Automated drill pipe defect intelligent detection system, method, terminal device and storage medium

Abstract

The application relates to the technical field of oil energy exploitation, and is an automatic drilling rod defect intelligent detection system and method, terminal equipment and storage medium, which comprise: a drilling tool moving assembly line, which is used for realizing automatic loading and unloading of drilling rods, ensuring that the drilling rods move to a detection area along a predetermined path; a drilling tool detection device, which is used for generating a strong and stable magnetic field, uniformly magnetizing the drilling rods, and controlling the distribution density of magnetic powder and recycling the magnetic powder that is not attached to defects; and a detection result collection and judgment device, which is used for shooting the surface of the drilling rods after magnetization treatment, processing image data in real time, accurately identifying and classifying defects, and generating a detailed detection report.The application realizes full automation and intelligence of drilling rod pipe end defect detection, significantly improves detection efficiency and accuracy, reduces dependence on skilled detection personnel, reduces human errors, and ensures the safety and continuity of drilling operations.

Description

Technical Field

[0001] This invention relates to the field of petroleum energy extraction technology, and is an automated intelligent detection system, method, terminal equipment, and storage medium for drill pipe defects. Background Technology

[0002] In modern oil drilling operations, drill pipe, as a key component connecting the drilling rig and drill bit, directly affects the safety and efficiency of the drilling operation. However, under long-term high-load operating conditions, drill pipe ends are highly susceptible to developing micro-cracks, wear, or other defects. If these defects are not detected and addressed in a timely manner, they can lead to serious safety accidents, such as drill pipe breakage and blowouts, causing significant economic losses to the company and even threatening the lives of workers. Therefore, regular and efficient defect inspection of drill pipe ends is of paramount importance.

[0003] Traditional defect detection methods, such as visual inspection and ultrasonic testing, each have their limitations. Visual inspection relies on the experience of the inspector and often fails to detect small or hidden defects. While ultrasonic testing can provide a deeper detection range, it requires a high level of operator skill, and its effectiveness is significantly reduced in complex working conditions, such as when the drill pipe surface is covered with contaminants. In addition, most of these methods require a large amount of manual intervention, resulting in low detection efficiency, high costs, and difficulty in meeting the rapid response requirements of modern drilling operations.

[0004] Therefore, magnetic particle testing, as a mature non-destructive testing technology, has gradually gained attention in the industry due to its extremely high sensitivity to surface and near-surface defects in ferromagnetic materials. This technology works by using an external magnetic field to cause magnetic particles to accumulate at material defects, forming a clear magnetic trace that visually reveals the location and size of the defect. However, the standard magnetic particle testing process is still quite cumbersome, requiring multiple manual steps, including magnetization, application of magnetic particles, observation, and final defect analysis and judgment. This is not only time-consuming and labor-intensive but also prone to misjudgment due to human factors. Summary of the Invention

[0005] This invention provides an automated intelligent drill pipe defect detection system, method, terminal equipment, and storage medium, which overcomes the shortcomings of the prior art and can effectively solve the problems of low efficiency, limited accuracy, and high dependence on manual operation in the detection of drill pipe end defects in existing drilling operations.

[0006] One of the technical solutions of this invention is achieved through the following measures: an automated intelligent drill pipe defect detection system, comprising: The drill string moving assembly line is used to realize the automatic loading and unloading of drill pipes, ensuring that the drill pipes move to the inspection area along a predetermined path; Drill tool inspection equipment is used to generate a strong and stable magnetic field to uniformly magnetize the drill rod, control the distribution density of magnetic powder, and recover magnetic powder that has not adhered to defects. The test result collection and judgment device is used to photograph the surface of the drill rod after magnetization, process the image data in real time, accurately identify and classify defects, and generate a detailed test report.

[0007] The following are further optimizations and / or improvements to one of the above-mentioned technical solutions: The aforementioned drilling tool moving production line may include a roller and steel frame mechanism, a loading and unloading mechanism, and a conveying and guiding mechanism; The roller and steel frame mechanism are used to ensure that the drill rod is transported smoothly and steadily, reducing frictional loss. The loading and unloading mechanism is used to automatically load and unload drill rods through precise positioning and gripping. The conveying and guiding mechanism is used to ensure that the drill pipe moves to the detection area along a predetermined path.

[0008] The aforementioned drill bit inspection equipment may include a flaw detection main unit and a magnetic particle application and collection system; Among them, the flaw detection host is used to generate a strong and stable magnetic field to uniformly magnetize the drill pipe; The magnetic powder application and collection system includes a magnetic powder spraying device and a magnetic powder collection box. The magnetic powder spraying device is used to precisely control the distribution density of the magnetic powder, and the magnetic powder collection box is used to recover the magnetic powder that has not adhered to the defect.

[0009] The aforementioned device for collecting and judging detection results may include an image acquisition unit, an edge computing unit, and a cloud-based intelligent judgment unit; The image acquisition unit is used to take pictures of the surface of the drill rod after magnetization from all angles to capture magnetic trace images. The edge computing unit is used to process the image data captured by the camera in real time, and to perform preliminary screening and preprocessing. The cloud-based intelligent judgment unit is used to accurately identify and classify defects, and automatically determines the severity of defects according to preset standards, and generates a detailed inspection report.

[0010] The second technical solution of the present invention is achieved through the following measures: an automated intelligent detection method for drill pipe defects, comprising the following steps: Step 1: For both defective and defect-free drill pipes, collect a large number of magnetic particle images using an image acquisition module; Step 2: Perform image data preprocessing on the acquired magnetic particle display images; Step 3: Train the unsupervised anomaly detection model based on the dataset of images with defects; Step 4: Using the trained drill string defect detection model, calculate the anomaly score for the input drill string image to be detected, and filter out the abnormal magnetic particle images. Step 5: Submit the abnormal magnetic particle images to the business personnel for secondary review and confirmation.

[0011] The following are further optimizations and / or improvements to the second technical solution of the above invention: Step 2 above may specifically include the following steps: Step 2.1, Contrast Enhancement: The collected magnetic powder image is enhanced in contrast using image sharpening methods; Step 2.2, Image Data Enlargement: Increase the amount of relevant data to improve the training effect of the image defect detection model.

[0012] Step 3 above may specifically include the following steps: Step 3.1: Based on the collected and preprocessed defect image data, construct a drill tool defect identification algorithm model using image recognition algorithms in the field of image processing; Step 3.2: Input the collected and preprocessed magnetic particle image of the drill pipe into the constructed model, and use gradient descent optimization to train the model until the loss function of the model gradually converges, and the model training ends.

[0013] The third technical solution of the present invention is achieved through the following measures: a terminal device, including a memory and a processor, wherein the memory stores a program that can run on the processor, and the processor executes the program to realize the above-mentioned automated drill pipe defect intelligent detection method.

[0014] The fourth technical solution of the present invention is achieved through the following measures: a storage medium storing one or more programs, which can be executed by one or more processors to realize the above-mentioned automated intelligent detection method for drill pipe defects.

[0015] This invention breaks through the technical bottlenecks of existing technologies, focusing on combining magnetic particle inspection technology with advanced automated control, machine vision, and artificial intelligence algorithms to develop an intelligent automated drill pipe defect detection system. It can automatically complete the entire process from drill pipe positioning, magnetization treatment, magnetic particle spraying, image acquisition, to defect identification and classification using deep learning algorithms, greatly improving detection efficiency and accuracy. Through an integrated intelligent judgment module, it can not only accurately locate defects but also automatically assess their severity according to preset standards, providing a scientific basis for subsequent maintenance decisions. This invention significantly reduces reliance on professional inspection personnel, minimizes human error, and greatly shortens the inspection cycle, improving the overall safety and economic benefits of drilling operations. This invention aims to achieve the automation and intelligent upgrade of drill pipe end defect detection in oil drilling operations, with the core being the construction of a highly efficient automated intelligent drill pipe defect detection system. Compared to traditional automated systems that rely on magnetic particle testing for drill pipe defect assessment, this invention adds an intelligent identification module. Leveraging advanced computer vision technology in the field of artificial intelligence, it assists frontline personnel in identification and assessment, reducing their workload. Through the integration of the above systems, the detection of drill pipe end defects is fully automated and intelligent, significantly improving detection efficiency and accuracy, reducing reliance on skilled personnel, minimizing human error, and ensuring the safety and continuity of drilling operations. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the architecture of an automated drill pipe defect intelligent detection system according to an embodiment of the present invention.

[0017] Figure 2 This is a flowchart illustrating the automated drill pipe defect intelligent detection method according to an embodiment of the present invention.

[0018] Figure 3 This is a magnetic particle image of a defective drill rod according to an embodiment of the present invention.

[0019] Figure 4 This is a magnetic particle image of a defect-free drill pipe according to an embodiment of the present invention.

[0020] Figure 5 This is a schematic diagram of the U-Net model network structure according to an embodiment of the present invention.

[0021] Figure 6 This is a schematic diagram of the overall assembly structure of the tube end magnetic particle detection according to an embodiment of the present invention.

[0022] Figure 7 This is a schematic diagram of the conveyor roller conveyor according to an embodiment of the present invention.

[0023] Figure 8This is a schematic diagram of the flaw detection host according to an embodiment of the present invention.

[0024] Figure 9 This is a schematic diagram of the loading and unloading platform structure according to an embodiment of the present invention.

[0025] As shown in the diagram: 101 is the loading platform, 102 is the conveyor roller conveyor, 103 is the rotating roller conveyor, 104 is the magnetic particle inspection main unit, 105 is the demagnetizer, 106 is the V-shaped bracket, and 107 is the unloading platform; 201 is the conveyor roller conveyor base, 202 is the end cover, 203 is the spacer, 204 is the self-aligning roller bearing, 205 is the spacer, 206 is the oil cup, 207 is the V-shaped roller, 208 is the spacer, 209 is the shaft, 210 is the double-row chain, 211 is the sprocket, 212 is the motor support, 213 is the geared motor, and 214 is the bearing housing; 301 is the bed, 302 is the linear sliding unit support, 303 is the roller, and 304 is the bearing seat. Spacer, 305 is a bracket, 306 is a deep groove ball bearing, 311 is a fixed seat, 312 is a pad, 313 is a movable seat, 314 is a fixed block, 315 is a magnetic plate, 316 is a fixed block, 317 is a fixed block, 318 is a circumferential induction coil, 319 is a cylinder connector, 320 is a cylinder fixing plate, 321 is a standard cylinder, 322 is a fixing plate, 323 is a lifting plate, 324 is a linear bearing, 325 is a guide shaft, 326 is a lifting mechanism, 327 is a shaft, 328 is a shaft, 329 is an outer spherical bearing, 330 is a pad, 331 is a handwheel; 401 is a loading / unloading rack, 402 is a stop block, 403 is an insert. Detailed Implementation

[0026] The present invention is not limited to the following embodiments, and the specific implementation can be determined according to the technical solution of the present invention and the actual situation.

[0027] The present invention will be further described below with reference to embodiments: Example 1: As shown in the attached document Figure 1 As shown, the automated drill pipe defect intelligent detection system includes: This drill bit moving assembly line is used to automatically load and unload drill rods, ensuring that the drill rods move to the inspection area along a predetermined path. In this embodiment, the drill bit moving assembly line includes a roller and steel frame mechanism, a loading and unloading mechanism, and a conveying and guiding mechanism. The roller and steel frame mechanism uses a custom-made steel frame made of high-strength steel to support the entire assembly line, equipped with wear-resistant and durable rollers to ensure smooth and stable transmission of the drill rods and reduce friction loss. The loading and unloading mechanism is an intelligent loading and unloading robot or robotic arm, used for precise positioning and gripping to automatically load and unload drill rods, significantly improving work efficiency and reducing manual labor. The conveying and guiding mechanism includes a speed-adjustable motor-driven conveyor belt and precision guide components to ensure that the drill rods move to the inspection area along a predetermined path; it can accommodate drill rods of different sizes and weights. This drill bit moving assembly line is ingeniously designed, integrating advanced transmission and control technologies to ensure seamless connection of the drill rod end inspection process.

[0028] Drill string inspection equipment is used to generate a strong and stable magnetic field to uniformly magnetize the drill pipe and control the distribution density of magnetic powder, recovering magnetic powder that is not attached to defects. In this embodiment, the drill string inspection equipment includes a flaw detection host and a magnetic powder application and collection system. The flaw detection host, as the core of magnetic powder inspection, is used to generate a strong and stable magnetic field to uniformly magnetize the drill pipe, ensuring that even tiny defects can be effectively excited. The magnetic powder application and collection system includes a magnetic powder spraying device and a magnetic powder collection box. The magnetic powder spraying device is used to precisely control the distribution density of magnetic powder, and the magnetic powder collection box is used to recover magnetic powder that is not attached to defects, reducing waste and keeping the working environment clean.

[0029] The detection result collection and judgment device is used to photograph the surface of the magnetized drill rod, process the image data in real time, accurately identify and classify defects, and generate a detailed inspection report. In this embodiment, the detection result collection and judgment device includes an image acquisition unit, an edge computing unit, and a cloud-based intelligent judgment unit. The image acquisition unit is equipped with a dedicated camera for magnetic particle display result acquisition, featuring high-resolution, high-speed imaging technology, used to capture magnetic trace images of the magnetized drill rod surface from all angles. The edge computing unit, deployed on-site, processes the image data captured by the camera in real time, performing preliminary screening and preprocessing, effectively reducing the pressure on the cloud server and improving response speed. The cloud-based intelligent judgment unit, based on a deep learning and image recognition technology, is located on the cloud server. It receives the image data processed by the edge computing unit, accurately identifies and classifies defects, automatically determines the severity of defects according to preset standards, and generates a detailed inspection report.

[0030] This invention specifically relates to pipeline inspection technology in drilling operations, particularly for detecting minute defects at the drill pipe ends of various wells in oilfields. It utilizes the principle of magnetization to achieve highly sensitive non-destructive testing. This invention overcomes the technical bottlenecks of existing technologies, combining magnetic particle inspection technology with advanced automated control, machine vision, and artificial intelligence algorithms to develop an intelligent automated drill pipe defect detection system. It can automatically complete the entire process from drill pipe positioning, magnetization treatment, magnetic particle spraying, image acquisition, to defect identification and classification using deep learning algorithms, greatly improving the efficiency and accuracy of inspection. Through an integrated intelligent judgment module, it can not only accurately locate defects but also automatically assess their severity according to preset standards, providing a scientific basis for subsequent maintenance decisions. This invention significantly reduces reliance on professional inspection personnel, minimizes human error, and greatly shortens the inspection cycle, improving the overall safety and economic benefits of drilling operations. Furthermore, this invention can be applied in other industrial fields.

[0031] This invention achieves full automation and intelligence in defect detection, which is reflected in the following aspects: (1) Improved detection efficiency: By integrating automated control technology, the entire process from magnetization and magnetic powder application to image acquisition and analysis is unmanned, greatly shortening the detection time and adapting to the needs of rapid response drilling operations. (2) Enhanced detection accuracy: Combining high-precision magnetic powder detection technology and advanced machine vision algorithms, even the smallest defects can be accurately identified and located, reducing missed detections and misjudgments, and ensuring the safe use of drill pipes. (3) Reduced reliance on manpower: The automated detection process reduces the reliance on the skills of professional inspection personnel, reduces labor costs, and avoids detection instability caused by human factors, improving the consistency and reliability of detection. (4) Intelligent judgment and management: An intelligent judgment module is introduced to automatically assess the defect level based on the detection data and provide immediate repair or replacement suggestions, providing scientific guidance for the maintenance and management of drilling equipment, extending the service life of drill pipes, and reducing unplanned downtime. (5) Promoted safe production: Through timely and accurate defect detection, downhole accidents caused by drill pipe damage are effectively prevented, ensuring the safe operation of drilling operations and protecting personnel safety and the environment from damage.

[0032] This invention aims to automate and intelligently upgrade the detection of drill pipe end defects in oil drilling operations. The core of this system is the construction of a highly efficient automated intelligent drill pipe defect detection system. It should be noted that, to meet diverse user needs and achieve rapid judgment, two types of computing units are simultaneously set up in the detection result collection and judgment device: an edge computing unit and a cloud-based intelligent judgment unit. The edge computing unit is deployed locally, without connecting to an internal network server, enabling rapid local judgment and facilitating quick screening. The screening results are then rapidly provided to frontline business users. The cloud-based intelligent judgment unit is deployed on the business unit's internal network service. This unit is specifically designed for scenarios where frontline business users have doubts about the judgment results of the screened images and require secondary confirmation. In this case, the magnetic particle images collected on-site can be transmitted to a remote server via the unit's internal network. Utilizing the powerful computing resources of the remote server, more accurate judgments can be made on some difficult magnetic particle images. The two modules complement each other, satisfying both the requirement for rapid judgment and the requirement to maximize system accuracy.

[0033] It is important to emphasize that, compared to traditional assembly line systems that rely on magnetic particle testing for drill pipe defect assessment, this invention adds an intelligent identification module. Leveraging advanced computer vision technology in the field of artificial intelligence, this module assists frontline personnel in identification and assessment, reducing their workload. Through the integration of the above systems, fully automated and intelligent drill pipe end defect detection is achieved, significantly improving detection efficiency and accuracy, reducing reliance on skilled personnel, minimizing human error, and ensuring the safety and continuity of drilling operations.

[0034] Example 2: Figures 2 to 5 As shown, this embodiment provides an automated intelligent detection method for drill pipe defects, including the following steps: Step 1: For both defective and defect-free drill pipes, a large number of magnetic particle images are collected using an image acquisition module. This invention requires collecting a large number of images of defective drill pipes after magnetic particle application to train the subsequent image recognition algorithm model. For defective drill pipes, the flaw detection host and magnetic particle application and collection system in the automated drill pipe defect intelligent detection system are used to apply magnetic particle, and then a dedicated camera for displaying magnetic particle results is used to collect the corresponding magnetic particle images. The collected magnetic particle images are as follows: Figure 3 As shown. Simultaneously, it is also necessary to collect a large number of magnetic particle images of defect-free drill pipes, such as... Figure 4 As shown.

[0035] Step 2: Perform image data preprocessing on the acquired magnetic particle display images. Since the number of defects in drill pipes is relatively small in reality, it is difficult to directly support the training data volume of the artificial intelligence model. Furthermore, the collected images cannot be directly input into the defect detection algorithm model. Therefore, preprocessing of the collected image data is necessary. In this embodiment, Step 2 specifically includes the following steps: Step 2.1, Contrast Enhancement: Using image sharpening methods, the collected magnetic particle image (e.g., ...) is enhanced. Figure 3 The image (as shown) undergoes contrast enhancement to make defects in the image more apparent. Many mature methods for image contrast enhancement exist in the field of automated image processing, including but not limited to histogram equalization and gamma correction. However, these are not the focus of this invention, and therefore, this invention does not limit the scope of the invention.

[0036] Step 2.2, Image Data Augmentation: Increase the amount of relevant data to improve the training effect of the image defect detection model. The image data augmentation methods include, but are not limited to: random cropping, rotation, flipping, scaling, translation, color transformation, noise injection, affine transformation, perspective transformation, etc. However, this is not the focus of this invention, and therefore, it is not limited thereto. It should be emphasized that image data augmentation not only increases the amount of relevant data but also improves the generalization performance of the image defect detection model during evaluation through various transformations, thereby increasing accuracy. Therefore, data augmentation should be performed on both defective and defect-free magnetic particle images. However, since there are relatively more defect-free drill pipe images, the augmentation can be performed with relatively less data.

[0037] Step 3: Train the unsupervised anomaly detection model based on the dataset of images with defects. In this embodiment, Step 3 specifically includes the following steps: Step 3.1: Based on the collected and preprocessed defect image data, an image recognition algorithm model for drill string defects is constructed using image recognition algorithms in the field of image processing. Defects such as cracks in drill strings are essentially surface damage problems. There is a clear difference between local surface damage areas and other areas. Therefore, this invention transforms it into an image segmentation problem. Image segmentation algorithms are used to identify the defective areas of the drill string. Related algorithms include, but are not limited to, U-net, V-net, Seg-net, Link-net, etc., and this invention does not impose any limitations. Specifically, preferably, this invention selects the universally applicable, highly generalizable, and easily improved deep learning semantic segmentation model U-net as the base model to train the drill string defect recognition segmentation model. U-Net is an image segmentation model widely used in fields such as medical imaging lesion recognition and material defect recognition. Compared with other common segmentation models mainly used to identify macroscopic objects in natural images, U-Net has unique advantages. The outputs of each layer of the encoder in U-Net are directly connected to the decoder, allowing the model to directly use high-resolution low-level visual information to assist in segmentation prediction. This advantage is crucial in medical image segmentation, material damage segmentation, and the apparent damage segmentation task in this project, as these all involve relatively subtle segmentation targets and often rely on low-level visual information for discrimination. Furthermore, U-Net uses convolutional layers instead of pooling layers to expand the receptive field, effectively avoiding the loss of local detail information. The overall model structure of U-Net is as follows: Figure 5 As shown. It should be noted that, in Figure 5 In the diagram, `conv3x3,ReLU` refers to a 3x3 convolutional kernel and a ReLU activation function (ReLU is a convolutional layer with a 3x3 kernel, activated by ReLU); `copy and crop` refers to copying and cropping (in the UNet method, the output size needs to be copied and centered for easier concatenation with the size generated by subsequent upsampling); `max pool 2x2` is a max pooling layer with a 2x2 pooling window; `up-conv 2x2` is deconvolutional upsampling with a 2x2 kernel; and `cony 1x1` is a 1x1 convolutional kernel.

[0038] Step 3.2: Input the collected and preprocessed magnetic particle image of the drill pipe into the constructed model, and use gradient descent optimization to train the model until the loss function of the model gradually converges, and the model training ends.

[0039] Step 4: Using the trained drill string defect detection model, calculate the anomaly score for the input drill string image to be inspected, and filter out abnormal magnetic particle images. Based on the trained drill string defect detection model, for any drill rod magnetic particle image collected in the previous stage of the integrated intelligent drill string defect detection system, the algorithm model will identify whether it is a defective drill rod and output an anomaly score. Users can set a threshold for the anomaly score to quickly filter out abnormal drill rod magnetic particle images and provide them to front-line business personnel, thereby reducing the workload of manual judgment by business personnel. It should be noted that the threshold can be flexibly set and adjusted by users according to their business rules and business objectives. Generally speaking, the higher the threshold is set, the more stringent the screening conditions are. Users can flexibly adjust the threshold here based on the results of the filtering and the effect of the secondary review in Step 5 to achieve the best effect in specific business scenarios. This invention does not impose any limitations.

[0040] Step 5: For the abnormal magnetic particle images identified in the initial screening, submit them to the business personnel for secondary review and confirmation. Based on the abnormal magnetic particle images initially screened by the system, the business personnel can conduct a focused secondary review and confirmation. If the business personnel have doubts about the results and believe that there are no drill pipe defects in the magnetic particle image, they can upload the image to a remote server. The remote cloud-based intelligent judgment unit will then perform a second, more precise judgment. Simultaneously, the judgment result from the cloud-based intelligent unit will be submitted to the next higher-level work team for further analysis, improving the accuracy of the judgment. It should be noted that the intelligent judgment function provides a scientific basis for subsequent maintenance decisions, helping to prevent potential equipment failures and extend the service life of the drill pipe. Furthermore, the modular design of this system provides flexibility for future technology upgrades and expanded applications, possessing broad application prospects and significant economic and social value.

[0041] In this embodiment, the drill rod loading and initial conveying process includes loading platform operation and production line startup. The loading platform operation process is as follows: First, an operator or automated robot places the drill rod to be inspected onto a specially designed loading platform. The platform design ensures stable loading of the drill rod, avoiding any potential damage. The production line startup process is as follows: Once the drill rod is correctly positioned, the control system activates the conveyor rollers, and the drill rod is automatically fed into the inspection production line. This production line uses a high-quality roller design to ensure smooth movement of the drill rod during conveying, reducing vibration and impact.

[0042] In this embodiment, the magnetization station and detection process includes a magnetization station, an observation station, and a demagnetization station. The magnetization station process is as follows: the drill rod reaches the first critical magnetization station; here, the drill rod is uniformly and strongly magnetized to facilitate subsequent magnetic particle inspection; the magnetization equipment can automatically adjust the magnetization intensity and duration according to the drill rod's material and specifications to ensure optimal inspection results. The observation station process is as follows: subsequently, the magnetized drill rod is transferred to the observation station. This station is equipped with a high-performance image acquisition device (such as a high-definition camera) to capture omnidirectional, high-resolution images of the magnetic powder sprayed on the drill rod surface under specific lighting conditions. The magnetic powder will accumulate at any surface or near-surface defects, forming a visible magnetic trace pattern. The demagnetization station process is as follows: after image acquisition, the drill rod enters the demagnetization station. Here, a specific demagnetization device eliminates residual magnetic fields to prevent interference with subsequent operations or the drill rod itself.

[0043] In this embodiment, intelligent analysis and decision-making includes image uploading and intelligent algorithm processing, as well as result feedback and operational decision-making. During image uploading and intelligent algorithm processing, the acquired magnetic trace images are uploaded to a cloud server in real time for analysis by a pre-trained intelligent algorithm. Based on a deep learning model, the algorithm accurately identifies and classifies different defect types, such as cracks and corrosion spots, and assesses the severity of the defects according to preset standards. During result feedback and operational decision-making, the algorithm analysis results are immediately fed back to the operator, who, combined with their experience, determines whether the drill pipe is qualified. For uncertain cases or those requiring further verification, the system supports secondary inspection of the drill pipe or direct manual review. The start and stop of the production line can be flexibly controlled via a central control console, ensuring the efficiency and flexibility of the inspection process.

[0044] In this embodiment, the material unloading and subsequent processing flow includes an unloading platform and a closed-loop process. During unloading, drill rods that have undergone magnetization, observation, and demagnetization at both ends are finally transported to the unloading platform. Qualified drill rods can be directly transported to subsequent use or storage stages, while drill rods marked as defective must be repaired or scrapped according to the inspection report. During the closed-loop process, the entire inspection process is managed in a closed loop, with each step recorded and traceable, ensuring transparency and auditability of quality control.

[0045] This invention not only significantly improves the production efficiency of the oil drilling industry but also greatly enhances safety performance, while promoting environmental protection and sustainable resource utilization. Firstly, from a production efficiency perspective, the automated operation of the drill string moving assembly line greatly shortens the drill pipe inspection cycle, reducing the time and labor costs consumed in traditional manual loading, unloading, and handling processes. The continuous operation capability of the assembly line means that the maintenance and preparation time of drilling equipment is significantly reduced, enabling drilling operations to respond more quickly to production needs and improving overall operational efficiency and capacity. Secondly, the combination of drill string inspection equipment and intelligent judgment devices ensures the accuracy and reliability of defect detection. Magnetic particle inspection technology, combined with precise image acquisition and analysis algorithms, detects even the smallest cracks or damage, effectively preventing accidents caused by drill string damage, ensuring the safety of personnel at the work site, avoiding costly equipment damage and production interruptions, thereby reducing operational risks. Thirdly, this invention also contributes to environmental protection. The intelligent magnetic particle collection and recycling mechanism reduces waste during the inspection process, avoids magnetic particle pollution of the environment, and embodies the concept of green production. Simultaneously, by optimizing energy use, such as employing energy-efficient motors and highly efficient energy management systems, energy consumption is reduced, meeting the urgent global demand for energy conservation and emission reduction. Finally, from a long-term development perspective, this invention sets a benchmark for technological upgrading in the oil drilling industry. Its modular and scalable design facilitates future technological iterations and the integration of new functions, laying the foundation for a transformation towards higher levels of automation and intelligence. This can enhance enterprise competitiveness, drive technological innovation across the entire industry, and create conditions for achieving digital and intelligent oilfield operation modes. The embodiments of this invention achieve full automation of drill pipe defect detection, incorporating intelligent analysis and decision-making, greatly improving detection efficiency and accuracy while reducing the labor intensity of operators, providing strong technical support for safe and efficient production in the oil drilling industry.

[0046] Example 3: This example provides a terminal device, which includes a memory, a processor, a communication interface, and a communication bus. The memory stores a program that can run on the processor. When the processor executes the program, it implements the automated drill pipe defect intelligent detection method in the above example.

[0047] The processor can be a central processing unit, or it can be other general-purpose processors, digital signal processors, application-specific integrated circuits, field-programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0048] The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and units, such as the program units corresponding to those in the above-described method embodiments of the present invention. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and modules stored in the memory, thereby realizing the automated drill pipe defect intelligent detection method described in the above embodiments.

[0049] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor, etc. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. The memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. One or more programs are stored in the memory and, when executed by the processor, perform the automated drill pipe defect intelligent detection method described in the above embodiments.

[0050] Example 4: This example provides a storage medium for computer-readable storage. The storage medium stores one or more programs, which can be executed by one or more processors to implement the automated drill pipe defect intelligent detection method as described in the above examples.

[0051] The storage medium can be an internal storage unit of the terminal device, such as the hard drive or memory of the terminal device. Alternatively, the storage medium can be an external storage device of the terminal device, such as a plug-in hard drive, smart memory card, secure digital card, or flash memory card installed on the terminal device.

[0052] The above technical features constitute the embodiments of the present invention, which have strong adaptability and implementation effect. Unnecessary technical features can be added or removed according to actual needs to meet the needs of different situations.

Claims

1. An automated intelligent detection system for drill pipe defects, characterized in that... include: The drill string moving assembly line is used to realize the automatic loading and unloading of drill pipes, ensuring that the drill pipes move to the inspection area along a predetermined path; Drill tool inspection equipment is used to generate a strong and stable magnetic field to uniformly magnetize the drill rod, control the distribution density of magnetic powder, and recover magnetic powder that has not adhered to defects. The test result collection and judgment device is used to photograph the surface of the drill rod after magnetization, process the image data in real time, accurately identify and classify defects, and generate a detailed test report.

2. The automated drill pipe defect intelligent detection system according to claim 1, characterized in that... The drilling tool moving production line includes a roller and steel frame mechanism, a loading and unloading mechanism, and a conveying and guiding mechanism; The roller and steel frame mechanism are used to ensure that the drill rod is transported smoothly and steadily, reducing frictional loss. The loading and unloading mechanism is used to automatically load and unload drill rods through precise positioning and gripping. The conveying and guiding mechanism is used to ensure that the drill pipe moves to the detection area along a predetermined path.

3. The automated drill pipe defect intelligent detection system according to claim 1 or 2, characterized in that... Drill bit inspection equipment includes a flaw detector and a magnetic particle application and collection system; Among them, the flaw detection host is used to generate a strong and stable magnetic field to uniformly magnetize the drill pipe; The magnetic powder application and collection system includes a magnetic powder spraying device and a magnetic powder collection box. The magnetic powder spraying device is used to precisely control the distribution density of the magnetic powder, and the magnetic powder collection box is used to recover the magnetic powder that has not adhered to the defect.

4. The automated drill pipe defect intelligent detection system according to claim 1 or 2, characterized in that... The detection result collection and judgment device includes an image acquisition unit, an edge computing unit, and a cloud-based intelligent judgment unit; The image acquisition unit is used to take pictures of the surface of the drill rod after magnetization from all angles to capture magnetic trace images. The edge computing unit is used to process the image data captured by the camera in real time, and to perform preliminary screening and preprocessing. The cloud-based intelligent judgment unit is used to accurately identify and classify defects, and automatically determines the severity of defects according to preset standards, and generates a detailed inspection report.

5. An automated intelligent detection method for drill pipe defects, characterized in that... Includes the following steps: Step 1: For both defective and defect-free drill pipes, collect a large number of magnetic particle images using an image acquisition module; Step 2: Perform image data preprocessing on the acquired magnetic particle display images; Step 3: Train the unsupervised anomaly detection model based on the dataset of images with defects; Step 4: Using the trained drill string defect detection model, calculate the anomaly score for the input drill string image to be detected, and filter out the abnormal magnetic particle images. Step 5: Submit the abnormal magnetic particle images to the business personnel for secondary review and confirmation.

6. The automated drill pipe defect intelligent detection method according to claim 5, characterized in that... Step 2 specifically includes the following steps: Step 2.1, Contrast Enhancement: The collected magnetic powder image is enhanced in contrast using image sharpening methods; Step 2.2, Image Data Enlargement: Increase the amount of relevant data to improve the training effect of the image defect detection model.

7. The automated intelligent detection method for drill pipe defects according to claim 5 or 6, characterized in that... Step 3 specifically includes the following steps: Step 3.1: Based on the collected and preprocessed defect image data, construct a drill tool defect identification algorithm model using image recognition algorithms in the field of image processing; Step 3.2: Input the collected and preprocessed magnetic particle image of the drill pipe into the constructed model, and use gradient descent optimization to train the model until the loss function of the model gradually converges, and the model training ends.

8. A terminal device, comprising a memory and a processor, wherein the memory stores a program executable on the processor, characterized in that, When the processor executes the program, it implements the automated drill pipe defect intelligent detection method as described in any one of claims 5 to 7.

9. A storage medium, characterized in that, The storage medium stores one or more programs, which can be executed by one or more processors to implement the automated drill pipe defect intelligent detection method as described in any one of claims 5 to 7.

Images