Geological exploration dynamic sounding device and method based on intelligent sensing system

Abstract

The application is suitable for the technical field of geological exploration, and provides a geological exploration dynamic sounding device and method based on an intelligent sensing system. The device comprises a rack, a power source fixed on the rack, a hitting assembly movably installed on the rack, and a detection assembly arranged below the hitting assembly. The hitting assembly comprises an electro-hydraulic hammer and a high-frequency electromagnetic proportional valve. The detection assembly comprises a detection rod body and a sensing head arranged at the lower end of the detection rod body. A three-axis accelerometer and an acoustic sensor are arranged in the sensing head. An electrode ring is arranged on the outer wall of the detection rod body. The hitting assembly is connected with the detection assembly through a transmission connecting sleeve. A controller is electrically connected with the high-frequency electromagnetic proportional valve and various sensors. The application can synchronously obtain multi-dimensional geological characteristics such as impact rebound, acoustics and electricity after single hitting. The geological type is accurately identified through fusion identification. The hitting strategy is adaptively and timely optimized according to the geological type. The accuracy and efficiency of exploration are improved.

Description

Technical Field

[0001] This invention belongs to the field of geological exploration technology, and in particular relates to a dynamic probing device and method for geological exploration based on an intelligent sensing system. Background Technology

[0002] In geotechnical engineering investigation, geological hazard assessment, and resource exploration, dynamic penetration testing (DPPT) is widely used as an in-situ testing technique. Existing DPPT equipment typically evaluates the hardness of strata by recording single physical indicators such as the number of standard penetration tests (SPT) blows or measuring the penetration rate. However, geological structures are often complex and heterogeneous. For example, wet soft clay and loose sand with similar physical and mechanical properties, or fractured rock masses and intact bedrock that are equally hard but have vastly different integrity, are difficult to accurately distinguish using only a single mechanical response indicator. Furthermore, traditional equipment often operates with constant impact energy and frequency, and cannot adaptively adjust the impact strategy according to real-time changes in the strata encountered. This may not only lead to over-penetration in soft strata, damaging the original soil sample and affecting the detection accuracy, but may also prolong the construction period and increase costs when encountering hard rock strata due to low rock breaking efficiency. Therefore, existing technologies are insufficient in terms of the ability to finely identify geological types and the intelligence and efficiency of the exploration process, making it difficult to meet the needs of high-precision geological exploration. Summary of the Invention

[0003] This invention provides a dynamic penetration test device and method for geological exploration based on an intelligent sensing system, aiming to solve the problem of excessive penetration in soft strata damaging the original soil sample and affecting the detection accuracy.

[0004] This invention is implemented as follows: a geological exploration power penetration device based on an intelligent sensing system, comprising: a frame; a power source fixedly mounted on the frame; an impact assembly movably mounted on a vertical guide rail on the frame and connected to the power source to obtain power; the impact assembly includes an electro-hydraulic hammer for generating impact energy and a high-frequency electromagnetic proportional valve for adjusting the impact energy and frequency; a detection assembly disposed below the impact assembly; the detection assembly includes a detection rod and a sensing head disposed at the lower end of the detection rod; the sensing head is provided with a triaxial accelerometer and an acoustic sensor, and an electrode ring is disposed on the outer wall of the detection rod; a transmission connecting sleeve for connecting the output end of the impact assembly to the upper end of the detection assembly to transmit impact energy; and a controller electrically connected to the high-frequency electromagnetic proportional valve, the triaxial accelerometer, the acoustic sensor, and the electrode ring respectively.

[0005] Furthermore, the transmission connecting sleeve is provided with a spline structure inside, so that the detection assembly can rotate circumferentially relative to the impact assembly while transmitting axial impact force.

[0006] Furthermore, the probe body is a hollow structure used to accommodate the signal lines of the triaxial accelerometer, the acoustic sensor, and the electrode ring; the signal lines are connected to the controller through a slip ring located on the top of the probe assembly.

[0007] Furthermore, the power source is a hydraulic pump station, which is connected to the impact assembly via a high-pressure oil pipe; the high-frequency electromagnetic proportional valve is installed at the oil inlet of the electro-hydraulic hammer to precisely control the flow rate and pressure of the hydraulic oil entering the piston cavity.

[0008] A dynamic probing method for geological exploration based on an intelligent sensing system includes the following steps: After a single impact, signals output by the triaxial accelerometer, the acoustic sensor, and the electrode ring are acquired, and a combined feature vector containing impact rebound characteristics, rock and soil acoustic characteristics, and stratum electrical characteristics is calculated and generated based on the signals. The combined feature vector is input into a preset geological classification model to calculate a geological classification result that represents the current geological type; Based on the geological classification results, a preset impact parameter mapping table is queried to determine the corresponding optimal impact strategy; The high-frequency electromagnetic proportional valve is controlled to execute the next strike based on the impact energy and frequency set according to the optimal striking strategy.

[0009] Furthermore, the impact rebound characteristics include: rebound peak value, rebound duration, and decay rate calculated based on the signal from the triaxial accelerometer.

[0010] Furthermore, the acoustic characteristics of the rock and soil are as follows: after frequency domain conversion of the signal from the acoustic sensor, the proportion of signal energy in a high-frequency band that has been adaptively calibrated to the total signal energy is calculated.

[0011] Furthermore, the preset steps of the geological classification model include: collecting physical samples of known geological types and calibrating them to obtain ground truth labels; using the impact device to perform impact tests on the physical samples to obtain corresponding combined feature vectors; and pairing the combined feature vectors with the ground truth labels to train and generate the geological classification model.

[0012] Furthermore, the preset steps of the impact parameter mapping table include: testing multiple combinations of impact energy and frequency for each calibrated geological type; simultaneously evaluating the penetration efficiency and signal quality of each test; selecting test groups that meet the preset signal quality threshold, and selecting the one with the highest penetration efficiency as the optimal impact strategy for that geological type and storing it in the mapping table.

[0013] Furthermore, it also includes: when the geological classification result undergoes a preset significant change during continuous impact, it is determined to have entered the stratigraphic boundary and a high-precision detection mode is triggered. The high-precision detection mode sets the impact energy to a lower limit that meets the minimum signal quality and increases the impact frequency to a preset upper limit to perform a fine scan of the stratigraphic interface.

[0014] Compared with the prior art, the embodiments of this application have the following main advantages: The present invention provides a geological exploration dynamic probing device and method based on an intelligent sensing system. By integrating a triaxial accelerometer and an acoustic sensor on the sensing head of the probe assembly and setting an electrode ring on the probe rod, it can simultaneously acquire impact rebound characteristics, rock and soil acoustic characteristics, and formation electrical characteristics after a single impact, and fuse these multi-dimensional information into a combined feature vector. The combined feature vector is input into a preset geological classification model for calculation, which can obtain more accurate geological classification results than a single physical index, thereby effectively distinguishing complex strata with similar physical properties but different origins. Furthermore, the impact parameter mapping table is queried according to the geological classification results to determine the optimal impact strategy, and the impact energy and frequency are adjusted in real time by controlling a high-frequency electromagnetic proportional valve through a controller, realizing adaptive optimization of the exploration process and improving exploration efficiency and data quality. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall frame structure; Figure 2 This is a schematic diagram of the power source structure; Figure 3 This is a schematic diagram of the impact assembly and its connecting structure; Figure 4 This is a schematic diagram of the detection assembly structure; Figure 5 This is a schematic diagram of the cross-sectional structure of the sensor head; Figure 6 This is a schematic diagram of the striking control method of the present invention.

[0016] In the diagram: 100, frame; 200, power source; 300, impact assembly; 310, electro-hydraulic hammer; 320, high-frequency electromagnetic proportional valve; 330, transmission connecting sleeve; 400, detection assembly; 410, detection rod; 420, sensor head; 421, triaxial accelerometer; 422, acoustic sensor; 430, electrode ring. Detailed Implementation

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0018] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0019] like Figure 1-6 As shown, this embodiment of the invention provides a geological exploration power penetration device based on an intelligent sensing system, comprising: a frame 100; a power source 200 fixedly mounted on the frame 100; an impact assembly 300 movably mounted on a vertical guide rail on the frame 100 and connected to the power source 200 to obtain power; the impact assembly 300 includes an electro-hydraulic hammer 310 for generating impact energy and a high-frequency electromagnetic proportional valve 320 for adjusting the impact energy and frequency; and a detection assembly 400 disposed below the impact assembly 300; The detection assembly 400 includes a detection rod 410 and a sensing head 420 disposed at the lower end of the detection rod 410; a triaxial accelerometer 421 and an acoustic sensor 422 are disposed inside the sensing head 420; an electrode ring 430 is disposed on the outer wall of the detection rod 410; a transmission connecting sleeve 330 is used to connect the output end of the impact assembly 300 to the upper end of the detection assembly 400 to transmit impact energy; and a controller, which is electrically connected to the high-frequency electromagnetic proportional valve 320, the triaxial accelerometer 421, the acoustic sensor 422 and the electrode ring 430 respectively.

[0020] Conventional geological exploration equipment often suffers from insufficient discrimination capabilities when facing complex and varied strata due to limited information, making it difficult to accurately distinguish media with similar physical properties but different geological origins. This embodiment provides a geological exploration dynamic penetration device based on an intelligent sensing system, which obtains more dimensions of geological information through collaborative structural design.

[0021] Preferably, the transmission connecting sleeve 330 is provided with a spline structure inside, so that the detection assembly 400 can rotate circumferentially relative to the impact assembly 300 while transmitting axial impact force.

[0022] In this embodiment, the transmission connecting sleeve 330 efficiently transmits the axial impact force generated by the impact assembly 300 to the detection assembly 400. To improve adaptability in complex strata, the transmission connecting sleeve 330 is equipped with a spline structure inside. While ensuring the effective transmission of axial impact force, it allows the detection assembly 400 to undergo a small-amplitude passive circumferential rotation and passive attitude adjustment when encountering inclined hard rocks or changes in stratum texture during penetration. This allows the sensor head 420 of the detection assembly 400 to find an easier angle for penetration, which can reduce the risk of jamming caused by excessive local force to a certain extent.

[0023] Preferably, the probe rod 410 has a hollow structure to accommodate the signal lines of the triaxial accelerometer 421, the acoustic sensor 422, and the electrode ring 430; the signal lines are connected to the controller through a slip ring located at the top of the probe assembly 400.

[0024] In this embodiment, considering that the detection assembly 400 may rotate circumferentially due to the spline structure, in order to ensure that the signal can be continuously and stably transmitted to the fixed controller during the rotation, a slip ring is provided on the top of the detection assembly 400. The signal line is first connected to the rotating part of the slip ring, and then the signal is transmitted to the fixed part of the slip ring through the contact structure such as brushes, and finally connected to the controller.

[0025] Preferably, the power source 200 is a hydraulic pump station, which is connected to the impact assembly 300 through a high-pressure oil pipe; the high-frequency electromagnetic proportional valve 320 is installed at the oil inlet of the electro-hydraulic hammer 310 and is used to precisely control the flow rate and pressure of the hydraulic oil entering the piston cavity.

[0026] In this embodiment, the power source 200 can take various forms. One specific, but non-limiting, implementation is a hydraulic pump station driven by an electric motor. This hydraulic pump station is connected to the impact assembly 300 via a high-pressure oil pipe to transmit hydraulic power. For precise control of the impact behavior, a high-frequency electromagnetic proportional valve 320 is directly installed at the oil inlet of the electro-hydraulic hammer 310. Its function is as a high-speed flow and pressure regulating actuator. It receives electrical signals from the controller and, according to the signal instructions, adjusts the flow and pressure of hydraulic oil entering the piston cavity inside the electro-hydraulic hammer 310 with a millisecond-level response speed. In this way, the controller can directly and accurately control the energy magnitude and frequency of each impact, providing a foundation for subsequent adaptive adjustment of the impact strategy based on geological type.

[0027] A dynamic probing method for geological exploration based on an intelligent sensing system includes the following steps: after a single impact, acquiring signals output by the triaxial accelerometer 421, the acoustic sensor 422, and the electrode ring 430 respectively, and calculating and generating a combined feature vector containing impact rebound characteristics, rock and soil acoustic characteristics, and formation electrical characteristics based on the signals; inputting the combined feature vector into a preset geological classification model to calculate a geological classification result characterizing the current geological type; querying a preset impact parameter mapping table according to the geological classification result to determine the corresponding optimal impact strategy; and controlling the high-frequency electromagnetic proportional valve 320 to execute the next impact according to the impact energy and frequency set by the optimal impact strategy.

[0028] In this embodiment, after each impact action, the controller synchronously acquires signals from the triaxial accelerometer 421, acoustic sensor 422, and electrode ring 430. These independent physical signals are then processed and calculated to generate a combined feature vector. This vector comprehensively describes the characteristics of the current stratum from the mechanical, acoustic, and electrical dimensions of the impact response. This combined feature vector is input into a preset geological classification model. The purpose of this model is to analyze this multidimensional information and output a clear geological classification result, such as determining whether the current stratum is "wet soft clay" or "intact basement". The classification result, "rock," serves as the basis for decision-making. A pre-defined impact parameter mapping table is used to find the optimal impact strategy matching the current geological type. This strategy specifies the impact energy and frequency to be used in the next impact. Based on the retrieved strategy, the controller sends specific instructions to the high-frequency electromagnetic proportional valve 320, causing it to adjust hydraulic parameters to execute the impact. Through this series of steps, the exploration method can be optimized in real time according to changes in the formation. The formation electrical characteristics are: the apparent resistivity calculated based on the voltage and current signals measured when the electrode ring 430 contacts the formation. The specific calculation method is as follows: An alternating current I of known frequency and amplitude is applied to the paired electrode rings 430 by the controller, while simultaneously measuring the potential difference U between the pair of electrode rings 430. The apparent resistivity ρs can then be calculated using the formula: ρs = K * (U / I), K is the electrode device coefficient, which is related to the geometric dimensions and spacing of the electrode rings 430 and can be determined through a pre-calibration experiment. This characteristic mainly reflects the water content, porosity, and ion concentration of the formation, and is effective in distinguishing between saturated sandy soils and dry, dense clay.

[0029] Preferably, the impact rebound characteristics include: rebound peak value, rebound duration, and decay rate calculated based on the signal from the triaxial accelerometer 421.

[0030] The impact rebound characteristic in this embodiment consists of three parameters, all calculated from the signals of the triaxial accelerometer 421: Rebound peak value, which refers to the maximum value of the acceleration amplitude in the rebound waveform; this value directly reflects the hardness of the formation, with harder formations typically producing larger rebound peak values; Rebound duration, calculated as the time from the moment of impact until the envelope amplitude of the rebound acceleration signal first decays and stabilizes below a preset multiple of the background noise level before the impact; this parameter reflects the stiffness characteristics of the formation, with greater stiffness resulting in faster energy transfer and typically shorter duration; and Decay rate, which describes the rate at which rebound energy dissipates in the formation. Its specific calculation process involves fitting an exponential decay model curve to the decreasing portion of the rebound acceleration signal's amplitude envelope from the peak value, as follows: A(t) = A0 * e^(-t / τ) + C, A(t) is the amplitude at time t; A0 is the initial amplitude; C represents the background noise level; The time constant τ is extracted from the fitting results; the reciprocal of τ, 1 / τ, is the decay rate, and the magnitude of this constant reflects the degree of fragmentation or plasticity of the formation.

[0031] Preferably, the acoustic characteristics of the rock and soil are as follows: after frequency domain conversion of the signal from the acoustic sensor 422, the proportion of signal energy in a high-frequency band that has been adaptively calibrated to the total signal energy is calculated.

[0032] The purpose of the acoustic characteristics of the soil and rock in this embodiment is to quantify the brittleness of the soil and rock medium under impact. The calculation process is as follows: A fast Fourier transform is performed on the time series of the signal collected by the acoustic sensor 422 during a single impact, converting it from the time domain to the frequency domain to obtain the signal spectrum. The purpose of this step is to reveal the distribution of signal energy at different frequencies. Then, a boundary point for dividing the high and low frequency bands is determined. This boundary point is determined using an adaptive calibration method, rather than using a fixed frequency value. This calibration process analyzes brittle fracture with known physical meaning. The spectral data of samples (such as intact bedrock) and plastically deformed samples (such as wet soft clay) are used to iteratively find the frequency points that can maximize the distinction between the energy distribution differences of these two types of samples, and these points are used as the final boundary. This process ensures the objectivity of frequency band division. After determining the range of the high-frequency band, the total signal energy within that frequency band is calculated, and its proportion of the total signal energy is obtained. This proportion is the final acoustic characteristics of the rock and soil. For example, the brittle sound produced when hitting hard rock will result in a higher proportion of high-frequency energy, while the muffled sound when hitting clay will result in a lower proportion.

[0033] Preferably, the preset steps of the geological classification model include: collecting physical samples of known geological types and calibrating them to obtain ground truth labels; using the impact device to perform impact tests on the physical samples to obtain corresponding combined feature vectors; and pairing the combined feature vectors with the ground truth labels to train and generate the geological classification model.

[0034] The geological classification model in this embodiment establishes a mapping relationship between multi-dimensional combined feature vectors and specific geological types. This requires building a sample library for training. This process involves systematically collecting physical samples (such as rock cores or soil samples) from different geological regions and sending these samples to the laboratory for standard physical and mechanical property tests to obtain their authoritative geological classification names. These names serve as the ground truth labels for model training. Then, the device in this technical solution is used to perform standardized impact tests on these physical samples that have obtained ground truth labels one by one. During the test, the combined feature vector corresponding to each sample is collected and calculated simultaneously. Each obtained combined feature vector is paired with its corresponding ground truth label to form a database containing a large number of feature and label data pairs. Using this database, a suitable machine learning algorithm, such as support vector machine, is selected for training. The training process uses the learning algorithm to find the decision boundary in the feature space that can best separate samples of different geological types and solidifies these learned boundary parameters, ultimately generating a geological classification model that can classify unknown geological samples.

[0035] Preferably, the preset steps of the impact parameter mapping table include: testing multiple combinations of impact energy and frequency for each calibrated geological type; simultaneously evaluating the penetration efficiency and signal quality of each test; selecting test groups that meet the preset signal quality threshold, and selecting the one with the highest penetration efficiency as the optimal impact strategy for that geological type and storing it in the mapping table.

[0036] The impact parameter mapping table in this embodiment is also pre-set before the equipment is used on-site. First, a series of parameter matrix tests are performed on physical samples of each geological type for which true value labels have been obtained. During the tests, different impact energies and impact frequency parameters are systematically traversed and combined. Each time a specific parameter combination is used for impact, two core indicators are simultaneously evaluated: first, penetration efficiency, defined as the penetration depth achieved per unit of energy consumption, reflecting the economy of the operation; and second, signal quality, quantified by analyzing the signal-to-noise ratio and saliency of the sensor signal, reflecting the accuracy of perception. After testing all parameter combinations for a geological sample, the optimal strategy is sought. The search process consists of two steps: First, based on the signal clarity requirements of the subsequent geological classification model, a minimum effective signal quality threshold is set, and all parameter combinations that can meet this threshold are selected to form an effective candidate strategy pool. Then, within this effective candidate strategy pool, the penetration efficiency is used as the sole criterion for comparison, and the parameter combination that brings the highest penetration efficiency is selected. The selected parameter combination is identified as the optimal striking strategy for this geological type and is stored in a mapping table, associated with the corresponding geological type label.

[0037] Preferably, when the geological classification result undergoes a predetermined significant change during continuous impact, it is determined that the geological boundary has been entered, and a high-precision detection mode is triggered. The high-precision detection mode sets the impact energy to a lower limit that meets the minimum signal quality and increases the impact frequency to a predetermined upper limit to perform a fine scan of the geological boundary.

[0038] In this embodiment, the controller detects a significant, pre-defined change in the geological classification results during continuous impacts. For example, the classification result stably changes from clay to sand within a short distance. At this point, the system determines that the stratigraphic boundary has been reached and triggers this mode. The impact strategy changes: the impact energy is no longer directly based on the values ​​in the mapping table, but is dynamically set through a real-time control process. This process uses an extremely low energy to make trial impacts and analyzes the signal-to-noise ratio (SNR) of the rebound signal. If the SNR is lower than the minimum usable signal quality threshold required to ensure the accuracy of subsequent discrimination, the impact energy is slightly increased and the test is repeated until the SNR first meets the threshold. At this point, the energy value is locked as the optimal tapping energy that ensures signal quality while avoiding over-penetration, i.e., the lower limit of the minimum signal quality. After locking this lower limit of energy, in order to collect as much information as possible at the boundary, the controller increases the impact frequency to the upper limit allowed by the system while reducing the overall lowering speed of the device. Through this combination of low energy and high frequency, high-density continuous sampling and fine scanning of the location and characteristics of the stratigraphic interface are achieved.

[0039] It should be noted that, for the sake of simplicity, the foregoing embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to the present invention. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0040] It should be understood that the disclosed apparatus can be implemented in other ways, given the several embodiments provided in this application. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units described above may be implemented in other ways in practice. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or communication connections shown or discussed may be through some interfaces; indirect coupling or communication connections between devices or units may be telecommunications or other forms.

[0041] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0042] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can still combine, add, delete, or otherwise adjust the features of the various embodiments of the present invention according to the circumstances without conflict or creative effort, thereby obtaining different technical solutions that do not fundamentally depart from the concept of the present invention. These technical solutions also fall within the scope of protection of the present invention.

Claims

1. A dynamic penetration test method for geological exploration based on an intelligent sensing system, characterized in that, The geological exploration dynamic penetration test is carried out using a geological exploration system based on an intelligent sensing system. The system includes: A frame (100); a power source (200) fixedly mounted on the frame (100); An impact assembly (300) is movably mounted on a vertical guide rail provided on the frame (100) and connected to the power source (200) to obtain power; the impact assembly (300) includes an electro-hydraulic hammer (310) for generating impact energy and a high-frequency electromagnetic proportional valve (320) for adjusting the impact energy and frequency. A detection assembly (400) is disposed below the impact assembly (300); the detection assembly (400) includes a detection rod (410) and a sensor head (420) disposed at the lower end of the detection rod (410); a triaxial accelerometer (421) and an acoustic sensor (422) are disposed inside the sensor head (420), and an electrode ring (430) is disposed on the outer wall of the detection rod (410); A transmission connecting sleeve (330) is used to connect the output end of the impact assembly (300) to the upper end of the detection assembly (400) to transmit impact energy; And a controller, which is electrically connected to the high-frequency electromagnetic proportional valve (320), the triaxial accelerometer (421), the acoustic sensor (422) and the electrode ring (430), respectively; The method includes the following steps: After a single impact, signals output by the triaxial accelerometer (421), the acoustic sensor (422), and the electrode ring (430) are acquired respectively, and a combined feature vector containing impact rebound characteristics, rock and soil acoustic characteristics, and stratum electrical characteristics is calculated and generated based on the signals. The combined feature vector is input into a preset geological classification model to calculate a geological classification result that represents the current geological type; Based on the geological classification results, a preset impact parameter mapping table is queried to determine the corresponding optimal impact strategy; The high-frequency electromagnetic proportional valve (320) is controlled to execute the next strike according to the impact energy and frequency set by the optimal striking strategy; It also includes: when the geological classification result undergoes a preset significant change during continuous impact, it is determined that it has entered the stratigraphic boundary and a high-precision detection mode is triggered. The high-precision detection mode sets the impact energy to a lower limit that can meet the minimum signal quality and increases the impact frequency to a preset upper limit to perform a fine scan of the stratigraphic interface.

2. The geological exploration dynamic penetration test method based on an intelligent sensing system according to claim 1, characterized in that, The transmission connecting sleeve (330) is provided with a spline structure inside, so that the detection assembly (400) can rotate circumferentially relative to the impact assembly (300) while transmitting axial impact force.

3. The geological exploration dynamic penetration test method based on an intelligent sensing system according to claim 1, characterized in that, The probe rod (410) is a hollow structure used to accommodate the signal lines of the triaxial accelerometer (421), the acoustic sensor (422), and the electrode ring (430); the signal lines are connected to the controller through a slip ring located on the top of the probe assembly (400).

4. The geological exploration dynamic penetration test method based on an intelligent sensing system according to claim 1, characterized in that, The power source (200) is a hydraulic pump station, which is connected to the impact assembly (300) via a high-pressure oil pipe; the high-frequency electromagnetic proportional valve (320) is installed at the oil inlet of the electro-hydraulic hammer (310) and is used to precisely control the flow rate and pressure of the hydraulic oil entering the piston cavity.

5. The geological exploration dynamic penetration test method based on an intelligent sensing system according to claim 1, characterized in that, The impact rebound characteristics include: rebound peak value, rebound duration and decay rate calculated based on the signal from the triaxial accelerometer (421).

6. The geological exploration dynamic penetration test method based on an intelligent sensing system according to claim 5, characterized in that, The acoustic characteristics of the soil and rock are as follows: after frequency domain conversion of the signal of the acoustic sensor (422), the proportion of signal energy in a high-frequency band that has been adaptively calibrated to the total signal energy is calculated.

7. The geological exploration dynamic penetration test method based on an intelligent sensing system according to claim 5, characterized in that, The preset steps of the geological classification model include: collecting physical samples of known geological types and calibrating them to obtain true value labels; using the impact assembly (300) to perform impact tests on the physical samples to obtain corresponding combined feature vectors; and pairing the combined feature vectors with the true value labels to train and generate the geological classification model.

8. The geological exploration dynamic penetration test method based on an intelligent sensing system according to claim 5, characterized in that, The preset steps of the impact parameter mapping table include: testing multiple combinations of impact energy and frequency for each calibrated geological type; simultaneously evaluating the penetration efficiency and signal quality of each test; selecting test groups that meet the preset signal quality threshold, and selecting the one with the highest penetration efficiency as the optimal impact strategy for that geological type and storing it in the mapping table.

Images