Rock-soil in-situ detection equipment and method

By designing a geotechnical in-situ detection equipment combining probe rods, combined probes and neural network models, the problems of inaccurate geotechnical test results in the existing technology are solved, and efficient, accurate and low-cost geotechnical in-situ detection effects are achieved.

CN120195166APending Publication Date: 2025-06-24JIANGSU UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510279303.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-11
Publication Date
2025-06-24

AI Technical Summary

Technical Problem

The existing geotechnical testing technology results are inaccurate, high cost, low efficiency, and rely on manual experience.

Method used

A geotechnical in-situ detection equipment is designed, including a probe rod and a data processing system. A combined probe is installed at the lower end of the probe rod. Combined with an imaging unit, a detection unit and a wireless transmission equipment, the geotechnical category judgment and mechanical performance parameters are obtained through pre-trained CNN and PINN neural network models.

Benefits of technology

It realizes efficient, accurate and low-cost in-situ detection of geotechnical soil, and can accurately measure the mechanical properties of stratigraphic maps and each layer of geotechnical soil, and is suitable for geotechnical engineering and other fields.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120195166A_ABST
    Figure CN120195166A_ABST
Patent Text Reader

Abstract

The invention discloses a rock-soil in-situ detection device and method, and the method comprises the steps: shooting a rock-soil layer image at a certain depth in an exploration hole in real time through a camera, inputting the image into a pre-trained CNN neural network model, and judging the type of rock-soil at the depth; if it is judged that the probe is a soil body, the probe is driven by a force loading strategy of a detection unit driving module to transversely pierce into the soil body, and the time of drilling into the soil body and a probe sinking depth signal are recorded; if the rock mass is judged, recording the time of drilling into the soil mass and a depth signal of the drill bit under the driving of a force loading strategy of a driving module of the detection unit by utilizing the drill bit; inputting the mechanical signals into a pre-trained PINN neural network model, and obtaining mechanical property parameters corresponding to the soil body or rock; moving the probe rod to change the detection depth, and repeating the process to complete the detection of the whole exploration hole; and outputting a stratigraphic map and the mechanical property of each layer of rock soil. The method has the advantages of high efficiency, accuracy and low cost.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a geotechnical in-situ detection device and method. Background Art

[0002] Currently, geotechnical testing mainly relies on in-situ tests on site combined with laboratory tests on samples taken on site. Its results are affected by factors such as instruments, test conditions, test methods, and operation skills. The judgment of soil properties and soil mechanical parameters still depends on manual experience, resulting in inaccurate test results, high costs, and low efficiency in the implementation process. Summary of the Invention

[0003] Object of the Invention: The object of the present invention is to provide a geotechnical in-situ detection device and method that are efficient, accurate, and low-cost.

[0004] Technical Solution: A geotechnical in-situ detection device of the present invention includes a drill rod and a data processing system. A combined probe is provided at the lower end of the drill rod. The combined probe includes a camera unit, a detection unit, and a power box;

[0005] The camera unit includes a camera with supplementary lighting. The camera is installed in the internal cavity of the combined probe, and the optical axis is perpendicular to the axis of the drill rod, for taking images of the rock and soil layers inside the exploration hole; The combined probe has a transparent outer wall, and the position of the transparent outer wall corresponds to that of the camera;

[0006] The detection unit includes a probe and a drill bit. Holes for the probe and the drill bit to pass through are provided on the outer wall of the combined probe; The power box includes a detection unit driving module. Under the force loading strategy of the detection unit driving module, the probe is laterally inserted into the soil, recording the time of drilling into the soil and the depth signal of the probe immersed; Under the force loading strategy of the detection unit driving module, the drill bit records the time of drilling into the soil and the depth signal of the drill bit immersed;

[0007] The data processing system is used to analyze and process the collected image signals and mechanical signals to obtain geotechnical test data.

[0008] Furthermore, the number of cameras is two, which are respectively arranged on the upper and lower surfaces outside the power box; The power box further includes a camera unit driving module. The camera unit driving module has dual output ends, and the dual output ends respectively extend out of the upper and lower surfaces of the power box; The two cameras are respectively connected to the dual output ends through connecting rods; The camera unit driving module can drive the two cameras to synchronously rotate continuously by 360° around the central axis of the drill rod to realize image acquisition of the rock and soil layer structure inside the exploration hole; The transparent outer wall is arranged circumferentially.

[0009] Further, there are two probes and two drill bits each, which are installed on the combined probe in a circumferentially alternating arrangement, and the central angle between adjacent probes and drill bits is 90°; there are four sets of detection unit drive modules, which are respectively used to drive the corresponding probe or drill bit to expand and contract.

[0010] Further, the drive module uses a high-precision servo motor as the power source.

[0011] Further, the probe is replaced with a CPT probe installed with sensors for measuring tip resistance, sidewall friction resistance, and pore water pressure.

[0012] Further, the combined probe further includes a wireless transmission device for realizing data interaction between the combined probe and the ground control system and the data processing system.

[0013] Further, the data processing system uses a computer.

[0014] A geotechnical in-situ detection method of the present invention includes:

[0015] S1. At a certain depth inside the exploration hole, the image of the rock and soil layer is taken in real time through a camera;

[0016] S2. The collected image signal is input into a pre-trained CNN neural network model to judge the type of rock and soil at this depth, and the types of rock and soil include soil and rock mass;

[0017] S3. If it is judged as soil, the probe is laterally inserted into the soil under the force loading strategy of the detection unit drive module, and the time of drilling into the soil and the depth signal of the probe sinking are recorded; if it is judged as rock mass, the drill bit is used to record the time of drilling into the soil and the depth signal of the drill bit sinking under the force loading strategy of the detection unit drive module;

[0018] S4. The mechanical signal collected in step S3 is input into a pre-trained PINN neural network model to obtain the mechanical property parameters corresponding to this type of soil or rock and store them;

[0019] S5. Move the drill rod to change the detection depth, and repeat steps S1 to S4 until the detection of the entire exploration hole is completed;

[0020] S6. Output the rock and soil layer information and corresponding mechanical parameters at different depths.

[0021] Further, the orifice of the exploration hole is shaded to avoid interference of external light on the image signal during camera shooting.

[0022] Further, the CNN neural network model and the PINN neural network model are trained according to the local geological exploration expert database.

[0023] Beneficial effects: Compared with the prior art, the present invention has the following remarkable advantages: The in-situ geotechnical detection equipment provided by the present invention is small, portable and flexible to use. By matching the collected images with a pre-trained CNN neural network model to judge the geotechnical categories, according to the judgment results, mechanical signals are obtained by using a probe or a drill bit, and then the corresponding mechanical property parameters of the soil or rock are obtained by matching with a pre-trained PINN neural network model. It can efficiently and accurately measure the stratigraphic map and the mechanical properties of each layer of geotechnical materials in-situ, providing accurate and reliable basic stratigraphic conditions and physical and mechanical parameters for engineering design, model tests and numerical simulations in geotechnical engineering, hydraulic engineering, bridge and tunnel engineering, mining engineering, etc., and has good application prospects. Brief Description of the Drawings

[0024] Figure 1 is a schematic structural diagram of the in-situ geotechnical detection equipment provided by an embodiment of the present invention;

[0025] Figure 2 is a schematic diagram of the lower end structure of the probe rod in an embodiment of the present invention;

[0026] Figure 3 is Figure 2 sectional view of;

[0027] Figure 4 is Figure 2 top view of;

[0028] Figure 5 is a flow block diagram of the in-situ geotechnical detection method provided by an embodiment of the present invention. Detailed Embodiments

[0029] The present invention will be further described below with reference to the accompanying drawings.

[0030] Attached Figures 1 to 5 The reference numerals in the drawings are as follows:

[0031] 1, probe rod; 12, lower end of the probe rod; 2, combined probe; 211, transparent outer wall; 212, camera; 221, probe; 222, drill bit; 23, wireless transmission device; 24, power box; 241, camera unit drive module; 242, detection unit drive module; 3, data processing system.

[0032] As Figure 1 and Figure 2 shown, an embodiment of the present invention provides an in-situ geotechnical detection equipment, including a probe rod 1 and a data processing system 3. A combined probe 2 is provided at the lower end 12 of the probe rod. The combined probe 2 can be sent to a specified depth of the exploration hole through the probe rod 1.

[0033] Combined with Figure 3 and Figure 4, the combined probe 2 includes a camera unit, a detection unit, a wireless transmission device 23, and a power box 24.

[0034] The camera unit includes two cameras 212, which are installed in the inner cavity of the combined probe 2. The two cameras 212 are respectively arranged on the upper and lower surfaces outside the power box 24, and the optical axes of the two cameras 212 are perpendicular to the axis of the probe rod 1. The combined probe 2 has a transparent outer wall 211, and the transparent outer wall 211 is circumferentially arranged corresponding to the positions of the cameras 212 to ensure that the cameras 212 can clearly collect external image information through the transparent outer wall 211 within a 360° angle range, realizing multi-angle image acquisition of the working area of the probe 221 or the drill bit 222. The transparent outer wall 211 can be made of high-strength transparent tempered glass.

[0035] The camera unit driving module 241 in the power box 24 has dual output ends, and the dual output ends respectively extend out of the upper and lower surfaces of the power box 24. The two cameras 212 are respectively connected to the dual output ends through connecting rods. The camera unit driving module 241 can drive the two cameras 212 to rotate continuously by 360° synchronously around the central axis of the probe rod 1.

[0036] The camera 212 is configured with a supplementary lighting device (such as an LED supplementary lighting device) for providing auxiliary lighting in low-light environments to achieve high-definition image acquisition of the rock and soil layer structure inside the exploration hole. The collected image data is transmitted to the data processing system 3 in real time through the wireless transmission device 23. In this embodiment, the camera 212 preferably uses a CMOS image sensor to improve the sensitivity and resolution of image acquisition.

[0037] The detection unit includes a pair of probes 221 and a pair of drill bits 222. The probes 221 and the drill bits 222 are installed on the combined probe 2 in a circumferentially alternating arrangement, and the central angle between adjacent probes 221 and drill bits 222 is 90°, thus forming a symmetrically distributed structural layout. This arrangement is beneficial to improving the detection and sampling efficiency and ensuring the force balance of the combined probe 2 during operation. To facilitate the telescopic movement of the probes 221 and the drill bits 222, the outer wall of the combined probe 2 is provided with holes matching the probes 221 and the drill bits 222 for them to pass through.

[0038] Four sets of detection unit driving modules 242 are provided in the power box 24, which are respectively used to drive the corresponding probes 221 or drill bits 222 to expand and contract. Specifically, the detection unit driving module 242 can be configured in the following two working modes:

[0039] Mode 1: Simultaneously drive a pair of drill bits 222 and a pair of probes 221 to expand and contract;

[0040] Mode 2: Individually drive a pair of drill bits 222 or a pair of probes 221 to perform telescopic motion.

[0041] During the driving process, the movements of the two drill bits 222 are synchronized to provide reaction forces to each other to ensure the stability of the combined probe 2; similarly, the movements of the two probes 221 are also synchronized to provide reaction forces to each other to ensure the stability of the combined probe 2. The force signals applied by the detection unit driving module 242 to the probes 221 and the drill bits 222 are transmitted by the ground control system through the wireless transmission device 23.

[0042] Both the camera unit driving module 241 and the detection unit driving module 242 use high-precision servo motors as the power source, and among them, the detection unit driving module 242 adopts a force loading strategy to drive the probes 221 and / or the drill bits 222 to perform preset detection actions.

[0043] During the detection process, the probe 221 is laterally inserted into the soil mass, and at each level of force value given by the detection unit driving module 242, the time of penetration into the soil mass and the depth signal of the probe 221 immersed are recorded. Similarly, the drill bit 222 is laterally drilled into the rock mass, and at each level of force value given by the detection unit driving module 242, the time of penetration into the soil mass and the depth signal of the probe 221 immersed are recorded. The wireless transmission device 23 wirelessly transmits the collected image signals and mechanical signals to the data processing system 3 in real time. The data processing system 3 analyzes and processes the received image signals and mechanical signals to generate geotechnical test data. In this embodiment, the data processing system 3 uses a computer to achieve efficient data calculation and analysis functions.

[0044] Optionally, the probe 221 can be replaced with a CPT probe equipped with sensors for measuring the cone tip resistance (q), sidewall friction resistance (f), and pore water pressure (u), and the wireless transmission device 23 wirelessly transmits the collected mechanical signals to the data processing system 3 in real time.

[0045] As Figure 5 shown, the embodiment of the present invention also provides a geotechnical in-situ detection method, which uses the geotechnical in-situ detection device described in the embodiment of the present invention, and specifically includes the following steps:

[0046] S1. At a certain depth inside the exploration hole, drive the camera 212 to rotate 360° along the axis of the drill rod 1 through the camera unit driving module 241 to capture the image of the rock and soil layers inside the exploration hole in real time; during the shooting process, shade the orifice of the exploration hole to avoid the interference of external light on the image signal during the shooting of the camera 212 and improve the accuracy of subsequent image processing.

[0047] S2. Input the collected image signals into a pre-trained CNN neural network model to determine the geotechnical category at this depth, and the geotechnical categories include soil mass and rock mass.

[0048] S3. If it is determined to be soil, the probe 221 is laterally inserted into the soil under the force loading strategy of the detection unit driving module 242, and the time of drilling into the soil and the depth signal of the probe 221 sinking into the soil are recorded; if it is determined to be rock, the drill bit 222 is used to record the time of drilling into the soil and the depth signal of the drill bit 222 sinking into the soil under the force loading strategy of the detection unit driving module 242;

[0049] S4. The mechanical signals collected in step S3 are input into the pre-trained PINN neural network model to obtain the mechanical property parameters corresponding to this type of soil or rock and store them. The mechanical property parameters include the deformation modulus, strength, allowable bearing capacity, etc. of the geotechnical material.

[0050] S5. Move the probe rod 1 to change the detection depth, and repeat steps S1 to S4 until the detection of the entire exploration hole is completed.

[0051] In this embodiment, the detection order of the exploration hole is from bottom to top and from deep to shallow. When the lower end 12 of the probe rod moves up to the exploration hole opening, the operation ends.

[0052] S6. Output the rock and soil stratum information and corresponding mechanical parameters at different depths, that is, the stratum map and the mechanical properties of each layer of rock and soil.

[0053] The above CNN neural network model and PINN neural network model are configured in the data processing system 3 and pre-trained according to the local geological exploration expert database.

Claims

1. A rock and soil in-situ detection device, characterized in that: It comprises a probe rod (1) and a data processing system (3); a combined probe (2) is arranged at the lower end (12) of the probe rod; the combined probe (2) comprises a camera unit, a detection unit and a power box (24); The camera unit comprises a camera (212) with fill light, the camera (212) being installed in the internal cavity of the combined probe (2), with the optical axis being perpendicular to the axis of the probe rod (1), and being used to capture images of rock and soil layers inside the exploration hole; the combined probe (2) having a transparent outer wall (211), the transparent outer wall (211) corresponding to the position of the camera (212); The detection unit comprises a probe (221) and a drill bit (222); the outer wall of the combined probe (2) is provided with a hole for the probe (221) and the drill bit (222) to pass through; the power box (24) comprises a detection unit driving module (242); the probe (221) is driven by the force loading strategy of the detection unit driving module (242) to penetrate into the soil laterally, and records the time of drilling into the soil and the depth signal of the probe (221) being immersed; the drill bit (222) is driven by the force loading strategy of the detection unit driving module (242) to record the time of drilling into the soil and the depth signal of the drill bit (222) being immersed; The data processing system (3) is used to analyze and process the collected image signals and mechanical signals to obtain geotechnical test data.

2. The rock and soil in-situ detection equipment according to claim 1, characterized in that: There are two cameras (212), which are respectively arranged on the upper and lower surfaces outside the power box (24); the power box (24) also includes a camera unit driving module (241), the camera unit driving module (241) has dual output ends, and the dual output ends extend out of the upper and lower surfaces of the power box (24) respectively; the two cameras (212) are respectively connected to the dual output ends through connecting rods; the camera unit driving module (241) can drive the two cameras (212) to synchronously rotate 360 ​​degrees continuously around the central axis of the probe rod (1), so as to realize image acquisition of the rock and soil layer structure inside the exploration hole; the transparent outer wall (211) is arranged in a circumferential direction.

3. The rock and soil in-situ detection equipment according to claim 2, characterized in that: There are two probes (221) and two drill bits (222), which are installed on the combined probe (2) in a circumferentially alternating arrangement, and the central angle between adjacent probes (221) and drill bits (222) is 90°; there are four sets of detection unit drive modules (242), which are respectively used to drive the corresponding probes (221) or drill bits (222) to extend and retract.

4. The rock and soil in-situ detection equipment according to claim 3, characterized in that: The drive module uses a high-precision servo motor as the power source.

5. The rock and soil in-situ detection equipment according to claim 1, characterized in that: The probe (221) is replaced with a CPT probe equipped with sensors for measuring cone tip resistance, side wall friction resistance and pore water pressure.

6. The rock and soil in-situ detection equipment according to claim 1, characterized in that: The combined probe (2) also includes a wireless transmission device (23) for realizing data interaction between the combined probe (2) and the ground control system and the data processing system (3).

7. The rock and soil in-situ detection equipment according to claim 1, characterized in that: The data processing system (3) uses a computer.

8. A method for in-situ detection of rock and soil, characterized in that: include: S1. At a certain depth inside the exploration hole, a camera (212) is used to take real-time images of rock and soil layers; S2, inputting the collected image signal into the pre-trained CNN neural network model to determine the rock and soil category at the depth, where the rock and soil category includes soil and rock; S3, if it is determined to be soil, the probe (221) is driven by the force loading strategy of the detection unit driving module (242) to penetrate the soil laterally, and the time of drilling into the soil and the depth signal of the probe (221) being immersed are recorded; If it is determined to be a rock mass, the drill bit (222) is driven by the force loading strategy of the detection unit driving module (242) to record the time of drilling into the soil and the depth signal of the drill bit (222) being immersed; S4, inputting the mechanical signal collected in step S3 into the pre-trained PINN neural network model to obtain and store the mechanical property parameters corresponding to the soil or rock; S5, moving the probe rod (1) to change the detection depth, and repeating steps S1 to S4 until the detection of the entire exploration hole is completed; S6. Output rock formation information and corresponding mechanical parameters at different depths.

9. The rock and soil in-situ detection method according to claim 8, characterized in that: The opening of the exploration hole is shielded from light to prevent external light from interfering with the image signal when the camera (212) is shooting.

10. The rock and soil in-situ detection method according to claim 8, characterized in that: The CNN neural network model and the PINN neural network model are trained based on the local geological exploration expert database.