decision-making based on a drilling system for gas drainage directional drilling risk self-perception and decision-making specifically includes the following steps:

[0015] S1 collects propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, borehole gas emission concentration, azimuth and dip angle in real time through the data acquisition unit, and performs feature extraction within the rolling time window to obtain drilling feature data;

[0016] S2, through the borehole risk assessment unit, the lumped disturbance of the corresponding loop is estimated by the rotary loop expansion state observer and the propulsion loop expansion state observer respectively; the stuck drill risk level is determined according to the amplitude of the lumped disturbance of the rotary loop and the speed drop rate, and the blowout risk level is determined according to the gas concentration change rate; when the gas concentration exceeds the over-limit threshold, the gas concentration over-limit flag is output.

[0017] S3, through the drilling risk assessment unit, determines the drilling trajectory deviation based on the azimuth and dip angle, and combines the drilling trajectory deviation, stuck drill risk level and blowout risk level to determine the dominant risk type and risk level;

[0018] S4 generates target propulsion hydraulic pressure, target slewing speed, and safety control signals based on the dominant risk type and risk level through the drilling parameter decision unit;

[0019] S5, through the parameter tracking control unit, performs closed-loop control based on the target propulsion hydraulic pressure, target slewing speed and actual feedback signal, and uses the propulsion loop lumped disturbance estimate and the slewing loop lumped disturbance estimate for feedforward compensation, and adjusts the action of the propulsion hydraulic cylinder (3) and the hydraulic slewing motor (4);

[0020] S6, update the risk assessment threshold based on the drilling process data, and optimize the initial risk assessment threshold for the next borehole based on the single-hole drilling data.

[0021] Furthermore, S1 is detailed as follows:

[0022] S1-1 Before drilling begins, input the design azimuth angle, design dip angle, reference propulsion hydraulic pressure, reference slewing speed, stuck drill disturbance threshold, slewing speed reduction rate threshold, blowout risk threshold, trajectory deviation threshold, and gas concentration exceeding limit threshold to the processor, and complete the signal check of each sensor. The processor outputs the initial control quantity to the propulsion circuit electro-hydraulic proportional valve and the slewing circuit electro-hydraulic proportional valve, so that the propulsion hydraulic cylinder and the hydraulic slewing motor can carry out drilling work according to the reference drilling parameters.

[0023] S1-2, During drilling, the data acquisition unit synchronously collects sampling times through the propulsion hydraulic pressure sensor, rotary hydraulic pressure sensor, rotary speed sensor, orifice gas concentration sensor, and borehole attitude measurement component. The propulsion hydraulic pressure at that time Rotary hydraulic pressure Rotation speed Concentration of gas emitted from boreholes Actual azimuth and actual tilt angle ;

[0024] S1-3, the processor performs range conversion, filtering, outlier removal, and time alignment on the acquired signal, and calculates the rate of decrease in rotational speed within the rolling time window. and the rate of change of borehole gas emission concentration This generates drilling characteristic data for risk assessment, as follows:

[0025] S1-3-1, Signal Filtering and Outlier Handling: For propulsion hydraulic pressure Rotary hydraulic pressure and rotational speed The signal is low-pass filtered; the concentration of gas emitted from the borehole is measured. The signal is filtered using median filtering or moving average filtering;

[0026] S1-3-2, Rotational speed reduction rate calculate:

[0027] In length of Within the rolling time window, the rate of decrease in rotational speed is calculated using the following formula. :

[0028]

[0029] In the formula: Indicates the sampling time Rotational speed at that time; Indicates the rotational speed at the start of the rolling time window; This refers to the duration of the scrolling time window. A positive value indicates a decrease in rotational speed;

[0030] S1-3-3, Borehole gas emission concentration change rate calculate:

[0031] The rate of change in borehole gas emission concentration is calculated using the following formula. :

[0032]

[0033] In the formula: Indicates the sampling time The concentration of gas emitted from the borehole at that time; The borehole gas emission concentration at the start of the rolling time window; A positive value indicates an increase in the gas concentration at the orifice.

[0034] S1-3-4, Time Alignment and Generation of Feature Data:

[0035] Timestamps are added to each data frame based on the processor's system master clock, ensuring that pressure, speed, gas concentration, and attitude data within the same control cycle correspond to each other, generating data including... , , , , , , and Drilling characteristic data.

[0036] Furthermore, S2 is detailed as follows:

[0037] S2-1, Lumped Disturbance Estimation: The drilling risk assessment unit inputs the control input of the electro-hydraulic proportional valve of the rotary loop and the rotary speed to the rotary loop expansion state observer to obtain the lumped disturbance estimate of the rotary loop; the drilling risk assessment unit inputs the control input of the propulsion loop and the propulsion hydraulic pressure to the propulsion loop expansion state observer to obtain the lumped disturbance estimate of the propulsion loop.

[0038] S2-2, Dual-channel risk assessment: The borehole risk assessment unit compares the amplitude of the lumped disturbance estimate of the rotary loop with the stuck drill disturbance threshold and the rotary speed reduction rate. The risk of stuck drill pipe is determined as low, medium, or high risk based on a comparison with the threshold for the rate of decrease in rotational speed; the borehole risk assessment unit is based on the rate of change in borehole gas emission concentration. Determine the nozzle risk level at the sampling time. Drilling gas concentration at that time Exceeding the gas concentration threshold At that time, the gas concentration exceeding the limit is displayed.

[0039] Furthermore, in S2-1, the process of constructing the extended state observer of the gyroscopic loop is as follows:

[0040] The dynamic equations of the rotary loop are expressed as follows:

[0041]

[0042] In the formula: Indicates the sampling time The rate of change of rotational speed at that time; Indicates the sampling time Control input of the electro-hydraulic proportional valve in the rotary circuit; Indicates the input gain of the slewing loop; Indicates the sampling time The total disturbance of the rotary loop is caused by borehole wall friction, drill cuttings accumulation, formation changes, and dynamics of the rotary loop not being modeled.

[0043] The rotational speed estimation error is defined as follows:

[0044]

[0045] In the formula: Indicates the sampling time The error in the estimated rotational speed at that time; Indicates the sampling time The estimated rotational speed at that time; Indicates the sampling time Rotational speed at that time;

[0046] The extended state observer of the gyroscopic loop is then constructed, as described below:

[0047]

[0048]

[0049] In the formula: Indicates the sampling time The estimated value of the lumped disturbance of the gyratory loop at that time; This represents the first gain coefficient of the rotary loop observer; This represents the second gain coefficient of the cyclotron loop observer; and They represent and The first derivative with respect to time;

[0050] The specific process of constructing the propulsion loop expansion state observer is as follows:

[0051] The dynamic equations of the propulsion loop are expressed as follows:

[0052]

[0053] In the formula: Indicates the sampling time The rate of change of propulsion hydraulic pressure over time; Indicates the sampling time The propulsion loop control input at that time; Indicates the input gain of the propulsion loop; Indicates the sampling time The propulsion loop lumped disturbance is caused by changes in propulsion load, changes in hydraulic parameters, and unmodeled dynamics.

[0054] The propulsion hydraulic pressure estimation error is defined as follows:

[0055]

[0056] In the formula: Indicates the sampling time Error in estimating propulsion hydraulic pressure at that time; Indicates the sampling time Estimated propulsion hydraulic pressure at that time;

[0057] The propulsion loop expansion state observer is then constructed, as described below:

[0058]

[0059]

[0060] In the formula: Indicates the sampling time Estimated value of lumped disturbance in the propulsion loop at that time; This represents the first gain coefficient of the propulsion loop observer; This represents the second gain coefficient of the propulsion loop observer; and They represent and The first derivative with respect to time.

[0061] Furthermore, in S2-2, the specific determination of the risk of stuck drill bit is as follows:

[0062] Define the amplitude of the slewing loop disturbance. The expression is as follows:

[0063]

[0064] Set the stuck drill disturbance threshold Threshold for rate of decrease in rotational speed Nozzle detection threshold High-risk assessment coefficient for stuck drill bit High risk assessment coefficient for nozzles ,in , ;

[0065] The risk of stuck drill pipes is classified according to the following rules:

[0066] ①When and At that time, the risk level of the stuck drill was low.

[0067] ②When and Sometimes, or Sometimes, or and At that time, the risk level of the stuck drill was medium risk;

[0068] ③When and At that time, the risk level of the stuck drill was high.

[0069] The risk channels for nozzles are classified according to the following rules:

[0070] ①When At that time, the risk level of the nozzle was low.

[0071] ②When At that time, the risk level of the nozzle was medium risk;

[0072] ③When At that time, the risk level of the nozzle was high.

[0073] Furthermore, S3 is detailed as follows:

[0074] S3-1, Drilling trajectory deviation correction:

[0075] Calculate the borehole trajectory deviation using the following formula. :

[0076]

[0077] In the formula: Indicates the sampling time The actual azimuth angle at that time; Indicates the sampling time The actual tilt angle at that time; Indicates the design azimuth angle; Indicates the design tilt angle; Indicates the weight of azimuth deviation. This represents the weight of the tilt angle deviation, and ;

[0078] When the drilling trajectory deviates When the preset trajectory deviation threshold is exceeded, the stuck drill risk level will be increased by one level, and the increased stuck drill risk level will not exceed the high risk level.

[0079] S3-2, Identification of the dominant risk:

[0080] The dominant risk type is determined according to the following rules: The processor compares the stuck drill risk level and the blowout risk level before and after the correction. When both the stuck drill risk level and the blowout risk level are low risk, the current drilling state is determined to be normal. When at least one type of risk is higher than the low risk, the risk type corresponding to the higher risk level is taken as the dominant risk type. When both types of risk levels are the same and both are higher than the low risk, the blowout risk is given priority as the dominant risk.

[0081] Furthermore, S4 is detailed as follows:

[0082] Define the reference propulsion hydraulic pressure and reference rotational speed ,in The value represents the statistical mean of the propulsion hydraulic pressure under low-risk, normal drilling conditions. The statistical mean of the rotational speed under low-risk normal drilling conditions;

[0083] The drilling parameter decision unit then generates differentiated instructions according to the following rules:

[0084] ① When both the stuck drill risk level and the blowout risk level are low risk levels, the current drilling status is determined to be normal. , ;

[0085] ② When the dominant risk type is stuck drill bit and the stuck drill bit risk level after trajectory deviation correction is medium risk level, , ,in The first pressure reduction coefficient, and , For acceleration coefficient, and ;

[0086] ③ When the dominant risk type is stuck drill bit and the stuck drill bit risk level after trajectory deviation correction is high risk level, , ,in The second pressure drop coefficient, and ;

[0087] ④ When the dominant risk type is nozzle and the nozzle risk level is medium risk, , ,in For the nozzle risk reduction coefficient, and ;

[0088] ⑤ When the dominant risk type is nozzle and the nozzle risk level is high risk, , It also outputs a pressure relief start signal;

[0089] When a gas concentration exceeding the limit is received, a power-off interlock signal is output. The execution priority of the power-off interlock signal and the pressure relief opening signal is higher than the tracking control of the target propulsion hydraulic pressure and the target rotation speed.

[0090] Furthermore, S5 is detailed as follows:

[0091] by Indicates the control channel. This indicates the advancement of the hydraulic pressure control channel. This indicates the rotational speed control channel;

[0092] definition , The corresponding target values ​​are respectively , The tracking error is defined according to the following formula:

[0093]

[0094] In the formula: Indicates the sampling time Tracking error at that time; Indicates the sampling time The actual feedback value at that time; Indicates the sampling time Time control channel The target value;

[0095] Construct the integral terminal sliding surface according to the following formula:

[0096]

[0097] In the formula: Indicates the sampling time The integral terminal sliding surface at that time; ,and It is the terminal nonlinear exponent, and 0 < < 1; τ is the time integration variable; sgn(·) is the sign function;

[0098] The following formula represents the approach law:

[0099]

[0100] In the formula: Indicates the sampling time The derivative of the sliding surface function value with respect to time; For adaptive switching gain; sat(·) is a continuous saturation function; The default boundary layer thickness is 100 mm, and 200 mm is 100 mm. > 0;

[0101] in, Given the boundary layer thickness, the saturation function is defined by the following formula:

[0102]

[0103] In the formula: Indicates the sampling time The value of the sliding surface function at time; sgn(·) is the sign function;

[0104] The switching gain is adjusted according to the dominant risk level using the following formula:

[0105]

[0106] In the formula: Indicates the sampling time Risk level scheduling switching gain at that time; Indicates the reference gain; This represents the risk adjustment coefficient, and ; Indicates the sampling time The risk level quantification value at that time, under normal or low risk conditions. When the risk level is medium High-risk level ;

[0107] Based on the sliding mode surface of the integrator terminal, the reaching law, and the lumped disturbance estimate of the corresponding control channel, the control command is generated according to the following formula:

[0108]

[0109] In the formula: For control commands; To control the channel The input gain when hour = ,when hour = ; The first derivative of the target value with respect to time is taken when the target value is set to a constant. =0, or generated by filtering with a tracking differentiator; For the lumped disturbance estimate, when hour = ,when hour = .

[0110] Furthermore, S6 is detailed as follows:

[0111] S6-1, Low-risk data window maintenance:

[0112] Within each control cycle, when both the stuck drill risk level and the blowout risk level are low, and the trajectory deviation does not exceed the limit and no pressure relief opening signal or power-off interlocking signal is triggered, the length is... The rolling time window is maintained to store the slewing disturbance amplitude sequence, slewing speed decrease rate sequence, and gas concentration change rate sequence;

[0113] S6-2, Updated Threshold for Drill Jam Risk:

[0114] The amplitude of the gyratory disturbance was respectively and the rate of decrease in rotational speed Calculate the empirical cumulative distribution function and The stuck drill disturbance threshold and the slewing speed reduction rate threshold are updated according to the following formulas:

[0115]

[0116]

[0117] In the formula: This indicates the threshold quantile level of the stuck drill disturbance. ; This indicates the threshold quantile level of the rate of decrease in rotational speed. ;

[0118] S6-3, Nozzle Risk Threshold Update:

[0119] Change rate of borehole gas concentration Calculate the empirical cumulative distribution function The nozzle detection threshold is updated according to the following formula:

[0120]

[0121] In the formula: Indicates the nozzle threshold quantile level. ;

[0122] S6-4, Joint optimization of initial threshold for the next borehole:

[0123] After a single hole is drilled, the processor extracts the full-hole rotational disturbance amplitude sequence, rotational speed decrease rate sequence, and gas concentration change rate sequence from memory. Using the stuck drill events and blowout events confirmed on-site as sample labels, and within a preset threshold range, it sets the initial stuck drill disturbance threshold for the next hole. Initial rotational speed reduction threshold and initial nozzle determination threshold The objective function for joint optimization is expressed as follows:

[0124]

[0125] In the formula: This represents the number of false alarms. This represents the number of missed reports; and These represent the total number of normal samples and the total number of abnormal samples, respectively. and These are the false positive weighting coefficient and the false negative weighting coefficient, respectively. ;

[0126] Search within a preset threshold range to achieve the objective function The minimum combination of thresholds is used as the initial stuck drill disturbance threshold, initial slewing speed reduction rate threshold, and initial blowout determination threshold for the next borehole.

[0127] Compared with existing technologies, the drilling system and method for risk self-sensing and decision-making in directional drilling for gas extraction achieves synchronous sensing of hydraulic conditions, drill string status, gas emission status, and borehole trajectory status by coordinating the acquisition of propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, orifice gas concentration, and borehole attitude. It estimates the lumped loop disturbance, which is difficult to measure directly, using an extended state observer, and combines the slewing disturbance amplitude and speed reduction rate for stuck drill risk assessment, and uses gas concentration and its rate of change for blowout risk assessment, thus enabling the classification and grading of stuck drill and blowout risks. Based on the dominant risk, it generates differentiated propulsion and slewing target parameters, and uses integral terminal sliding mode control with disturbance feedforward compensation to drive the actuators, achieving the entire process from risk perception and parameter decision-making to execution control. This system establishes a closed-loop control mechanism; it also updates the judgment threshold online using low-risk data and optimizes the initial threshold for the next borehole based on single-hole data, enhancing the system's adaptability to different coal seam conditions, borehole sections, and drill bit status changes. Without relying on a large-scale labeled training set, it utilizes existing hydraulic and rotational speed signals of the drilling rig, combined with gas concentration and borehole attitude information, to achieve real-time identification of stuck drill and blowout risks, generate differentiated drilling parameters, and complete safety interlocks. This solves the problems of existing technologies such as single monitoring parameters, difficulty in distinguishing between stuck drill and blowout risks, reliance on manual adjustment of drilling parameters, and insufficient adaptability of fixed risk thresholds. It enables real-time perception, classification decision-making, and closed-loop safety control of risks in directional drilling for gas drainage, and is particularly suitable for the construction of directional drilling for gas drainage in underground coal mines. Attached Figure Description

[0128] Figure 1 This is a schematic diagram of the structure and hardware layout of the gas extraction directional drilling risk self-sensing and decision-making drilling system of the present invention;

[0129] Figure 2 This is a flowchart of the gas extraction directional drilling risk self-sensing and decision-making drilling method of the present invention.

[0130] In the diagram: 1. Drilling rig frame; 2. Propulsion guide rail; 3. Propulsion hydraulic cylinder; 4. Hydraulic rotary motor; 5. Drill rod; 6. Drill bit; 7. Drill hole opening device; 8. Coal seam; 9. Hydraulic pump station; 10. Propulsion circuit electro-hydraulic proportional valve; 11. Rotation circuit electro-hydraulic proportional valve; 12. Propulsion hydraulic pressure sensor; 13. Rotation hydraulic pressure sensor; 14. Rotation speed sensor; 15. Orifice gas concentration sensor; 16. Drilling attitude measurement component; 17. Control device. Detailed Implementation

[0131] The present invention will be further described below with reference to the accompanying drawings.

[0132] like Figure 1As shown, the drilling system for risk self-sensing and decision-making in directional drilling for gas extraction includes a drilling rig frame 1, a propulsion guide rail 2, a propulsion hydraulic cylinder 3, a hydraulic rotary motor 4, a drill rod 5, a drill bit 6, a borehole opening device 7, a hydraulic pump station 9, an electro-hydraulic proportional valve for the propulsion circuit 10, an electro-hydraulic proportional valve for the rotary circuit 11, and also includes a propulsion hydraulic pressure sensor 12, a rotary hydraulic pressure sensor 13, a rotary speed sensor 14, a borehole gas concentration sensor 15, a borehole attitude measurement component 16, and a control device 17.

[0133] The drilling rig frame 1 supports the feed guide rail 2 and the power head slide. The cylinder end of the feed hydraulic cylinder 3 is fixedly connected to the drilling rig frame 1, and the piston end is fixedly connected to the power head slide. When the feed hydraulic cylinder 3 extends or retracts, it can drive the power head slide forward or backward along the feed guide rail 2. The hydraulic rotary motor 4 is fixedly mounted on the power head slide. The power output end of the hydraulic rotary motor 4 is installed and connected to the rear end of the drill rod 5, and the front end of the drill rod 5 is installed and connected to the drill bit 6. The hydraulic rotary motor 4 provides rotational power to the drill rod 5 and the drill bit 6.

[0134] The pressure oil output port of the hydraulic pump station 9 is connected to the propulsion hydraulic circuit and the slewing hydraulic circuit respectively. In the propulsion hydraulic circuit, the hydraulic pump station 9 is connected to the propulsion hydraulic cylinder 3 through the propulsion circuit electro-hydraulic proportional valve 10. The oil inlet state and propulsion hydraulic pressure of the propulsion hydraulic cylinder 3 are adjusted by changing the control amount of the propulsion circuit electro-hydraulic proportional valve 10. In the slewing hydraulic circuit, the hydraulic pump station 9 is connected to the hydraulic slewing motor 4 through the slewing circuit electro-hydraulic proportional valve 11. The oil supply state and slewing speed of the hydraulic slewing motor 4 are adjusted by changing the control amount of the slewing circuit electro-hydraulic proportional valve 11.

[0135] The propulsion hydraulic pressure sensor 12 is installed on the pressure line between the propulsion circuit electro-hydraulic proportional valve 10 and the propulsion hydraulic cylinder 3, with its detection end connected to this pressure line. The rotary hydraulic pressure sensor 13 is installed on the pressure line between the rotary circuit electro-hydraulic proportional valve 11 and the hydraulic rotary motor 4, with its detection end connected to this pressure line. The signal output ends of the two pressure sensors are respectively connected to the data acquisition port of the control device 17 to generate feedback signals for propulsion hydraulic pressure and rotary hydraulic pressure, respectively.

[0136] The rotation speed sensor 14, which is electrically connected to the control device 17, is installed at the power output end of the hydraulic rotary motor 4. It is used to collect the rotation speed at the power output end of the hydraulic rotary motor 4 and transmit the rotation speed signal to the control device 17.

[0137] A borehole opening device 7 is installed at the borehole opening to collect and discharge methane gas emitted during drilling. A borehole methane concentration sensor 15 is installed inside the methane discharge channel of the borehole opening device 7, with its detection end connected to the methane discharge channel to obtain the concentration of methane gas emitted from the borehole. Its signal output terminal is electrically connected to the control device 17. The installation position of the orifice gas concentration sensor 15 avoids the rotational movement area of ​​the drill rod 5 so as to continuously collect gas concentration signals during drilling.

[0138] The borehole attitude measurement assembly 16 is installed within the measurement-while-drilling (MWD) sub, which is located in the non-magnetic drill string section. The borehole attitude measurement assembly 16 includes a MWD device, an azimuth measurement element, and an inclination measurement element, used to acquire actual azimuth angles during drilling. and actual tilt angle The attitude data is transmitted to the control device 17 through the data communication port of the drilling measurement device; the design azimuth angle and design dip angle are pre-stored in the control device 17.

[0139] Control device 17 includes a data acquisition port, a processor, a memory, a control output port, a data acquisition unit, a borehole risk assessment unit, a drilling parameter decision unit, and a parameter tracking control unit. The data acquisition port receives data on propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, borehole gas emission concentration, actual azimuth angle, and actual dip angle. The memory stores the design trajectory and the reference propulsion hydraulic pressure. Reference rotational speed Risk assessment thresholds and control parameters; the processor runs drilling risk assessment, drilling parameter decision and parameter tracking control programs; the control output ports are electrically connected to the pressure relief actuator in the propulsion circuit electro-hydraulic proportional valve 10, the rotary circuit electro-hydraulic proportional valve 11, the hydraulic pump station 9 and the power-off interlocking actuator in the drilling rig power supply circuit.

[0140] The drilling risk assessment unit operates the rotary loop expansion state observer, the propulsion loop expansion state observer, the stuck drill risk channel, and the blowout risk channel based on the collected data; the drilling parameter decision unit generates the target propulsion hydraulic pressure based on the dominant risk type and dominant risk level. Target rotational speed Safety control signals include a pressure relief opening signal for high-risk nozzle conditions and a power-off interlock signal when the gas concentration exceeds the over-limit threshold. The pressure relief opening signal is used to drive the pressure relief actuator in the hydraulic pump station 9 to depressurize the propulsion hydraulic circuit and the slewing hydraulic circuit. The power-off interlock signal is used to drive the power-off interlock actuator in the drilling rig power supply circuit to stop the power supply to the drilling rig power circuit and keep it locked. The parameter tracking control unit calculates the control quantity of the propulsion circuit and the slewing circuit according to the target value and the feedback value, and outputs the control quantity to the corresponding electro-hydraulic proportional valve.

[0141] When using a drilling system with self-sensing and decision-making capabilities for risk assessment in directional drilling for gas drainage, the following steps should be followed:

[0142] S1, system initialization, data acquisition and feature extraction.

[0143] After inputting the baseline drilling parameters to the control device 17, drilling is performed according to the baseline drilling parameters. The propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, borehole gas concentration, actual azimuth angle, and actual dip angle are synchronously and in real-time collected by the propulsion hydraulic pressure sensor 12, slewing hydraulic pressure sensor 13, slewing speed sensor 14, borehole gas concentration sensor 15, and borehole attitude measurement component 16. The collected signals undergo range conversion, filtering, outlier removal, and time alignment. Within the rolling time window, the slewing speed decrease rate and gas concentration change rate are calculated to form drilling characteristic data. Specifically:

[0144] S1-1, Before drilling begins, input the design azimuth, design dip angle, and reference propulsion hydraulic pressure into the control device 17. Reference rotational speed Thresholds for stuck drill disturbance, slewing speed reduction rate, blowout risk, trajectory deviation, and excessive gas concentration. The control device 17 outputs initial control values ​​to the propulsion circuit electro-hydraulic proportional valve 10 and the rotary circuit electro-hydraulic proportional valve 11, so that the propulsion hydraulic cylinder 3 and the hydraulic rotary motor 4 perform drilling operations according to the reference drilling parameters.

[0145] S1-2, During drilling, the data acquisition unit synchronously collects sampling data through the data acquisition port and the propulsion hydraulic pressure sensor 12, the rotary hydraulic pressure sensor 13, the rotary speed sensor 14, the orifice gas concentration sensor 15, and the drilling attitude measurement component 16. The propulsion hydraulic pressure at that time Rotary hydraulic pressure Rotation speed Concentration of gas emitted from boreholes Actual azimuth and actual tilt angle .

[0146] S1-3, the processor performs range conversion, filtering, outlier removal, and time alignment on the acquired signal, and calculates the rate of decrease in rotational speed within the rolling time window. and the rate of change of borehole gas emission concentration This generates drilling characteristic data for risk assessment. The specific process is as follows:

[0147] S1-3-1, Signal Filtering and Outlier Handling: For propulsion hydraulic pressure Rotary hydraulic pressure and rotational speed The signal is low-pass filtered to suppress hydraulic pulsation and high-frequency electrical noise; the concentration of borehole gas emission is monitored. The signal uses median filtering or moving average filtering to suppress single-point glitches; when the signal exceeds the range, is disconnected, or jumps abruptly, the processor marks the corresponding data as invalid. Invalid values ​​are not written into the low-risk threshold update window (i.e., they are not used for risk threshold updates) to avoid abnormal sampling values ​​changing the risk judgment threshold.

[0148] S1-3-2, Rotational speed reduction rate calculate:

[0149] In length of Within the rolling time window, the rate of decrease in rotational speed is calculated using the following formula. :

[0150]

[0151] In the formula: Indicates the sampling time Rotational speed at that time; Indicates the rotational speed at the start of the rolling time window; This refers to the duration of the scrolling time window. A positive value indicates a decrease in rotational speed.

[0152] S1-3-3, Borehole gas emission concentration change rate calculate:

[0153] The rate of change in borehole gas emission concentration is calculated using the following formula. :

[0154]

[0155] In the formula: Indicates the sampling time The concentration of gas emitted from the borehole at that time; The borehole gas emission concentration at the start of the rolling time window; A positive value indicates an increase in the concentration of gas at the orifice.

[0156] S1-3-4, Time Alignment and Feature Data Formation:

[0157] Because the sampling frequencies of pressure, speed, gas concentration, and attitude data differ, timestamps are added to each data frame based on the processor's system master clock. Interpolation, hold, or resampling methods are used to unify various signals to the same control cycle, ensuring that pressure, speed, gas concentration, and attitude data within the same control cycle correspond to each other, forming a comprehensive set of data. , , , , , , and Drilling characteristic data.

[0158] S2, lumped disturbance estimation and dual-channel risk assessment.

[0159] The processor-controlled borehole risk assessment unit uses the rotary loop expansion state observer and the propulsion loop expansion state observer to estimate the lumped disturbance of the rotary loop and the propulsion loop, respectively. It determines the stuck drill risk level based on the amplitude of the lumped disturbance of the rotary loop and the rate of rotational speed decrease, and determines the blowout risk level based on the rate of change of gas concentration. When the gas concentration exceeds the limit threshold, it outputs a gas concentration exceeding the limit flag. This dual-channel risk assessment does not rely on a pre-constructed large-scale labeled training set. Specifically:

[0160] S2-1, Lumped Disturbance Estimation: The drilling risk assessment unit inputs the control input of the rotary loop electro-hydraulic proportional valve and the rotary speed to the rotary loop expansion state observer to obtain the lumped disturbance estimate of the rotary loop. The lumped disturbance estimate of the rotary loop is used for stuck drill risk determination. The drilling risk assessment unit inputs the propulsion loop control input and the propulsion hydraulic pressure to the propulsion loop expansion state observer to obtain the lumped disturbance estimate of the propulsion loop. The lumped disturbance estimate of the propulsion loop is used for feedforward compensation of the propulsion hydraulic pressure control channel.

[0161] The specific process of constructing the extended state observer for the gyroscopic loop is as follows:

[0162] The dynamic equation of the rotary loop can be expressed as follows:

[0163]

[0164] In the formula: Indicates the sampling time The rate of change of rotational speed at that time; Indicates the sampling time Control input of the electro-hydraulic proportional valve 11 in the rotary circuit; Indicates the input gain of the slewing loop; Indicates the sampling time The total disturbance of the rotary loop is caused by borehole wall friction, drill cuttings accumulation, formation changes, and the dynamic formation of the rotary loop without modeling.

[0165] The rotational speed estimation error is defined as follows:

[0166]

[0167] In the formula: Indicates the sampling time The error in the estimated rotational speed at that time; Indicates the sampling time The estimated rotational speed at that time; Indicates the sampling time Rotational speed at that time.

[0168] Construct an extended state observer for the rotary loop, as described below:

[0169]

[0170]

[0171] In the formula: Indicates the sampling time The estimated value of the lumped disturbance of the gyratory loop at that time; This represents the first gain coefficient of the rotary loop observer; This represents the second gain coefficient of the cyclotron loop observer; and They represent and The first derivative with respect to time. The first gain coefficient of the gyroscopic loop observer. Second gain coefficient The settings are adjusted based on the dynamic bandwidth of the rotary loop and the sensor noise level.

[0172] The specific process of constructing the propulsion loop expansion state observer is as follows:

[0173] The dynamic equations of the propulsion loop can be expressed as follows:

[0174]

[0175] In the formula: Indicates the sampling time The rate of change of propulsion hydraulic pressure over time; Indicates the sampling time The propulsion loop control input at that time; Indicates the input gain of the propulsion loop; Indicates the sampling time The propulsion loop lumped disturbance is caused by changes in propulsion load, changes in hydraulic parameters, and unmodeled dynamics.

[0176] The propulsion hydraulic pressure estimation error is defined as follows:

[0177]

[0178] In the formula: Indicates the sampling time Error in estimating propulsion hydraulic pressure at that time; Indicates the sampling time The estimated value of the propulsion hydraulic pressure at that time.

[0179] Construct a propulsion loop expansion state observer, as described below:

[0180]

[0181]

[0182] In the formula: Indicates the sampling time Estimated value of lumped disturbance in the propulsion loop at that time; This represents the first gain coefficient of the propulsion loop observer; This represents the second gain coefficient of the propulsion loop observer; and They represent and The first derivative with respect to time.

[0183] S2-2, Dual-channel risk assessment:

[0184] S2-2-1, the drilling risk assessment unit compares the amplitude of the lumped disturbance estimate of the rotary loop with the stuck drill disturbance threshold and the rotary speed reduction rate. Based on a comparison with the threshold for the rate of decrease in rotational speed, the risk of stuck drill bit is determined as low-risk, medium-risk, or high-risk. Details are as follows:

[0185] Define the amplitude of the slewing loop disturbance. The expression is as follows:

[0186]

[0187] Set the stuck drill disturbance threshold Threshold for rate of decrease in rotational speed Nozzle detection threshold High-risk assessment coefficient for stuck drill bit High risk assessment coefficient for nozzles ,in , .

[0188] The risk of stuck drill pipes is classified according to the following rules:

[0189] ①When and At that time, the risk level of the stuck drill was low.

[0190] ②When and Sometimes, or Sometimes, or and At that time, the risk level of the stuck drill was medium risk;

[0191] ③When and At that time, the risk level of the stuck drill was high.

[0192] S2-2-2, The borehole risk assessment unit is based on the rate of change in borehole gas emission concentration. Determine the risk level of the nozzle. Details are as follows:

[0193] The risk channels for nozzles are classified according to the following rules:

[0194] ①When At that time, the risk level of the nozzle was low.

[0195] ②When At that time, the risk level of the nozzle was medium risk;

[0196] ③When At that time, the risk level of the nozzle was high.

[0197] When sampling time Drilling gas concentration at that time Exceeding the gas concentration threshold At that time, the gas concentration exceeding the limit is displayed.

[0198] The borehole risk assessment unit updates the stuck drill risk level, blowout risk level, and gas concentration exceedance indicator in each control cycle.

[0199] S3, Drilling trajectory deviation correction and dominant risk determination.

[0200] The processor determines the actual azimuth angle. and actual tilt angle Relative to the design azimuth angle and design tilt angle The deviation is calculated to determine the borehole trajectory deviation. When the borehole trajectory deviation exceeds a set threshold, the stuck drill risk level is increased. The stuck drill risk level is compared with the blowout risk level, and the risk type corresponding to the higher risk level is taken as the dominant risk type. When both risk levels are the same and both are higher than the low risk level, the blowout risk takes priority. Details are as follows:

[0201] S3-1, Drilling trajectory deviation correction:

[0202] Calculate the borehole trajectory deviation using the following formula. :

[0203]

[0204] In the formula: Indicates the sampling time The actual azimuth angle at that time; Indicates the sampling time The actual tilt angle at that time; Indicates the design azimuth angle; Indicates the design tilt angle; Indicates the weight of azimuth deviation. This represents the weight of the tilt angle deviation, and Azimuth deviation is calculated using the minimum circumferential angle difference between the actual azimuth and the design azimuth, while tilt deviation is calculated using the minimum circumferential angle difference between the actual tilt and the design tilt.

[0205] When the drilling trajectory deviates When the preset trajectory deviation threshold is exceeded, the stuck drill risk level is increased by one level, and the increased stuck drill risk level does not exceed high risk (if the original stuck drill risk was low risk, it is corrected to medium risk; if the original stuck drill risk was medium risk, it is corrected to high risk; if the original stuck drill risk was high risk, it remains high risk). Drill trajectory deviation It is only used to correct the stuck drill risk level and does not replace the independent determination of the stuck drill risk channel and the blowout risk channel.

[0206] S3-2, Identification of the dominant risk:

[0207] The dominant risk type is determined according to the following rules: The processor compares the stuck drill risk level and the blowout risk level before and after the correction. When both the stuck drill risk level and the blowout risk level are low risk, the current drilling state is determined to be normal. When at least one type of risk is higher than the low risk, the risk type corresponding to the higher risk level is taken as the dominant risk type. When both types of risk levels are the same and both are higher than the low risk, the blowout risk is given priority as the dominant risk.

[0208] S4, Drilling parameter decision and generation of safety signals.

[0209] The drilling parameter decision unit generates the target propulsion hydraulic pressure based on the dominant risk type and dominant risk level. Target rotational speed And safety control signals. Baseline drilling parameters are used during normal or low-risk levels; during medium-risk, high-risk, or medium-risk levels of stuck drill bit, the propulsion hydraulic pressure is reduced according to the corresponding coefficient, and the slewing speed is increased within the allowable range; during high-risk levels of blowout, the propulsion and slewing target values ​​are set to zero and a pressure relief open signal is output; a power-off interlock signal is output when the gas concentration exceeds the limit. Details are as follows:

[0210] Define the reference propulsion hydraulic pressure and reference rotational speed ,in The value represents the statistical mean of the propulsion hydraulic pressure under low-risk, normal drilling conditions. This represents the statistical average rotational speed under normal drilling conditions with low risk.

[0211] The drilling parameter decision unit then generates differentiated instructions according to the following rules:

[0212] ① When both the stuck drill risk level and the blowout risk level are low risk levels, the current drilling status is determined to be normal. , ;

[0213] ② When the dominant risk type is stuck drill bit and the stuck drill bit risk level after trajectory deviation correction is medium risk level, , ,in The first pressure reduction coefficient, and , For acceleration coefficient, and ;

[0214] ③ When the dominant risk type is stuck drill bit and the stuck drill bit risk level after trajectory deviation correction is high risk level, , ,in The second pressure drop coefficient, and ;

[0215] ④ When the dominant risk type is nozzle and the nozzle risk level is medium risk, , ,in For the nozzle risk reduction coefficient, and ;

[0216] ⑤ When the dominant risk type is nozzle and the nozzle risk level is high risk, , It also outputs a pressure relief start signal.

[0217] Upon receiving a gas concentration exceeding the limit warning, a power-off interlock signal is output. The execution priority of the power-off interlock signal and the pressure relief opening signal is higher than that of the target propulsion hydraulic pressure and target rotation speed tracking control. That is, if parameter tracking control commands and safety control signals are generated simultaneously in the same control cycle, the processor will prioritize outputting the pressure relief opening signal or the power-off interlock signal; after executing the safety control action, the corresponding actuators will no longer be driven by the target propulsion hydraulic pressure and target rotation speed generated in that cycle.

[0218] S5, closed-loop tracking control of drilling parameters.

[0219] The parameter tracking control unit performs closed-loop tracking control of the propulsion hydraulic pressure and slewing speed, based on the target propulsion hydraulic pressure. Target rotational speed The tracking error is calculated based on the corresponding actual feedback signal. An integral terminal sliding mode control law is used to generate control quantities for the propulsion and slewing loops. Feedforward compensation is performed using the corresponding lumped disturbance estimate to generate propulsion control commands. These commands, after being constrained by amplitude limiting and rate of change, are output to the electro-hydraulic proportional valve 10 in the propulsion loop and the electro-hydraulic proportional valve 11 in the slewing loop, adjusting the actions of the propulsion hydraulic cylinder 3 and the hydraulic slewing motor 4. Details are as follows:

[0220] by Indicates the control channel. This indicates the advancement of the hydraulic pressure control channel. This indicates the rotational speed control channel;

[0221] definition , The corresponding target values ​​are respectively , The tracking error is defined according to the following formula:

[0222]

[0223] In the formula: Indicates the sampling time Tracking error at that time; Indicates the sampling time The actual feedback value at that time; Indicates the sampling time Time control channel The target value.

[0224] Construct the integral terminal sliding surface according to the following formula:

[0225]

[0226] In the formula: Indicates the sampling time The integral terminal sliding surface at that time; ,and It is the terminal nonlinear exponent, and 0 < < 1; τ is the time integration variable; sgn(·) is the sign function.

[0227] The following formula represents the approach law:

[0228]

[0229] In the formula: Indicates the sampling time The derivative of the sliding surface function value with respect to time; For adaptive switching gain; sat(·) is a continuous saturation function; The default boundary layer thickness is 100 mm, and 200 mm is 100 mm. > 0.

[0230] in, Given the boundary layer thickness, the saturation function is defined by the following formula:

[0231]

[0232] In the formula: Indicates the sampling time The value of the sliding surface function at time; sgn(·) is the sign function.

[0233] The switching gain is adjusted according to the dominant risk level using the following formula:

[0234]

[0235] In the formula: Indicates the sampling time Risk level scheduling switching gain at that time; Indicates the reference gain; This represents the risk adjustment coefficient, and ; Indicates the sampling time The risk level quantification value at that time, under normal or low risk conditions. When the risk level is medium High-risk level .

[0236] Based on the sliding mode surface of the integrator terminal, the reaching law, and the lumped disturbance estimate of the corresponding control channel, the control command is generated according to the following formula:

[0237]

[0238] In the formula: For control commands; To control the channel The input gain when hour = ,when hour = ; The first derivative of the target value with respect to time is taken when the target value is set to a constant. =0, or generated by filtering with a tracking differentiator; For the lumped disturbance estimate, when hour = ,when hour = .

[0239] when hour (That is, the corresponding hydraulic pressure control channel). For the lumped disturbance estimate of the propulsion loop ;when hour (i.e., the corresponding rotational speed control channel). The lumped disturbance estimate for the rotary loop The control commands, after being constrained by amplitude limiting and rate of change, are output to the electro-hydraulic proportional valve 10 in the propulsion circuit and the electro-hydraulic proportional valve 11 in the slewing circuit, respectively.

[0240] The switching gain of the two control channels adjusts according to the dominant risk level: the gain corresponding to the low risk level is used in normal or low-risk conditions, the control response strength is increased in medium-risk conditions, and the control response strength is further increased in high-risk conditions. The processor collects the propulsion hydraulic pressure again in the next control cycle. and rotational speed The new tracking error is calculated and the valve control quantity is updated, thereby achieving closed-loop tracking of the target parameter.

[0241] S6, online risk threshold update and hole-to-hole optimization.

[0242] When both the stuck drill risk and the blowout risk are at a low risk level, and no pressure relief opening signal or power failure lockout signal is triggered, the stuck drill disturbance threshold, slewing speed reduction rate threshold, and blowout determination threshold are updated online using the empirical quantiles of the low-risk drilling data. After a single hole is completed, the initial risk determination threshold for the next hole is jointly optimized using full-hole data and confirmed risk events, forming a closed loop of "online sensing and control - low-risk threshold update - inter-hole parameter optimization". Specifically:

[0243] S6-1, Low-risk data window maintenance:

[0244] Within each control cycle, when both the stuck drill risk level and the blowout risk level are low, and the trajectory deviation does not exceed the limit and no pressure relief opening signal or power-off interlocking signal is triggered, the length is... The rolling time window is maintained to store the slewing disturbance amplitude sequence, slewing speed decrease rate sequence, and gas concentration change rate sequence.

[0245] S6-2, Updated Threshold for Drill Jam Risk:

[0246] The amplitude of the gyratory disturbance was respectively and the rate of decrease in rotational speed Calculate the empirical cumulative distribution function and The stuck drill disturbance threshold and the slewing speed reduction rate threshold are updated according to the following formulas:

[0247]

[0248]

[0249] In the formula: This indicates the threshold quantile level of the stuck drill disturbance. ; This indicates the threshold quantile level of the rate of decrease in rotational speed. .

[0250] S6-3, Nozzle Risk Threshold Update:

[0251] Change rate of borehole gas concentration Calculate the empirical cumulative distribution function The nozzle detection threshold is updated according to the following formula:

[0252]

[0253] In the formula: Indicates the nozzle threshold quantile level. .

[0254] After online update , and Limiting and gradual change processing are adopted to avoid threshold abrupt changes caused by short-term operating condition fluctuations.

[0255] S6-4, Joint optimization of initial threshold for the next borehole:

[0256] After a single hole is drilled, the processor extracts the full-hole rotational disturbance amplitude sequence, rotational speed decrease rate sequence, and gas concentration change rate sequence from memory. Using the stuck drill events and blowout events confirmed on-site as sample labels, and within a preset threshold range, it sets the initial stuck drill disturbance threshold for the next hole. Initial rotational speed reduction threshold and initial nozzle determination threshold The objective function for joint optimization is expressed as follows:

[0257]

[0258] In the formula: This represents the number of false alarms. This represents the number of missed reports; and These represent the total number of normal samples and the total number of abnormal samples, respectively. and These are the false positive weighting coefficient and the false negative weighting coefficient, respectively. .

[0259] Search within a preset threshold range to achieve the objective function The minimum combination of thresholds is used as the initial stuck drill disturbance threshold, initial slewing speed reduction rate threshold, and initial blowout determination threshold for the next borehole.

[0260] To verify the ability of the method of the present invention to characterize the risk of stuck drill bit by the rate of decrease in slewing speed and the change in slewing load, the following offline verification was performed using field data from the directional drilling of the transport roadway of the 1223(3) working face of a mine in Huainan. The field used a ZDY23000LDK high-power electro-hydraulic intelligent directional drilling rig, with 4 directional boreholes arranged in the drilling site. The designed borehole depth was 505-520 m, and the sampling interval for drilling parameters was 0.5 s. The published samples are represented by the number 1 to indicate normal working conditions and the number 2 to indicate stuck drill conditions. The specific data segments of samples 364 to 367 are shown in Table 1 below.

[0261]

[0262] 1. Normal drilling phase (samples 364→365)

[0263] During the normal drilling phase of samples 364 to 365, the rate of decrease in rotary speed was:

[0264]

[0265] 2. Stuck drill transition stage (sample 365→366)

[0266] During the stuck drill phase when transitioning from sample 365 to 366, the power head speed dropped sharply from 115 r / min to 80 r / min, with the speed decrease rate being:

[0267]

[0268] Simultaneously, the slewing pressure and feed pressure increased significantly: the slewing pressure rose from 5.151 MPa to 12.400 MPa, with the increase being:

[0269]

[0270] The relative increase in slewing pressure is:

[0271]

[0272] The feed pressure increased from 1.112 MPa to 1.526 MPa, with a relative increase of:

[0273]

[0274] The relative decrease in the power head speed is:

[0275]

[0276] The field measurement data above shows that under normal operating conditions, the rotational speed decrease rate between adjacent samples is 2.0 r / min / s, while it reaches 70.0 r / min / s during the stuck drill transition stage, accompanied by a significant increase in rotational pressure. Therefore, in this data segment, the initial rotational speed decrease rate threshold can be selected between 2.0 and 70.0 r / min / s, and corrected according to the empirical quantile update method of this invention during subsequent drilling.

[0277] The above results indicate that when a stuck drill occurs, it is characterized by a significant increase in rotary load and a rapid decrease in the power head speed, which is consistent with the technical approach of this invention to determine the risk of stuck drill by using the amplitude of the lumped disturbance in the rotary loop and the rate of decrease in rotary speed.

[0278] The drilling system and method for risk self-sensing and decision-making in directional drilling for gas extraction achieves coordinated monitoring of drilling rig hydraulic conditions, drill string rotation status, wellhead gas emission status, and borehole trajectory status by synchronously acquiring propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, wellhead gas concentration, and borehole azimuth and inclination angles. The system estimates lumped disturbances caused by coal and rock load, borehole wall friction, drill cuttings accumulation, formation changes, and unmodeled dynamics through propulsion loop expansion state observers and slewing loop expansion state observers, and classifies risks through stuck drill risk channels and blowout risk channels respectively. Based on the two risk levels and borehole trajectory deviation, the dominant risk type is determined, generating target propulsion hydraulic pressure, target slewing speed, pressure relief opening signal, and power-off interlocking signal. Closed-loop regulation is achieved by using integral terminal sliding mode control with lumped disturbance feedforward compensation to drive the propulsion hydraulic cylinder and hydraulic slewing motor. The drilling system and method for risk self-sensing and decision-making in directional drilling for gas drainage can, without relying on a large-scale labeled training set, utilize the existing hydraulic and rotational signals of the drilling rig, combined with gas concentration and borehole attitude information, to achieve real-time identification of stuck drill and blowout risks, generate differentiated drilling parameters, and complete safety interlocks. It can solve the problems of existing technologies such as single monitoring parameters, difficulty in distinguishing between stuck drill and blowout risks, reliance on manual adjustment of drilling parameters, and insufficient adaptability of fixed risk thresholds. It realizes real-time perception, classification decision-making, and safety closed-loop control of risks in directional drilling for gas drainage, and is particularly suitable for the construction of directional drilling for gas drainage in underground coal mines.

Claims

1. A drilling system for risk self-sensing and decision-making in directional drilling for gas extraction, comprising a drilling rig frame (1), a propulsion guide rail (2), a propulsion hydraulic cylinder (3), a hydraulic rotary motor (4), a drill rod (5), a drill bit (6), a borehole opening device (7), a hydraulic pump station (9), a propulsion circuit electro-hydraulic proportional valve (10), and a rotary circuit electro-hydraulic proportional valve (11). The cylinder end of the propulsion hydraulic cylinder (3) is fixedly connected to the drilling rig frame (1), and the piston end is fixedly connected to the power head slide. The hydraulic rotary motor (4) is fixedly installed on the power head slide. The power output end of the hydraulic rotary motor (4) is installed and connected to the rear end of the drill rod (5), and the front end of the drill rod (5) is installed and connected to the drill bit (6). The hydraulic pump station (9) is connected to the propulsion hydraulic cylinder (3) through the propulsion circuit electro-hydraulic proportional valve (10), and the hydraulic pump station (9) is connected to the hydraulic rotary motor (4) through the rotary circuit electro-hydraulic proportional valve (11). characterized in that It also includes a propulsion hydraulic pressure sensor (12), a rotary hydraulic pressure sensor (13), a rotary speed sensor (14), a borehole gas concentration sensor (15), a borehole attitude measurement assembly (16), and a control device (17). The propulsion hydraulic pressure sensor (12) is installed on the pressure line between the propulsion loop electro-hydraulic proportional valve (10) and the propulsion hydraulic cylinder (3); the rotary hydraulic pressure sensor (13) is installed on the pressure line between the rotary loop electro-hydraulic proportional valve (11) and the hydraulic rotary motor (4); the rotary speed sensor (14) is installed at the power output end of the hydraulic rotary motor (4); the orifice gas concentration sensor (15) is installed in the gas discharge channel of the borehole orifice device (7); the borehole attitude measurement assembly (16) is installed in the measurement while drilling sub, which is arranged in the non-magnetic drill string section. The borehole attitude measurement assembly (16) includes a measurement while drilling device, an azimuth measurement element, and an inclination measurement element. The control device (17) includes a processor and a memory, a data acquisition unit, a borehole risk assessment unit, a drilling parameter decision unit, and a parameter tracking control unit, all electrically connected to the processor. The data acquisition unit is electrically connected to the propulsion hydraulic pressure sensor (12), the slewing hydraulic pressure sensor (13), the slewing speed sensor (14), the orifice gas concentration sensor (15), and the borehole attitude measurement component (16) through a data acquisition port. The borehole risk assessment unit includes a slewing loop expansion state observer, a propulsion loop expansion state observer, a stuck drill risk channel, and a blowout risk channel, used to determine the dominant risk type and dominant risk level. The drilling parameter decision unit is used to generate a target propulsion hydraulic pressure according to the dominant risk type and the dominant risk level , a target rotary speed , and a safety control signal; the parameter tracking control unit is electrically connected with a propulsion circuit electro-hydraulic proportional valve (10), a rotary circuit electro-hydraulic proportional valve (11), a pressure relief actuator in a hydraulic pump station (9), and a power-off locking actuator in a rig power supply circuit through control output ports, and is used to perform closed-loop control according to the target propulsion hydraulic pressure , the target rotary speed , and actual feedback signals.

2. A gas extraction directional drilling risk self-perception and decision-making drilling method based on the gas extraction directional drilling risk self-perception and decision-making drilling system according to claim 1, characterized in that, Specifically, the following steps are included: S1 collects propulsion hydraulic pressure, slewing hydraulic pressure, slewing speed, borehole gas emission concentration, azimuth and dip angle in real time through the data acquisition unit, and performs feature extraction within the rolling time window to obtain drilling feature data; S2, through the borehole risk assessment unit, the lumped disturbance of the corresponding loop is estimated by the rotary loop expansion state observer and the propulsion loop expansion state observer respectively; the stuck drill risk level is determined according to the amplitude of the lumped disturbance of the rotary loop and the speed drop rate, and the blowout risk level is determined according to the gas concentration change rate; when the gas concentration exceeds the over-limit threshold, the gas concentration over-limit flag is output. S3, through the drilling risk assessment unit, determines the drilling trajectory deviation based on the azimuth and dip angle, and combines the drilling trajectory deviation, stuck drill risk level and blowout risk level to determine the dominant risk type and risk level; S4 generates target propulsion hydraulic pressure, target slewing speed, and safety control signals based on the dominant risk type and risk level through the drilling parameter decision unit; S5, through the parameter tracking control unit, performs closed-loop control based on the target propulsion hydraulic pressure, target slewing speed and actual feedback signal, and uses the propulsion loop lumped disturbance estimate and the slewing loop lumped disturbance estimate for feedforward compensation, and adjusts the action of the propulsion hydraulic cylinder (3) and the hydraulic slewing motor (4); S6, update the risk assessment threshold based on the drilling process data, and optimize the initial risk assessment threshold for the next borehole based on the single-hole drilling data.

3. The drilling method for risk self-perception and decision-making in directional drilling for gas drainage according to claim 2, characterized in that, S1 is as follows: S1-1 Before drilling begins, input the design azimuth angle, design dip angle, reference propulsion hydraulic pressure, reference slewing speed, stuck drill disturbance threshold, slewing speed reduction rate threshold, blowout risk threshold, trajectory deviation threshold and gas concentration exceeding limit threshold to the processor, and complete the signal check of each sensor. The processor outputs the initial control quantity to the propulsion circuit electro-hydraulic proportional valve (10) and the slewing circuit electro-hydraulic proportional valve (11) so that the propulsion hydraulic cylinder (3) and the hydraulic slewing motor (4) perform drilling work according to the reference drilling parameters. S1-2, During the drilling process, the data acquisition unit synchronously collects sampling times through the propulsion hydraulic pressure sensor (12), the rotary hydraulic pressure sensor (13), the rotary speed sensor (14), the orifice gas concentration sensor (15), and the drilling attitude measurement component (16). The propulsion hydraulic pressure at that time Rotary hydraulic pressure Rotation speed Concentration of gas emitted from boreholes Actual azimuth and actual tilt angle ; S1-3, the processor performs range conversion, filtering, outlier removal, and time alignment on the acquired signal, and calculates the rate of decrease in rotational speed within the rolling time window. and the rate of change of borehole gas emission concentration This generates drilling characteristic data for risk assessment, as follows: S1-3-1, Signal Filtering and Outlier Handling: For propulsion hydraulic pressure Rotary hydraulic pressure and rotational speed The signal is low-pass filtered; the concentration of gas emitted from the borehole is measured. The signal is filtered using median filtering or moving average filtering; S1-3-2, Rotational speed reduction rate calculate: In length Within the rolling time window, the rate of decrease in rotational speed is calculated using the following formula. : In the formula: Indicates the sampling time Rotational speed at that time; Indicates the rotational speed at the start of the rolling time window; This refers to the duration of the scrolling time window. A positive value indicates a decrease in rotational speed; S1-3-3, Borehole gas emission concentration change rate calculate: The rate of change in borehole gas emission concentration is calculated using the following formula. : In the formula: Indicates the sampling time The concentration of gas emitted from the borehole at that time; The borehole gas emission concentration at the start of the rolling time window; A positive value indicates an increase in the gas concentration at the orifice. S1-3-4, Time Alignment and Feature Data Generation: Timestamps are added to each data frame based on the processor's system master clock, ensuring that pressure, speed, gas concentration, and attitude data within the same control cycle are correlated, generating data including... , , , , , , and Drilling characteristic data.

4. The drilling method for risk self-perception and decision-making in directional drilling for gas drainage according to claim 3, characterized in that, S2 is as follows: S2-1, Lumped Disturbance Estimation: The drilling risk assessment unit inputs the control input of the electro-hydraulic proportional valve of the rotary loop and the rotary speed to the rotary loop expansion state observer to obtain the lumped disturbance estimate of the rotary loop; the drilling risk assessment unit inputs the control input of the propulsion loop and the propulsion hydraulic pressure to the propulsion loop expansion state observer to obtain the lumped disturbance estimate of the propulsion loop. S2-2, Dual-channel risk assessment: The borehole risk assessment unit compares the amplitude of the lumped disturbance estimate of the rotary loop with the stuck drill disturbance threshold, and the rotary speed reduction rate... By comparing the risk of stuck drill bit with the threshold of the rate of decrease in rotational speed, the risk of stuck drill bit is determined as low risk, medium risk or high risk. The borehole risk assessment unit is based on the rate of change of borehole gas emission concentration. Determine the nozzle risk level at the sampling time. Drilling gas concentration at that time Exceeding the gas concentration threshold At that time, the gas concentration exceeding the limit is displayed.

5. The drilling method for risk self-perception and decision-making in directional drilling for gas drainage according to claim 4, characterized in that, In S2-1, the specific process of constructing the extended state observer of the gyroscopic loop is as follows: The dynamic equations of the rotary loop are expressed as follows: In the formula: Indicates the sampling time The rate of change of rotational speed at that time; Indicates the sampling time The control input of the electro-hydraulic proportional valve (11) in the rotary circuit; Indicates the input gain of the slewing loop; Indicates the sampling time The total disturbance of the rotary loop is caused by borehole wall friction, drill cuttings accumulation, formation changes, and dynamics of the rotary loop not being modeled. The rotational speed estimation error is defined as follows: In the formula: Indicates the sampling time The error in the estimated rotational speed at that time; Indicates the sampling time The estimated rotational speed at that time; Indicates the sampling time Rotational speed at that time; The extended state observer of the gyroscopic loop is then constructed, as described below: In the formula: Indicates the sampling time The estimated value of the lumped disturbance of the gyratory loop at that time; This represents the first gain coefficient of the rotary loop observer; This represents the second gain coefficient of the cyclotron loop observer; and They represent and The first derivative with respect to time; The specific process of constructing the propulsion loop expansion state observer is as follows: The dynamic equations of the propulsion loop are expressed as follows: In the formula: Indicates the sampling time The rate of change of propulsion hydraulic pressure over time; Indicates the sampling time The propulsion loop control input at that time; Indicates the input gain of the propulsion loop; Indicates the sampling time The propulsion loop lumped disturbance is caused by changes in propulsion load, changes in hydraulic parameters, and unmodeled dynamics. The propulsion hydraulic pressure estimation error is defined as follows: In the formula: Indicates the sampling time Error in estimating propulsion hydraulic pressure at that time; Indicates the sampling time Estimated propulsion hydraulic pressure at that time; The propulsion loop expansion state observer is then constructed, as described below: In the formula: Indicates the sampling time Estimated value of lumped disturbance in the propulsion loop at that time; This represents the first gain coefficient of the propulsion loop observer; This represents the second gain coefficient of the propulsion loop observer; and They represent and The first derivative with respect to time.

6. The drilling method for risk self-perception and decision-making in directional drilling for gas drainage according to claim 4, characterized in that, In S2-2, the specific determination of the risk of stuck drill bit is as follows: Define the amplitude of the slewing loop disturbance. The expression is as follows: Set the stuck drill disturbance threshold Threshold for rate of decrease in rotational speed Nozzle detection threshold High-risk assessment coefficient for stuck drill bit High risk assessment coefficient for nozzles ,in , ; The risk of stuck drill pipes is classified according to the following rules: ①When and At that time, the risk level of the stuck drill was low. ②When and Sometimes, or Sometimes, or and At that time, the risk level of the stuck drill was medium risk; ③When and At that time, the risk level of the stuck drill was high. The risk channels for nozzles are classified according to the following rules: ①When At that time, the risk level of the nozzle was low. ②When At that time, the risk level of the nozzle was medium risk; ③When At that time, the risk level of the nozzle was high.

7. The drilling method for risk self-perception and decision-making in directional drilling for gas drainage according to claim 4, characterized in that, S3 is as follows: S3-1, Drilling trajectory deviation correction: Calculate the borehole trajectory deviation using the following formula. : In the formula: Indicates the sampling time The actual azimuth angle at that time; Indicates the sampling time The actual tilt angle at that time; Indicates the design azimuth angle; Indicates the design tilt angle; Indicates the weight of azimuth deviation. This represents the weight of the tilt angle deviation, and ; When the drilling trajectory deviates When the preset trajectory deviation threshold is exceeded, the stuck drill risk level will be increased by one level, and the increased stuck drill risk level will not exceed the high risk level. S3-2, Identification of the dominant risk: The dominant risk type is determined according to the following rules: The processor compares the stuck drill risk level with the blowout risk level before and after the correction. When both the stuck drill risk level and the blowout risk level are low risk, the current drilling status is determined to be normal. When at least one type of risk is higher than the low risk, the risk type corresponding to the higher risk level shall be taken as the dominant risk type. When two types of risk levels are the same and both are higher than low risk, the dominant risk should be identified as the nozzle risk.

8. The drilling method for risk self-perception and decision-making in directional drilling for gas extraction according to claim 7, characterized in that, S4 is as follows: Define the reference propulsion hydraulic pressure and reference rotational speed ,in The value represents the statistical mean of the propulsion hydraulic pressure under low-risk, normal drilling conditions. The statistical mean of the rotational speed under low-risk normal drilling conditions; The drilling parameter decision unit then generates differentiated instructions according to the following rules: ① When both the stuck drill risk level and the blowout risk level are low risk levels, the current drilling status is determined to be normal. , ; ② When the dominant risk type is stuck drill bit and the stuck drill bit risk level after trajectory deviation correction is medium risk level, , ,in The first pressure reduction coefficient, and , For acceleration coefficient, and ; ③ When the dominant risk type is stuck drill bit and the stuck drill bit risk level after trajectory deviation correction is high risk level, , ,in The second pressure drop coefficient, and ; ④ When the dominant risk type is nozzle and the nozzle risk level is medium risk, , ,in For the nozzle risk reduction coefficient, and ; ⑤ When the dominant risk type is nozzle and the nozzle risk level is high risk, , It also outputs a pressure relief start signal; When a gas concentration exceeding the limit is received, a power-off interlock signal is output. The execution priority of the power-off interlock signal and the pressure relief opening signal is higher than the tracking control of the target propulsion hydraulic pressure and the target rotation speed.

9. The drilling method for risk self-perception and decision-making in directional drilling for gas drainage according to claim 8, characterized in that, S5 is detailed below: by Indicates the control channel. This indicates the advancement of the hydraulic pressure control channel. This indicates the rotational speed control channel; definition , The corresponding target values ​​are respectively , The tracking error is defined according to the following formula: In the formula: Indicates the sampling time Tracking error at that time; Indicates the sampling time The actual feedback value at that time; Indicates the sampling time Time control channel Target value; Construct the integral terminal sliding surface according to the following formula: In the formula: Indicates the sampling time The integral terminal sliding surface at that time; ,and It is the terminal nonlinear exponent, and 0 < < 1; τ is the time integration variable; sgn(·) is the sign function; The following formula represents the approach law: In the formula: Indicates the sampling time The derivative of the sliding surface function value with respect to time; For adaptive switching gain; sat(·) is a continuous saturation function; The default boundary layer thickness is 100 mm, and 200 mm is 100 mm. > 0; in, Given the boundary layer thickness, the saturation function is defined by the following formula: In the formula: Indicates the sampling time The value of the sliding surface function at time; sgn(·) is the sign function; The switching gain is adjusted according to the dominant risk level using the following formula: In the formula: Indicates the sampling time Risk level scheduling switching gain at that time; Indicates the reference gain; This represents the risk adjustment coefficient, and ; Indicates the sampling time The risk level quantification value at that time, under normal or low risk conditions. When the risk level is medium High-risk level ; Based on the sliding mode surface of the integrator terminal, the reaching law, and the lumped disturbance estimate of the corresponding control channel, the control command is generated according to the following formula: In the formula: For control commands; To control the channel The input gain when hour = ,when hour = ; The first derivative of the target value with respect to time is taken when the target value is set to a constant. =0, or generated by filtering with a tracking differentiator; For the lumped disturbance estimate, when hour = ,when hour = .

10. The drilling method for risk self-perception and decision-making in directional drilling for gas drainage according to claim 9, characterized in that, S6 is detailed below: S6-1, Low-risk data window maintenance: Within each control cycle, when both the stuck drill risk level and the blowout risk level are low, and the trajectory deviation does not exceed the limit and no pressure relief opening signal or power-off interlocking signal is triggered, the length is... The rolling time window is maintained to store the slewing disturbance amplitude sequence, slewing speed decrease rate sequence, and gas concentration change rate sequence; S6-2, Updated Threshold for Drill Jam Risk: The amplitude of the gyratory disturbance was respectively and the rate of decrease in rotational speed Calculate the empirical cumulative distribution function and The stuck drill disturbance threshold and the slewing speed reduction rate threshold are updated according to the following formulas: In the formula: This indicates the threshold quantile level of the stuck drill disturbance. ; This indicates the threshold quantile level of the rate of decrease in rotational speed. ; S6-3, Nozzle Risk Threshold Update: Change rate of borehole gas concentration Calculate the empirical cumulative distribution function The nozzle detection threshold is updated according to the following formula: In the formula: Indicates the nozzle threshold quantile level. ; S6-4, Joint optimization of initial threshold for the next borehole: After a single hole is drilled, the processor extracts the full-hole rotational disturbance amplitude sequence, rotational speed decrease rate sequence, and gas concentration change rate sequence from memory. Using the stuck drill events and blowout events confirmed on-site as sample labels, and within a preset threshold range, it sets the initial stuck drill disturbance threshold for the next hole. Initial rotational speed reduction threshold and initial nozzle determination threshold The objective function for joint optimization is expressed as follows: In the formula: This represents the number of false alarms. This represents the number of missed reports; and These represent the total number of normal samples and the total number of abnormal samples, respectively. and These are the false positive weighting coefficient and the false negative weighting coefficient, respectively. ; Search within a preset threshold range to achieve the objective function The minimum combination of thresholds is used as the initial stuck drill disturbance threshold, initial slewing speed reduction rate threshold, and initial blowout determination threshold for the next borehole.