Search

CN-121993167-A - Coal mine drilling track analysis device and method

CN121993167ACN 121993167 ACN121993167 ACN 121993167ACN-121993167-A

Abstract

The invention belongs to the technical field of coal mine drilling, in particular to a coal mine drilling track analysis device and a method, wherein the device comprises a drill bit, a drill rod and a pipeline; the drill bit is fixedly connected with a connecting section, an empty slot is formed in the connecting section, a damping component is connected with the inner wall of the empty slot, the damping component is connected with an outer frame, a position sensor, a depth sensor, an angle sensor and a speed sensor which are electrically connected with a drilling equipment control processor are arranged in the outer frame, a buffer body is connected with the outer surfaces of the position sensor, the depth sensor, the angle sensor and the speed sensor, drilling data are collected in real time by the position sensor, the depth sensor, the angle sensor and the speed sensor, the drilling data comprise position information, drilling depth, drill bit angles and drilling speeds at all time points, and the drilling data are transmitted to the drilling equipment control processor, and the drilling equipment control processor adjusts drilling operation according to the obtained data so as to follow optimized tracks.

Inventors

  • YU XUELEI
  • Jin Weihu
  • CHEN HUITING
  • WANG HAOXING
  • Pan Yanqi

Assignees

  • 淮北矿业股份有限公司

Dates

Publication Date
20260508
Application Date
20251222

Claims (10)

  1. 1. The coal mine drilling track analysis device comprises a drill bit (1), a drill rod (2) and a pipeline (4), and is characterized in that a connecting section (3) is fixedly connected between the drill bit (1) and the drill rod (2), and the connecting section (3) is provided with a hollow groove (5); the damping component is connected with the inner wall of the empty groove (5); the damping component is connected with an outer frame (7), and a position sensor (10), a depth sensor (11), an angle sensor (12) and a speed sensor (13) which are electrically connected with the drilling equipment control processor are arranged in the outer frame (7); the buffer body (92), the buffer body (92) is connected with the outer surfaces of the position sensor (10), the depth sensor (11), the angle sensor (12) and the speed sensor (13).
  2. 2. The coal mine drilling track analysis device according to claim 1, wherein the damping components are provided with a plurality of groups and are uniformly distributed on the outer surface of the outer frame (7), the damping components comprise springs (81) connected with the inner wall of the empty groove (5), fixed rods (82) fixedly connected with the inner wall of the empty groove (5) are sleeved in the springs (81), limiting grooves (83) are formed in the fixed rods (82), movable rods (84) are clamped in the limiting grooves (83), and one ends, far away from the inner wall of the empty groove (5), of the movable rods (84) are fixedly connected with the outer frame (7).
  3. 3. The coal mine drilling track analysis device according to claim 2, wherein the limiting groove (83) is arranged in a stepped mode, one end, away from the outer frame (7), of the movable rod (84) is fixedly connected with a limiting plate (85), and the limiting plate (85) is arranged inside the limiting groove (83).
  4. 4. A coal mine borehole trajectory analysis device as claimed in claim 3, wherein said buffer body (92) is of soft rubber material.
  5. 5. The coal mine drilling track analysis device according to claim 4, wherein the buffer body (92) is hollow, the drill bit (1), the connecting section (3) and the drill rod (2) are connected with the pipeline (4) together, the connecting section (3) is symmetrically provided with the through groove (6), the top of the buffer body (92) is connected with the drainage tube (91) communicated with the pipeline (4), the drainage tube (91) is connected with the water pump (94), and the bottom of the buffer body (92) is connected with the return pipe (93) communicated with the pipeline (4).
  6. 6. A coal mine borehole trajectory analysis method employing the coal mine borehole trajectory analysis device according to any one of claims 1 to 5, comprising the steps of: S1, acquiring position information, drilling depth, drill bit angle and drilling speed of each time point of drilling in real time from a position sensor (10), a depth sensor (11), an angle sensor (12) and a speed sensor (13) in the working process of coal mine drilling equipment, and transmitting the position information, the drilling depth, the drill bit angle and the drilling speed to a drilling equipment control processor to construct a drilling track data set; s2, preprocessing an acquired drilling track data set, including data cleaning, missing value filling and noise filtering; s3, constructing a dynamic three-dimensional geological model of the drilling area by combining the preprocessed drilling track data set and geological exploration data acquired in advance through a three-dimensional geological modeling technology; S4, designing a multi-layer self-adaptive neural network model: The self-adaptive drilling path prediction module is used for providing geological environment information through a dynamic three-dimensional geological model, performing path prediction on real-time drilling data by utilizing a multi-layer long-short-term memory network and combining an attention mechanism, and predicting an optimal drilling path under the geological environment; the anomaly detection and processing module is used for designing an anomaly detection algorithm based on generation of an countermeasure network, detecting anomaly data possibly occurring in the drilling process in real time, and providing a countermeasure strategy by generating an countermeasure network simulation anomaly scene; s5, inputting the preprocessed drilling track data set into a multi-layer self-adaptive neural network model for training, and optimizing the prediction precision and the real-time response capability of the multi-layer self-adaptive neural network model through repeated iteration and optimization; S6, optimizing a drilling path through an optimization algorithm based on an output result of the multi-layer self-adaptive neural network model, and generating an optimized drilling track; and S7, feeding the optimized drilling track back to a drilling equipment control processor in real time, and adjusting the drilling operation to follow the optimized track.
  7. 7. The method for analyzing coal mine drilling trajectories according to claim 6, wherein S1 comprises the steps of: S11, acquiring drilling data in real time through the position sensor (10), the depth sensor (11), the angle sensor (12) and the speed sensor (13) in the working process of the coal mine drilling equipment, wherein the drilling data comprise position information, drilling depth, drill bit angle and drilling speed at each time point: Drilling position information (x (t), y (t), z (t)) represents the spatial coordinates of the drilling apparatus at a point in time t, where x (t) and y (t) are horizontal coordinates and z (t) is a vertical depth coordinate; The drilling depth d (t) is the vertical depth of the drill bit at the time point t, and the unit is meter; angle of drill bit The inclination angle of the drill bit relative to the vertical direction at the time point t is in degrees; The drilling speed v (t) is the feeding speed of the drill bit at the time point t, and the unit is meter/second; S12, organizing drilling data into a drilling track data set according to time sequence: ; where T represents a time series set of drilling operations and D represents a drilling trajectory dataset.
  8. 8. The method for analyzing coal mine drilling trajectories according to claim 7, wherein S3 comprises the steps of: S31, acquiring geological exploration data collected in advance and formatting, wherein the geological feature parameters are obtained The geologic feature parameters are described in terms of multiple dimensions, including spatial location, material properties, and stress states, with the following expressions being constructed: ; Wherein, the Representing geological feature parameters The coordinates of the spatial location where it is located, Is a geological feature parameter Comprising m material properties, the material properties comprising density, porosity and permeability, Is a geological feature parameter Describing stress distribution of the geological feature parameters in x, y and z directions; S32, matching the drilling track data set with the geological data set to enable the geological model to reflect the relation between the actual drilling track and the geological environment at different time points and different positions, wherein the matching process is described by the following formula, and corresponding mapping is carried out between the drilling data and the geological feature parameters: ; Wherein M (x, y, z, t) represents a geologic model value at a time t, in three-dimensional space coordinates (x, y, z), For time-dependent weighting functions, for adjusting the degree of matching of borehole trajectory data and geologic features, Is an interpolation kernel function for determining the position of a geological parameter according to the coordinates (x (t), y (t), z (t)) of a drilling track The distance between them to calculate the effect of the geologic model, A vector representing the current borehole trajectory spatial location; s33, constructing a dynamic three-dimensional geological model based on an adaptive grid modeling technology, wherein the adaptive grid modeling technology dynamically adjusts the grid density of the model, and the adjusting process is combined with drilling track data: ; Wherein, the Represents the grid density adjustment value on three-dimensional space coordinates (x (t), y (t), z (t)) at time t, Is the scaling factor of the adaptive mesh adjustment, Representing the gradient of the geologic model at the spatial location of the borehole trajectory, the greater the gradient, the corresponding increase in grid density should be, Is a spatially distributed function representing grid density requirements in a spatial region of the borehole trajectory, Is a decay function based on key geological structures, A position vector representing a key geological structure, The decay rate is controlled.
  9. 9. The method for analyzing coal mine drilling trajectories according to claim 8, wherein S4 comprises the steps of: s41, designing an adaptive drilling path prediction module, wherein the adaptive drilling path prediction module provides geological environment information through a dynamic three-dimensional geological model M (x, y, z and t), and performs path prediction on real-time drilling data by utilizing a multi-layer long-short-term memory network and combining an attention mechanism; s42, designing an abnormality detection and processing module, wherein the abnormality detection and processing module is used for detecting and processing abnormal data possibly occurring in the drilling process based on the generation countermeasure network: The generation network G (z) generates simulated borehole data by inputting noise z Where z is a random variable sampled from the noise distribution; Discriminating between actual borehole data and generated simulation data by the network D (x) As an input, outputting a probability P (D) of distinguishing actual data from analog data; Generating a loss function of the network G (z) and the discrimination network D (x) by optimizing such that the generated analog data The method can approximate actual drilling data so as to identify and process abnormal data possibly occurring in the drilling process, and the abnormal detection standard is as follows: ; Wherein, the In order to generate a loss function against the network, For the distribution of the actual borehole data, Is the noise distribution; S43, generating possible abnormal scenes of the countermeasure network simulation according to the detected abnormal data, and providing corresponding countermeasure strategies; The path prediction comprises the steps of taking real-time drilling data and a dynamic three-dimensional geological model as inputs, inputting the inputs into a multi-layer long-short-term memory network, and capturing the geological change mode in a time sequence and the historical influence of drilling operation; Applying attention mechanisms Hidden state for multi-layer long-short-term memory network output Weighting to obtain context vector of predicted drilling path : ; Wherein, the The attention weight of the kth time point reflects the contribution of the current time point to the path prediction; Context-based vector And current drilling data to generate the optimal drilling path prediction result of the next step ; The generation of the optimal drilling path prediction result comprises the following steps: context vector And drilling data at the current point in time t Inputting the drill path information into a path prediction function, and predicting the next drill path by combining the time sequence and the spatial position information through a nonlinear mapping relation : ; Wherein, the And Respectively context vectors And borehole data Representing the weights of the path prediction function in the time series and spatial dimensions, b being a bias term for adjusting the output of the function to increase flexibility, Is a time-dependent amplitude adjustment parameter, is associated with periodic fluctuations in borehole path prediction, As a function of the corresponding time period, Is an superposition of the historical spatial location vectors, for introducing the influence of the historical path on the current prediction, Is an activation function, for introducing a nonlinear characteristic, Is a time dependent noise term used to simulate uncertainty in the actual drilling environment; nonlinear transformation is carried out on the output result of the path prediction function by utilizing the activation function, so that the optimal drilling path prediction result of the next step is obtained : ; Wherein, the And For the adjustment coefficients in the nonlinear transformation, respectively corresponding to the tanh and ReLU activation functions, for balancing the contributions of the two activation functions to the path prediction, Is the weight coefficient of the hidden layer output, For the hidden state output at the point in time t, And ReLU is an activation function for adding nonlinear processing.
  10. 10. The method for analyzing coal mine drilling trajectories according to claim 9, wherein S7 comprises the steps of: S71, predicting an optimal drilling path by using a multi-layer adaptive neural network model and an optimization algorithm Converting into executable drilling control instructions, particularly comprising predicting optimal drilling path results Conversion to bit angle Feed speed And a drilling depth adjustment value Control parameters: Wherein, the The initial bit angle, initial feed rate and initial drilling depth, For the corresponding adjustment coefficient(s), 、 And Gradients of the optimal path in x, y and z directions, respectively; S72, sending the calculated control instruction to a drilling equipment control processor through a real-time control interface, and regulating and controlling the rotation, feeding speed and drilling depth units of the drill bit (1) by the drilling equipment control processor; s73, after receiving a control instruction, the drilling equipment control processor adjusts the angle, the feeding speed and the drilling depth of the drill bit to enable the actual drilling track Approaching the optimal drilling path prediction result after optimization : Wherein, the Is the position vector of the actual drilling trajectory at time t+1; s74, monitoring feedback data of the drilling equipment in real time, and dynamically adjusting a control instruction of the next step according to the feedback data to ensure continuity of drilling operation and accuracy of tracks, wherein the feedback data comprises actual drilling angle, feeding speed and depth change values: ; Wherein, the Representing the difference between the feedback data and the desired control command, In order to feed back the data, For the calculated control instruction; s75, according to the difference And correspondingly adjusting the control parameters to continuously optimize the drilling operation.

Description

Coal mine drilling track analysis device and method Technical Field The invention belongs to the technical field of coal mine drilling, and particularly relates to a coal mine drilling track analysis device and method. Background In the coal mine exploitation process, the accuracy and efficiency of the drilling track are directly related to the exploitation quality and safety of the coal mine, and the prior art relies on manual experience and basic measuring tool combined part automation means in the coal mine exploitation process and on a preset fixed path and operation mode. However, the technology often has the following defects that potential safety hazards possibly existing in drilling operation cannot be accurately pre-warned in real time, such as deviation of a drill bit from a preset track, and operators often have difficulty in timely finding and processing the potential safety problems due to lack of effective real-time monitoring and data analysis means, so that the risk of drilling operation is increased, and serious safety accidents can be even caused. Disclosure of Invention In order to make up for the deficiencies of the prior art, at least one technical problem presented in the background art is solved. The technical scheme includes that the coal mine drilling track analysis device comprises a drill bit and a drill rod, a connecting section is fixedly connected between the drill bit and the drill rod, a hollow groove is formed in the connecting section, a shock absorption component is connected with the inner wall of the hollow groove, an outer frame is connected with the shock absorption component, a position sensor, a depth sensor, an angle sensor and a speed sensor which are electrically connected with a drilling equipment control processor are arranged in the outer frame, and a buffer body is connected with the outer surfaces of the position sensor, the depth sensor, the angle sensor and the speed sensor. Preferably, the damping components are provided with a plurality of groups and are uniformly distributed on the outer surface of the outer frame, the damping components comprise springs connected with the inner wall of the empty groove, fixing rods fixedly connected with the inner wall of the empty groove are sleeved in the springs, limiting grooves are formed in the fixing rods, movable rods are clamped in the limiting grooves, one ends, far away from the inner wall of the empty groove, of the movable rods are fixedly connected with the outer frame, the limiting grooves are arranged in a stepped mode, one ends, far away from the outer frame, of the movable rods are fixedly connected with limiting plates, and the limiting plates are arranged inside the limiting grooves. Preferably, the buffer body is made of soft rubber, the buffer body is hollow, the drill bit, the connecting section and the drill rod are connected with a pipeline together, the connecting section is symmetrically provided with through grooves, the top of the buffer body is connected with a drainage tube communicated with the pipeline, the drainage tube is connected with a water pump, and the bottom of the buffer body is connected with a return tube communicated with the pipeline. A coal mine drilling track analysis method comprises the following steps: S1, acquiring position information, drilling depth, drill bit angle and drilling speed of each time point of drilling in real time from the position sensor, the depth sensor, the angle sensor and the speed sensor in the working process of the coal mine drilling equipment, and transmitting the position information, the drilling depth, the drill bit angle and the drilling speed to a drilling equipment control processor to construct a drilling track data set; s2, preprocessing an acquired drilling track data set, including data cleaning, missing value filling and noise filtering; s3, constructing a dynamic three-dimensional geological model of the drilling area by combining the preprocessed drilling track data set and geological exploration data acquired in advance through a three-dimensional geological modeling technology; S4, designing a multi-layer self-adaptive neural network model: The self-adaptive drilling path prediction module is used for providing geological environment information through a dynamic three-dimensional geological model, performing path prediction on real-time drilling data by utilizing a multi-layer long-short-term memory network and combining an attention mechanism, and predicting an optimal drilling path under the geological environment; the anomaly detection and processing module is used for designing an anomaly detection algorithm based on generation of an countermeasure network, detecting anomaly data possibly occurring in the drilling process in real time, and providing a countermeasure strategy by generating an countermeasure network simulation anomaly scene; s5, inputting the preprocessed drilling track data set into a multi-layer self-adapt