Tunnel face borehole intelligent projection positioning and drilling precision control system

Abstract

The application discloses a tunnel face borehole intelligent projection positioning and drilling precision control system, belongs to the tunnel drilling and blasting construction technical field, and aims to solve the technical problems of inaccurate tunnel face borehole positioning, large drilling direction deviation and low construction efficiency. Technical scheme points are as follows: the control system comprises a structured light projector, at least two industrial cameras, a fiber shape demodulation module and a controller which are installed on a tunnel drilling and blasting vehicle, a fiber shape sensing unit is arranged in a drill rod at the end of a drill arm of the drilling and blasting vehicle, the structured light projector projects structured light, the industrial cameras collect binocular images, the fiber shape demodulation module calculates the curvature and the twist angle sequence of the sensing unit, and the controller controls the drill arm driving mechanism to correct the position and direction of the drill bit through three-dimensional reconstruction, coordinate comparison and real-time deviation adjustment. The application realizes intelligent positioning of the tunnel face borehole and drilling precision control, and improves the accuracy and efficiency of the tunnel drilling and blasting construction.

Description

Technical Field

[0001] This invention relates to the field of tunnel drilling and blasting construction technology. More specifically, this invention relates to an intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face. Background Technology

[0002] During tunnel drilling and blasting construction, the positioning accuracy of the blast holes and the accuracy of the drilling direction directly affect the drilling and blasting effect, which in turn relates to the safety and efficiency of tunnel construction. Currently, the positioning of blast holes and drilling operations at the tunnel face mostly rely on manual operation. Construction workers mark the positions of the blast holes on the tunnel face according to the design drawings, using tools such as measuring tapes and total stations, and then rely on experience to adjust the direction of the drill arm to carry out the drilling operation.

[0003] Current construction methods lack effective real-time monitoring and precise control, making it impossible to achieve automated and intelligent linkage between borehole positioning and the drilling process. On the one hand, there is a lack of technical means to quickly and accurately acquire the three-dimensional topography of the tunnel face and the actual spatial coordinates of the designed borehole positions. Manual measurement and marking are difficult to avoid environmental interference and operational errors, easily leading to deviations between the actual drilling positions and the designed positions, affecting subsequent blasting effects and potentially increasing subsequent support workload and construction costs. On the other hand, it is impossible to monitor the spatial attitude of the drill rod and the position and direction of the drill bit in real time. During drilling, the drill rod is easily affected by factors such as rock resistance and drill arm vibration, causing deviations that are difficult for construction personnel to detect and correct in a timely manner, potentially reducing blasting efficiency and even causing safety hazards. Therefore, there is an urgent need to provide an intelligent projection positioning and drilling accuracy control system for tunnel face boreholes. Summary of the Invention

[0004] Another objective of this invention is to provide an intelligent projection positioning and drilling accuracy control system for blast holes at tunnel faces. Traditional manual positioning drilling suffers from large errors in blast hole location marking and difficulty in accurately controlling the drilling direction. Furthermore, it lacks real-time and accurate monitoring methods for the spatial attitude of the drill rod, making it impossible to achieve real-time closed-loop correction of the drill bit position and drilling direction. This results in low drilling accuracy, severe over-excavation and under-excavation, affecting construction quality and efficiency. This invention aims to achieve intelligent projection positioning of blast holes, real-time perception of drill rod attitude, and automatic control of drilling accuracy, thereby improving the precision of drilling positioning and direction control.

[0005] Relying solely on hydraulic pressure and propulsion speed to determine rock strata interfaces is susceptible to equipment operating conditions and construction interference, resulting in insufficient identification accuracy and potential misjudgments of rock strata interface location, thus affecting the accuracy of advance compensation. This invention introduces vibration sensing signals to assist in the identification of rock strata interfaces. By utilizing sudden changes in vibration energy, the reliability of interface identification is improved. Furthermore, by fusing and comparing the results with hardness identification, the positioning accuracy of rock strata interfaces is enhanced, ensuring the reliable execution of the advance compensation strategy.

[0006] Single vibration energy characteristics can still lead to misjudgments under complex geological conditions, making it difficult to accurately distinguish rock strata interfaces from interference such as local fracture zones and isolated boulders. This invention acquires vibration wave velocity changes through multiple vibration sensors, further improves interface identification accuracy by utilizing the differences in wave velocity propagation characteristics in different rock strata, and enhances the robustness of position determination through weighted fusion, providing a more accurate basis for borehole pre-correction.

[0007] The undulating terrain and local unevenness of the tunnel face can lead to significant errors in the direct mapping of borehole design coordinates. Traditional simple projection methods are difficult to adapt to complex curved surfaces, easily resulting in unreasonable borehole placement and uneven borehole spacing. This invention improves the rationality of borehole placement and accuracy on complex rock surfaces by constructing a curved surface model of the tunnel face, calculating the terrain complexity, and processing the borehole mapping positions in a hierarchical manner.

[0008] When optimizing boreholes individually in areas with complex terrain, the actual spacing between boreholes often deviates significantly from the design value, affecting the uniformity of blasting. This invention uses iterative optimization of geodesic distances to ensure that the surface spacing between boreholes meets the design requirements, achieving adaptive correction of borehole positions and making the actual borehole layout more consistent with the blasting design intent.

[0009] Single-scale neighborhood calculations of terrain complexity are easily affected by local concavity and convexity noise, failing to accurately reflect the true terrain undulations of the working face and leading to errors in borehole location judgment. This invention employs multi-scale nested neighborhood analysis of curvature variance changes to improve the stability and accuracy of terrain complexity evaluation, providing a more reliable basis for borehole location judgment.

[0010] Drill pipes undergo elastic deformation under the influence of thrust and rock resistance. Correcting this solely based on the current drill bit position is prone to subsequent trajectory deviation due to the release of elastic deformation, resulting in control lag. This invention establishes a drill pipe elastic deformation deflection model to predict future trajectory deviation trends and implement proactive corrections, thereby improving the long-term accuracy of borehole trajectory control.

[0011] To achieve these objectives and other advantages according to the present invention, a smart projection positioning and drilling accuracy control system for blast holes at a tunnel face is provided, comprising:

[0012] The structured light projector, at least two cameras, demodulation module and controller are all mounted on the drilling and blasting trolley; the drill arm of the drilling and blasting trolley is connected to the drill rod, and the drill rod is equipped with a sensing unit, which is connected to the demodulation module.

[0013] The structured light projector is used to project structured light onto the drill pipe and the working face, the camera is used to acquire binocular images of the surface including the drill pipe and the working face, and the demodulation module is used to calculate the curvature value sequence and torsion angle value sequence of each point along the sensing unit in real time.

[0014] The controller has pre-stored the design coordinate data of the borehole and the design drilling direction vector data. The controller is used to perform the following operations:

[0015] Three-dimensional reconstruction is performed based on the binocular images acquired before the drill bit contacts the working face to obtain the initial spatial curve of the drill rod center axis and the three-dimensional shape of the working face, and the actual spatial coordinates of the designed borehole position on the working face are calculated; the initial installation orientation parameters of the sensing unit are calibrated based on the initial spatial curve.

[0016] After the drill bit contacts the face, based on the initial installation orientation parameters, curvature value sequence, and torsion angle value sequence, a continuous spatial curve of the drill pipe center axis from the near end of the drill pipe to the drill bit position is reconstructed.

[0017] The real-time three-dimensional spatial coordinates and drilling direction vector values ​​of the drill bit center point are calculated based on the continuous spatial curve, and compared with the actual spatial coordinates and the designed drilling direction vector values ​​to calculate the real-time radial deviation and angular deviation values. When the real-time radial deviation or angular deviation value exceeds the preset threshold range, the controller sends an adjustment command to the drive mechanism of the drill arm to make the drill bit center point return to the preset threshold range.

[0018] Preferably, the controller is also configured to:

[0019] Real-time oil pressure data and drill pipe advance speed data of each joint of the drill arm are continuously collected according to the preset sampling cycle;

[0020] Based on oil pressure and propulsion speed data, the oil pressure change rate and propulsion speed change rate are calculated in real time. If the oil pressure change rate and propulsion speed change rate are negatively correlated and both exceed their respective set thresholds in the current sampling period, it is determined that the rock formation hardness has changed in the current sampling period, and the ratio of oil pressure to propulsion speed in the current sampling period is recorded as the rock formation hardness characterization value. If the oil pressure change rate and propulsion speed change rate are not negatively correlated in the current sampling period, or although they are negatively correlated but do not exceed their respective set thresholds, it is determined that the rock formation hardness is stable in the current sampling period, and the controller maintains the current control parameters of the drill arm.

[0021] Based on the rock hardness characterization values ​​recorded in multiple consecutive sampling periods, combined with drill pipe advance displacement data, a continuous distribution curve of rock hardness along the drilling direction is established; when it is determined that the rock hardness changes during the current sampling period, the drill bit position corresponding to that sampling period is identified as the interface between soft and hard rock layers, and the degree of hardness difference is estimated based on the hardness characterization values ​​on both sides of the interface.

[0022] Based on the degree of hardness difference on both sides of the rock layer interface, and combined with the drilling mechanics model, the direction and amount of deviation that the drill bit will produce when crossing the interface are predicted.

[0023] While identifying the rock interface, the controller calculates the advance compensation amount and compensation timing based on the predicted deviation direction and amount, combined with the mechanical response delay parameters of the drill arm drive mechanism; the compensation direction is set to be opposite to the predicted deviation direction, and the compensation timing is set to ensure that the compensation action is completed before the drill bit completely crosses the interface;

[0024] When the calculated compensation timing is reached, the controller will send an adjustment command containing the advance compensation amount to the drill arm drive mechanism to suppress the drilling direction deviation caused by the sudden change in lithology when the drill bit crosses the rock interface.

[0025] Preferably, a vibration sensing unit is also installed inside the drill pipe, which is connected to the demodulation module, and the controller is further configured as follows:

[0026] After the drill bit contacts the working face, the axial vibration signal of the drill pipe collected by the vibration sensing unit is acquired in real time.

[0027] The axial vibration signal is transformed in the time and frequency domain to extract the vibration energy characteristic value within the preset frequency band, and the change curve of the vibration energy characteristic value with the displacement of the drill pipe is recorded.

[0028] When the rate of change of the vibration energy characteristic value exceeds the preset abrupt change threshold, the current position of the drill bit is determined to be the interface between soft and hard rock layers, and the coordinates of the interface position along the drill rod axis are recorded.

[0029] The interface position identified by the vibration energy feature value is fused and compared with the interface position identified based on the rock layer hardness distribution curve. When the deviation between the two exceeds the preset deviation threshold, the interface position identified by the vibration energy feature value is taken as the standard. The hardness characterization values ​​corresponding to both sides of the new interface position are extracted from the continuous distribution curve, and the estimated hardness difference between the two sides of the interface is recalculated based on the extracted hardness characterization values.

[0030] Preferably, the vibration sensing unit includes multiple vibration sensors spaced apart along the drill pipe axis;

[0031] The controller is also configured as follows:

[0032] The axial vibration signals collected synchronously by multiple vibration sensors are segmented according to a preset time window length. Cross-correlation analysis is performed on the vibration signals within each time window to calculate the arrival time difference of the vibration signals between two adjacent vibration sensors.

[0033] Based on the arrival time difference and the interval distance between adjacent vibration sensors, the average wave velocity of the vibration wave propagating along the drill pipe within the current time window is calculated, and the average wave velocity is recorded as the real-time wave velocity as the drill pipe advances, forming a real-time wave velocity change curve.

[0034] Slope analysis is performed on the real-time wave velocity change curve. When the absolute value of the slope exceeds the preset wave velocity change rate threshold, the location of the drill bit is determined to be the interface between soft and hard rock layers, and the coordinates of the interface location along the drill rod axis are recorded.

[0035] The interface location identified based on real-time wave velocity is weighted and fused with the interface location identified based on vibration energy feature value. The fused location is then used as the updated interface location identified based on vibration energy feature value, and compared with the interface location identified based on rock layer hardness distribution curve.

[0036] Preferably, when the controller calculates the actual spatial coordinates of the designed borehole positions on the working face, it performs the following operations:

[0037] Based on the 3D topography of the tunnel face obtained by 3D reconstruction, a triangular mesh surface model of the tunnel face is constructed, and the local principal curvature and normal vector of each mesh node of the triangular mesh surface model are calculated.

[0038] The spatial points corresponding to the borehole design coordinate data are used as initial projection points and projected onto the triangular mesh surface model along the preset initial projection direction to obtain preliminary mapped coordinates.

[0039] Using the initial mapped coordinates as the center, a local neighborhood is defined on the triangular mesh surface model. The mean curvature of each mesh node in the local neighborhood is calculated as a local tilt characterization, and the curvature variance is calculated as a local fluctuation characterization.

[0040] The local terrain complexity index is calculated based on the weighted fusion of local tilt and local undulation characteristics.

[0041] The actual spatial coordinates of the boreholes are determined based on the local terrain complexity index: when the local terrain complexity index is lower than a preset first threshold, the preliminary mapped coordinates are used as the actual spatial coordinates of the boreholes; when the local terrain complexity index is higher than a preset second threshold, a conformal mapping correction procedure is performed to obtain corrected coordinates, which are then used as the actual spatial coordinates of the boreholes; when the local terrain complexity index is between the preset first threshold and the preset second threshold, the preliminary mapped coordinates and the corrected coordinates are linearly interpolated, and the interpolation coefficients are determined according to the proportional relationship between the local terrain complexity index and the preset first and second thresholds to obtain the fused actual spatial coordinates of the boreholes.

[0042] The final determined actual spatial coordinates of all boreholes are stored and used for real-time comparison during subsequent drilling processes.

[0043] Preferably, the conformal mapping correction procedure includes: using the initial mapping coordinates as initial values, employing a geodesic distance iterative optimization algorithm on the triangular mesh surface model to search for the optimal position that satisfies the constraint conditions as the corrected coordinates; the constraint condition is that the difference between the surface distance between the optimal position and the adjacent determined actual spatial coordinate points of the boreholes and the designed borehole spacing is less than a preset allowable error value.

[0044] Preferably, when the controller calculates the local terrain complexity index, it performs the following operations:

[0045] Using the initial mapped coordinates as the center, nested sub-neighborhoods are defined within the local neighborhood using multiple radii of different scales.

[0046] Calculate the curvature variance of each grid node in each sub-neighborhood to obtain the sequence of curvature variance as a function of scale;

[0047] Perform linear fitting on the sequence to obtain the slope of the fitted line, and use the slope as a scale dependence index.

[0048] The scale dependence index is weighted and fused with the local tilt and local fluctuation characteristics to obtain the local terrain complexity index.

[0049] Preferably, when the controller sends an adjustment command to the drive mechanism of the drill arm, it specifically performs the following:

[0050] Based on the real-time calculated continuous space curve and the material stiffness coefficient of the drill pipe, an elastic deformation deflection model of the drill pipe under the lateral resistance of the rock strata is established.

[0051] Based on the elastic deformation deflection model and the real-time distance between the drill bit center point and the drill arm holder, the future trajectory deviation trend of the drill bit after continuing to drill a unit depth is predicted under the condition of maintaining the current thrust.

[0052] When the real-time radial deviation value exceeds the set deviation threshold range, the controller calculates the advance correction amount based on the future trajectory offset trend and sends the adjustment command containing the advance correction amount to the drill arm drive mechanism, so that the adjustment action of the drill arm can compensate in advance for the subsequent displacement offset caused by the release of the elastic potential energy of the drill pipe.

[0053] The present invention has at least the following beneficial effects:

[0054] This invention achieves initial three-dimensional reconstruction of the tunnel face and drill rod through structured light projection and binocular vision, and combines fiber optic shape sensing to achieve continuous real-time perception of the drill rod axis. It can accurately calculate the spatial position and drilling direction of the drill bit, realize automatic closed-loop control of borehole positioning and drilling deviation, significantly improve the accuracy of borehole position and direction, reduce over-excavation and under-excavation, improve the quality and automation level of tunnel drilling and blasting construction, and reduce errors caused by manual intervention.

[0055] This invention introduces a vibration sensing unit and uses the sudden change in vibration energy to identify rock strata interfaces. It can be cross-checked and integrated with the oil pressure velocity identification results to improve the accuracy and reliability of interface position identification, reduce misjudgments caused by errors in single sensing information, make the advance compensation strategy more in line with actual geological conditions, and improve the adaptability and correction accuracy of the control system in complex strata.

[0056] This invention uses multiple vibration sensors to acquire axial vibration and calculate the change in vibration wave velocity, which can further improve the recognition accuracy of the interface between soft and hard rock layers. By weighted fusion of multi-sensor information, the misjudgment rate caused by local geological interference is reduced, making the interface positioning more stable and reliable, providing more accurate input for advanced correction, and improving the drilling trajectory control accuracy and system robustness.

[0057] This invention constructs a triangular mesh surface on the tunnel face and calculates the terrain complexity. It can adaptively select a borehole positioning strategy based on the degree of undulation of the tunnel face, automatically correct the borehole position in complex terrain areas, ensure that the borehole layout is more in line with the actual rock surface morphology, improve the accuracy and rationality of borehole position, improve blasting uniformity, and reduce construction disturbance.

[0058] This invention employs iterative optimization of geodesic distances for conformal mapping correction, ensuring that the actual surface spacing between boreholes matches the design value. This avoids abnormal borehole spacing caused by terrain undulations, making the borehole layout scheme more in line with blasting design requirements, improving blasting effect, reducing surrounding rock disturbance, and enhancing construction safety and economy.

[0059] This invention analyzes terrain complexity through multi-scale nested neighborhood analysis, which can suppress the interference of local minor bumps and depressions on terrain evaluation, more accurately reflect the real terrain features of the working face, improve the stability of complexity index calculation, provide a more reliable basis for the selection of borehole positioning strategies, and further improve the accuracy and rationality of borehole layout under complex rock surface conditions.

[0060] This invention establishes a drill pipe elastic deformation deflection model and predicts future trajectory deviation trends, realizing advanced correction based on deformation prediction. This can effectively compensate for the hysteresis deviation caused by the elastic deformation of the drill pipe, avoid the continuous trajectory deviation caused by relying solely on real-time position correction, and improve the trajectory control accuracy and stability throughout the drilling process.

[0061] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0062] Figure 1 This is a schematic diagram of the intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face, as described in one of the technical solutions of the present invention. Detailed Implementation

[0063] The present invention will be further described in detail below with reference to embodiments, so that those skilled in the art can implement it based on the description.

[0064] According to one embodiment of the present invention, such as Figure 1 As shown, the intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face includes:

[0065] The structured light projector, at least two cameras (the cameras are industrial cameras), fiber optic shape demodulation module (the adjustment module is a fiber optic shape adjustment module), and controller (a computer controller) are all mounted on the tunnel drilling and blasting trolley. The drilling and blasting trolley has multiple drill arms, the ends of which are connected to drill rods. The drill rods contain sensing units (fiber optic shape sensors) that are connected to the demodulation module.

[0066] The structured light projector is used to project structured light onto the drill pipe and the working face before the drill bit contacts the working face. The camera is used to acquire binocular images of the surfaces (drill pipe surface and working face surface) including the drill pipe and the working face and transmit the binocular images to the controller. The demodulation module is used to calculate the curvature value sequence and twist angle value sequence of each point along the fiber shape sensing unit in real time and transmit the data to the controller.

[0067] The controller stores pre-defined borehole coordinate data and borehole direction vector data. The controller is used to perform the following operations to achieve borehole positioning and borehole accuracy control:

[0068] Three-dimensional reconstruction is performed based on the binocular images acquired before the drill bit contacts the working face to obtain the initial spatial curve of the drill rod center axis and the three-dimensional shape of the working face, and the actual spatial coordinates of the designed borehole position on the working face are calculated; the initial installation orientation parameters of the sensing unit are calibrated based on the initial spatial curve.

[0069] After the drill bit contacts the face, based on the initial installation orientation parameters, the real-time received curvature value sequence, and the torsion angle value sequence, the continuous spatial curve of the drill pipe center axis from the near end of the drill pipe to the drill bit position is reconstructed.

[0070] The real-time three-dimensional spatial coordinates and drilling direction vector values ​​of the drill bit center point are calculated based on the continuous spatial curve. The real-time three-dimensional spatial coordinates and drilling direction vector values ​​are compared with the actual spatial coordinates and the designed drilling direction vector values ​​to calculate the real-time radial deviation and angular deviation values. When the real-time radial deviation or angular deviation value exceeds the preset threshold range, the controller sends an adjustment command to the drive mechanism of the drill arm to make the drill bit center point return to the preset threshold range.

[0071] In this technical solution, the structured light projector can be a blue light structured light projection device, the camera can be a gigabit network port industrial area array camera, the demodulation module can be a distributed fiber optic demodulation device, the controller can be an embedded industrial controller, the sensing unit can be a multi-core fiber optic sensing component, the drill rod can be made of alloy structural steel, and the drill arm can be made of high-strength alloy steel. The structured light projector and industrial camera can be mounted on the front frame of the drilling and blasting trolley, facing the working face. The demodulation module can be mounted in the electrical installation area of ​​the drilling and blasting trolley, the controller can be mounted in the electrical control cabinet of the drilling and blasting trolley, the drill arm can be mounted on both sides of the main body of the drilling and blasting trolley, and the drill rod can be mounted at the end of the drill arm through a clamping mechanism. The sensing unit can be embedded inside the drill rod along the axis of the drill rod. 3D reconstruction can be achieved through a binocular stereo matching algorithm, and the initial installation orientation parameters can be calibrated by least squares fitting. Before the drill bit contacts the working face, the structured light projector projects structured light onto the drill rod and the working face, and the two cameras simultaneously acquire binocular images containing the surfaces of the drill rod and the working face. During data acquisition, ensure that the frame rate of the two industrial cameras matches the projection frame rate of the structured light projector. The frame rate can be set to 30fps. The acquired binocular images are transmitted to the controller in real time via a gigabit network port. The controller uses a binocular stereo matching algorithm to perform 3D reconstruction on the binocular images, obtaining the initial spatial curve of the drill pipe's central axis and the 3D shape of the face. The binocular stereo matching algorithm can be implemented using the SIFT feature matching algorithm. During the 3D reconstruction process, the images need to be denoised, which can be done using a Gaussian filtering algorithm. After the binocular stereo matching algorithm completes the matching, the 3D point cloud is downsampled. The downsampling resolution is set to 0.01m to reduce the amount of data and ensure reconstruction accuracy.

[0072] To ensure the accuracy and consistency of the above coordinate comparison process, all coordinate data involved in the calculation must be unified under the same global reference coordinate system. In this system, the global reference coordinate system is the tunnel construction coordinate system, which is typically established with the tunnel centerline as the axis and the center of the tunnel floor at a specific mileage station as the origin. All borehole design coordinate data are pre-stored in this coordinate system. To achieve coordinate system unification, the controller executes the following calibration procedure during the system initialization phase:

[0073] Camera coordinate system and global coordinate system calibration: After the drill and blasting rig is positioned at a predetermined location in front of the tunnel face, multiple calibration targets with known precise coordinates in the tunnel construction coordinate system are first deployed at stable locations on the drill and blasting rig body or tunnel sidewall. Subsequently, a structured light projector and an industrial camera are used to acquire images and perform 3D reconstruction of the calibration targets, obtaining the coordinates of these target points in the camera coordinate system. By solving the rigid body transformation relationship (rotation matrix R1 and translation vector T1) between two sets of corresponding point sets, a transformation model from the camera coordinate system to the tunnel construction coordinate system can be established. Afterward, all point cloud data obtained based on binocular vision 3D reconstruction (such as the tunnel face topography and initial drill pipe curve) can be transformed to the tunnel construction coordinate system using this model.

[0074] Drill arm base coordinate system and global coordinate system calibration: The drill arm's drive mechanism operates in its own base coordinate system. By installing positioning markers on the drill arm base and end effector, and using a calibrated vision system or total station to measure the coordinates of these markers in the tunnel construction coordinate system, combined with the drill arm's forward kinematics model, the transformation relationship from the drill arm base coordinate system to the tunnel construction coordinate system (rotation matrix R2 and translation vector T2) can be solved.

[0075] The drill pipe's servo coordinate system is linked to the global coordinate system: The continuous spatial curve of the drill pipe's central axis, reconstructed by the sensing unit, is initially defined in a servo coordinate system with the near end of the drill pipe (the connection point with the drill arm) as the origin. At the borehole starting point, the absolute coordinates of the near end of the drill pipe in the tunnel construction coordinate system are obtained through the aforementioned calibrated vision system. Combined with the geometric parameters of the initial spatial curve of the drill pipe, the coordinates of all points on the curve in this servo coordinate system can be transformed to the tunnel construction coordinate system in real time.

[0076] Through the above calibration steps, the controller can unify the pre-stored borehole design coordinates, the tunnel face terrain and actual borehole coordinates obtained by 3D reconstruction, the drill bit coordinates and direction calculated in real time by fiber optic sensing, and the adjustment commands that need to be sent to the drill arm drive mechanism into the tunnel construction coordinate system for calculation and comparison, thereby ensuring the high precision and reliability of the control system.

[0077] The working process is as follows: Before the drill bit contacts the working face, a structured light projector projects structured light onto the drill pipe and the working face. Two industrial cameras simultaneously acquire binocular images containing the drill pipe and the working face and transmit them to the controller. The controller uses a binocular stereo matching algorithm to perform 3D reconstruction of the binocular images, obtaining the initial spatial curve of the drill pipe's central axis and the 3D shape of the working face. Simultaneously, it calculates the actual spatial coordinates of the designed borehole position on the working face and, based on the initial spatial curve, uses the least squares method to fit and calibrate the initial installation orientation parameters of the sensing unit. After the drill bit contacts the working face, the sensing unit collects the curvature and torsion angle information of each point along its own line in real time and transmits it to the demodulation module. The demodulation module calculates the curvature value sequence and torsion angle value sequence and transmits them to the controller. The controller, combined with the initial installation orientation parameters, reconstructs the continuous spatial curve of the drill pipe's central axis from the near end to the drill bit position. Then, based on this curve, it calculates the real-time 3D spatial coordinates of the drill bit's center point and the real-time drilling direction vector, and compares them with the borehole position. The actual spatial coordinates are compared with the designed drilling direction vector to calculate the real-time radial deviation and angular deviation values. The preset threshold for radial deviation can be set to 5mm, and the preset threshold for angular deviation can be set to 0.5°. When the real-time deviation exceeds the threshold, the controller sends an adjustment command to the hydraulic drive mechanism of the drill arm to bring the drill bit back to the threshold range. Two industrial cameras simultaneously acquire binocular images of the drill rod and the face. During acquisition, it is ensured that the acquisition frame rate of the two industrial cameras is consistent with the projection frame rate of the structured light projector. The frame rate can be set to 30fps. The acquired binocular images are transmitted to the controller in real time through a gigabit network port. The controller uses a binocular stereo matching algorithm to perform three-dimensional reconstruction of the binocular images to obtain the initial spatial curve of the drill rod center axis and the three-dimensional shape of the face. The binocular stereo matching algorithm can be implemented using the SIFT feature matching algorithm. During the three-dimensional reconstruction process, the image needs to be denoised. The denoising process can use a Gaussian filtering algorithm.

[0078] The radial deviation threshold and angular deviation threshold are determined based on the accuracy requirements of tunnel drilling and blasting construction and combined with on-site construction experience. The initial installation orientation parameters are determined by fitting the initial spatial curve of the drill rod and the positional relationship of the fiber optic sensing unit using the least squares method. Functional testing can be performed using a tunnel drilling and blasting construction scenario as the experimental object. The experimental method involves starting the system to perform borehole positioning and drilling operations under different tunnel face terrain conditions, recording the borehole position deviation and direction deviation for each drilling operation, repeating the experiment 30 times, and performing statistical analysis on the experimental data to calculate the average and standard deviation of the deviations to verify the stability and accuracy of the system.

[0079] By employing this technical solution, the present invention can complete the three-dimensional reconstruction of the tunnel face topography and the initial posture of the drill rod through structured light projection and binocular vision. In conjunction with the sensing unit, the continuous spatial curve of the drill rod's central axis is calculated in real time, accurately obtaining the spatial coordinates of the drill bit and the drilling direction vector. This enables real-time comparison and closed-loop correction between the designed borehole position and the actual borehole posture, effectively improving the accuracy of borehole positioning and drilling direction control, reducing over-excavation and under-excavation in tunnel construction, increasing the standardization and automation of drill and blast construction, reducing errors and uncertainties caused by manual operation, and ensuring the stability and safety of the construction process.

[0080] According to yet another embodiment of the present invention, the controller is further configured to:

[0081] Real-time oil pressure data and drill pipe advance speed data of each joint of the drill arm are continuously collected according to the preset sampling cycle;

[0082] Based on oil pressure and propulsion speed data, the oil pressure change rate and propulsion speed change rate are calculated in real time. If the oil pressure change rate and propulsion speed change rate are negatively correlated and both exceed their respective set thresholds in the current sampling period, it is determined that the rock formation hardness has changed in the current sampling period, and the ratio of oil pressure to propulsion speed in the current sampling period is recorded as the rock formation hardness characterization value. If the oil pressure change rate and propulsion speed change rate are not negatively correlated in the current sampling period, or although they are negatively correlated but do not exceed their respective set thresholds, it is determined that the rock formation hardness is stable in the current sampling period, and the controller maintains the current control parameters of the drill arm.

[0083] Based on the rock hardness characterization values ​​recorded in multiple consecutive sampling periods, combined with drill pipe advance displacement data, a continuous distribution curve of rock hardness along the drilling direction is established; when it is determined that the rock hardness changes during the current sampling period, the drill bit position corresponding to that sampling period is identified as the interface between soft and hard rock layers, and the degree of hardness difference is estimated based on the hardness characterization values ​​on both sides of the interface.

[0084] Based on the degree of hardness difference on both sides of the rock layer interface, and combined with the drilling mechanics model, the direction and amount of deviation that the drill bit will produce when crossing the interface are predicted.

[0085] While identifying the rock interface, the controller calculates the advance compensation amount and compensation timing based on the predicted deviation direction and amount, combined with the mechanical response delay parameters of the drill arm drive mechanism; the compensation direction is set to be opposite to the predicted deviation direction, and the compensation timing is set to ensure that the compensation action is completed before the drill bit completely crosses the interface;

[0086] When the calculated compensation timing is reached, the controller will send an adjustment command containing the advance compensation amount to the drill arm drive mechanism to suppress the drilling direction deviation caused by the sudden change in lithology when the drill bit crosses the rock interface.

[0087] In the above technical solution, the preset sampling period can be set to 100ms, the accuracy of hydraulic pressure data acquisition at each joint of the drill arm can be controlled within ±0.1MPa, and the accuracy of drill rod propulsion speed data acquisition can be controlled within ±0.1mm / s. Hydraulic pressure data can be acquired through a hydraulic pressure sensor, and propulsion speed data can be acquired through a displacement sensor. The acquired data can be transmitted to the controller via an analog input interface with a transmission accuracy of not less than 0.5%. The hydraulic pressure sensor can be a strain gauge type, and the displacement sensor can be a laser displacement sensor. The controller can be an industrial PLC controller with multi-channel analog input functionality. The housing of the hydraulic pressure sensor can be made of stainless steel, and the strain gauge can be a constantan strain gauge. The housing of the displacement sensor can be made of engineering plastic, and the laser emitter head can be made of optical glass. The hydraulic pressure sensor can be installed at the hydraulic cylinder inlet of each joint of the drill arm and fixed by threaded connection. The displacement sensor can be installed on the propulsion mechanism of the drill arm, with the lens facing the propulsion direction of the drill rod and parallel to the drill rod. The connection lines between the hydraulic pressure sensor, displacement sensor, and controller can be arranged inside the drill arm bracket and properly protected.

[0088] The rate of change of oil pressure can be calculated by dividing the oil pressure difference between two adjacent sampling periods by the sampling period, with a calculation accuracy controlled within ±0.1 MPa / ms; the rate of change of propulsion speed can be calculated by dividing the propulsion speed difference between two adjacent sampling periods by the sampling period, with a calculation accuracy controlled within ±0.01 mm / s. 2 The threshold values ​​for the oil pressure change rate can be set to 0.5 MPa / ms, and the threshold values ​​for the propulsion speed change rate can be set to 0.05 mm / s. 2 The calculation accuracy of the rock layer hardness characterization value can be controlled within ±0.1MPa·s / mm. The controller can be an industrial PLC controller with high-speed data processing capability. The computing chip can be a silicon-based chip. The controller needs to be arranged in close proximity to the oil pressure sensor and displacement sensor.

[0089] Multiple consecutive sampling periods can be set to 10-20 sampling periods, i.e., sampling data within 1s-2s; the continuous distribution curve of rock hardness along the drilling direction can be obtained using a polynomial fitting algorithm, with the fitting order selectable to be 3, and the fitting error not exceeding ±0.1MPa·s / mm; the acquisition accuracy of drill rod advance displacement data can be controlled within ±0.1mm; the identification accuracy of the soft and hard rock layer interface position can be controlled within ±1mm; the calculation of the estimated degree of hardness difference can be done by using the absolute value of the difference between the average values ​​of hardness characterization values ​​on both sides of the interface within 5 sampling periods, and the calculation accuracy can be controlled within ±0.1MPa·s / mm;

[0090] The drilling mechanics model can adopt the drill rod bending mechanics model, and the calculation error of the model should not exceed ±0.1mm. The prediction of the deviation direction can be determined based on the degree of hardness difference. If the hardness of the rock layer on the other side of the interface increases, the drill bit is likely to deviate towards the side with lower hardness, and vice versa. The prediction accuracy of the deviation amount can be controlled within ±0.3mm. The mechanical response delay parameter of the drill arm drive mechanism can be set to 200ms. The calculation of the advance compensation amount can be 1.2 times the predicted deviation amount, and the calculation accuracy can be controlled within ±0.1mm. The calculation of the compensation timing can be determined based on the current drill bit advance speed and the interface thickness. The interface thickness can be preset to 50mm-100mm, and the compensation timing is set when the drill bit is 10mm-20mm away from the edge of the interface. The drill arm drive mechanism can be a hydraulic drive mechanism, and the hydraulic valve group can be made of stainless steel. The control line between the controller and the drill arm drive mechanism must ensure reliable connection.

[0091] The working process is as follows: the controller collects oil pressure data and drill pipe advance displacement data at each joint of the drill arm through oil pressure sensors and displacement sensors according to a preset sampling period of 100ms. It calculates the drill pipe advance speed data, filters the collected data, and stores it. It calculates the rate of change of oil pressure and the rate of change of advance speed in real time, determining whether the two are negatively correlated and both exceed a set threshold. If so, it determines that the rock hardness has changed and records the hardness characterization value; otherwise, it determines that the rock hardness is stable and maintains the current control parameters of the drill arm. It continuously stores the hardness characterization values ​​for 10-20 sampling periods, and combines them with the advance displacement data using a cubic polynomial. The fitting algorithm establishes a continuous distribution curve of rock hardness along the drilling direction. When a change in rock hardness is detected, the corresponding drill bit position is identified as the interface between soft and hard rock layers. Hardness characterization values ​​of 5 sampling periods on each side of the interface are extracted, and the absolute value of the difference between the average values ​​is calculated as the estimated degree of hardness difference. Combined with the drill pipe bending mechanics model, the deviation direction and amount when the drill bit crosses the interface are predicted. The mechanical response delay parameter of the drill arm drive mechanism is called to calculate the advance compensation amount and compensation timing. The compensation direction is opposite to the predicted deviation direction. When the compensation timing is reached, an adjustment command containing the advance compensation amount is sent to the drill arm drive mechanism to suppress drill bit deviation.

[0092] Regarding parameter setting methods, the sampling period is set based on the drill pipe advance speed and the response speed to changes in rock hardness. The acquisition accuracy of oil pressure and displacement sensors is calibrated using calibration equipment. The threshold values ​​for oil pressure change rate and advance speed change rate are set based on the parameter variation patterns under different rock hardness conditions combined with construction experience. The number of continuous sampling periods, the number of polynomial fitting iterations, and the number of sampling periods on both sides of the interface are all determined through debugging. The borehole mechanics model parameters are calibrated based on the drill pipe material properties, dimensional parameters, and actual borehole mechanics test data. The mechanical response delay parameter is determined by testing the action response time of the drive mechanism. The advance compensation coefficient, interface thickness, and compensation timing distance parameters are set based on the actual compensation effect and common interface conditions. Regarding algorithm implementation, the rate of change calculation and negative correlation judgment are programmed in C language. The polynomial fitting algorithm is programmed in MATLAB and then embedded in the controller firmware. The borehole mechanics model calculation and compensation parameter calculation are programmed in C++ language. Functional testing can be conducted in the laboratory to simulate rock layers of different hardness and the interface between soft and hard rock layers. The experimental objects include simulated rock layers, drill pipes, drill arms, and drilling and blasting trolleys. The experimental methods include changing the hardness of the simulated rock layers and having the drill pipe pass through the simulated interface. The test results include the accuracy of data acquisition, the accuracy of hardness change judgment, the effect of curve establishment, the accuracy of interface recognition, the effect of deviation prediction and compensation, and the error and accuracy are statistically analyzed through multiple tests.

[0093] This technical solution enables real-time acquisition of drill arm oil pressure and drill rod advance speed data, accurate judgment of rock hardness changes, establishment of rock hardness distribution curves, identification of the interface between soft and hard rock layers and calculation of the degree of hardness difference, prediction of deviation when the drill bit crosses the interface, and adjustment of drill arm posture through advance compensation. This effectively suppresses drilling direction deviation caused by abrupt changes in lithology, ensures that drill arm control parameters adapt to changes in rock hardness, and maintains and further improves drilling accuracy.

[0094] According to another embodiment of the present invention, a vibration sensing unit is further provided inside the drill pipe, the vibration sensing unit is connected to the demodulation module, and the controller is further configured to:

[0095] After the drill bit contacts the working face, the axial vibration signal of the drill pipe collected by the vibration sensing unit is acquired in real time.

[0096] The axial vibration signal is transformed in the time and frequency domain to extract the vibration energy characteristic value within the preset frequency band, and the change curve of the vibration energy characteristic value with the displacement of the drill pipe is recorded.

[0097] When the rate of change of the vibration energy characteristic value exceeds the preset abrupt change threshold, the current position of the drill bit is determined to be the interface between soft and hard rock layers, and the coordinates of the interface position along the drill rod axis are recorded.

[0098] The interface position identified by the vibration energy feature value is fused and compared with the interface position identified based on the rock layer hardness distribution curve. When the deviation between the two exceeds the preset deviation threshold, the interface position identified by the vibration energy feature value is taken as the standard. The hardness characterization values ​​corresponding to both sides of the new interface position are extracted from the continuous distribution curve, and the estimated hardness difference between the two sides of the interface is recalculated based on the extracted hardness characterization values.

[0099] In the above technical solution, the preset frequency band can be set to 10Hz–500Hz, the threshold for the sudden change rate of vibration energy characteristic value can be set to 15%, and the preset deviation threshold can be set to 0.1m. The vibration sensing unit can be a piezoelectric vibration sensor, and the time-frequency domain transformation can be implemented using a Fourier transform algorithm. The vibration sensor can be mounted along the drill pipe axis on the inner wall of the drill pipe, spaced apart from the sensing unit, to ensure accurate acquisition of the axial vibration signal of the drill pipe. The vibration signal is amplified and filtered before transmission; the filtering method is bandpass filtering, and the passband range is consistent with the vibration energy extraction frequency band.

[0100] It should be noted that there is an inherent slight delay in identifying rock strata interfaces using vibration signals. When the drill bit crosses the interface between soft and hard rock strata, the cutting vibration characteristics change, but the vibration signal needs to accumulate for a certain period of time or energy change before it can be identified as an abrupt change. Therefore, when the rate of change of the vibration energy characteristic value exceeds the threshold and is identified as an interface, the drill bit has actually already crossed that interface location. Based on this, the interface location identified by vibration is located in the already drilled area and must be within the coverage of the continuous distribution curve of this scheme. Extracting the hardness value from this curve is logically reasonable.

[0101] The working process is as follows: After the drill bit contacts the working face and begins drilling, the vibration sensing unit collects the axial vibration signal of the drill rod in real time and transmits the signal to the demodulation module. The demodulation module preprocesses the signal and then transmits it to the controller. The controller uses the Fourier transform algorithm to perform time-frequency domain transformation on the axial vibration signal and extracts the vibration energy feature value in the 10Hz–500Hz frequency band (the extraction method is to calculate the sum of the energy of all frequency components in this frequency band as the vibration energy feature value at the current sampling time). The change curve of this feature value with the displacement of the drill rod is recorded. When the change rate of the vibration energy feature value exceeds the abrupt threshold of 15%, the current position of the drill bit is determined to be the interface position of soft and hard rock layers, and the coordinates of this interface along the axial direction of the drill rod are recorded. The controller compares this interface position with the interface position identified based on the rock layer hardness distribution curve. When the difference between the two exceeds 0.1m, the interface position identified by the vibration energy feature value is taken as the standard, and the hardness characterization values ​​on both sides of this position are extracted from the hardness distribution curve. The estimated hardness difference between the two sides of the interface is recalculated to provide an accurate basis for advance compensation.

[0102] The preset frequency band is determined based on the common frequency range of drill pipe vibration, the threshold for sudden change in vibration energy change rate is determined based on the differences in vibration characteristics during drilling of different rock strata, and the preset deviation threshold is determined based on the accuracy of the two interface identification methods.

[0103] This technical solution can collect axial vibration signals of the drill pipe through vibration sensing units, use changes in vibration energy characteristics to help identify the interface between soft and hard rock strata, and cross-verify and merge the rock strata identification results based on oil pressure and propulsion speed to improve the accuracy and reliability of rock strata interface location judgment, reduce misjudgments caused by construction interference from single monitoring methods, provide more realistic and reliable geological basis for advanced compensation control, enhance the adaptability of the overall control system, and ensure the stability of the borehole trajectory at the rock strata interface.

[0104] According to another embodiment of the present invention, the vibration sensing unit includes a plurality of vibration sensors arranged at intervals along the drill pipe axis;

[0105] The controller is also configured as follows:

[0106] The axial vibration signals collected synchronously by multiple vibration sensors are segmented according to a preset time window length. Cross-correlation analysis is performed on the vibration signals within each time window to calculate the arrival time difference of the vibration signals between two adjacent vibration sensors.

[0107] Based on the arrival time difference and the interval distance between adjacent vibration sensors, the average wave velocity of the vibration wave propagating along the drill pipe within the current time window is calculated, and the average wave velocity is recorded as the real-time wave velocity as the drill pipe advances, forming a real-time wave velocity change curve.

[0108] Slope analysis is performed on the real-time wave velocity change curve. When the absolute value of the slope exceeds the preset wave velocity change rate threshold, the location of the drill bit is determined to be the interface between soft and hard rock layers, and the coordinates of the interface location along the drill rod axis are recorded.

[0109] The interface position identified based on real-time wave velocity is weighted and fused with the interface position identified based on vibration energy feature value. The fused position is used as the updated interface position identified by vibration energy feature value and compared with the interface position identified based on rock layer hardness distribution curve. That is, when the deviation between the two exceeds the preset deviation threshold, the interface position identified by the updated vibration energy feature value is taken as the standard. The hardness characterization values ​​on both sides of the position are extracted from the continuous distribution curve, and the estimated hardness difference between the two sides of the interface is recalculated based on the extracted hardness characterization values.

[0110] In the above technical solution, the spacing between adjacent vibration sensors can be set to 0.2m, the time window length can be set to 20ms, and the wave velocity change rate threshold can be set to 50m / (s·m). Miniature accelerometers can be used as vibration sensors. Cross-correlation analysis can be implemented using digital signal processing algorithms, and weighted fusion can be achieved using an arithmetic average method. Multiple vibration sensors can be mounted at equal intervals along the drill pipe axis on the inner wall of the drill pipe, with an adjacent sensor spacing of 0.2m to ensure accurate acquisition of the vibration wave propagation signal. Cross-correlation analysis is performed on the vibration signals within each time window, specifically using a cross-correlation algorithm based on Fast Fourier Transform. During the analysis, the vibration signal is first detrended to remove the DC component and linear trend. Then, the cross-correlation function between two adjacent vibration sensors is calculated. The arrival time difference of the vibration signal is determined based on the peak position of the cross-correlation function. Then, based on the arrival time difference and the interval distance between adjacent vibration sensors, the average wave velocity of the vibration wave propagating along the drill pipe within the current time window is calculated. Weighted fusion uses an arithmetic average method, averaging the wave velocity identification interface position and the vibration energy identification interface position as the final fusion result.

[0111] The working process is as follows: Multiple miniature accelerometers synchronously collect axial vibration signals from the drill pipe and transmit them to the demodulation module. The controller segments the received vibration signals into 20ms time windows, performs cross-correlation analysis on each segment, calculates the arrival time difference of the vibration signals between two adjacent sensors, and calculates the average wave velocity of the vibration wave propagating along the drill pipe within the current time window based on the time difference and the 0.2m sensor spacing. This average wave velocity is recorded as the drill pipe advances, forming a real-time wave velocity change curve. The controller performs slope analysis on the wave velocity change curve. When the absolute value of the slope exceeds the threshold of 50m / (s·m), the location of the drill bit is determined to be the interface between soft and hard rock layers, and the axial coordinates of this interface are recorded. The controller performs arithmetic mean weighted fusion of the interface location identified by the wave velocity and the interface location identified by the vibration energy characteristic value. The fused location is used as the updated vibration energy identification interface location and compared with the interface location identified based on the rock layer hardness distribution curve. When the deviation exceeds 0.1m, the updated location is used as the standard, the hardness characterization value is re-extracted, and the degree of hardness difference is calculated.

[0112] The sensor spacing is determined based on the drill pipe diameter and vibration wave propagation characteristics; the time window length is determined based on the stability of the vibration signal; and the wave velocity change rate threshold is determined based on the differences in vibration wave velocity in different rock strata. Functional testing uses soft and hard rock strata containing localized fracture zones as the experimental object. The experimental method involves starting the system and drilling, recording the wave velocity change curve and vibration energy change curve, comparing the fused interface location with the actual rock strata interface location, repeating the experiment 30 times, and performing statistical analysis to verify the robustness of the interface recognition.

[0113] This technical solution can acquire the propagation information of vibration waves by multiple vibration sensors arranged along the drill pipe axis. Based on the changes in vibration wave velocity, the location of rock strata interfaces can be more accurately identified. The wave velocity identification results are then weighted and fused with the vibration energy identification results to improve the accuracy and robustness of interface positioning. This reduces the interference of complex geological conditions such as local fracturing and isolated boulders on the identification results, making the advanced correction strategy more consistent with the actual geological conditions, improving the accuracy of borehole trajectory control, and ensuring the quality of boreholes under complex geological conditions.

[0114] According to another embodiment of the present invention, when the controller calculates the actual spatial coordinates of the designed borehole position on the working face, it specifically performs the following operations:

[0115] Based on the 3D topography of the tunnel face obtained by 3D reconstruction, a triangular mesh surface model of the tunnel face is constructed, and the local principal curvature and normal vector of each mesh node of the triangular mesh surface model are calculated.

[0116] The spatial points corresponding to the borehole design coordinate data are used as initial projection points and projected onto the triangular mesh surface model along the preset initial projection direction to obtain preliminary mapped coordinates.

[0117] Using the initial mapped coordinates as the center, a local neighborhood is defined on the triangular mesh surface model. The mean curvature of each mesh node in the local neighborhood is calculated as a local tilt characterization, and the curvature variance is calculated as a local fluctuation characterization.

[0118] The local terrain complexity index is calculated based on the weighted fusion of local tilt and local undulation characteristics.

[0119] The actual spatial coordinates of the boreholes are determined based on the local terrain complexity index: when the local terrain complexity index is lower than a preset first threshold, the preliminary mapped coordinates are used as the actual spatial coordinates of the boreholes; when the local terrain complexity index is higher than a preset second threshold, a conformal mapping correction procedure is performed to obtain corrected coordinates, which are then used as the actual spatial coordinates of the boreholes; when the local terrain complexity index is between the preset first threshold and the preset second threshold, the preliminary mapped coordinates and the corrected coordinates are linearly interpolated, and the interpolation coefficients are determined according to the proportional relationship between the local terrain complexity index and the preset first and second thresholds to obtain the fused actual spatial coordinates of the boreholes.

[0120] The final determined actual spatial coordinates of all boreholes are stored and used for real-time comparison during subsequent drilling processes.

[0121] In the above technical solution, the local neighborhood radius used for terrain complexity analysis can be set to 0.1m, the first threshold can be set to 0.3, and the second threshold can be set to 0.7. Based on the 3D topography of the tunnel face obtained from 3D reconstruction, the controller uses the Poisson reconstruction algorithm to construct a triangular mesh surface model of the tunnel face. During model construction, the mesh resolution is set to 0.01m to ensure that the model accurately reflects the actual topography of the tunnel face. When calculating the local principal curvature and normal vector of each mesh node in the triangular mesh surface model, a neighborhood least squares fitting algorithm is used. The neighborhood radius used for curvature fitting is set to 0.05m to improve the calculation accuracy of curvature and normal vectors. Weighted fusion can be implemented using a weight allocation method, and linear interpolation can be implemented using a linear weighted interpolation method. The controller's computation module can integrate the above algorithms to ensure the real-time performance and accuracy of the computation.

[0122] The working process is as follows: The controller performs 3D reconstruction on the binocular images acquired by the industrial camera to obtain the 3D topographic data of the tunnel face. A 3D point cloud reconstruction algorithm is used to construct a triangular mesh surface model of the tunnel face. The local principal curvature and normal vector of each mesh node in the model are calculated by neighborhood fitting. The spatial points corresponding to the borehole design coordinate data are used as initial projection points and projected onto the triangular mesh surface model along the initial projection direction perpendicular to the tunnel face design plane to obtain preliminary mapped coordinates. With the preliminary mapped coordinates as the center, a local neighborhood with a radius of 0.1m is defined on the surface model. The mean curvature of each mesh node in this neighborhood is calculated as a local tilt characterization quantity, and the curvature variance is calculated as a local fluctuation. The local terrain complexity index is calculated by weighting and fusing the local tilt and local undulation indices. When the complexity index is below 0.3, the initial mapped coordinates are used as the actual spatial coordinates of the boreholes. When the complexity index is above 0.7, a conformal mapping correction procedure is executed to obtain corrected coordinates, which are used as the actual spatial coordinates of the boreholes. When the complexity index is between 0.3 and 0.7, the initial mapped coordinates and the corrected coordinates are linearly interpolated, and the interpolation coefficients are determined according to the ratio between the complexity index and two thresholds to obtain the fused actual spatial coordinates of the boreholes. The controller stores the actual spatial coordinates of all boreholes for real-time comparison during subsequent drilling processes.

[0123] This technical solution enables the construction of a triangular mesh surface model based on the three-dimensional topography of the tunnel face. The terrain complexity is calculated through local curvature and fluctuation characteristics. The borehole positioning method is adaptively selected according to the terrain complexity, making the actual spatial coordinates of the boreholes more consistent with the real shape of the tunnel face. This improves the rationality and standardization of the borehole layout, ensures uniform distribution of blasting holes, improves the blasting effect, reduces unnecessary disturbance to the surrounding rock, and provides a good hole position basis for subsequent blasting operations.

[0124] According to another embodiment of the present invention, the conformal mapping correction procedure includes: using the preliminary mapping coordinates as initial values, employing a geodesic distance iterative optimization algorithm on the triangular mesh surface model to search for the optimal position that satisfies the constraint conditions as the corrected coordinates; the constraint condition is that the difference between the surface distance between the optimal position and the adjacent determined actual spatial coordinate points of the boreholes and the designed borehole spacing is less than a preset allowable error value.

[0125] In the above technical solution, the allowable error value can be set to 3mm. The geodesic distance can be calculated using a curved mesh path search algorithm, and iterative optimization can be achieved using the gradient descent method. The number of iterations can be set to 50 to ensure the stability of the optimization results. During the iterative optimization process, the geodesic distance is recalculated after each iteration until the hole spacing error requirement is met.

[0126] During operation, when executing the conformal mapping correction procedure, the initial mapped coordinates are used as the initial optimization values. On the triangular mesh surface model, a surface mesh path search algorithm is used to calculate the geodesic distance between this initial value and the actual spatial coordinates of adjacent determined boreholes. This distance is compared with the designed borehole spacing, and the difference is calculated. Iterative optimization is performed using the gradient descent method. In each iteration, the position of the initial value is adjusted, and the difference between the geodesic distance and the designed borehole spacing is recalculated until the difference is less than the allowable error value of 3mm. This position is the optimal position and is used as the correction coordinates. The parameter setting method is as follows: the allowable error value is determined according to the accuracy requirements of the borehole layout, and the number of iterations is determined according to the convergence speed of the optimization algorithm. Functional testing uses a working face with complex terrain as the experimental object. The experimental method involves performing conformal mapping correction on multiple boreholes, comparing the difference between the corrected borehole spacing and the designed borehole spacing, repeating the experiment 20 times, and performing statistical analysis to verify the effectiveness of the correction procedure.

[0127] This technical solution can perform conformal mapping correction of the borehole position through geodesic distance iterative optimization algorithm, ensuring that the surface spacing between adjacent boreholes is consistent with the design borehole spacing, reducing the borehole spacing deviation caused by terrain undulations, making the borehole layout scheme more in line with blasting design requirements, improving the regularity of the blasting profile, which is beneficial to subsequent support operations and construction safety, and reducing the problem of uneven blasting energy distribution caused by borehole spacing deviation.

[0128] According to another embodiment of the present invention, when the controller performs the calculation of the local terrain complexity index, it specifically performs the following operations:

[0129] Using the initial mapped coordinates as the center, nested sub-neighborhoods are defined within the local neighborhood using multiple radii of different scales.

[0130] Calculate the curvature variance of each grid node in each sub-neighborhood to obtain the sequence of curvature variance as a function of scale;

[0131] Perform linear fitting on the sequence to obtain the slope of the fitted line, and use the slope as a scale dependence index.

[0132] The scale dependence index is weighted and fused with the local tilt and local fluctuation characteristics to obtain the local terrain complexity index.

[0133] In the above technical solution, multiple different scale radii can be set to 0.05m, 0.1m, and 0.15m respectively. Linear fitting can be achieved using the least squares method. The weights of the weighted fusion can be determined according to the influence of each indicator. The weight of the local tilt characterization quantity can be set to 0.4, the weight of the local fluctuation characterization quantity can be set to 0.3, and the weight of the scale dependence index can be set to 0.3.

[0134] During operation, when the controller calculates the local terrain complexity index, it uses the initial mapped coordinates as the center and, within the previously defined 0.1m local neighborhood, delineates three nested sub-neighborhoods with scale radii of 0.05m, 0.1m, and 0.15m respectively, expanding the range from the inside out. It calculates the curvature variance of each grid node within each sub-neighborhood, obtaining three curvature variance values, forming a sequence of curvature variance varying with scale radius. The least squares method is used to linearly fit this sequence (during the fitting process, the linear relationship between the curvature variance corresponding to each scale radius and the scale radius is calculated, and the slope of the fitted line is used as a scale dependence index; the goodness of fit R is also calculated during the fitting process). 2 Ensure R 2 To ensure the reliability of the fitting results, the index was weighted and fused with the local slope and local fluctuation metrics (≥0.9) to obtain the slope of the fitted line, which was then used as the scale dependence index. The scale dependence index was then weighted and fused with the previously calculated local slope and local fluctuation metrics at weights of 0.4, 0.3, and 0.3 respectively to calculate the final local terrain complexity index. The radii of different scales were determined based on the size of the local neighborhood to ensure reasonable nesting of sub-neighborhoods. The weights for the weighted fusion were determined based on the influence of each index on terrain complexity, combined with construction experience. Functional testing used a tunnel face with local minor irregularities as the experimental object. The experimental method involved calculating the terrain complexity index using single-scale and multi-scale neighborhoods respectively, comparing the fit between the results of the two methods and the actual terrain complexity, repeating the experiment 25 times, and performing statistical analysis to verify the superiority of multi-scale analysis.

[0135] This technical solution can reduce the impact of local minor bumps and depressions on terrain evaluation by analyzing terrain complexity through multi-scale nested neighborhood analysis, more accurately reflect the overall terrain characteristics of the working face, improve the stability and reliability of terrain complexity index calculation, provide a more reasonable judgment basis for borehole adaptive positioning, further improve the accuracy and reliability of borehole layout under complex rock surface conditions, and avoid borehole positioning deviation caused by local terrain interference.

[0136] According to another embodiment of the present invention, when the controller sends an adjustment command to the drive mechanism of the drill arm, it specifically performs the following:

[0137] Based on the real-time calculated continuous space curve and the material stiffness coefficient of the drill pipe, an elastic deformation deflection model of the drill pipe under the lateral resistance of the rock strata is established.

[0138] Based on the elastic deformation deflection model and the real-time distance between the drill bit center point and the drill arm holder, the future trajectory deviation trend of the drill bit after continuing to drill a unit depth is predicted under the condition of maintaining the current thrust.

[0139] When the real-time radial deviation value exceeds the set deviation threshold range, the controller calculates the advance correction amount based on the future trajectory offset trend and sends the adjustment command containing the advance correction amount to the drill arm drive mechanism, so that the adjustment action of the drill arm can compensate in advance for the subsequent displacement offset caused by the release of the elastic potential energy of the drill pipe.

[0140] In the above technical solution, the drill pipe can be made of alloy structural steel. The material stiffness coefficient can be obtained from a table of material mechanics parameters, with the unit being Pa. The unit depth can be set to 0.1m, and the real-time radial deviation threshold is set to 5mm. An elastic deformation deflection model of the drill pipe under the lateral resistance of the rock strata is established, based on the Euler-Bernoulli beam bending theory in mechanics of materials. The model needs to incorporate parameters such as the moment of inertia of the drill pipe section and the elastic modulus of the material. The moment of inertia of the section is calculated based on the diameter and wall thickness of the drill pipe, and the elastic modulus of the material is obtained from a table based on the alloy structural steel material selected for the drill pipe. After the model is established, it is verified using finite element analysis software to ensure that the model can accurately predict the elastic deformation of the drill pipe. A drill arm holder is provided at the end of the drill arm. This holder is a conventional component in the field of rock drilling rigs, used to hold and guide the drill pipe to drill in a predetermined direction. The drill pipe passes through the drill arm holder and can rotate and advance under the constraint of the holder. The controller stores the initial relative position parameters between the drill arm holder and the drill bit center point. During the drilling process, based on the drill rod advance displacement data, the real-time axial distance between the drill bit center point and the drill arm holder can be calculated in real time. This distance is one of the key boundary conditions for establishing the drill rod elastic deformation deflection model.

[0141] During operation, the controller calculates the continuous spatial curve of the drill pipe's central axis in real time. Combining this with the drill pipe's material stiffness coefficient, it establishes an elastic deformation deflection model of the drill pipe under the lateral resistance of the rock strata, based on beam bending theory. The controller acquires the real-time distance between the drill bit's center point and the drill arm holder using sensors, and substitutes this distance into the elastic deformation deflection model to predict the future trajectory deviation trend after the drill bit continues drilling to a depth of 0.1m while maintaining the current thrust. When the real-time radial deviation exceeds a set threshold of 5mm, the controller calculates a lead-in correction based on the predicted future trajectory deviation trend and the mechanical response speed of the drill arm drive mechanism. An adjustment command containing this lead-in correction is sent to the drill arm drive mechanism, enabling the drill arm to complete its adjustment action ahead of schedule, compensating for the subsequent displacement caused by the release of the drill pipe's elastic potential energy. The unit depth is determined based on the influence range of the drill pipe's elastic deformation, the material stiffness coefficient is obtained from a table based on the alloy structural steel material selected for the drill pipe, and the lead-in correction is calculated based on the deviation trend and the drive mechanism's response speed. Functional testing used drill rods of different lengths as experimental objects. The experimental method was to start the system to perform drilling operations, record the drilling trajectory offset with and without advance correction, repeat the experiment 20 times, and perform statistical analysis to verify the effect of advance correction.

[0142] This technical solution enables the establishment of an elastic deformation deflection model based on the drill rod's spatial curve and material stiffness. It predicts the trajectory deviation trend of the drill bit during subsequent drilling, and implements advance correction based on the prediction results to compensate for the hysteresis deviation caused by the drill rod's elastic deformation. This avoids the accumulation of deviations caused by relying solely on real-time position correction, improves the trajectory control accuracy throughout the drilling process, ensures stable and reliable drilling quality, reduces drilling deviations caused by drill rod deformation, and improves construction efficiency.

[0143] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. A smart projection positioning and drilling accuracy control system for blast holes at tunnel face, characterized in that, include: The structured light projector, at least two cameras, demodulation module and controller are all mounted on the drilling and blasting trolley; the drill arm of the drilling and blasting trolley is connected to the drill rod, and the drill rod is equipped with a sensing unit, which is connected to the demodulation module. The structured light projector is used to project structured light onto the drill pipe and the working face, the camera is used to acquire binocular images of the surface including the drill pipe and the working face, and the demodulation module is used to calculate the curvature value sequence and torsion angle value sequence of each point along the sensing unit in real time. The controller has pre-stored the design coordinate data of the borehole and the design drilling direction vector data. The controller is used to perform the following operations: Three-dimensional reconstruction is performed based on the binocular images acquired before the drill bit contacts the working face to obtain the initial spatial curve of the drill rod center axis and the three-dimensional shape of the working face, and the actual spatial coordinates of the designed borehole position on the working face are calculated; the initial installation orientation parameters of the sensing unit are calibrated based on the initial spatial curve. After the drill bit contacts the face, based on the initial installation orientation parameters, curvature value sequence, and torsion angle value sequence, a continuous spatial curve of the drill pipe center axis from the near end of the drill pipe to the drill bit position is reconstructed. The real-time three-dimensional spatial coordinates and drilling direction vector values ​​of the drill bit center point are calculated based on the continuous spatial curve, and compared with the actual spatial coordinates and the designed drilling direction vector values ​​to calculate the real-time radial deviation and angular deviation values. When the real-time radial deviation or angular deviation value exceeds the preset threshold range, the controller sends an adjustment command to the drive mechanism of the drill arm to make the drill bit center point return to the preset threshold range. When the controller calculates the actual spatial coordinates of the designed borehole positions on the working face, it performs the following operations: Based on the 3D topography of the tunnel face obtained by 3D reconstruction, a triangular mesh surface model of the tunnel face is constructed, and the local principal curvature and normal vector of each mesh node of the triangular mesh surface model are calculated. The spatial points corresponding to the borehole design coordinate data are used as initial projection points and projected onto the triangular mesh surface model along the preset initial projection direction to obtain preliminary mapped coordinates. Using the initial mapped coordinates as the center, a local neighborhood is defined on the triangular mesh surface model. The mean curvature of each mesh node in the local neighborhood is calculated as a local tilt characterization, and the curvature variance is calculated as a local fluctuation characterization. The local terrain complexity index is calculated based on the weighted fusion of local tilt and local undulation characteristics. The actual spatial coordinates of the boreholes are determined based on the local terrain complexity index: when the local terrain complexity index is lower than the preset first threshold, the preliminary mapped coordinates are used as the actual spatial coordinates of the boreholes. When the local terrain complexity index is higher than the preset second threshold, a conformal mapping correction procedure is executed to obtain the corrected coordinates, and the corrected coordinates are used as the actual spatial coordinates of the boreholes; when the local terrain complexity index is between the preset first threshold and the preset second threshold, the preliminary mapping coordinates and the corrected coordinates are linearly interpolated, and the interpolation coefficient is determined according to the ratio of the local terrain complexity index to the preset first threshold and the preset second threshold to obtain the fused actual spatial coordinates of the boreholes. The final determined actual spatial coordinates of all boreholes are stored and used for real-time comparison during subsequent drilling processes; When the controller calculates the local terrain complexity index, it performs the following operations: Using the initial mapped coordinates as the center, nested sub-neighborhoods are defined within the local neighborhood using multiple radii of different scales. Calculate the curvature variance of each grid node in each sub-neighborhood to obtain the sequence of curvature variance as a function of scale; Perform linear fitting on the sequence to obtain the slope of the fitted line, and use the slope as a scale dependence index. The scale dependence index is weighted and fused with the local tilt and local fluctuation characteristics to obtain the local terrain complexity index.

2. The intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face as described in claim 1, characterized in that, The controller is also configured as follows: Real-time oil pressure data and drill pipe advance speed data of each joint of the drill arm are continuously collected according to the preset sampling cycle; Based on oil pressure and propulsion speed data, the oil pressure change rate and propulsion speed change rate are calculated in real time. If the oil pressure change rate and propulsion speed change rate are negatively correlated and both exceed their respective set thresholds in the current sampling period, it is determined that the rock formation hardness has changed in the current sampling period, and the ratio of oil pressure to propulsion speed in the current sampling period is recorded as the rock formation hardness characterization value. If the oil pressure change rate and propulsion speed change rate are not negatively correlated in the current sampling period, or although they are negatively correlated but do not exceed their respective set thresholds, it is determined that the rock formation hardness is stable in the current sampling period, and the controller maintains the current control parameters of the drill arm. Based on the rock hardness characterization values ​​recorded in multiple consecutive sampling periods, combined with drill pipe advance displacement data, a continuous distribution curve of rock hardness along the drilling direction is established; when it is determined that the rock hardness changes during the current sampling period, the drill bit position corresponding to that sampling period is identified as the interface between soft and hard rock layers, and the degree of hardness difference is estimated based on the hardness characterization values ​​on both sides of the interface. Based on the degree of hardness difference on both sides of the rock layer interface, and combined with the drilling mechanics model, the direction and amount of deviation that the drill bit will produce when crossing the interface are predicted. While identifying the rock interface, the controller calculates the advance compensation amount and compensation timing based on the predicted deviation direction and amount, combined with the mechanical response delay parameters of the drill arm drive mechanism; the compensation direction is set to be opposite to the predicted deviation direction, and the compensation timing is set to ensure that the compensation action is completed before the drill bit completely crosses the interface; When the calculated compensation timing is reached, the controller will send an adjustment command containing the advance compensation amount to the drill arm drive mechanism to suppress the drilling direction deviation caused by the sudden change in lithology when the drill bit crosses the rock interface.

3. The intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face as described in claim 2, characterized in that, The drill pipe is also equipped with a vibration sensing unit, which is connected to the demodulation module. The controller is also configured as follows: After the drill bit contacts the working face, the axial vibration signal of the drill pipe collected by the vibration sensing unit is acquired in real time. The axial vibration signal is transformed in the time and frequency domain to extract the vibration energy characteristic value within the preset frequency band, and the change curve of the vibration energy characteristic value with the displacement of the drill pipe is recorded. When the rate of change of the vibration energy characteristic value exceeds the preset abrupt change threshold, the current position of the drill bit is determined to be the interface between soft and hard rock layers, and the coordinates of the interface position along the drill rod axis are recorded. The interface position identified by the vibration energy feature value is fused and compared with the interface position identified based on the rock layer hardness distribution curve. When the deviation between the two exceeds the preset deviation threshold, the interface position identified by the vibration energy feature value is taken as the standard. The hardness characterization values ​​corresponding to both sides of the new interface position are extracted from the continuous distribution curve, and the estimated hardness difference between the two sides of the interface is recalculated based on the extracted hardness characterization values.

4. The intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face as described in claim 3, characterized in that, The vibration sensing unit includes multiple vibration sensors spaced apart along the drill pipe axis; The controller is also configured as follows: The axial vibration signals collected synchronously by multiple vibration sensors are segmented according to a preset time window length. Cross-correlation analysis is performed on the vibration signals within each time window to calculate the arrival time difference of the vibration signals between two adjacent vibration sensors. Based on the arrival time difference and the interval distance between adjacent vibration sensors, the average wave velocity of the vibration wave propagating along the drill pipe within the current time window is calculated, and the average wave velocity is recorded as the real-time wave velocity as the drill pipe advances, forming a real-time wave velocity change curve. Slope analysis is performed on the real-time wave velocity change curve. When the absolute value of the slope exceeds the preset wave velocity change rate threshold, the location of the drill bit is determined to be the interface between soft and hard rock layers, and the coordinates of the interface location along the drill rod axis are recorded. The interface location identified based on real-time wave velocity is weighted and fused with the interface location identified based on vibration energy feature value. The fused location is then used as the updated interface location identified based on vibration energy feature value, and compared with the interface location identified based on rock layer hardness distribution curve.

5. The intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face as described in claim 1, characterized in that, The conformal mapping correction procedure includes: using the initial mapping coordinates as the initial values, a geodesic distance iterative optimization algorithm is used on the triangular mesh surface model to search for the optimal position that satisfies the constraint conditions as the corrected coordinates; the constraint condition is that the difference between the surface distance between the optimal position and the adjacent determined actual spatial coordinates of the borehole and the designed borehole spacing is less than a preset allowable error value.

6. The intelligent projection positioning and drilling accuracy control system for blast holes at the tunnel face as described in claim 1, characterized in that, When the controller sends adjustment commands to the drive mechanism of the drill arm, it specifically performs the following: Based on the real-time calculated continuous space curve and the material stiffness coefficient of the drill pipe, an elastic deformation deflection model of the drill pipe under the lateral resistance of the rock strata is established. Based on the elastic deformation deflection model and the real-time distance between the drill bit center point and the drill arm holder, the future trajectory deviation trend of the drill bit after continuing to drill a unit depth is predicted under the condition of maintaining the current thrust. When the real-time radial deviation value exceeds the set deviation threshold range, the controller calculates the advance correction amount based on the future trajectory offset trend and sends the adjustment command containing the advance correction amount to the drill arm drive mechanism, so that the adjustment action of the drill arm can compensate in advance for the subsequent displacement offset caused by the release of the elastic potential energy of the drill pipe.

Images