Deep overburden large-diameter deep hole sampling method

Abstract

The application discloses a deep thick overburden large-diameter deep hole sampling method, and relates to the technical field of deep hole sampling, which comprises the following steps: first and second sound wave receivers are arranged at two ends of a connecting structure in a drilling machine sampling device, a gap is kept between the two receivers, and the connecting lines of the two receivers are perpendicular to the drilling axis; during drilling, the two receivers synchronously acquire a composite signal composed of structure propagation interference and overburden interface reflection; the stable delay characteristics of the interference signal during structure propagation are utilized to identify and remove the structure propagation interference, and the pure reflection signal is extracted from the composite signal; the overburden interface distance is calculated according to the pure reflection signal; and after the interface position is determined, the drilling machine is continuously pushed down into the overburden to implement soil sampling. The application can determine the position of the overburden while drilling, thereby improving the positioning accuracy and sampling efficiency of deep hole sampling.

Description

Technical Field

[0001] This invention relates to the field of deep hole sampling technology, and more specifically, to a method for sampling large-diameter deep holes in thick overburden layers. Background Technology

[0002] Large-diameter deep-hole sampling in thick overburden typically requires the drilling rig to advance steadily over a considerable depth range, accurately identifying the bottom interface of the overburden before entering the bottom for soil sampling. However, in the increasingly deep drilling environment, borehole wall friction, the interaction between the drill string and the soil, and structural vibrations continuously generate complex acoustic interference, making the acoustic environment inside the deep borehole highly mixed. In practical engineering, to detect the location of the overburden interface, acoustic pulses are often emitted and the reflected signals are received to estimate the interface depth. However, in thick overburden, these reflected signals are often covered or superimposed by structural noise and airborne noise, increasing the difficulty of detecting the interface echo. As the drilling depth increases, the energy and form of various interferences also change dynamically, making traditional acoustic judgments relying on single-channel reception highly unstable. Therefore, how to detect the location of the overburden interface while drilling has become a problem that needs to be solved. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method for sampling large-diameter deep holes in thick overburden layers, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A method for sampling large-diameter deep holes in thick overburden layers includes the following steps:

[0006] A drilling rig sampling device is provided, including a housing and a connection structure disposed inside the housing and perpendicularly connected to the housing at one end;

[0007] A first acoustic receiver is provided at the connection between the connecting structure and the outer shell, and a second acoustic receiver is provided at the other end of the connecting structure. A gap is left between the first acoustic receiver and the second acoustic receiver, and the line connecting them is perpendicular to the drilling axis of the drilling rig.

[0008] An acoustic transmitter is installed inside the drilling rig to emit sound pulse signals along the drilling direction to excite the echo response in the overburden layer;

[0009] The first and second acoustic receivers synchronously receive composite signals during drilling. The composite signals include interference signals and reflected signals from the overburden interface. The interference signals are noise generated by the interaction between the drilling rig and the soil.

[0010] The interference signal is detected based on the time delay characteristics of the interference signal propagating along the connection structure, and the interference signal is removed from the composite signal to obtain a pure reflected signal. The distance to the cover layer interface is calculated based on the pure reflected signal, and the soil is extracted by drilling into the interior after determining that the cover layer interface has been reached.

[0011] Preferably, the process of detecting the interference signal includes: decomposing the composite signal to obtain multiple signal components, calculating the time-domain cross-correlation function between the signals received by the first acoustic receiver and the second acoustic receiver, and extracting the signal component with a fixed time delay Δt; when the fixed time delay Δt is within a preset time delay range, determining that the corresponding signal component is an interference signal propagating along the connection structure.

[0012] Preferably, the method further includes:

[0013] Among multiple signal components, the interference signal propagating along the connection structure is used as a reference template, and similarity matching is performed on the remaining signal components. When the waveform similarity between a certain signal component and the reference template is higher than a preset threshold, the corresponding signal component is determined to be a homogeneous interference signal propagating along the air.

[0014] The pure reflected signal is obtained by simultaneously removing interference signals propagating along the connection structure and interference signals of the same origin propagating along the air from the composite signal.

[0015] Preferably, the preset delay range is obtained through a calibration phase, which includes:

[0016] The drilling rig casing was made to vibrate with a known waveform, and the structural propagation interference signals received by the first and second acoustic receivers were recorded.

[0017] Multiple sets of structural propagation delay samples are calculated based on the time-domain cross-correlation function, and the minimum delay Δt is determined based on the statistical distribution of the structural propagation delay samples. min With the maximum time delay Δt max , Δt min With Δt max The range between these values ​​is used as the preset delay range.

[0018] Preferably, the decomposition of the composite signal includes segmenting the composite signal according to a preset time window, performing a Fourier transform on each segment, and dividing each frequency band signal into several independent signal components according to the energy distribution of different frequency bands, so as to obtain multiple signal components constituting the composite signal.

[0019] Preferably, the method further includes:

[0020] During the drilling process, the measured distance of the overburden interface is acquired as a function of the drilling distance. If the slope of the curve is positive, an abnormal warning is issued.

[0021] Preferably, the method further includes: during the drilling process, acquiring the advance distance ΔL between two adjacent measurements and the change in the measurement distance ΔD at the overburden interface;

[0022] When the absolute value of the change ΔD does not fall within the ratio range k1·ΔL to k2·ΔL set for ΔL, it is determined that there is an anomaly in the measurement of the cover layer interface, where k1 and k2 are preset ratio coefficients.

[0023] Preferably, the value range of k1 is set as: k1 = 0.5 to 0.9.

[0024] Preferably, the value range of k2 is set as: k2 = 1.1 to 1.5.

[0025] Preferably, the acoustic transmitter is positioned on a reference line parallel to the drilling rig's propulsion axis, wherein the reference line passes through the midpoint between the first acoustic receiver and the second acoustic receiver.

[0026] The advantage of this invention over existing technologies lies in its arrangement of two sets of acoustic receivers between the drilling rig's outer shell and internal connecting structure, with the line connecting them positioned laterally relative to the drilling axis. This arrangement creates a stable propagation time delay difference between the two receivers for the structurally propagated interference signal. Because this lateral arrangement ensures that both receivers face the drilling direction simultaneously in space, when the interface reflected wave propagates back along the drilling axis, both receivers receive the reflected signal almost simultaneously. Meanwhile, the interference signal propagating along the drilling rig's metal structure inevitably creates a time delay difference between the two receivers, thus forming a clear distinguishing feature. This structural arrangement naturally separates the interface reflected signal from the structurally propagated interference signal in the time domain, providing a stable foundation for subsequent interference removal and interface identification.

[0027] Based on the aforementioned spatial layout, this invention utilizes the time delay characteristics of interference signals propagating along connecting structures to identify structural propagation noise from composite signals. Then, it removes structural propagation interference from the composite signal to obtain a pure reflected signal free of structural noise, thereby enabling accurate calculation of the overburden interface distance. Interface arrival determination is no longer affected by structural noise, allowing the drilling rig to promptly enter the target layer for soil sampling when approaching the bottom interface of the overburden, thus improving the accuracy of sampling locations.

[0028] By decomposing the composite signal and extracting interference components with stable time delays, this invention constructs a reference template for structure-propagated interference and further introduces an airborne interference identification function based on waveform similarity, enabling the removal of both types of co-originating interference in the composite signal. The reflected echo thus maintains sufficient clarity even in multi-interference environments, making interface location identification more stable and reliable. The time delay range obtained through calibration provides a verifiable basis for interference identification, enabling the system to adapt to different drilling tools and different formation environments.

[0029] During the dynamic drilling process, this invention establishes a curve showing the change in the interface distance of the overburden layer as a function of drilling distance, monitors the changing trend of the interface measurement results, and promptly provides an anomaly alert when the changing trend is inconsistent with the drilling direction. Furthermore, it enhances the reliability of the judgment by combining the quantitative correspondence between the drilling distance and the interface distance change. Through these error detection and self-verification mechanisms, this invention significantly improves the stability and reliability of acoustic detection in deep hole sampling of thick overburden layers, effectively avoiding sampling deviations caused by interface misjudgment.

[0030] This invention can locate the overburden interface in real time during drilling, enabling the determination of the target depth while drilling, thereby significantly improving sampling efficiency and operational accuracy. Attached Figure Description

[0031] Figure 1 This is a structural diagram of the present invention;

[0032] Figure 2 This is a schematic diagram of signal transmission in this invention. Detailed Implementation

[0033] The specific embodiments of the present invention will now be described with reference to the accompanying drawings.

[0034] Figure 1 shows an overall schematic diagram of the invention. This invention uses a drilling rig to drill holes, reaching the bottom of a deep overburden layer before soil sampling. The goal of this invention is to measure the distance to the deep overburden layer while drilling. In some specific embodiments, the drilling rig casing is made of high-strength alloy steel, capable of withstanding the high pressure and strong vibrations of deep-hole operations with diameters of 800–1500 mm. A transverse connecting structure is fixed inside the casing. One end of this connecting structure is welded perpendicularly to the inner wall of the casing or rigidly connected by high-strength bolts, while the other end extends suspended towards the center of the casing, forming a beam-like support. The total length of the connecting structure is typically between 400 and 800 mm, precisely designed according to the diameter and rigidity requirements of the drilling rig casing to provide sufficient transverse baseline length while ensuring the stability of the structural propagation path.

[0035] In this invention, the drilling device is not merely a mechanical structure, but rather the physical carrier of the entire ranging algorithm. The lateral connection structure provides a structural propagation path with controllable length and stiffness, allowing interference signals propagating along the drilling rig's steel structure to form a stable time difference between the two acoustic receivers. This provides reliable features for subsequent interference identification based on cross-correlation peak time delay. In other words, the drilling rig shell, the lateral connection structure, and the paired acoustic receivers together constitute a spatial filter. First, at the structural level, interference and useful reflections are differentiated along the propagation path. Then, at the computational level, these differences are amplified and utilized through cross-correlation and template matching. Therefore, the structural arrangement of the drilling device is tightly coupled with the subsequent interface distance calculation and interference removal, rather than being independent modules.

[0036] At the end of the connecting structure closest to the outer shell, a first acoustic receiver is installed close to the shell. This first receiver is a high-sensitivity piezoelectric hydrophone with a frequency response range of 20Hz to 12kHz and a sensitivity better than -185dB, capable of simultaneously capturing Maud's low-frequency structural vibrations and high-frequency reflected echoes. A second acoustic receiver, identical in model and installation orientation, is installed at the other end of the connecting structure. A fixed gap of 300–600 mm is maintained between the two receivers, and their connection line is strictly perpendicular to the drilling axis. This lateral symmetrical arrangement is one of the core innovations of this invention. The reflected waves from the bottom interface of the overburden layer can be considered approximately plane waves over a distance of 30–100 meters. When returning parallel to the drilling axis, the two receivers, being on the same lateral plane and facing the drilling direction simultaneously, receive the reflected signals almost at the same moment, with a time difference of less than 20 microseconds. The cross-correlation peak stably appears near zero time delay. The structural noise generated by the friction and impact between the drilling rig and the soil must propagate along the fixed, rigid path of the metal connection structure. The propagation path length is basically equal to or slightly longer than the baseline length (considering the installation of supports and detours). Under the dominance of low-frequency bending waves and shell modes, the equivalent group velocity is typically 800–2000 m / s, resulting in a highly stable propagation time delay difference Δt, usually between 0.1 and 2.0 milliseconds, which is precisely determined during the calibration phase. This time delay difference fluctuates by less than 5% throughout the drilling process, becoming the most reliable identification feature for subsequent interference removal.

[0037] In a further embodiment, an acoustic transmitter is installed at the bottom of the drilling rig near the drill bit, along a reference line parallel to the drilling axis. This reference line is specifically designed to pass through the midpoint of the line connecting the first and second acoustic receivers, ensuring that the main lobe of the emitted sound beam is directly facing the interface of the underlying overburden layer, and that the amplitude responses of the two receivers to the reflected signals are essentially identical, with an amplitude difference of less than 3 dB. The acoustic transmitter uses a high-power piezoelectric transducer with a center frequency of 2.5–4 kHz, emitting a sinusoidal envelope pulse signal with a width of 3–5 milliseconds each time, achieving a peak sound pressure level of over 192 dB, sufficient to excite detectable interface reflection echoes in saturated overburden layers 50–120 meters deep.

[0038] As shown in Figure 2, during drilling, the acoustic transmitter emits a probe pulse every 1.5 to 3 seconds. The first and second acoustic receivers acquire the composite signal synchronously, with a sampling rate set above 50 kHz to ensure microsecond-level time delay resolution. The composite signal actually contains three main components: structural noise propagating along the metal structure, interference from the same source propagating directly through the borehole fluid (air or mud), and the truly useful echo reflected from the bottom interface of the overburden. Structural noise has the largest amplitude and longest duration, making it difficult to separate using traditional methods.

[0039] In traditional deep-hole sampling projects, the common practice is to rely entirely on changes in drilling parameters such as drill torque, current, and drilling pressure, combined with operator experience, to determine whether the approach to the bottom interface of the overburden layer is possible. Alternatively, a single-channel longitudinal acoustic time measurement method is used to roughly read the echo arrival time under strong vibration conditions. These methods have two typical problems: one is that the interface reflection echo is completely submerged under strong structural noise, resulting in the system being unable to provide any effective interface distance; the other is misjudgment, where certain structural noise or direct fluid interference forms seemingly reasonable wave packets in the time domain, which are misjudged as interface reflections, causing interface distance deviations of more than two meters. In contrast, this invention uses two transversely arranged acoustic receivers, utilizing the difference in time delay between the structural propagation path and the fluid propagation path, to classify all interference strongly related to the drill rig's steel structure into common-source interference. First, a stable time delay range is determined during the calibration phase. Then, during actual drilling, it is jointly identified through cross-correlation peak time delay and waveform similarity, and after alignment in the time domain, it is subtracted from the original composite signal. In this way, the interface distance calculation is no longer based on dirty signals with strong interference, but on clean reflected echo envelopes, which greatly reduces the probability of misjudgment and ensures that every update of the interface distance is supported by clear physical basis and traceable signal characteristics.

[0040] To cleanly extract the reflected signal from the composite signal, this invention first decomposes the composite signal from the two channels. Specifically, the composite signal, acquired in each session with a duration of 300–600 milliseconds, is segmented into fixed time windows of 15–25 milliseconds. A Fast Fourier Transform (FFT) is performed on the signal in each time segment to obtain the frequency domain energy distribution. Based on the energy concentration, multiple frequency bands, such as 20Hz–400Hz, 400Hz–2kHz, and 2kHz–7kHz, are extracted. Then, an Inverse Fourier Transform (IFT) is used to reconstruct several time-independent signal components, typically yielding 10–18 components. These components include wideband low-frequency structural noise, narrowband mid-frequency reflected echoes, and some mid-frequency fluid-induced direct-source interference.

[0041] Next, a time-domain cross-correlation operation is performed on the two channel signal components within the same time window. The cross-correlation function will form a sharp peak on the time delay axis; the time delay corresponding to the peak position is the propagation delay of that component between the two receivers. Specifically, the received signal of the first acoustic receiver is denoted as... The received signal of the second sound wave receiver is denoted as By constructing a cross-correlation function between the two It can be achieved by shifting the amount of time. Next, compare the similarity between two signals to find their best matching time offset. The cross-correlation function is usually written as:

[0042] ;

[0043] exist When it is a specific value, It will reach its maximum value, and the corresponding τ is the time offset required to align the two signals. This time offset is defined as the cross-correlation peak delay Δt.

[0044] For genuine interface reflection signals, since they arrive almost simultaneously, the cross-correlation peak appears at a position where Δt≈0; while for structural noise propagating along the connection structure, the peak stably appears at a fixed non-zero time delay Δt0. This Δt0 will only fluctuate slightly throughout the drilling process and will not change abruptly.

[0045] In some embodiments, to accurately determine which components belong to structural propagation interference, a reliable preset time delay range needs to be obtained through a calibration phase. The calibration phase is conducted on the ground: the drilling rig does not drill; instead, it generates structural vibrations of known waveforms on the outer shell using an electromagnetic hammer or hydraulic impactor, repeating this process 150–300 times, and recording the pure structural propagation signals received by two receivers. For each recording, the cross-correlation peak time delay is calculated, resulting in a set of time delay samples. The distribution of these samples is statistically analyzed; in some embodiments, it is found that the time delay basically follows a normal distribution, and the mean ± 3σ is taken as the minimum time delay Δt. min and maximum time delay Δt max In some embodiments, Δt is taken. min ≈0.15 milliseconds, Δt max ≈1.8 milliseconds. This preset delay range can accommodate minor differences caused by different drill bit lengths, materials, and installation methods.

[0046] During the actual drilling, for each signal component extracted, its cross-correlation peak delay Δt is calculated. If Δt falls within the calibrated Δt... min ~Δt max If the signal is within the specified range, the component is immediately identified as structural interference propagating along the connection structure and saved as a reference template. The reference template is typically selected from the structural interference component with the highest signal-to-noise ratio.

[0047] In a further embodiment, this reference template is used to perform waveform similarity matching on the remaining signal components within the same window. Specifically, the normalized cross-correlation coefficient or cosine similarity is calculated. In some embodiments, when the similarity exceeds a preset threshold of, for example, 0.7 to 0.8, the component is considered to be a homologous interference to structural vibrations that propagates directly through borehole fluid (air or mud), and is also marked as interference. In this way, both structural propagation and direct fluid propagation homologous interferences can be completely identified.

[0048] Finally, by aligning all components marked as interference in the time domain and subtracting them directly from the original composite signal, a nearly interference-free pure reflected signal is obtained. The pure reflected signal will exhibit a clear echo envelope highly similar to the transmitted pulse waveform. The envelope is extracted using Hilbert transform, and the arrival time t corresponding to the envelope peak is automatically searched. peak The distance D between the overlay interface and the surface layer is equal to the speed of sound v multiplied by t. peak Half of that. In saturated overburden, the sound velocity v is usually taken as 1460–1540 m / s (which can be corrected in real time to take into account the effects of temperature, density, and moisture content), and the calculation accuracy can reach ±0.15 meters.

[0049] This invention does not introduce additional complex empirical models in the calculation of the interface distance D. Instead, it uses the familiar principle of acoustic wave round-trip ranging, which uses the product of propagation time and sound speed to represent the propagation distance. It only ensures, through the aforementioned time delay feature extraction and dual interference removal process, that tpeak corresponds to the reflected echo from the bottom interface of the overburden layer, rather than any arbitrary high-amplitude interference envelope. In other words, the innovation of this invention focuses on determining which wave packet is considered the interface echo, rather than changing the fundamental physical formula of sound speed ranging. This makes it easier for field personnel to understand and accept, and also facilitates integration with existing ultrasonic testing equipment.

[0050] To further improve the reliability of deep hole sampling, this invention continuously records the change of the interface distance D with the drilling depth L throughout the entire drilling process, forming a DL curve. Under normal circumstances, as the drilling rig advances downward, the interface distance D should decrease monotonically, and the slope of the curve should always be negative. If, in a certain segment, the slope becomes positive, meaning the interface distance actually increases, this clearly violates the laws of physics and is highly likely a false echo caused by residual interference. The system immediately issues an audible and visual alarm, prompting the operator to pause the drilling and remeasure.

[0051] In addition to slope monitoring, a more stringent quantitative self-verification mechanism is implemented. The drilling rig advance distance ΔL between each pulse transmission is precisely measured using a high-precision encoder, typically between 0.1 and 0.6 meters. Simultaneously, the change in interface distance ΔD between two adjacent valid measurements is calculated. Theoretically, if the formation interface is approximately horizontal and uncompressed, ΔD should be close to -ΔL. However, actual overburden layers may have some compression or inclination, therefore the absolute value of ΔD should fall between 0.5–0.9 times ΔL and 1.1–1.5 times ΔL. This proportionality range was obtained through extensive field testing: the lower limit of 0.5–0.9 considers soil compression and sound velocity measurement errors, while the upper limit of 1.1–1.5 allows for a certain interface dip angle or measurement noise. If the absolute value of ΔD falls outside this range, the current measurement is considered abnormal, and the system automatically discards the result, transmits multiple pulses consecutively to take the median, or prompts the operator to manually check.

[0052] When multiple consecutive sets of measurements show that the distance from the interface to D is consistently less than 0.4 meters and the trend is reliable, the system determines that the bottom interface of the overburden has been reached. It then controls the drilling rig to switch to low-speed propulsion, enters the bottom of the bottom interface of the overburden, and extends the sampling tube to take soil samples from the bottom of the deep overburden.

[0053] In practical engineering, the sampling location accuracy has been improved from the traditional ±2 to 3 meters to within ±0.2 meters, which greatly improves the success rate and sample representativeness of deep borehole sampling in thick overburden layers.

[0054] In a flood control dike reinforcement project in a riverside city, multiple deep boreholes with a diameter of 1200 mm and a depth of approximately 70 meters were required to sample the bottom of a 35-meter-thick saturated fine sand overburden layer. Using traditional experience-based judgment and single-channel acoustic timing, three boreholes experienced premature penetration, yielding samples still containing significant fine sand layers, necessitating subsequent drilling corrections. The average rework time per borehole exceeded six hours. In the borehole section using the method of this invention, the system continuously provided a stable interface distance D when drilling to approximately 1.1 meters from the bottom interface of the overburden layer. When D decreased to within the range of 0.35–0.4 meters, it automatically triggered low-speed advancement and sampling commands. Ultimately, the error between the measured interface position and the design value for each borehole was controlled within 0.2 meters, eliminating the need for rework. The construction unit reported that this method not only reduced mud consumption and drill bit wear but also significantly lowered the risk of re-drilling due to misjudged interfaces, transforming deep overburden bottom sampling from an experience-based process into a quantifiable and verifiable engineering procedure.

[0055] In actual construction, the entire signal processing flow is completed in real time on the embedded industrial computer on the drilling rig. The algorithm is written in C++, and the processing time for a single operation is less than 250 milliseconds, fully meeting the real-time requirements of measurement while drilling. The acoustic receiver is coated with a waterproof and mud-proof layer, and the transmitter is protected by wear-resistant ceramic, ensuring long-term reliable operation in deep holes up to 100 meters deep and in high-pressure mud environments. Through the organic combination of the above-mentioned series of lateral receiver layout, time delay feature extraction, dual interference elimination, calibration adaptation, and multiple anomaly detection mechanisms, this invention achieves high-precision overburden interface identification in strong interference environments, providing stable and reliable technical support for sampling large-diameter deep holes with thick overburden.

[0056] Overall, the special structural arrangement of the drilling equipment is responsible for creating distinguishable propagation paths at the physical level, the interference elimination algorithm is responsible for converting these propagation differences into calculable time delays and waveform characteristics at the signal level, and the interface distance calculation provides the position parameters of the bottom interface of the overburden layer that can be directly used for construction decisions. The three constitute a complete closed loop from mechanical structure and signal processing to engineering decision-making.

[0057] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for sampling large-diameter deep holes in thick overburden layers, characterized in that, Includes the following steps: A drilling rig sampling device is provided, including a housing and a connection structure disposed inside the housing and perpendicularly connected to the housing at one end; A first acoustic receiver is provided at the connection between the connecting structure and the outer shell, and a second acoustic receiver is provided at the other end of the connecting structure. A gap is left between the first acoustic receiver and the second acoustic receiver, and the line connecting them is perpendicular to the drilling axis of the drilling rig. An acoustic transmitter is installed inside the drilling rig to emit sound pulse signals along the drilling direction to excite the echo response in the overburden layer; The first and second acoustic receivers synchronously receive composite signals during drilling. The composite signals include interference signals and reflected signals from the overburden interface. The interference signals are noise generated by the interaction between the drilling rig and the soil. The interference signal is detected based on the time delay characteristics of the interference signal propagating along the connection structure, and the interference signal is removed from the composite signal to obtain a pure reflected signal. The distance to the cover layer interface is calculated based on the pure reflected signal, and the soil is extracted by drilling into the cover layer after determining that the cover layer interface has been reached.

2. The method for sampling large-diameter deep holes in thick overburden layers according to claim 1, characterized in that, The process of detecting the interference signal includes: decomposing the composite signal to obtain multiple signal components, calculating the time-domain cross-correlation function between the signals received by the first acoustic receiver and the second acoustic receiver, and extracting the signal component with a fixed time delay Δt; when the fixed time delay Δt is within a preset time delay range, the corresponding signal component is determined to be an interference signal propagating along the connection structure.

3. The method for sampling large-diameter deep holes in thick overburden layers according to claim 2, characterized in that, The method further includes: Among multiple signal components, the interference signal propagating along the connection structure is used as a reference template, and similarity matching is performed on the remaining signal components. When the waveform similarity between a certain signal component and the reference template is higher than a preset threshold, the corresponding signal component is determined to be a homogeneous interference signal propagating along the air. The pure reflected signal is obtained by simultaneously removing interference signals propagating along the connection structure and interference signals of the same origin propagating along the air from the composite signal.

4. The method for sampling large-diameter deep holes in thick overburden layers according to claim 2, characterized in that, The preset delay range is obtained through a calibration phase, which includes: The drilling rig casing was made to vibrate with a known waveform, and the structural propagation interference signals received by the first and second acoustic receivers were recorded. Multiple sets of structural propagation delay samples are calculated based on the time-domain cross-correlation function, and the minimum delay Δt is determined based on the statistical distribution of the structural propagation delay samples. min With the maximum time delay Δt max , Δt min With Δt max The range between these values ​​is used as the preset delay range.

5. The method for sampling large-diameter deep holes in thick overburden layers according to claim 2, characterized in that, The decomposition of the composite signal includes segmenting the composite signal according to a preset time window, performing a Fourier transform on each segment, and dividing each frequency band signal into several independent signal components according to the energy distribution of different frequency bands, so as to obtain multiple signal components constituting the composite signal.

6. The method for sampling large-diameter deep holes in thick overburden layers according to claim 1, characterized in that, The method further includes: During the drilling process, the measured distance of the overburden interface is acquired as a function of the drilling distance. If the slope of the curve is positive, an abnormal warning is issued.

7. The method for sampling large-diameter deep holes in thick overburden layers according to claim 6, characterized in that, The method further includes: during the drilling process, acquiring the advance distance ΔL between two adjacent measurements and the change in the measurement distance ΔD at the overburden interface; When the absolute value of the change ΔD does not fall within the ratio range k1·ΔL to k2·ΔL set for ΔL, it is determined that there is an anomaly in the measurement of the cover layer interface, where k1 and k2 are preset ratio coefficients.

8. The method for sampling large-diameter deep holes in thick overburden layers according to claim 7, characterized in that, The value range of k1 is set as: k1 = 0.5 to 0.

9.

9. The method for sampling large-diameter deep holes in thick overburden layers according to claim 7, characterized in that, The range of values ​​for k2 is set to: k2 = 1.1 to 1.

5.

10. The method for sampling large-diameter deep holes in thick overburden layers according to claim 1, characterized in that, The acoustic transmitter is positioned on a reference line parallel to the drilling rig's propulsion axis, wherein the reference line passes through the midpoint between the first acoustic receiver and the second acoustic receiver.

Images