Efficient drilling system for coal seam gas extraction drilling and regulation and control method

Abstract

The invention discloses an efficient drilling system for coal seam gas extraction drilling and a regulation and control method, fine crushing of particles in a chip returning process and efficient transportation and blockage removal of returned chip particles are realized by utilizing the crushing effect and the transportation effect of a spiral blade, and meanwhile, working condition parameters of a drill rod and the particle size and the quality after returned chips are crushed can be dynamically monitored; according to the method, key parameters are determined by analyzing GRA through the grey correlation degree through quantitative evaluation of the returned chip crushing effect, and then the returned chip crushing effect is predicted through a GA-BP neural network according to different drilling parameter combinations. And finally, an automatic control signal is output according to the returned chip crushing effect evaluation index error to dynamically regulate and control the drilling parameters so that the returned chip effect can reach the expected target, and efficient crushing and intelligent regulation and control in the drilling process can be achieved on the premise that the probability that a drill hole is blocked by coal rock returned chips in the drilling process is reduced. And the problem caused by blockage of the drill hole due to returned chips can be fundamentally solved, so that the hole forming efficiency of the gas extraction drill hole is improved.

Description

Technical Field

[0001] The present invention relates to a borehole drilling system and a control method, in particular to an efficient drilling system and a control method for coal seam gas extraction drilling, and belongs to the technical field of mining engineering. Background Art

[0002] As the main energy source in my country, coal plays a dominant role in industrial production and energy security. However, coal mine gas disasters seriously restrict the efficiency of coal mining, and gas extraction drilling technology has become the key to gas control. The currently widely used directional drilling process for coal seams is easily affected by factors such as coal rock structure fragmentation and drilling trajectory deviation under complex geological conditions. During the drilling process, large pieces of coal rock chips can easily clog the drill hole and cause poor chip removal, which can easily lead to problems such as drill sticking, drilling stoppage, and even drill sticking, and these problems occur frequently. On the one hand, traditional solutions such as reverse rotation of the drill bit and the use of lubricants have limited effects and are difficult to fundamentally solve the problem. On the other hand, traditional drill rods lack the ability to finely crush the chips, and large pieces of chips are difficult to effectively discharge, resulting in low drilling efficiency. On the other hand, drilling parameters have an important influence on the chip crushing effect, and the traditional drilling parameter control method often lags behind the dynamic changes of the formation, especially in coal seams with high gas content. It is often impossible to achieve real-time matching of drilling pressure, rotation speed and slag discharge efficiency, thereby reducing the drilling efficiency of gas extraction drilling, thereby restricting the gas extraction efficiency of single hole. Summary of the Invention

[0003] In response to the problems existing in the above-mentioned prior art, the present invention provides an efficient drilling system and control method for coal seam gas extraction drilling holes, which can realize dynamic regulation and intelligent control of drilling parameters while reducing the probability of coal rock chips clogging the drill hole during the drilling process, fundamentally solving the problems caused by coal rock chips clogging the drill hole, and thus improving the drilling efficiency of gas extraction drilling holes.

[0004] To achieve the above purpose, this coal seam gas extraction drilling system includes a drilling rig part, a material guide part, a power torque detection part, a return chip quantitative detection part and a centralized electronic control part;

[0005] The drilling rig part includes a drill rod, on the rod body of the drill rod is fixed a spiral blade spirally arranged along the axial direction of the rod body, the thickness of the spiral blade gradually decreases from the root to the edge, and the outer edge of the spiral blade is provided with a crushing tooth extending forward, the spiral blade includes at least one group of spiral blade composite structures in the axial direction of the rod body, the spiral blade composite structure includes a coarse crushing section, a transition section and a dense conveying section from front to back, and the leads of the coarse crushing section, the transition section and the dense conveying section are arranged in descending order, and the rod body of the drill rod is also provided with a pressure gas channel arranged along the axial direction of the drill rod, and at least the rod body of the dense conveying section is provided with a pneumatic boosting hole that penetrates the pressure gas channel;

[0006] The material guiding part includes a sealing material guiding device which can be sleeved on the drill rod and a chip collecting box arranged on the frame;

[0007] The power torque detection part includes a torque sensor installed on the drill pipe;

[0008] The return chip quantitative detection part includes a particle size analyzer arranged on the receiving port of the return chip collection box and a quality sensor arranged on the return chip collection box;

[0009] The centralized electronic control part includes an electronic control box, a central processing unit, a drilling control circuit, a power torque monitoring and analysis circuit, and a particle size quality analysis circuit. The central controller is electrically connected to the rotary drive device, torque sensor, particle size analyzer, and quality sensor respectively.

[0010] As a further improvement of the present invention, the outer diameter of the spiral blades in the transition section or the dense conveying section gradually decreases from the back to the front.

[0011] As a further improvement of the present invention, the spiral blade may be provided with a radial flow-guiding structure from the root toward the edge.

[0012] As a further improvement of the present invention, the spiral blades of the transition section may be provided with turbulent grooves with their openings facing forward.

[0013] As a further improvement of the present invention, the crushing teeth on the outer edge of the spiral blade have a forward rake angle and an obtuse-angle tooth top structure.

[0014] A method for efficiently drilling and controlling coal seam gas extraction drilling holes based on a coal seam gas extraction drilling hole efficient drilling system specifically comprises the following steps:

[0015] Step 1: quantitatively evaluate the effect of the returned chips after crushing;

[0016] Step 2: Based on the evaluation results of Step 1, the drill pipe working parameters are used as the variable sequence and the chip crushing evaluation index is used as the reference sequence. Grey relational analysis (GRA) is used to determine the key parameters.

[0017] Step 3: Use GA-BP neural network to predict the chip crushing effect according to different drilling parameter combinations;

[0018] Step 4: Based on the prediction model of Step 3, the prediction result of chip return crushing is output, and the chip return crushing effect evaluation index error e(t) and torque error are calculated. Then, the central processing unit adjusts the coefficients of proportion, integration and differentiation according to the chip return crushing effect evaluation index error e(t), and outputs the automatic control signal u(t) to dynamically adjust the drilling parameters so that the chip return effect reaches the expected target.

[0019] Furthermore, Step 1 is as follows:

[0020] Step 1-1, quantitative evaluation of particle size: Use a particle size analyzer to dynamically scan the drill cuttings sample to obtain particle size spectrum data. The central processing unit divides the particle size spectrum into m statistical windows and records the particle count N in each window. m , and the relative frequency method is used to calculate the particle size distribution probability, the formula is as follows:

[0021]

[0022] Where: p m is the frequency of particles in the mth statistical window; N is the total number of particles in the returned chip sample; d * m is the normalized particle size value of the mth statistical window; d i is the average particle size of the mth statistical window; μ d is the mean particle size; σ d is the particle size standard deviation;

[0023] The parameters λ and β of the Rosin-Rammler distribution are fitted by the measured data, and the measured cumulative distribution value F is calculated. ovs (d j ), define the residual function R, and use the optimization algorithm to solve λ and β. The formula is as follows:

[0024]

[0025] Where: R(λ,β) is the residual function; F obs (d j ) is the measured cumulative distribution value; d m is the representative particle size value of the mth particle size interval; k is the total number of particle size intervals; λ is the characteristic particle size parameter of the Rosin-Rammler distribution; β is the shape parameter of the Rosin-Rammler distribution; is the optimal parameter obtained by the optimization algorithm to minimize the residual function;

[0026] The Rosin-Rammler distribution function is used to describe the particle size distribution after particle crushing. The cumulative distribution function F(d) represents the proportion of particles with a particle size less than or equal to d, and the probability density function f(d) represents the probability density of the occurrence of particles with a particle size of d. The formula is as follows:

[0027]

[0028] d x =λ(-ln(1-x / 100)) 1 / β

[0029] Where: d is the particle size of the returned chips; x is the quantile percentage; d x is the particle size value corresponding to the cumulative probability of x%, indicating that the particle size is smaller than d x The proportion of particles is x%;

[0030] Calculate d 10 d 50 d 90 The particle size uniformity index U is obtained by the following formula:

[0031]

[0032] Where: U is the uniformity index; d 10 d 50 d 90 They represent the particle sizes corresponding to the 10%, 50%, and 90% quantiles in the cumulative distribution, respectively;

[0033] Step 1-2, quality assessment: The mass sensor measures the total mass of the returned chip sample, and the returned chip mass ratio ε is calculated to quantify the matching degree between the actual returned chip mass and the theoretical crushing mass. The returned chip mass ratio is analyzed to obtain the returned chip transport efficiency;

[0034] The calculation formula of the return chip mass ratio ε is as follows:

[0035]

[0036] Where: ρ is the density of the returned cuttings sample; V is the theoretical drilling volume; M is the total mass of the returned cuttings sample;

[0037] Step 1-3, comprehensive evaluation: 50 , particle size uniformity index U and return chip mass ratio ε are normalized, and the formula is as follows:

[0038]

[0039] Where: D 50 * d 50 Normalized index value; U * is the normalized value of the particle size uniformity index U; ε * is the normalized index value of the returned chip mass ratio ε; D 50min and D 50max are the historical experimental data 50 The minimum and maximum values ​​of U min and U nax are the minimum and maximum values ​​of the particle size uniformity index U in the historical experimental data; ε min and ε max are the minimum and maximum values ​​of the returned chip mass ratio ε in the historical experimental data;

[0040] The weights are adaptively adjusted according to the lithology, and the modified power function weighting method is used to construct a comprehensive evaluation index of return chip crushing through the crushing efficiency index CEI. The formula is as follows:

[0041] CEI=(D * 50 ) ω1 (U * ) ω2 (ε * ) ω3

[0042] Where: CEI is the crushing efficiency index; ω1, ω2, ω3 are d 50 , the weight constraints of particle size uniformity index U and return chip mass ratio ε satisfy ω1+ω2+ω3=1.

[0043] Furthermore, Step 2 is as follows:

[0044] Step 2-1, perform linear normalization on the operating parameters. The formula for linear normalization is as follows:

[0045]

[0046] Where: X * is the data after normalization; X i is the input data; X min is the minimum value of the input sample data; X max is the maximum value of the input sample data;

[0047] Step 2-2: Take the chip crushing evaluation index as the reference sequence, and calculate the correlation between the working condition parameters and the reference sequence, the correlation coefficient η ij The calculation formula is as follows:

[0048]

[0049] Where: δ ij is the absolute difference of the i-th parameter sequence at point j; Δ min is the global minimum difference; Δ max is the global maximum difference; τ is the dynamic sensitivity adjustment factor;

[0050] Correlation The calculation formula is as follows:

[0051]

[0052] Where: n is the length of the data sequence;

[0053] Step 2-3, determine the weight of each working condition parameter on the chip crushing evaluation index, and obtain the influence degree of each index on the roof lane layout according to the sum of the weights of each physical index. The weight H i The calculation formula is as follows:

[0054]

[0055] Where: a is the number of operating parameters.

[0056] Furthermore, Step 3 is as follows:

[0057] Key drilling parameters are used as the input layer of the GA-BP neural network, and the return chip crushing evaluation index is used as the output layer of the GA-BP neural network. Two hidden layers are set in the middle to determine the BP neural network structure, and then the number of weight and bias combinations of each parameter is determined. A genetic algorithm is used to optimize the initial weights and biases of the BP neural network. The MSE is selected as the fitness function to calculate the fitness value of the individual. The genetic algorithm finds the individual corresponding to the optimal fitness value by selecting crossover and mutation operations. Finally, the BP neural network uses the genetic algorithm to obtain the optimal individual to assign the network initial weights and biases. The collected data is used for training and prediction. By gradually optimizing the model, a prediction model based on drilling parameters is finally formed to feedback the return chip crushing evaluation results.

[0058] The calculation formula of fitness function MSE is as follows:

[0059]

[0060] Where: y i is the measured value; is the predicted value; M is the total number of training samples.

[0061] Furthermore, in Step 4, the calculation formula of the chip crushing effect evaluation index error e(t) is as follows:

[0062] e(t)=CEI(t) * -CEI(t)

[0063] Where: CEI(t) * is the target value of chip crushing effect; CEI(t) is the predicted value of neural network;

[0064] The calculation formula of the automatic control signal u(t) is as follows:

[0065]

[0066] Where: u(t) is the automatic control signal, where t is time; e(t) is the error of the chip crushing effect evaluation index, where t is time; K P , KI , K D are the proportional coefficient, integral coefficient, and differential coefficient respectively.

[0067] Compared with the existing technology, this coal seam gas extraction drilling system uses the crushing and transportation effects of spiral blades to achieve fine crushing of particles in the chip return process and efficient transportation and unblocking of chip return particles. Due to the power torque detection part and the chip return quantitative detection part, the drill pipe working parameters, the particle size and quality of the chip return after crushing can be dynamically monitored. By analyzing the particle size and quality of the chip return after crushing, a comprehensive evaluation index of the chip return crushing effect is constructed, which can achieve quantitative evaluation of the chip return crushing effect. Based on the quantitative evaluation results, gray relational analysis GRA is used to determine the key Parameters, and then the GA-BP neural network is used to predict the chip crushing effect according to different drilling parameter combinations. Finally, the chip crushing effect evaluation index error is calculated, and the coefficients of proportion, integration and differentiation are adjusted according to the chip crushing effect evaluation index error. The automatic control signal is output to dynamically adjust the drilling parameters so that the chip return effect reaches the expected goal. It can achieve efficient crushing and intelligent control of the drilling process under the premise of reducing the probability of coal and rock chip return blocking the borehole during the drilling process, and can fundamentally solve the problem caused by chip return blocking the borehole, thereby improving the drilling efficiency of gas extraction drilling. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a structural diagram of the efficient drilling system for coal seam gas extraction drilling;

[0069] Figure 2 This is a structural diagram of the drill rod of the high-efficiency drilling system for coal seam gas extraction;

[0070] Figure 3 This is a flow chart of the efficient drilling control method for coal seam gas extraction drilling.

[0071] In the figure: 1. Torque sensor, 2. Particle size analyzer, 3. Mass sensor, 4. Central controller, 5. Electric control box. DETAILED DESCRIPTION

[0072] The present invention will be further described below with reference to the accompanying drawings.

[0073] like Figure 1 As shown, the high-efficiency drilling system for coal seam gas extraction drilling includes a drilling rig part, a material guide part, a power torque detection part, a return chip quantitative detection part and a centralized electronic control part.

[0074] The drilling rig includes a frame and a rotary drive device, a drilling propulsion device and an elevation and pitch angle control device installed on the frame. The rotary drive device is connected to the drill rod in a transmission manner. The drilling propulsion device is used to apply axial pressure to the drill rod during drilling. The elevation and pitch angle control device is used to control the drilling direction of the drill rod. A rock breaking drill bit is installed at the front end of the drill rod. The above technical solution is an existing technology and will not be described in detail here. Figure 2 As shown, a spiral blade spirally arranged along the axial direction of the rod body is fixed on the rod body of the drill rod, the thickness of the spiral blade gradually decreases from the root to the edge, and a crushing tooth (not shown in the figure) extending forward is provided on the outer edge of the spiral blade, and the surface of the spiral blade can be welded with a wear-resistant coating, and the spiral blade includes at least one group of spiral blade composite structures in the axial direction of the rod body, and the spiral blade composite structure includes a coarse crushing section, a transition section and a dense conveying section from front to back, and the lead of the coarse crushing section, the transition section and the dense conveying section are arranged in descending order, and a pressure gas channel arranged along the axial direction of the drill rod is also provided in the rod body of the drill rod, and at least a pneumatic boosting hole communicating with the pressure gas channel is provided on the rod body of the dense conveying section.

[0075] The material guide part includes a hole-sealing material guide device that can be sleeved on the drill rod and a chip collection box arranged on the frame. The hole-sealing material guide device is used to seal the hole end of the drill hole and guide the chips discharged during the drilling process to the chip collection box.

[0076] The power torque detection part includes a torque sensor 1 installed on the drill rod.

[0077] The return chip quantitative detection part includes a particle size analyzer 2 arranged on the receiving port of the return chip collection box and a quality sensor 3 arranged on the return chip collection box. The particle size analyzer is preferably a laser particle size analyzer based on the MIE scattering principle.

[0078] The centralized electronic control part includes an electric control box 5, a central processing unit 4, a drilling control circuit, a power torque monitoring and analysis circuit, and a particle size quality analysis circuit. The central controller 4 is electrically connected to the rotary drive device, the torque sensor 1, the particle size analyzer 2, and the quality sensor 3 respectively. A PID controller electrically connected to the central controller 4 can be set in the electric control box 5.

[0079] When drilling a gas extraction borehole using the present coal seam gas extraction drilling high-efficiency drilling system, the pressure gas channel of the drill pipe is connected to the pressure gas source. During the drilling process through the drill pipe, the returned chips generated by the rock breaking drill bit can be discharged through the spiral blades. Since the thickness of the spiral blades is gradually reduced from the root to the edge, a wedge-shaped extrusion surface can be formed on the surface of the spiral blades during the rotation of the drill pipe. Since crushing teeth extending forward are provided on the outer edge of the spiral blades, the crushing teeth can cut the returned chips during the rotation of the drill pipe. The combined action of rotary cutting and extrusion can make the layers of large returned chips smaller. The rough crushing section with large lead can cooperate with the cutting teeth of the rock breaking drill bit to form a primary crushing cavity, and can crush large pieces of returning chips by generating axial extrusion force through the spiral rise angle of the blade, so as to realize preliminary rough crushing of large pieces of returning chips and realize rapid chip removal through its large lead to avoid the returning chips being blocked behind the rock breaking drill bit. The transition section with medium lead can further crush the returning chips, and the dense conveying section with small lead can finely grind the returning chips, and form a dual-mode dense conveying of spiral mechanical conveying and pneumatic boosting through the blade gap and the pneumatic boosting hole, which can greatly reduce the chance of coal and rock returning chips clogging the drill hole.

[0080] In order to further reduce the probability of coal rock cuttings blocking the drill hole, as a further improvement of the present invention, Figure 2 As shown, the outer diameter of the spiral blades of the transition section or dense conveying section gradually decreases from back to front, so that the entire transition section or dense conveying section has a conical structure with a small front and a large back. With such a setting, local vortices can be generated through cross-sectional mutations, thereby destroying the accumulation structure of the returned chips, and further reducing the probability of coal rock returned chips clogging the drill hole.

[0081] For the return chips with high viscosity, in order to reduce the probability of coal rock return chips clogging the drill hole, as a further improvement scheme of the present invention, the spiral blades can be provided with a radial guide structure from the root to the edge. The guide structure can be a groove structure or a ridge structure. With such a setting, the rotating centrifugal force of the drill rod can be used to achieve directional stripping of the return chips with high viscosity, thereby reducing the probability of coal rock return chips clogging the drill hole.

[0082] In order to further reduce the probability of coal rock chips clogging the drill hole, as a further improvement of the present invention, the spiral blades of the transition section can be provided with turbulent grooves with the opening facing forward. With this arrangement, the turbulent effect in the turbulent grooves can be utilized to cause local high-frequency collisions of the chips, thereby allowing the chips with larger particle sizes to be crushed for a second time.

[0083] In order to reduce the crushing energy consumption of the spiral blades when crushing the returned chips, as a further improvement of the present invention, the crushing teeth on the outer edge of the spiral blades can be made of carbide, and the crushing teeth are tilted forward (that is, have a forward tilt angle) and have an obtuse-angle tooth top structure. Such a setting can enable the crushing teeth to apply directional pressure along the cracks of large pieces of returned chips, thereby effectively reducing the crushing energy consumption.

[0084] The crushed chips can enter the chip collection box through the sealing and guiding device. During this process, the particle size analyzer 2 and the quality sensor 3 measure the particle size and quality of the crushed chips in real time. During the drilling process, the central processing unit 4 monitors and records the drilling parameters in real time, and analyzes the data fed back by the particle size analyzer 2 and the quality sensor 3 in real time. Then, the working parameters of the drill pipe are adjusted according to the data analysis results. The details are as follows:

[0085] Step 1: Quantitatively evaluate the effect of the returned chips after crushing: The central controller analyzes the crushing effect and mass transfer efficiency of the returned chips based on the particle size and mass of the returned chips to quantitatively evaluate the effect of the returned chips after crushing, as follows:

[0086] Step 1-1, quantitative evaluation of particle size:

[0087] The particle size analyzer 2 is used to dynamically scan the drill cuttings sample to obtain the particle size spectrum data. The central processing unit 4 divides the particle size spectrum into m statistical windows and records the particle count N in each window. m , and the relative frequency method is used to calculate the particle size distribution probability, the formula is as follows:

[0088]

[0089] Where: p m is the frequency of particles in the mth statistical window; N is the total number of particles in the returned chip sample; d * m is the normalized particle size value of the mth statistical window; d i is the average particle size of the mth statistical window; μ d is the mean particle size; σ d is the particle size standard deviation.

[0090] The parameters λ and β of the Rosin-Rammler distribution are fitted by the measured data, and the measured cumulative distribution value F is calculated. ovs (d j ), define the residual function R, and use the optimization algorithm to solve λ and β. The formula is as follows:

[0091]

[0092] Where: R(λ,β) is the residual function, which is used to measure the difference between the measured and theoretical distributions; Fobs (d j ) is the measured cumulative distribution value; d m is the representative particle size value of the mth particle size interval; k is the total number of particle size intervals; λ is the characteristic particle size parameter of the Rosin-Rammler distribution; β is the shape parameter of the Rosin-Rammler distribution; is the optimal parameter obtained by the optimization algorithm to minimize the residual function.

[0093] The Rosin-Rammler distribution function is used to describe the particle size distribution after particle crushing. The cumulative distribution function F(d) represents the proportion of particles with a particle size less than or equal to d, and the probability density function f(d) represents the probability density of the occurrence of particles with a particle size of d. The formula is as follows:

[0094]

[0095] d x =λ(-ln(1-x / 100)) 1 / β

[0096] Where: d is the particle size of the returned chips; x is the quantile percentage; d x is the particle size value corresponding to the cumulative probability of x%, indicating that the particle size is smaller than d x The proportion of particles is x%.

[0097] Calculate d 10 d 50 d 90 The particle size uniformity index U is obtained by the following formula:

[0098]

[0099] Where: d 10 d 50 d 90 They represent the particle sizes corresponding to the 10%, 50%, and 90% quantiles in the cumulative distribution respectively; U is the uniformity index. The smaller the U value, the more uniform the particle distribution after crushing and the better the crushing effect of the returned chip particle size.

[0100] Step 1-2, quality assessment:

[0101] The mass sensor 3 measures the total mass of the returned chip sample, calculates the returned chip mass ratio ε to quantify the matching degree between the actual returned chip mass and the theoretical crushing mass, and analyzes the returned chip mass ratio to obtain the returned chip transport efficiency.

[0102] The calculation formula of the return chip mass ratio ε is as follows:

[0103]

[0104] Where: ρ is the density of the returned cuttings sample; V is the theoretical drilling volume; M is the total mass of the returned cuttings sample.

[0105] Step 1-3, comprehensive evaluation:

[0106] To eliminate the dimension difference, d 50 , particle size uniformity index U and return chip mass ratio ε are normalized, and the formula is as follows:

[0107]

[0108] Where: D 50 * d 50 Normalized index value; U * is the normalized value of the particle size uniformity index U; ε * is the normalized index value of the returned chip mass ratio ε; D 50min and D 50max are the historical experimental data 50 The minimum and maximum values ​​of U min and U max are the minimum and maximum values ​​of the particle size uniformity index U in the historical experimental data; ε min and ε max are the minimum and maximum values ​​of the returned chip mass ratio ε in the historical experimental data;

[0109] The weights are adaptively adjusted according to the lithology, and the modified power function weighting method is used to construct a comprehensive evaluation index of return chip crushing through the crushing efficiency index CEI. The formula is as follows:

[0110] CEI=(D * 50 ) ω1 (U * ) ω2 (ε * ) ω3

[0111] Where: CEI is the crushing efficiency index; ω1, ω2, ω3 are d 50 , the weight constraints of particle size uniformity index U and return chip mass ratio ε satisfy ω1+ω2+ω3=1, and the default initial values ​​ω1=0.4, ω2=0.3, ω3=0.3.

[0112] Step 2, based on the evaluation results of Step 1, taking the working condition parameters of the drill rod as the variable sequence and the broken chip evaluation index as the reference sequence, the grey correlation degree analysis (GRA) is used to determine the key parameters. First, linear normalization processing is performed on the working condition parameters, and the broken chip evaluation index is taken as the reference sequence. On this basis, the correlation degree of the working condition parameters and the reference sequence is calculated, and then the influence weight of each working condition parameter on the broken chip evaluation index is determined, and then the influence degree of each index on the broken chip effect is obtained according to the sum of the weights of each physical index.

[0113] The formula of linear normalization processing is as follows:

[0114]

[0115] In the formula, X * is the normalized data; X i is the input data; X min is the minimum value in the input sample data; and X max is the maximum value in the input sample data.

[0116] The calculation formula of the correlation coefficient η ij is as follows:

[0117]

[0118] In the formula, δ ij is the absolute difference value of the i-th parameter sequence at the j-th point; Δ min is the global minimum difference; Δ max is the global maximum difference; and τ is the dynamic sensitivity adjustment factor.

[0119] The calculation formula of the correlation degree H is as follows:

[0120]

[0121] In the formula, n is the length of the data sequence.

[0122] The calculation formula of the weight H i is as follows:

[0123]

[0124] In the formula, a is the number of working condition parameters.

[0125] Step 3: A GA-BP neural network, using a genetic algorithm (GA) to optimize the initial weights and biases of the BP neural network based on different drilling parameter combinations, predicts the chip crushing effect. Key drilling parameters serve as the input layer, chip crushing assessment indicators serve as the output layer, and two hidden layers are set in the middle to determine the BP neural network structure. The number of weight and bias combinations for each parameter is then determined, and a genetic algorithm is used to optimize the initial weights and biases of the BP neural network. The MSE is chosen as the fitness function to calculate the fitness of the individual. The genetic algorithm finds the individual with the optimal fitness through selection, crossover, and mutation operations. Finally, the BP neural network uses the genetic algorithm to obtain the optimal individual, assigns the initial weights and biases to the network, and uses the collected data for training and prediction. By gradually optimizing the model, a prediction model is formed that provides feedback on chip crushing assessment results based on drilling parameters.

[0126] The calculation formula of fitness function MSE is as follows:

[0127]

[0128] Where: y i is the measured value; is the predicted value; M is the total number of training samples.

[0129] Step 4: Based on the prediction model of Step 3, the prediction result of chip crushing is output, and the chip crushing effect evaluation index error e(t) and torque error are calculated. Then, the central processing unit 4 adjusts the coefficients of proportion, integration and differentiation according to the chip crushing effect evaluation index error e(t), and outputs the automatic control signal u(t) to dynamically adjust the drilling parameters so that the chip crushing effect reaches the expected target.

[0130] Torque error can serve as an independent safety threshold monitoring signal. Excessive torque can cause the drill bit to jam or the equipment to overload, affecting chip return efficiency; insufficient torque may indicate insufficient drilling pressure, resulting in inadequate chip return. When the torque exceeds the limit, protective actions such as speed reduction and shutdown are directly triggered without waiting for e(t) feedback. Dynamic monitoring of torque error changes assists in optimization and adjustment. When the torque error is high, the differential term is temporarily increased to suppress overshoot; when the torque error is low, the integral term is emphasized to eliminate steady-state error.

[0131] The chip crushing effect evaluation index (CEI) error is the difference between the theoretical value and the actual value. The calculation formula of the chip crushing effect evaluation index error e(t) is as follows:

[0132] e(t)=CEI(t) * -CEI(t)

[0133] Where: CEI(t) * is the target value of the chip crushing effect; CEI(t) is the predicted value of the neural network.

[0134] The calculation formula of the automatic control signal u(t) is as follows:

[0135]

[0136] Where: u(t) is the automatic control signal, where t is time; e(t) is the error of the chip crushing effect evaluation index, where t is time; K P , K I , K D are the proportional coefficient, integral coefficient, and differential coefficient respectively.

[0137] A multivariable decoupled PID controller electrically connected to the central controller 4 can be installed within the electrical control box 5. Independent PID controllers can be designed for each parameter, and interactions can be eliminated through a coupled compensation matrix. To address the delay between parameter adjustment and the change in chip return effect, a Smith predictor is added before the PID controller to precalculate the error after the delay and compensate for the GA-BP model prediction delay. To address issues such as drill sticking, the torque error is dynamically monitored. When torque anomalies occur, the control mode is switched and drilling parameters are adjusted to prevent drill sticking in advance. A neural network prediction model is combined with a PID controller to achieve efficient fragmentation and intelligent regulation of the drilling process.

[0138] This coal seam gas extraction drilling efficient drilling system utilizes the crushing and transportation effects of spiral blades to achieve fine crushing of particles in the chip return process and efficient transportation and unblocking of chip particles. At the same time, it can dynamically monitor the drill rod working parameters, the particle size and quality of the chip return after crushing. Through quantitative evaluation of the chip return crushing effect, grey correlation analysis GRA is used to determine the key parameters. Then, GA-BP neural network is used to predict the chip return crushing effect according to different drilling parameter combinations. Finally, the chip return crushing effect evaluation index error is calculated, and the coefficients of proportion, integration and differentiation are adjusted according to the chip return crushing effect evaluation index error. Automatic control signals are output to dynamically adjust the drilling parameters so that the chip return effect reaches the expected target. It can achieve efficient crushing and intelligent regulation of the drilling process under the premise of reducing the probability of coal rock chip return blocking the borehole during the drilling process. It can fundamentally solve the problems caused by chip return blocking the borehole, thereby improving the drilling efficiency of gas extraction drilling.

Claims

1. An efficient drilling system for coal seam gas extraction, comprising a drilling rig, a material guide, a power torque detection unit, a return chip quantitative detection unit, and a centralized electronic control unit; characterized in that: The drilling rig part includes a drill rod, on the rod body of the drill rod is fixed a spiral blade spirally arranged along the axial direction of the rod body, the thickness of the spiral blade gradually decreases from the root to the edge, and the outer edge of the spiral blade is provided with a crushing tooth extending forward, the spiral blade includes at least one group of spiral blade composite structures in the axial direction of the rod body, the spiral blade composite structure includes a coarse crushing section, a transition section and a dense conveying section from front to back, and the leads of the coarse crushing section, the transition section and the dense conveying section are arranged in descending order, and the rod body of the drill rod is also provided with a pressure gas channel arranged along the axial direction of the drill rod, and at least the rod body of the dense conveying section is provided with a pneumatic boosting hole that penetrates the pressure gas channel; The material guiding part includes a sealing material guiding device which can be sleeved on the drill rod and a chip collecting box arranged on the frame; The power torque detection part includes a torque sensor (1) installed on the drill rod; The return chip quantitative detection part comprises a particle size analyzer (2) arranged on the receiving port of the return chip collection box and a quality sensor (3) arranged on the return chip collection box; The centralized electric control part includes an electric control box (5), a central processing unit (4), a drilling control circuit, a power torque monitoring and analysis circuit, and a particle size and quality analysis circuit. The central controller (4) is electrically connected to the rotary drive device, the torque sensor (1), the particle size analyzer (2), and the quality sensor (3).

2. The high-efficiency drilling system for coal seam gas extraction according to claim 1 is characterized in that: The outer diameter of the spiral blades in the transition section or the dense conveying section gradually decreases from the back to the front.

3. The high-efficiency drilling system for coal seam gas extraction according to claim 1 is characterized in that: The spiral blade is provided with a radial flow guide structure from the root to the edge.

4. The high-efficiency drilling system for coal seam gas extraction according to claim 1 is characterized in that: The spiral blades of the transition section are provided with turbulent grooves with openings facing forward.

5. The high-efficiency drilling system for coal seam gas extraction according to claim 1 is characterized in that: The crushing teeth on the outer edge of the spiral blade have a forward tilt angle and an obtuse tooth top structure.

6. A method for controlling efficient drilling of coal seam gas extraction drilling holes based on the efficient drilling system for coal seam gas extraction drilling holes according to claim 1, characterized in that: The specific steps include: Step 1: quantitatively evaluate the effect of the returned chips after crushing; Step 2: Based on the evaluation results of Step 1, the drill pipe working parameters are used as the variable sequence and the chip crushing evaluation index is used as the reference sequence. Grey relational analysis (GRA) is used to determine the key parameters. Step 3: Use GA-BP neural network to predict the chip crushing effect according to different drilling parameter combinations; Step 4, based on the prediction model of Step 3, outputs the prediction result of chip crushing, calculates the error of chip crushing effect evaluation index e(t) and torque error, and then the central processing unit (4) adjusts the coefficients of proportion, integration and differentiation according to the error of chip crushing effect evaluation index e(t), outputs the automatic control signal u(t) to dynamically adjust the drilling parameters, so that the chip crushing effect reaches the expected target.

7. The method for efficiently controlling and drilling coalbed gas extraction drilling according to claim 6, characterized in that: Step 1 is as follows: Step 1-1, quantitative evaluation of particle size: Use the particle size analyzer (2) to dynamically scan the drill cuttings sample to obtain particle size spectrum data. The central processing unit (4) divides the particle size spectrum into m statistical windows and records the particle count N in each window. m , and the relative frequency method is used to calculate the particle size distribution probability, the formula is as follows: Where: p m is the frequency of particles in the mth statistical window; N is the total number of particles in the returned chip sample; d * m is the normalized particle size value of the mth statistical window; d i is the average particle size of the mth statistical window; μ d is the mean particle size; σ d is the particle size standard deviation; The parameters λ and β of the Rosin-Rammler distribution are fitted by the measured data, and the measured cumulative distribution value F is calculated. obs (d j ), define the residual function R, and use the optimization algorithm to solve λ and β. The formula is as follows: Where: R(λ,β) is the residual function; F obs (d j ) is the measured cumulative distribution value; d m is the representative particle size value of the mth particle size interval; k is the total number of particle size intervals; λ is the characteristic particle size parameter of the Rosin-Rammler distribution; β is the shape parameter of the Rosin-Rammler distribution; is the optimal parameter obtained by the optimization algorithm to minimize the residual function; The Rosin-Rammler distribution function is used to describe the particle size distribution after particle crushing. The cumulative distribution function F(d) represents the proportion of particles with a particle size less than or equal to d, and the probability density function f(d) represents the probability density of the occurrence of particles with a particle size of d. The formula is as follows: d x =λ(-ln(1-x / 100)) 1 / β Where: d is the particle size of the returned chips; x is the quantile percentage; d x is the particle size value corresponding to the cumulative probability of x%, indicating that the particle size is smaller than d x The proportion of particles is x%; Calculate d 10 d 50 d 90 The particle size uniformity index U is obtained by the following formula: Where: U is the uniformity index; d 10 d 50 d 90 They represent the particle sizes corresponding to the 10%, 50%, and 90% quantiles in the cumulative distribution, respectively; Step 1-2, quality assessment: The mass sensor (3) measures the total mass of the returned chip sample, and calculates the returned chip mass ratio ε to quantify the matching degree between the actual returned chip mass and the theoretical crushing mass. The returned chip mass ratio is analyzed to obtain the returned chip transport efficiency; The calculation formula of the return chip mass ratio ε is as follows: Where: ρ is the density of the returned cuttings sample; V is the theoretical drilling volume; M is the total mass of the returned cuttings sample; Step 1-3, comprehensive evaluation: 50 , particle size uniformity index U and return chip mass ratio ε are normalized, and the formula is as follows: Where: D 50 * d 50 Normalized index value; U * is the normalized value of the particle size uniformity index U; ε * is the normalized index value of the returned chip mass ratio ε; and are the historical experimental data 50 The minimum and maximum values ​​of U min and U nax are the minimum and maximum values ​​of the particle size uniformity index U in the historical experimental data; ε min and ε max are the minimum and maximum values ​​of the returned chip mass ratio ε in the historical experimental data; The weights are adaptively adjusted according to the lithology, and the modified power function weighting method is used to construct a comprehensive evaluation index of return chip crushing through the crushing efficiency index CEI. The formula is as follows: CEI=(D * 50 ) ω1 (U * ) ω2 (ε * ) ω3 Where: CEI is the crushing efficiency index; ω1, ω2, ω3 are d 50 , the weight constraints of particle size uniformity index U and return chip mass ratio ε satisfy ω1+ω2+ω3=1.

8. The method for efficiently controlling and drilling coalbed gas extraction drilling according to claim 7, characterized in that: Step 2 is as follows: Step 2-1, perform linear normalization on the operating parameters. The formula for linear normalization is as follows: Where: X * is the data after normalization; X i is the input data; X min is the minimum value of the input sample data; X max is the maximum value of the input sample data; Step 2-2: Take the chip crushing evaluation index as the reference sequence, and calculate the correlation between the working condition parameters and the reference sequence, the correlation coefficient η ij The calculation formula is as follows: Where: δ ij is the absolute difference of the i-th parameter sequence at point j; Δ min is the global minimum difference; Δ max is the global maximum difference; τ is the dynamic sensitivity adjustment factor; Correlation The calculation formula is as follows: Where: n is the length of the data sequence; Step 2-3, determine the weight of each working condition parameter on the chip crushing evaluation index, and obtain the influence degree of each index on the roof lane layout according to the sum of the weights of each physical index. The weight H i The calculation formula is as follows: Where: a is the number of operating parameters.

9. The method for efficiently controlling and drilling coalbed gas extraction drilling according to claim 8, characterized in that: Step 3 is as follows: Key drilling parameters are used as the input layer of the GA-BP neural network, and the return chip crushing evaluation index is used as the output layer of the GA-BP neural network. Two hidden layers are set in the middle to determine the BP neural network structure, and then the number of weight and bias combinations of each parameter is determined. A genetic algorithm is used to optimize the initial weights and biases of the BP neural network. The MSE is selected as the fitness function to calculate the fitness value of the individual. The genetic algorithm finds the individual corresponding to the optimal fitness value by selecting crossover and mutation operations. Finally, the BP neural network uses the genetic algorithm to obtain the optimal individual to assign the network initial weights and biases. The collected data is used for training and prediction. By gradually optimizing the model, a prediction model based on drilling parameters is finally formed to feedback the return chip crushing evaluation results. The calculation formula of fitness function MSE is as follows: Where: y i is the measured value; is the predicted value; M is the total number of training samples.

10. The method for efficiently controlling and drilling coalbed gas extraction drilling according to claim 9, characterized in that: In Step 4, the calculation formula of the chip crushing effect evaluation index error e(t) is as follows: e(t)=CEI(t) * -CEI(t) Where: CEI(t) * is the target value of chip crushing effect; CEI(t) is the predicted value of neural network; The calculation formula of the automatic control signal u(t) is as follows: Where: u(t) is the automatic control signal, where t is time; e(t) is the error of the chip crushing effect evaluation index, where t is time; K P , K I , K D are the proportional coefficient, integral coefficient, and differential coefficient respectively.

Images