A method and device for advanced detection and intelligent obstacle avoidance of deep-hole blasting drilling

Abstract

This disclosure provides a method and apparatus for advanced detection and intelligent obstacle avoidance in deep-hole blasting drilling, relating to the field of deep-hole blasting engineering technology. The method includes: transmitting and receiving elastic wave signals to achieve advanced detection of geological structures ahead of the borehole; real-time acquisition of visualized image information of the borehole wall and surrounding rock to accurately capture geological details; collaborative processing and fusion analysis of the elastic wave signals and borehole television images to improve geological identification accuracy; and intelligent identification of unfavorable geological bodies and generation of the optimal obstacle-avoidance drilling path based on the fusion analysis results. This invention can adapt to the complex underground environment of deep-hole blasting, achieving accurate identification of unfavorable geological bodies and intelligent obstacle avoidance of the drilling path, significantly improving the safety, efficiency, and blasting effect of deep-hole blasting construction.

Description

Technical Field

[0001] This disclosure relates to the field of deep hole blasting engineering technology, and more specifically, to a method and device for advanced detection and intelligent obstacle avoidance of deep hole blasting boreholes. Background Technology

[0002] Currently, deep-hole blasting technology is widely used in large-scale engineering projects such as mining and tunnel excavation. The drilling quality directly affects the blasting effect, construction safety, and project progress. During deep-hole blasting drilling, the underground geological conditions are complex and variable. The presence of unfavorable geological bodies such as faults, karst caves, and weak interlayers can easily lead to safety accidents such as borehole collapse, drill rod jamming, and drill bit damage. At the same time, it can also cause problems such as uneven distribution of blasting energy and incomplete rock fragmentation, which seriously affect the quality and efficiency of the project.

[0003] Existing deep-hole blasting drilling detection technologies have many shortcomings: traditional geological exploration methods (such as ground seismic exploration and post-drilling coring) cannot obtain geological information in real time during the drilling process, resulting in delayed early warning; although single elastic wave detection technology can achieve advanced detection, it lacks sufficient accuracy in identifying details of geological bodies and is easily affected by environmental noise; in-hole television technology can only observe local conditions of the borehole wall and cannot detect unexposed geological structures ahead of the borehole; in addition, existing technologies lack effective data fusion and analysis methods, making it difficult to accurately identify adverse geological bodies, and a systematic intelligent obstacle avoidance control scheme has not been formed. Drilling path adjustment relies on manual experience, resulting in slow response speed and poor obstacle avoidance effect.

[0004] Therefore, there is an urgent need for a deep-hole blasting borehole detection method that integrates multi-source detection technology, has high-precision identification capabilities, and intelligent obstacle avoidance functions. This method would solve problems such as unreal-time geological information acquisition, low identification accuracy, and insufficient intelligent obstacle avoidance in existing technologies, thereby ensuring the safety and efficiency of deep-hole blasting construction. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this disclosure provides a method and apparatus for advanced detection and intelligent obstacle avoidance in deep-hole blasting drilling.

[0006] In a first aspect, embodiments of this disclosure provide a method for advanced detection and intelligent obstacle avoidance in deep-hole blasting drilling, including:

[0007] The original elastic wave signal emitted inside the hole is detected, and the reflected elastic wave echo signal is received.

[0008] Real-time acquisition of borehole image data, which includes at least: image data of the borehole wall and surrounding rock;

[0009] The raw and echo signals of the elastic wave and the image data inside the borehole are preprocessed respectively, and the geological bodies inside the borehole are analyzed using a preset data fusion algorithm;

[0010] Based on the fusion analysis results, the category of the adverse geological body is identified according to preset standards; and based on the detection parameters of the adverse geological body and the preset drilling and blasting requirements, the optimal obstacle avoidance drilling path is automatically generated.

[0011] In one optional embodiment, the detection of the emitted elastic wave signal within the aperture and the received reflected elastic wave echo signal includes:

[0012] The high-frequency elastic wave raw signal is transmitted into the surrounding rock inside the borehole by a transmitting sensor, and the elastic wave echo signal reflected by the geological body is collected by a receiving sensor.

[0013] The high-frequency elastic wave is penetrable.

[0014] In one optional implementation, the real-time acquisition of in-hole image data includes:

[0015] Image data of borehole wall and surrounding rock are acquired through a preset acquisition method, and the brightness of the supplementary light is adaptively adjusted during the acquisition process;

[0016] The preset acquisition method is rotation, with a rotation angle of 360 degrees.

[0017] In one optional implementation, the preprocessing of the elastic wave echo signal and the in-hole image data includes:

[0018] The original and echo signals of the elastic wave are normalized sequentially to obtain the first original elastic wave signal and the first elastic wave echo signal.

[0019] Fast Fourier transform is performed on the original and echo signals of the first elastic wave to obtain the single-sided amplitude spectrum and frequency domain signal of the original and echo signals of the first elastic wave.

[0020] A fifth-order Chebyshev type I bandpass filter is used to filter and reduce noise in the frequency domain of the first elastic wave original and echo signals to obtain the second elastic wave original signal and the second elastic wave echo signal.

[0021] An adaptive histogram equalization algorithm is used to enhance the image data to obtain the first image data;

[0022] The noise in the first image data is removed by a combination of median filtering and Gaussian filtering to obtain the second image data;

[0023] The geological body contour features in the second image data are extracted based on the Canny edge detection algorithm to generate the third image data;

[0024] The geological body contour features include: crack length, surrounding rock interface smoothness, and unfavorable geological body boundary.

[0025] In one optional implementation, identifying the category of the adverse geological body based on the fusion analysis results and according to preset criteria includes:

[0026] Based on the second elastic wave raw and echo signals and the third image data, a multi-dimensional feature vector is established through a preset weighted fusion strategy. The multi-dimensional feature vector includes at least: the wave velocity change and reflection energy characteristics of the second elastic wave raw and echo signals and the texture features and contour parameters of the third image data.

[0027] Based on preset identification criteria, a support vector machine model is used to identify the geological bodies and obtain their categories.

[0028] The weighting coefficients of the weighted fusion strategy are dynamically adjusted based on the reliability of the data.

[0029] In one optional implementation, the preset identification criteria include:

[0030] The categories of the adverse geological bodies are identified, and the detection parameters of the adverse geological bodies include at least: characteristic parameters of elastic wave velocity variation, characteristic parameters of reflected wave signal, and characteristic parameters of borehole television images;

[0031] The categories of adverse geological bodies include: faults, karst caves, and weak interlayers;

[0032] Based on the characteristic parameters of the elastic wave velocity change, the reduction value of the elastic wave velocity of the adverse geological body is compared with the preset elastic wave velocity reduction threshold.

[0033] If the reduction value of the elastic wave velocity reaches or exceeds the first reduction threshold, then the first type of discrimination process is entered.

[0034] If the reduction value of the elastic wave velocity does not reach the first reduction threshold but reaches or exceeds the second reduction threshold, then the second type of discrimination process is entered.

[0035] Wherein, the first reduction threshold is greater than the second reduction threshold;

[0036] Based on the first type of discrimination process, when the reflected wave signal features show that the S-wave energy disappears or attenuates, and the borehole television image features show that the cavity morphology, the unfavorable geological body is identified as a karst cave.

[0037] When the reflected wave signal features a sudden change in S-wave energy and the borehole television image features a fracture surface or fracture zone, the unfavorable geological body is identified as a fault.

[0038] Based on the second type of discrimination process, if the reflected wave signal features energy attenuation and the borehole television image features blurred rock layer interfaces and the presence of weak filling material, then the unfavorable geological body is identified as a weak interlayer.

[0039] In one optional implementation, the automatic generation of the optimal obstacle avoidance drilling path based on the detection parameters of the adverse geological body and the preset drilling and blasting requirements includes:

[0040] Based on the results of the fusion analysis, the spatial location and geometric boundaries of the unfavorable geological body are determined, which are defined as the extent of the unfavorable geological body.

[0041] Based on preset constraints, with the borehole design target point as the objective, the range of the adverse geological body is set as the obstacle zone in the path planning.

[0042] The A* algorithm is used to optimize the obstacle avoidance drilling path and generate the optimal obstacle avoidance drilling path.

[0043] In one optional implementation, the preset constraints include:

[0044] The constraint values ​​of the preset constraint conditions are set, including: a first preset distance, a first preset angle, and a first preset precision range;

[0045] The obstacle avoidance drilling path maintains a distance of not less than the first preset distance from the boundary of the obstacle area;

[0046] The local curvature of the obstacle avoidance drilling path must not exceed the first preset angle;

[0047] The deviation between the endpoint of the obstacle avoidance drilling path and the drilling design target point must not exceed the first preset accuracy range.

[0048] In one optional implementation, generating the optimal obstacle avoidance drilling path includes:

[0049] The information, including detection data, identification results of adverse geological bodies, and the optimal obstacle avoidance drilling path, will be transmitted to the ground control center.

[0050] The status of the drilling operation is displayed in real time at the ground control center.

[0051] In response to manual intervention commands issued through the ground control center, the optimal obstacle avoidance drilling path or drilling operation parameters are adjusted.

[0052] Secondly, this disclosure also provides a deep-hole blasting borehole advanced detection and intelligent obstacle avoidance device, including: an elastic wave detection module, an in-hole television imaging module, an elastic image fusion processing module, and an intelligent decision-making obstacle avoidance module, wherein;

[0053] The elastic wave detection module is used to detect the elastic wave signals emitted inside the hole and to receive the reflected elastic wave echo signals.

[0054] The borehole television imaging module is used to acquire borehole image data in real time, and the borehole image data includes at least: image data of the borehole wall and surrounding rock;

[0055] The elastic image fusion processing module is used to preprocess the elastic wave echo signal and the borehole image data respectively, and to analyze the geological information inside the borehole using a preset data fusion algorithm;

[0056] The intelligent decision-making obstacle avoidance module is used to identify the category of the adverse geological body based on the fusion analysis results and according to preset standards; and to automatically generate the optimal obstacle avoidance drilling path based on the detection parameters of the adverse geological body and preset drilling and blasting requirements.

[0057] Compared with existing technologies, this invention achieves real-time detection, accurate identification, and intelligent obstacle avoidance of adverse geological bodies during deep-hole blasting drilling through the coordinated operation of an elastic wave detection module, an in-hole television imaging module, an elastic image fusion processing module, and an intelligent decision-making and obstacle avoidance module. This technology can effectively adapt to the complex underground construction environment of deep-hole blasting, significantly reduce safety risks such as borehole collapse and drill rod jamming, improve drilling efficiency and blasting effect stability, and provide reliable advanced geological guarantees and technical support for the safe and efficient advancement of deep-hole blasting projects. Attached Figure Description

[0058] Figure 1 A flowchart of a deep-hole blasting borehole advance detection and intelligent obstacle avoidance method provided in this embodiment of the disclosure;

[0059] Figure 2 A flowchart for identifying unfavorable geological bodies in deep-hole blasting boreholes is provided as an embodiment of this disclosure;

[0060] Figure 3 A schematic diagram of a deep-hole blasting borehole advance detection and intelligent obstacle avoidance device provided in an embodiment of this disclosure;

[0061] Figure 4 This is a schematic diagram of the structure of a deep-hole blasting borehole advance detection device provided in an embodiment of the present disclosure;

[0062] Figure 5 This is a schematic diagram illustrating the usage state of a deep-hole blasting borehole advance detection device provided in an embodiment of this disclosure. Detailed Implementation

[0063] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0064] See Figure 1 The diagram shows a flowchart of a deep-hole blasting borehole advance detection and intelligent obstacle avoidance method provided in this embodiment of the present disclosure. The method includes steps S101 to S104, wherein:

[0065] S101: Detects the original elastic wave signal emitted inside the hole and receives the reflected elastic wave echo signal;

[0066] S102: Real-time acquisition of borehole image data, wherein the borehole image data includes at least: image data of the borehole wall and surrounding rock;

[0067] S103: The raw and echo signals of the elastic wave and the image data inside the borehole are preprocessed respectively, and the geological bodies inside the borehole are analyzed using a preset data fusion algorithm;

[0068] S104: Based on the fusion analysis results, identify the category of the adverse geological body according to preset standards; and automatically generate the optimal obstacle avoidance drilling path based on the detection parameters of the adverse geological body and the preset drilling and blasting requirements.

[0069] In practice, a high-frequency elastic wave raw signal is transmitted to the surrounding rock inside the borehole by a transmitting sensor, and the elastic wave echo signal reflected by the geological body is collected by a receiving sensor, and the signal acquisition timestamp is recorded simultaneously.

[0070] As an optional implementation, an elastic wave transmitting sensor with a center frequency of 50 kHz and an elastic wave receiving sensor with a sampling frequency of 8 MHz are selected, wherein the elastic wave transmitting sensor is used to excite high-frequency elastic waves to penetrate the surrounding rock, and the elastic wave receiving sensor is used to collect the original echo signal reflected by the geological body.

[0071] In practice, image data of the borehole wall and surrounding rock are acquired through a preset acquisition method;

[0072] As an optional implementation, a high-definition waterproof camera with a resolution of 1080P or higher is selected to capture images by rotating 360 degrees, and an LED fill light is used to adaptively adjust the fill light brightness during the acquisition process.

[0073] The original echo signal of the elastic wave and the in-hole television image data are transmitted synchronously through a signal transmission line to ensure that the two types of data are time-aligned.

[0074] During the actual transmission process, the data is initially verified to remove abnormal data.

[0075] For example, check whether the timestamps of consecutive data packets are continuous and whether they are reasonably synchronized with the system clock. If timestamps are disordered, repeated, or severely jump, it indicates that the data acquisition or transmission sequence may be disordered, and the relevant data packets will be regarded as abnormal and the abnormal data will be directly removed.

[0076] The original and echo signals of the elastic wave are normalized sequentially to obtain the first original elastic wave signal and the first elastic wave echo signal.

[0077] For example, the original echo signal of the elastic wave is normalized, assuming the data received by the sensor is [x n The normalized data is represented as [x] i ],make =β, α= Then the normalization formula is:

[0078] (1)

[0079] Performing a Fast Fourier Transform on the normalized signal yields the single-sided amplitude spectrum and frequency domain signal of the elastic wave echo signal, as shown in the formula:

[0080] (2)

[0081] Where f(t) is the normalized elastic wave signal, T is the signal period, ω0=2π / T is the fundamental frequency, nω0 is the harmonic frequency, and C is the fundamental frequency. n This represents the spectrum of the signal.

[0082] As an optional implementation, a fifth-order Chebyshev type I bandpass filter is used to filter and reduce noise in the frequency domain of the first elastic wave original and echo signals to obtain the second elastic wave original signal and the second elastic wave echo signal.

[0083] The fifth-order Chebyshev type I bandpass filter is an electronic filter that has equal ripple fluctuations in the passband frequency response and steep attenuation characteristics in the transition band.

[0084] Among them, "fifth order" represents the design complexity and filtering sharpness, "Chebyshev Type I" means that specific ripple is allowed in the passband in exchange for faster frequency cutoff characteristics, and "bandpass" means that only a specific frequency band of signals is allowed to pass through;

[0085] In this embodiment, the filter is used to accurately extract and enhance the target frequency band components from the original elastic wave signal, while effectively suppressing out-of-band noise.

[0086] For example, assuming a sampling frequency of 800Hz, the filter amplitude-frequency function is:

[0087] (3)

[0088] Where H(jΩ) is the amplitude-frequency function, Ω is the analog angular frequency, Ω0 is the center angular frequency of the passband, Ω1 is the lower boundary angular frequency of the passband, Ω2 is the upper boundary angular frequency of the passband, ε is the maximum ripple coefficient of the passband, and N=5 is the filter order.

[0089] The analog filter is converted into a digital filter using a bilinear transform, and the filtering result is calculated using a difference equation.

[0090] (4)

[0091] Where M is the number of input data, N is the number of output data, x is the filter input, y is the filter output, and a k and b k These are the filter coefficients.

[0092] As an optional implementation, an adaptive histogram equalization algorithm is used to enhance the image data. This is achieved by dividing the entire image into multiple overlapping or non-overlapping sub-regions and performing histogram equalization on each sub-region independently to obtain the first image data.

[0093] As an optional implementation, noise in the first image data is removed by a combination of median filtering and Gaussian filtering.

[0094] In a specific implementation, median filtering is applied to the first image data. Taking each pixel as the center, all pixel values ​​in the surrounding preset neighborhood are selected, sorted by size, and the median value is taken. This median value is then used to replace the original pixel value to filter out isolated noise points and impulse interference in the image.

[0095] The image processed by median filtering is then subjected to Gaussian filtering, and the image is smoothed by weighted averaging to obtain the second image data;

[0096] The weight distribution follows a Gaussian function (normal distribution), with neighboring pixels closer to the center pixel having higher weights.

[0097] As an optional implementation, the geological body contour features in the second image data are extracted based on the Canny edge detection algorithm to generate the third image data;

[0098] The geological body contour features include: crack length, surrounding rock interface smoothness, and unfavorable geological body boundary.

[0099] In practice, a Gaussian filter is used to lightly smooth the second image data to further reduce the interference of residual noise on edge detection.

[0100] The brightness gradient intensity and direction of each pixel in the image are calculated. The gradient magnitude is examined along the gradient direction, and only the local gradient maximum point is retained, thereby refining the edge and obtaining candidate edges with a width of one pixel.

[0101] Two thresholds are set: high and low. Pixels with gradient strengths higher than the high threshold are identified as strong edges, and those with gradient strengths lower than the low threshold are discarded. Pixels with gradient strengths in between are considered weak edges.

[0102] The high and low thresholds are determined based on the actual requirements for drilling accuracy.

[0103] For example, it is checked whether weak edges are connected to strong edges. Connected weak edges are preserved and connected to the edge profile, ultimately forming a continuous and complete geological body boundary.

[0104] The raw elastic wave signal, the echo signal, and the image data inside the hole are preprocessed respectively.

[0105] Based on the second elastic wave raw and echo signals and the third image data, a multi-dimensional feature vector is established through a preset weighted fusion strategy. The multi-dimensional feature vector includes at least: the wave velocity change and reflection energy characteristics of the second elastic wave raw and echo signals and the texture features and contour parameters of the third image data.

[0106] As an optional implementation, the elastic wave propagation velocity is recorded at each measurement point along the drilling depth direction at preset intervals;

[0107] For example, the average wave velocity of each measurement point and several adjacent points before and after it is calculated using the adjacent point smoothing method, thereby obtaining a baseline that can reflect the gradual change trend of wave velocity.

[0108] The elastic wave propagation velocity at each measurement point is compared with the baseline value at the corresponding location, and the reduction ratio is calculated. The reduction ratio is used as a parameter for the change in elastic wave velocity.

[0109] As an optional implementation, the echo signal at each measurement point is analyzed, and the contour line of the echo amplitude is plotted. By identifying significant peaks on the contour line, the location of the underground anomalous reflection interface is determined.

[0110] For the identified abnormal reflection interfaces, the reflection coefficient is calculated by comparing the amplitude of the reflected wave and the incident wave. This coefficient measures the degree of difference between the materials on both sides of the interface. The larger the reflection coefficient, the clearer the interface and the greater the difference in the properties of the materials on both sides.

[0111] In practical implementation, the first arriving longitudinal wave and the later arriving transverse wave in the elastic wave are analyzed;

[0112] Calculate the energy carried by the longitudinal wave and the transverse wave in the signal respectively, and determine the energy ratio between the two.

[0113] The energy ratio is extremely sensitive to the state of the geological body; for example, in cavities or fluids, the energy of transverse waves is significantly reduced.

[0114] As an optional implementation, the third image data is processed using a homomorphic filtering method, treating the image as a product of the illumination component (low frequency) and the reflection component (high frequency);

[0115] By performing logarithmic and Fourier transforms on the image, and using a filter function in the frequency domain that emphasizes high frequencies and suppresses low frequencies, the low-frequency brightness gradient components caused by uneven illumination are separated and weakened, thereby enhancing image details and achieving equalization of brightness across the entire image.

[0116] In practice, the median filtering method can be used to define a rectangular window of a specific size (such as 3×3 pixels) centered on the pixel to be processed, extract the gray values ​​of all pixels in the matrix window, sort them by size, and take the value in the middle after sorting as the new gray value of the center pixel.

[0117] Using the linear contrast stretching method, we statistically analyze the grayscale values ​​of all pixels in a single frame image or multiple frames of images at the same depth segment, and find out the minimum and maximum grayscale values.

[0118] By using a linear transformation formula, the gray value of each pixel in the original image is mapped to a new, uniform gray range (e.g., the entire available range from 0 to 255), resulting in preprocessed third image data.

[0119] The gray-level co-occurrence statistical method is used to perform texture quantization analysis on the preprocessed third image data. By statistically analyzing the gray-level relationship between pixels, four texture indicators are calculated, including: contrast, which reflects the strength of texture contrast; energy, which reflects the uniformity and fineness of texture; entropy, which reflects the complexity and disorder of texture; and homogeneity, which reflects the smoothness of local texture changes.

[0120] The entire hole wall image is spatially divided into numerous regular sub-blocks, and the above four texture indices are calculated for each sub-block.

[0121] The global texture features of an image at a given depth point are represented by the central tendency (such as average level) and dispersion of texture metrics across all sub-blocks.

[0122] As an optional implementation, image edge detection technology is used to identify and delineate the boundary shapes of geological anomalies such as caves and cracks in the borehole wall image, and geometric analysis is performed on these boundary shapes to extract their morphological description parameters, such as the smoothness and flatness of the contour.

[0123] For the measurement points to be analyzed, elastic wave and image features are collected and standardized.

[0124] As an alternative implementation, for each feature value, linear scaling is used to map the original value to a uniform range of values ​​(e.g., between zero and one).

[0125] The standardized feature values ​​are arranged in a predefined and fixed order and combined into a one-dimensional array as a multi-dimensional feature vector.

[0126] The geological bodies inside the borehole are analyzed using a preset data fusion algorithm, and the weight coefficients of the weighted fusion strategy are dynamically adjusted according to the reliability of the data.

[0127] In practice, the credibility of elastic wave features or image features at the current measurement point is quantitatively evaluated and used for weight allocation in the weighted fusion strategy.

[0128] As an optional implementation method, the acquired raw and echo signals of elastic waves are divided into a main segment containing effective waves and a tail segment containing only background noise. The average energy of the two segments is calculated to obtain the signal-to-noise ratio.

[0129] The signal-to-noise ratio (SNR) is the ratio of the energy of the effective main wave segment to the energy of the background noise segment. The higher the ratio, the clearer the signal, the less interference it is subject to, and the higher its reliability.

[0130] As an optional implementation, the Laplacian operator is used to perform convolution calculations on the image features, and the variance of the resulting image is statistically analyzed. The larger the variance, the clearer the image edges, the richer the details, and the higher the reliability.

[0131] Based on the confidence level obtained from the evaluation, weights are assigned to the multidimensional feature vectors. If the confidence score of the elastic wave data is high, a higher initial weight is assigned to it, and a lower initial weight is assigned to the image data. The sum of the two is 1.

[0132] When constructing multi-dimensional feature vectors, the standardized feature values ​​are multiplied by their corresponding weights and then concatenated.

[0133] For example, all standardized elastic wave feature values ​​(wave velocity variation, reflection coefficient, S / P energy ratio) are multiplied by the weighting coefficient of the elastic wave data respectively; all standardized image feature values ​​(each texture parameter, each contour parameter) are multiplied by the weighting coefficient of the image data respectively.

[0134] The weighted feature values ​​are concatenated in a preset order to form a weighted multi-dimensional feature vector, which is then used as the input feature to the support vector machine model.

[0135] The weighting coefficients are not fixed, but are dynamically optimized based on the feedback from the recognition results.

[0136] Based on preset identification criteria, a support vector machine model is used to identify the geological bodies and obtain their categories.

[0137] See Figure 2 As shown in the flowchart, this disclosure provides a flowchart for identifying unfavorable geological bodies in deep-hole blasting drilling, wherein:

[0138] In specific implementation, the types of the adverse geological bodies are identified, and the detection parameters of the adverse geological bodies include at least: characteristic parameters of elastic wave velocity variation, characteristic parameters of reflected wave signal, and characteristic parameters of borehole television images;

[0139] The categories of adverse geological bodies include: faults, karst caves, and weak interlayers.

[0140] As an optional implementation, based on the characteristic parameters of the elastic wave velocity change, the reduction value of the elastic wave velocity of the adverse geological body is compared with a preset elastic wave velocity reduction threshold.

[0141] If the reduction in elastic wave velocity reaches or exceeds the first reduction threshold, then the first type of discrimination process is initiated.

[0142] If the reduction value of the elastic wave velocity does not reach the first reduction threshold but reaches or exceeds the second reduction threshold, then the second type of discrimination process is entered.

[0143] Wherein, the first reduction threshold is greater than the second reduction threshold;

[0144] Based on the first type of discrimination process, when the reflected wave signal features show that the S-wave energy disappears or attenuates, and the borehole television image features show that the cavity morphology, the unfavorable geological body is identified as a karst cave.

[0145] When the reflected wave signal features a sudden change in S-wave energy and the borehole television image features a fracture surface or fracture zone, the unfavorable geological body is identified as a fault.

[0146] Based on the second type of discrimination process, if the reflected wave signal features energy attenuation and the borehole television image features blurred rock layer interfaces and the presence of weak filling material, then the unfavorable geological body is identified as a weak interlayer.

[0147] When outputting the geological body category, the support vector machine model also outputs a confidence score indicating the degree of certainty in this judgment;

[0148] For example, the support vector machine model calculates the distance between the input feature vector and the classification decision boundary, and uses a probability calibration method (such as Pratt scaling) to convert the distance into a confidence probability value between 0 and 1, thereby quantifying the confidence score of this class judgment.

[0149] If the confidence score reaches or exceeds the preset confidence score threshold, the identification result of the unfavorable geological body can be adopted and output.

[0150] And through local feature perturbation test, the features that play a key role in this discrimination are identified and sorted, thereby determining the feature that contributes the most and increasing its fusion weight by a preset small step size;

[0151] If the confidence score is lower than the preset confidence score threshold, check whether the current weight allocation deviates significantly from the initial weights based on the original quality (signal-to-noise ratio, image sharpness) of the current measurement point and re-evaluate them;

[0152] As an optional implementation, in the above-mentioned abnormal situation, the system will simultaneously trigger an alarm and request manual verification to ensure that the process is under control.

[0153] Based on the results of the fusion analysis, the spatial location and geometric boundaries of the unfavorable geological body are determined, which are defined as the extent of the unfavorable geological body.

[0154] Based on preset constraints, with the borehole design target point as the objective, the range of the adverse geological body is set as the obstacle zone in the path planning.

[0155] In practice, for each borehole, the vertical position range of the geological body on the borehole axis is determined based on its depth coordinates and the top and bottom depths of the identified adverse geological body.

[0156] By comparing the identification results of the same geological body between adjacent boreholes and combining the planar coordinates of the boreholes, the lateral extension trend of the unfavorable geological body between the boreholes can be inferred.

[0157] As an optional implementation, multiple spatial location points from different boreholes belonging to the same adverse geological body are used to generate a preliminary, continuous three-dimensional spatial surface as the extent of the adverse geological body using the Kriging interpolation spatial interpolation algorithm.

[0158] Furthermore, by connecting the top and bottom boundary points of the geological body in each borehole, a three-dimensional closed boundary model that encloses the entire body is constructed; the model is expressed in the form of a triangular mesh, thereby clearly defining the spatial occupancy range of the unfavorable geological body, i.e., the obstacle zone.

[0159] As an optional implementation, the constraint values ​​of the preset constraint conditions are set, including: a first preset distance, a first preset angle, and a first preset precision range;

[0160] In practice, the first preset distance is to ensure that there is sufficient safety buffer space between the drilling tool and the boundary of the unfavorable geological body;

[0161] The specific value of the first preset distance is determined based on engineering experience and drilling tool characteristics. The main factors considered include: the size of the drill bit and drill rod, possible drilling trajectory control errors, and the uncertainty that may exist at the boundary of the geological body.

[0162] The first preset angle constrains the severity of local deflection of the borehole path to match the drilling tool's directional drilling capability and ensure drilling safety.

[0163] The specific value of the first preset angle constraint is calculated and set according to the maximum build-up rate of the selected drill bit. When planning the path, the turning angle between adjacent line segments on the path will be calculated and ensured that it does not exceed the preset maximum allowable angle.

[0164] The first preset accuracy range is the maximum spatial tolerance allowed for the endpoint of the planned path to deviate from the design target point;

[0165] The specific value of the first preset accuracy range is determined according to the engineering design requirements. For example, it is a spherical region with the target point as the center and a specified length as the radius, and the planned path endpoint is ensured to fall within this spherical region.

[0166] In specific implementation, the obstacle avoidance drilling path and the boundary of the obstacle area maintain a distance of not less than the first preset distance;

[0167] The local curvature of the obstacle avoidance drilling path must not exceed the first preset angle;

[0168] The deviation between the endpoint of the obstacle avoidance drilling path and the drilling design target point must not exceed the first preset accuracy range.

[0169] The A* algorithm is used to optimize the obstacle avoidance drilling path and generate the optimal obstacle avoidance drilling path.

[0170] In practice, the three-dimensional engineering space, including the starting point, target point and obstacle zone, is discretized into a search space composed of many small cubes (grids);

[0171] The A* search algorithm starts from the starting grid, iteratively explores its neighboring grids, and calculates the estimated total cost (actual cost incurred plus estimated cost to the target grid) to reach the target grid from the starting grid through the current grid.

[0172] During the exploration process, the algorithm checks in real time whether the path segment violates the trajectory curvature limit. By calculating the angle between the direction vector of the new path segment and the previous path segment each time the extension path is considered, if the angle exceeds the first preset angle, the extension direction is abandoned.

[0173] The A* search algorithm always prioritizes expanding the grid with the lowest estimated total cost until the endpoint grid within the target accuracy range is found. By backtracking the parent-child grid relationships recorded during the search process, a node sequence that completely passes through the safe zone grid from the starting point to the target point is obtained.

[0174] The original path obtained by the A* search algorithm is a polyline formed by connecting the center points of the grid;

[0175] As an optional implementation, curve fitting technology (such as B-spline curve) is used to smooth the polyline, and under the premise of satisfying the curvature constraint, a smooth, continuous and directional three-dimensional drilling trajectory curve is generated, which is the optimal obstacle avoidance drilling path.

[0176] Output the coordinate sequence, attitude angle, and key parameters (such as total length and maximum curvature) of the optimal obstacle avoidance drilling path for subsequent drilling navigation.

[0177] The information, including detection data, identification results of adverse geological bodies, and the optimal obstacle avoidance drilling path, will be transmitted to the ground control center.

[0178] The status of the drilling operation is displayed in real time at the ground control center.

[0179] The ground control center receives the incoming data stream, parses, fuses, and reconstructs it; the monitoring interface displays the data in real time using multiple views and layered displays.

[0180] For example, on the main view, the real-time location of the borehole, the drilled trajectory, and the geological model around the borehole reconstructed based on the detection data are dynamically displayed in three-dimensional or cross-sectional form, with undesirable geological bodies highlighted.

[0181] The side panel displays key detection parameters (such as wave velocity and amplitude attenuation), the credibility score of the recognition results, and detailed parameters of the optimal obstacle avoidance path in the form of charts and lists.

[0182] The status area continuously updates the device's own operating information, such as propulsion speed, deflection angle, and sensor status.

[0183] In response to manual intervention commands issued through the ground control center, the optimal obstacle avoidance drilling path or drilling operation parameters are adjusted.

[0184] For example, the optimal obstacle avoidance drilling path is adjusted based on the aforementioned manual intervention command, taking into account both safety and economy.

[0185] Based on the same inventive concept, this disclosure also provides a deep hole blasting borehole advance detection and intelligent obstacle avoidance device for use with the deep hole blasting borehole advance detection and intelligent obstacle avoidance method. Since the principle of the control method in this disclosure is similar to the deep hole blasting borehole advance detection and intelligent obstacle avoidance method described above in this disclosure, the implementation of the deployment device can refer to the implementation of the method, and the repeated parts will not be described again.

[0186] See Figure 3 The diagram shown is a schematic of a deep-hole blasting borehole advanced detection and intelligent obstacle avoidance device provided in an embodiment of this disclosure. The device includes an elastic wave detection module 10, an in-hole television imaging module 20, an elastic image fusion processing module 30, and an intelligent decision-making obstacle avoidance module 40, wherein:

[0187] The elastic wave detection module 10 is used to detect the elastic wave signal emitted inside the hole and to receive the reflected elastic wave echo signal.

[0188] The borehole television imaging module 20 is used to acquire borehole image data in real time. The borehole image data includes at least the image data of the borehole wall and the surrounding rock.

[0189] The elastic image fusion processing module 30 is used to preprocess the elastic wave echo signal and the borehole image data respectively, and to analyze the geological information inside the borehole using a preset data fusion algorithm;

[0190] The intelligent decision-making obstacle avoidance module 40 is used to identify the category of the adverse geological body based on the fusion analysis results and according to preset standards; and to automatically generate the optimal obstacle avoidance drilling path based on the detection parameters of the adverse geological body and the preset drilling and blasting requirements.

[0191] See Figure 4 The diagram shown is a schematic representation of a deep-hole blasting borehole advance detection device provided in an embodiment of this disclosure, wherein:

[0192] As an optional implementation, an elastic wave transmitting / receiving sensor, a high-definition waterproof camera, and an LED supplementary lighting unit are fixedly installed 50-80cm away from the drill bit on the drill rod, allowing time for the drilling reaction.

[0193] In practice, each detection component is connected to the signal and image fusion processing module and the intelligent decision-making obstacle avoidance module via signal transmission lines to complete the debugging of sensor detection accuracy, image clarity and control command response speed, and ensure stable system operation.

[0194] As an optional implementation, a switchable dust cover is movably mounted on the outside of the elastic wave transmitting sensor, receiving sensor, and high-definition waterproof camera, and linked with the drill pipe or sensor bracket to ensure that it is closed during non-detection periods to isolate rock cuttings and mud, and opened during detection to ensure unobstructed signal and image acquisition.

[0195] See Figure 5 The diagram shown is a schematic representation of the usage state of a deep-hole blasting borehole advance detection device provided in this embodiment of the present disclosure, wherein:

[0196] During the drilling process, the device identifies and avoids obstacles in the borehole in real time through the intelligent decision-making obstacle avoidance module; the elastic image fusion processing module simultaneously collects and processes elastic wave signals around the borehole to generate a fused geological image; the intelligent decision-making module analyzes and judges based on the image data and real-time working conditions.

[0197] In this operating state, the device can detect and clearly identify adverse geological structures in front of and around the borehole, such as faults, weak interlayers and karst caves, thus providing accurate geological basis for the charge design and safe construction of deep hole blasting.

[0198] Those skilled in the art will understand that, in the methods described above in the specific embodiments, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic. It should be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information.

[0199] In the description of this specification, the terms "exemplary," "for example," "specifically," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

Claims

1. A method for advanced detection and intelligent obstacle avoidance in deep-hole blasting drilling, characterized in that, The method includes: The original elastic wave signal emitted inside the hole is detected, and the reflected elastic wave echo signal is received. Real-time acquisition of borehole image data, which includes at least: image data of the borehole wall and surrounding rock; The raw and echo signals of the elastic wave and the image data inside the borehole are preprocessed respectively, and the geological bodies inside the borehole are analyzed using a preset data fusion algorithm; Based on the fusion analysis results, the category of the adverse geological body is identified according to preset standards; and based on the detection parameters of the adverse geological body and the preset drilling and blasting requirements, the optimal obstacle avoidance drilling path is automatically generated.

2. The method according to claim 1, characterized in that, The detection of the emitted elastic wave signal within the hole and the received reflected elastic wave echo signal includes: The high-frequency elastic wave raw signal is transmitted into the surrounding rock inside the borehole by a transmitting sensor, and the elastic wave echo signal reflected by the geological body is collected by a receiving sensor. The high-frequency elastic wave is penetrable.

3. The method according to claim 1, characterized in that, The real-time acquisition of in-hole image data includes: Image data of borehole wall and surrounding rock are acquired through a preset acquisition method, and the brightness of the supplementary light is adaptively adjusted during the acquisition process; The preset acquisition method is rotation, with a rotation angle of 360 degrees.

4. The method according to claim 1, characterized in that, The preprocessing of the elastic wave echo signal and the in-hole image data includes: The original and echo signals of the elastic wave are normalized sequentially to obtain the first original elastic wave signal and the first elastic wave echo signal. Fast Fourier transform is performed on the original and echo signals of the first elastic wave to obtain the single-sided amplitude spectrum and frequency domain signal of the original and echo signals of the first elastic wave. A fifth-order Chebyshev type I bandpass filter is used to filter and reduce noise in the frequency domain of the first elastic wave original and echo signals to obtain the second elastic wave original signal and the second elastic wave echo signal. An adaptive histogram equalization algorithm is used to enhance the image data to obtain the first image data; The noise in the first image data is removed by a combination of median filtering and Gaussian filtering to obtain the second image data; The geological body contour features in the second image data are extracted based on the Canny edge detection algorithm to generate the third image data; The geological body contour features include: crack length, surrounding rock interface smoothness, and unfavorable geological body boundary.

5. The method according to claim 1, characterized in that, The identification of the unfavorable geological bodies based on the fusion analysis results and according to preset standards includes: Based on the second elastic wave raw and echo signals and the third image data, a multi-dimensional feature vector is established through a preset weighted fusion strategy. The multi-dimensional feature vector includes at least: the wave velocity change and reflection energy characteristics of the second elastic wave raw and echo signals and the texture features and contour parameters of the third image data. Based on preset identification criteria, a support vector machine model is used to identify the geological bodies and obtain their categories. The weighting coefficients of the weighted fusion strategy are dynamically adjusted based on the reliability of the data.

6. The method according to claim 5, characterized in that, The preset identification criteria include: The categories of the adverse geological bodies are identified, and the detection parameters of the adverse geological bodies include at least: characteristic parameters of elastic wave velocity variation, characteristic parameters of reflected wave signal, and characteristic parameters of borehole television images; The categories of adverse geological bodies include: faults, karst caves, and weak interlayers; Based on the characteristic parameters of the elastic wave velocity change, the reduction value of the elastic wave velocity of the adverse geological body is compared with the preset elastic wave velocity reduction threshold. If the reduction value of the elastic wave velocity reaches or exceeds the first reduction threshold, then the first type of discrimination process is entered. If the reduction value of the elastic wave velocity does not reach the first reduction threshold but reaches or exceeds the second reduction threshold, then the second type of discrimination process is entered. Wherein, the first reduction threshold is greater than the second reduction threshold; Based on the first type of discrimination process, when the reflected wave signal features show that the S-wave energy disappears or attenuates, and the borehole television image features show that the cavity morphology, the unfavorable geological body is identified as a karst cave. When the reflected wave signal features a sudden change in S-wave energy and the borehole television image features a fracture surface or fracture zone, the unfavorable geological body is identified as a fault. Based on the second type of discrimination process, if the reflected wave signal features energy attenuation and the borehole television image features blurred rock layer interfaces and the presence of weak filling material, then the unfavorable geological body is identified as a weak interlayer.

7. The method according to claim 1, characterized in that, The automatic generation of the optimal obstacle avoidance drilling path based on the detection parameters of the adverse geological body and the preset drilling and blasting requirements includes: Based on the results of the fusion analysis, the spatial location and geometric boundaries of the unfavorable geological body are determined, which are defined as the extent of the unfavorable geological body. Based on preset constraints, with the borehole design target point as the objective, the range of the adverse geological body is set as the obstacle zone in the path planning. The A* algorithm is used to optimize the obstacle avoidance drilling path and generate the optimal obstacle avoidance drilling path.

8. The method according to claim 7, characterized in that, The preset constraints include: The constraint values ​​of the preset constraint conditions are set, including: a first preset distance, a first preset angle, and a first preset precision range; The obstacle avoidance drilling path maintains a distance of not less than the first preset distance from the boundary of the obstacle area; The local curvature of the obstacle avoidance drilling path must not exceed the first preset angle; The deviation between the endpoint of the obstacle avoidance drilling path and the drilling design target point must not exceed the first preset accuracy range.

9. The method according to claim 7, characterized in that, The process of generating the optimal obstacle avoidance drilling path includes: The information, including detection data, identification results of adverse geological bodies, and the optimal obstacle avoidance drilling path, will be transmitted to the ground control center. The status of the drilling operation is displayed in real time at the ground control center. In response to manual intervention commands issued through the ground control center, the optimal obstacle avoidance drilling path or drilling operation parameters are adjusted.

10. A deep-hole blasting borehole advanced detection and intelligent obstacle avoidance device, characterized in that, The device includes: The elastic wave detection module is used to detect the elastic wave signals emitted inside the hole and to receive the reflected elastic wave echo signals. The borehole television imaging module is used to acquire borehole image data in real time, and the borehole image data includes at least: image data of the borehole wall and surrounding rock; The elastic image fusion processing module is used to preprocess the elastic wave echo signal and the borehole image data respectively, and to analyze the geological information inside the borehole using a preset data fusion algorithm; The intelligent decision-making obstacle avoidance module is used to identify the category of the adverse geological body based on the fusion analysis results and according to preset standards; and to automatically generate the optimal obstacle avoidance drilling path based on the detection parameters of the adverse geological body and preset drilling and blasting requirements.

Images