An ultrasonic guided wave detection robot for oil under water-sealed oil storage caverns and a detection method thereof

Abstract

The application discloses a kind of water seal oil cavern oil under ultrasonic guided wave detection robot and its detection method.It includes the robot main body of oil-resistant waterproof material, the adsorption device (vacuum chuck or magnetic chuck) suitable for concrete or steel structure cave wall, the ultrasonic guided wave detection module of adjustable detection depth, artificial intelligence data acquisition analysis evaluation module and power module.Detection method is through adsorption fixation, ultrasonic guided wave detection, intelligent data processing generates three-dimensional damage cloud picture, risk classification evaluation and other steps, realize the stable detection of oil under cave wall damage, intelligent identification and visual display.The application effectively improves the adsorption stability of detection, depth coverage accuracy, data processing intelligent level and risk closed-loop management ability, significantly meet the engineering needs of automation, precision and intelligence of water seal oil cavern oil under detection.

Description

Technical Field

[0001] This invention relates to the field of non-destructive testing technology for oil and gas storage facilities, and in particular to an ultrasonic guided wave testing robot and its testing method for oil under water-sealed oil caverns. Background Technology

[0002] Water-sealed oil caverns, as core facilities of national strategic petroleum reserves, are widely used in the oil and gas storage field due to their advantages such as large storage capacity, high safety, and excellent economic efficiency. However, their cavern walls are constantly exposed to a complex corrosive environment of oil and groundwater, making them prone to damage such as cracks, corrosion, and voids. Failure to detect and address these issues promptly can lead to major safety accidents such as oil leaks and cavern structural instability, seriously threatening energy security and public safety. Therefore, accurate detection of damage to the underground cavern walls is a crucial step in ensuring the safe operation of water-sealed oil caverns.

[0003] Currently, underwater inspection of water-sealed oil caverns mainly employs ultrasonic guided wave technology. This technology boasts advantages such as long detection distance (up to tens of meters), strong penetration, and the ability to achieve non-contact underwater inspection, making it the mainstream solution for underwater non-destructive testing. However, existing inspection equipment suffers from the following key shortcomings, making it difficult to meet practical engineering needs:

[0004] 1. Insufficient adsorption and fixation capacity, resulting in poor detection stability.

[0005] Existing equipment is mostly portable or semi-automated, relying on manual operation for fixation. In complex environments with high pressure, oil flow disturbances, and changes in tunnel wall curvature, the equipment struggles to maintain a stable fit against the tunnel wall, easily shifting or detaching, leading to detection interruptions, data gaps, and preventing continuous, full-coverage detection. Furthermore, the adjustment of detection depth (such as shallow cracks and deep corrosion) lacks precise control, making it difficult to cover multiple layers of damage to the tunnel wall.

[0006] 2. The data processing is not highly intelligent, resulting in insufficient efficiency and accuracy.

[0007] The detection data relies on manual interpretation, requiring professionals to analyze the ultrasonic echo signals with experience. This results in low identification efficiency (a single detection can take several hours to several days) and limited accuracy (susceptible to noise interference leading to misjudgments). Current technology cannot transform discrete detection point data into an intuitive spatial distribution map of damage, making it difficult to quantify the location, extent, and severity of damage, thus lacking a scientific basis for risk assessment.

[0008] 3. Lack of an integrated visualization and risk assessment system

[0009] The existing detection process is isolated and has not established a closed loop of "detection-analysis-early warning". Damage results are presented only as raw data or simple charts, which cannot achieve three-dimensional spatial visualization; risk assessment relies on human experience and lacks quantitative grading standards based on damage parameters (such as depth, area, and location), making it impossible to dynamically generate treatment suggestions (such as emergency repair, time-limited rectification, and regular monitoring), which makes it difficult to support proactive prevention and control decisions.

[0010] In summary, existing technologies suffer from systemic deficiencies in adsorption stability, detection accuracy, intelligence level, and risk closed-loop management, failing to meet the urgent needs for automation, precision, and intelligence in underwater detection of water-sealed oil caverns. A solution is urgently needed that can adapt to the complex underwater environment, achieve depth-adjustable detection, intelligently analyze damage, and provide automatic early warnings. Summary of the Invention

[0011] To address the shortcomings of existing technologies, this invention provides an ultrasonic guided wave detection robot and its detection method for underwater oil in water-sealed oil caverns.

[0012] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is as follows:

[0013] An ultrasonic guided wave detection robot for underwater oil in a water-sealed oil cavern includes: a robot body 1, an adsorption device 2, a guide rail fixing device, an ultrasonic guided wave detection module 3, a data acquisition and analysis module 4, and a power module 5.

[0014] The adsorption device 2 is installed on the robot body 1 and is used to adsorb and fix the robot body 1 to the subsealed oil cave wall 15 of the water-sealed oil cavern.

[0015] The guide rail fixing device is installed on the robot body 1 and is used to fix the robot body 1 on the guide rail 16 of the water-sealed oil cavern 13.

[0016] The robot body 1 is supported and protected by the support structure 12, which can cooperate with the guide rail 16 to allow the robot body 1 to slide on the guide rail.

[0017] The ultrasonic guided wave detection module 3 is mounted on the robot body 1 and is used to emit ultrasonic guided waves to the hole wall 15 and receive echo signals, and the detection depth can be adjusted.

[0018] The data acquisition and analysis module 4 is electrically connected to the ultrasonic guided wave detection module 3. It is used to acquire the echo signal of the ultrasonic guided wave, process the acquired signal, extract damage features and identify damage, generate a three-dimensional damage cloud map, and perform risk classification assessment of the damage cloud map for cavern damage.

[0019] Power module 5 is used to drive the robot to move along the cave wall 15.

[0020] Furthermore, the adsorption device 2 includes at least one set of adsorption execution units, which are connected to an adsorption control unit for adjusting the magnitude of the adsorption force.

[0021] Furthermore, the adsorption execution unit is a vacuum suction cup or a magnetic suction cup, wherein the vacuum suction cup is adapted to the concrete wall of the cavity, and the magnetic suction cup is adapted to the steel structure wall of the cavity.

[0022] Furthermore, the ultrasonic guided wave detection module 3 includes an ultrasonic guided wave sensor, an excitation unit 6, and a receiving unit 7. The ultrasonic guided wave sensor can adjust the contact angle with the cavity wall 15 and the detection depth.

[0023] Furthermore, the data acquisition and analysis module 4 includes:

[0024] The data preprocessing unit is used to perform noise reduction and filtering on the echo signal;

[0025] The damage feature extraction unit is used to extract the wave velocity change, amplitude attenuation, reflection signal delay, and amplitude characteristics of the ultrasonic guided wave.

[0026] The damage identification unit uses machine learning algorithms to identify damage.

[0027] The cloud map generation unit is used to generate a three-dimensional damage cloud map based on the damage identification results.

[0028] The risk grading unit is used to classify damage into three levels: low risk, medium risk, and high risk, based on the location, size, and depth of the damage. When the risk level exceeds the standard, an early warning signal is issued.

[0029] This invention also discloses an ultrasonic guided wave detection method for underwater oil in a water-sealed oil cavern, applied to the aforementioned detection robot, comprising the following steps:

[0030] S1. Robot Deployment: Deploy the detection robot to the sub-oil area of ​​the water-sealed oil storage cavern 14;

[0031] S2. Adsorption and fixation: Using the adsorption device 2 and the guide rail fixing device, the robot is adsorbed and fixed to the area to be detected on the cavity wall 15 and fixed on the guide rail 16.

[0032] S3. Detection execution: The ultrasonic guided wave detection module 3 emits ultrasonic guided waves, which propagate along the tunnel wall 15 and receive echo signals;

[0033] S4. Data Acquisition, Analysis, and Evaluation: The data acquisition module acquires echo signals and transmits them to the artificial intelligence analysis module; the artificial intelligence analysis module preprocesses the signals, extracts damage features, and identifies damage; it generates a three-dimensional damage cloud map by combining the three-dimensional coordinate data of the tunnel wall; and it performs a risk classification assessment based on the damage cloud map, dividing the damage into three levels: low risk, medium risk, and high risk.

[0034] S5. Output Results: Displays damage cloud map and risk assessment results. If the risk level exceeds the standard, an early warning will be activated.

[0035] S6. Moving Inspection: Power module 5 drives the robot to move to the next inspection area, repeating steps S2-S5 to complete the inspection of the entire cave 14.

[0036] Furthermore, in step S3, the operating parameters of the ultrasonic guided wave detection module 3 include the frequency of the ultrasonic guided wave, the detection depth, the excitation voltage, and the sampling frequency.

[0037] Furthermore, in step S4, the damage feature extraction includes extracting the wave velocity change, amplitude attenuation, reflection signal delay, and amplitude features of the ultrasonic guided wave.

[0038] Furthermore, in step S4, the risk classification assessment includes classifying cavern damage into three levels: low risk, medium risk, and high risk.

[0039] Compared with the prior art, the advantages of the present invention are as follows:

[0040] (1) Significantly improved adsorption and fixation stability: The adsorption device (including vacuum suction cup or magnetic suction cup) dynamically adjusts the adsorption force to achieve stable adhesion of the robot to the surface of the oil under high pressure, oil flow disturbance and complex curvature cavity wall, avoiding displacement or falling off during the detection process, significantly ensuring the continuity of detection and the reliability of operation, and adapting to the complex environment of oil under high pressure, corrosiveness and dynamic interference.

[0041] (2) Precise coverage of detection depth: The ultrasonic guided wave detection module supports flexible adjustment of detection depth (controlled by adjustable fitting angle and sensor depth), realizing layered coverage detection of shallow (such as surface cracks and corrosion) and deep (such as internal holes and structural peeling) damage of the hole wall, greatly improving the comprehensiveness and accuracy of detection, and overcoming the limitation of existing technologies that can only cover a single depth.

[0042] (3) Intelligent upgrade of data processing: The artificial intelligence analysis module (integrating data preprocessing, damage feature extraction and machine learning recognition unit) automatically completes the noise reduction, feature extraction (wave velocity change, amplitude attenuation, reflection delay, etc.) and intelligent damage recognition of echo signals, replacing traditional manual interpretation, significantly improving the efficiency and accuracy of damage recognition, and eliminating the risk of human misjudgment.

[0043] (4) Enhanced damage visualization and quantification capabilities: The damage cloud map generation function transforms discrete detection point data into an intuitive three-dimensional spatial damage cloud map, clearly presenting the location, range, depth and severity of damage, supporting rapid positioning, parameter quantification and spatial distribution analysis, significantly improving the understandability and engineering applicability of the detection results.

[0044] (5) Risk assessment closed-loop management: The visualization and risk assessment module automatically classifies risk levels based on key damage parameters (location, size, depth), generates targeted treatment suggestions, and triggers audible and visual warnings when the risk exceeds the standard, realizing an active prevention and control closed loop of "detection-analysis-early warning-treatment", providing accurate decision-making basis for the safe operation and maintenance of caves.

[0045] In summary, this invention systematically solves the problems of unstable adsorption, insufficient detection depth, low intelligence, lack of visualization, and crude risk assessment in existing technologies through five core innovations: adsorption stability, detection accuracy, intelligent processing, visualization, and risk closed-loop management. It fully meets the engineering requirements of automation, precision, and intelligence for underwater detection in water-sealed oil caverns. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a schematic diagram of the device structure of the detection robot of the present invention;

[0048] Figure 2 This is a schematic diagram of the ultrasonic guided wave detection device of the present invention;

[0049] Figure 3 This is a schematic diagram of the structure of the ultrasonic guided wave detection device of the present invention;

[0050] Figure 4 This is a schematic diagram of the suction cup device of the present invention;

[0051] Figure 5 This is a schematic diagram of the support structure of the detection robot of the present invention;

[0052] Figure 6 A schematic diagram of the caverns in a water-sealed oil storage facility;

[0053] Figure 7 This is a schematic diagram of the internal structure of the cavern.

[0054] Figure 8 This is a schematic diagram of the architecture of the artificial intelligence analysis module of the present invention;

[0055] Figure 9 This is a schematic diagram of the workflow of the visualization and risk assessment module of the present invention;

[0056] The attached diagram lists the components represented by each number as follows:

[0057] In the diagram: 1. Robot body; 2. Adsorption device; 3. Ultrasonic guided wave detection module; 4. Data acquisition and analysis module; 5. Power module; 6. Excitation unit; 7. Receiving unit; 8. Telescopic connection device; 9. Fixing block; 10. Fixing device; 11. Pump; 12. Support structure; 13. Rock mass; 14. Cave; 15. Cave wall; 16. Guide rail. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Please see Figures 1-9 As shown, the present invention is an ultrasonic guided wave detection robot and its detection method for oil in a water-sealed oil cavern.

[0060] Firstly, this invention provides an implementation scheme for an ultrasonic guided wave inspection robot for underwater operations in water-sealed oil caverns. This scheme is fully adaptable to the complex operating environment of high pressure, oil resistance, and waterproofing under oil conditions. Each functional module is designed collaboratively according to the inspection process to ensure the accuracy, stability, and operability of the cavern wall damage detection. The specific structure and functions are as follows:

[0061] The robot body 1 is integrally molded using special engineering materials that are oil-resistant, waterproof, and pressure-resistant. The structural design takes into account both lightweight and mechanical strength, effectively resisting the erosion of underwater corrosive media and pressure impact, providing a stable installation and bearing foundation for each functional module, and ensuring the stability of the robot's overall operating posture.

[0062] like Figure 4 As shown, the adsorption device 2 is integrated into the key stress-bearing parts of the robot body 1. Its core function is to achieve stable adsorption and fixation of the robot on the surface of the submerged wall 15 of the water-sealed oil cavern 14, preventing displacement or detachment due to oil flow disturbance, gravity, or other factors. The device is equipped with at least one set of adsorption execution units, each of which is connected to an adsorption control unit. It can dynamically adjust the adsorption force in real time according to the material characteristics of the cavern wall 15 and the stress conditions at the working position, ensuring the firmness of the adsorption while avoiding damage to the surface of the cavern wall 15. The adsorption execution unit adopts a scene-adaptive design: for concrete cavern walls 15, a vacuum suction cup is used to achieve adhesion and fixation through the principle of negative pressure adsorption; for steel structure cavern walls 15, a magnetic suction cup is used to ensure the stability of the adhesion by relying on the strong magnetic adsorption characteristics.

[0063] The robot body 1 is equipped with a fixing device 10, which is used to connect the adsorption device 2 and the pump 11 to prevent them from falling off in the oily environment. The pump 11 is used to remove impurities from the adsorption device 2, so that the adsorption device 2 can be adsorbed onto the cavern 14.

[0064] like Figure 2 and 3 As shown, the ultrasonic guided wave detection module 3 is installed at the detection end of the robot body 1 facing the cavity wall 15. Its core function is to emit ultrasonic guided wave signals into the cavity wall 15 and receive the echo signals reflected and transmitted through the cavity wall 15, thereby realizing the detection of the internal structural integrity of the cavity wall 15. This module consists of an ultrasonic guided wave sensor, an excitation unit, and a receiving unit. The excitation unit is responsible for providing a stable excitation signal to the sensor, driving the sensor to emit ultrasonic guided waves that meet the detection requirements; the receiving unit is responsible for capturing the echo signals and completing the preliminary signal conversion. To adapt to the detection requirements of cavity walls 15 with different areas and thicknesses, the ultrasonic guided wave sensor adopts an adjustable mounting structure, which can flexibly adjust the contact angle and detection depth with the cavity wall 15 to ensure that the guided wave signal effectively penetrates the cavity wall 15 and accurately captures the echo characteristics related to damage.

[0065] like Figure 5 As shown, the robot body 1 is supported and protected by the support structure 12. The support structure 12 can cooperate with the guide rail 16 to allow the robot body 1 to slide on the guide rail. The core function of the support structure 12 is to ensure safety and stability during inspection. To reach the next inspection point, the robot can be driven forward on the guide rail 16 by the power module 5.

[0066] The guide rail fixing device is installed on the robot body 1 and is used to fix the robot body 1 on the guide rail 16 of the water-sealed oil cavern 13;

[0067] The ultrasonic guided wave detection module 3 includes an excitation device 6, a receiving unit 7, and a retractable connecting device 8. The excitation device 6 generates an excitation signal that penetrates the cavity wall 15, and the receiving unit 7 receives the signal to check for damage to the cavity wall 15. The retractable connecting device 8 can detect different positions and directions of the cavity wall 15, enabling omnidirectional detection.

[0068] The robot body 1 is connected to the retractable connecting device 8 via the fixing block 9, which fixes the entire ultrasonic guided wave detection module 3 for convenient detection.

[0069] The data acquisition and analysis module 4 establishes a dedicated signal transmission link and electrical connection with the ultrasonic guided wave detection module 3. Its main responsibility is to acquire and buffer the ultrasonic guided wave echo signals generated during the detection process in real time. This module features an anti-interference design, effectively filtering electromagnetic interference and noise interference in the underwater environment to ensure the authenticity and reliability of the acquired raw signals, providing high-quality data support for subsequent signal processing and damage analysis. The artificial intelligence analysis and data acquisition module 4 achieves bidirectional data exchange. Its core responsibility is to systematically process the acquired raw signals, extract damage features, accurately identify damage, and generate an intuitive three-dimensional damage cloud map based on the identification results. This module integrates four core sub-units: the data preprocessing unit is responsible for processing the original echo signal by noise reduction, filtering, and signal enhancement, removing invalid interference information and restoring the essential characteristics of the signal; the damage feature extraction unit accurately extracts feature parameters related to damage such as cracks, peeling, and breakage in the tunnel wall from the preprocessed signal; the damage identification unit adopts machine learning algorithms adapted to engineering structure damage detection scenarios, and through training and learning on massive damage sample data, it can quickly and accurately identify the existence, type, and core features of damage; and the cloud map generation unit transforms the damage identification results into a three-dimensional visualized damage cloud map, clearly presenting the spatial distribution of damage.

[0070] The risk assessment module and the artificial intelligence analysis module establish data linkage. First, a signal is transmitted through the excitation unit 6. Later, it can realize two core functions: visualization of damage information and risk classification assessment of damage to cavern 14. This module includes three key sub-units: the display unit can present core information such as three-dimensional damage cloud map and damage characteristic parameters in real time, and supports inspection personnel to zoom in and view damage details; the risk classification unit, combined with the specific location of the damage, such as key load-bearing parts, sealing parts, damage range, damage depth and other core indicators, classifies the damage to cavern 14 into risk levels such as low risk, medium risk and high risk according to industry engineering risk assessment standards. When the damage risk level is detected to exceed the preset safety threshold, an audible and visual warning signal can be issued immediately, and the warning information can be fed back to the receiving unit 7 at the same time to provide timely reminders for subsequent emergency response.

[0071] The core function of the power module 5 is to provide stable power output for the robot to move to the next inspection point on the surface of the tunnel wall 15. This module adopts a power drive scheme adapted to the underwater working environment. It can accurately adjust the magnitude and direction of the driving force according to the movement command issued by the control module, so as to drive the robot to move smoothly along the surface of the tunnel wall 15. It adapts to the movement requirements of tunnel walls 15 with different curvatures and materials, ensuring that the robot can complete the full coverage inspection of the tunnel 14 according to the preset inspection path.

[0072] In use, the robot achieves stable adhesion to the underground cavern wall through an adsorption device, providing a stable operating basis for ultrasonic guided wave detection. The ultrasonic guided wave detection module 3 emits guided waves to the cavern wall. When the guided waves encounter damage, their propagation characteristics change. The data acquisition module 4 collects echo signals. Artificial intelligence analysis preprocesses the signals, extracts damage features, and intelligently identifies damage information through machine learning algorithms. It generates a three-dimensional damage cloud map by combining the three-dimensional coordinates of the cavern wall. Based on the damage parameters, risk classification is performed, and an early warning is activated when the risk exceeds the standard. The power module 5 drives the robot to move, achieving continuous detection of the entire cavern. The entire process is adapted to the complex underground environment, realizing stable detection, intelligent analysis, visualization, and risk warning of cavern wall damage.

[0073] Secondly, based on the first aspect, the present invention also provides an embodiment: an ultrasonic guided wave detection method for oil submerged in a water-sealed oil cavern, comprising the following steps:

[0074] S1. Robot Deployment: Deploy the detection robot to the sub-oil area of ​​the water-sealed oil cavern 14;

[0075] S2. Adsorption and Fixation: Activate the adsorption device 2 to adsorb and fix the robot to the area to be tested on the cavity wall 15 and to the guide rail.

[0076] S3. Detection execution: The ultrasonic guided wave detection module 3 emits ultrasonic guided waves, which propagate along the tunnel wall 15 and receive echo signals;

[0077] S4. Data Acquisition, Analysis, and Evaluation: Data acquisition module 4 acquires echo signals and transmits them to artificial intelligence analysis; such as... Figure 8 and Figure 9 The artificial intelligence analysis module preprocesses the signal, extracts damage features, and identifies damage. Damage feature extraction includes extracting the wave velocity change, amplitude attenuation, reflection signal delay, and amplitude characteristics of the ultrasonic guided wave. The artificial intelligence analysis combines the three-dimensional coordinate data of the tunnel wall 15 to generate a three-dimensional damage cloud map. Based on the damage cloud map, a risk classification assessment is conducted, which divides the damage of the tunnel 14 into three levels: low risk, medium risk, and high risk, with different treatment recommendations corresponding to different levels.

[0078] S5. Output Results: Displays damage cloud map and risk assessment results. If the risk exceeds the standard, an early warning will be activated.

[0079] This method achieves stable adsorption through adsorption device 2, adjusts the detection depth through ultrasonic guided wave detection module 3, completes intelligent data processing and damage cloud map generation through artificial intelligence analysis, and realizes damage risk classification through visualization and risk assessment, thus solving the problems of unstable adsorption, inaccurate detection and low level of intelligence in existing detection technologies.

[0080] Example 1

[0081] This embodiment is for the subsurface oil detection of the concrete walls of a water-sealed oil cavern: the robotic adsorption device 2 uses a vacuum suction cup assembly adapted to the concrete surface, and the ultrasonic guided wave detection module 3 uses a low-frequency sensor to cover both shallow and deep layers of the concrete cavern wall.

[0082] After the robot is deployed, its vacuum suction cups adhere to the tunnel wall, and its bottom is placed on the guide rail. After setting the detection parameters, it emits ultrasonic guided waves to collect echo signals. Artificial intelligence analyzes and extracts features such as wave velocity attenuation and reflection delay. Through neural networks, it identifies damage such as cracks and holes, generates a three-dimensional damage cloud map, and classifies the risk level according to the damage area and depth. High-risk damage triggers an early warning, providing a basis for the maintenance of concrete tunnel walls.

[0083] Example 2

[0084] This embodiment is for underwater inspection of the steel structure wall of a water-sealed oil cavern: the robot adsorption device 2 uses a magnetic chuck to fit the surface of the steel structure, and the ultrasonic guided wave detection module 3 uses a high-frequency sensor to accurately detect shallow damage such as corrosion and cracks in the steel structure.

[0085] After the robot is deployed, its magnetic suction cup adheres to the tunnel wall. After setting the detection parameters, it emits ultrasonic guided waves to collect echo signals. Artificial intelligence analyzes and extracts parameters such as amplitude attenuation and reflection characteristics. Support vector machine is used to identify damage such as corrosion and cracks, generating a three-dimensional damage cloud map. Combined with the damage location and the oil storage pressure of the tunnel, risk classification is performed, and treatment suggestions are generated for high-risk damage to ensure the safe operation of the steel structure tunnel wall.

[0086] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An ultrasonic guided wave detection robot for submerged oil in a water-sealed oil cavern, characterized in that, include: The robot body (1), adsorption device (2), guide rail fixing device, ultrasonic guided wave detection module (3), data acquisition and analysis module (4), power module (5) and support structure (12); The adsorption device (2) is installed on the robot body (1) to adsorb and fix the robot body (1) to the subsealed oil cave wall (15) of the water-sealed oil cave. The guide rail fixing device is installed on the robot body (1) and is used to fix the robot body (1) on the guide rail (16) of the water-sealed oil cavern (13); The robot body (1) is supported and protected by the support structure (12), which can cooperate with the guide rail (16) to allow the robot body 1 to slide on the guide rail; The ultrasonic guided wave detection module (3) is mounted on the robot body (1) and is used to transmit ultrasonic guided waves to the hole wall (15) and receive echo signals, and the detection depth can be adjusted. The data acquisition and analysis module (4) is electrically connected to the ultrasonic guided wave detection module (3) and is used to acquire the echo signal of the ultrasonic guided wave, process the acquired signal, extract damage features and identify damage, generate a three-dimensional damage cloud map, and conduct risk classification assessment of the damage cloud map for cavern damage. The power module (5) is used to drive the robot to move along the cave wall (15).

2. The ultrasonic guided wave detection robot for submerged oil in a water-sealed oil cavern according to claim 1, characterized in that, The adsorption device (2) includes at least one set of adsorption execution units, which are connected to an adsorption control unit for adjusting the magnitude of the adsorption force.

3. The ultrasonic guided wave detection robot for submerged oil in a water-sealed oil cavern according to claim 2, characterized in that, The adsorption execution unit is a vacuum suction cup or a magnetic suction cup. The vacuum suction cup is adapted to the concrete wall of the cavity, and the magnetic suction cup is adapted to the steel structure wall of the cavity.

4. The ultrasonic guided wave detection robot for submerged oil in a water-sealed oil cavern according to claim 1, characterized in that, The ultrasonic guided wave detection module (3) includes an ultrasonic guided wave sensor, an excitation unit (6) and a receiving unit (7). The ultrasonic guided wave sensor can adjust the contact angle and detection depth with the hole wall (15).

5. The ultrasonic guided wave detection robot for underwater oil in a water-sealed oil cavern according to claim 1, characterized in that, The data acquisition and analysis module (4) includes: The data preprocessing unit is used to perform noise reduction and filtering on the echo signal; The damage feature extraction unit is used to extract the wave velocity change, amplitude attenuation, reflection signal delay, and amplitude characteristics of the ultrasonic guided wave. The damage identification unit uses machine learning algorithms to identify damage. The cloud map generation unit is used to generate a three-dimensional damage cloud map based on the damage identification results. The risk grading unit is used to classify damage into three levels: low risk, medium risk, and high risk, based on the location, size, and depth of the damage. When the risk level exceeds the standard, an early warning signal is issued.

6. A method for ultrasonic guided wave detection of oil beneath a water-sealed oil cavern, characterized in that, The method applied to the inspection robot as described in any one of claims 1 to 5 includes the following steps: S1. Robot deployment: Deploy the detection robot to the subsurface area of ​​the water-sealed oil cavern (14); S2. Adsorption and fixation: Using the adsorption device (2) and the guide rail fixing device, the robot is adsorbed and fixed to the area to be tested on the hole wall (15) and fixed on the guide rail (16); S3. Detection execution: The ultrasonic guided wave detection module (3) emits ultrasonic guided waves, which propagate along the cave wall (15) and receive echo signals; S4. Data acquisition, analysis and evaluation: The data acquisition module acquires the echo signal and transmits it to the artificial intelligence analysis module; the artificial intelligence analysis preprocesses the signal, extracts damage features and identifies damage; a three-dimensional damage cloud map is generated by combining the three-dimensional coordinate data of the tunnel wall (15); the damage is classified into three levels: low risk, medium risk and high risk based on the damage cloud map. S5. Output Results: Displays damage cloud map and risk assessment results. If the risk level exceeds the standard, an early warning will be activated. S6. Moving detection: The power module (5) drives the robot to move to the next detection area, repeating steps S2-S5 to complete the detection of the entire cave (14).

7. The detection method according to claim 6, characterized in that, In step S3, the operating parameters of the ultrasonic guided wave detection module (3) include the frequency of the ultrasonic guided wave, the detection depth, the excitation voltage, and the sampling frequency.

8. The detection method according to claim 6, characterized in that, In step S4, the damage feature extraction includes extracting the wave velocity change, amplitude attenuation, reflection signal delay, and amplitude features of the ultrasonic guided wave.

9. The detection method according to claim 6, characterized in that, In step S4, the risk classification assessment includes classifying cavern damage into three levels: low risk, medium risk, and high risk.

Images