Well control blowout preventer field installation centering method and device based on stereoscopic vision identification

Abstract

The invention belongs to the technical field of oil and gas exploitation, and discloses a well control blowout preventer field installation centering method and device based on stereoscopic vision identification, four cameras are distributed around a blowout preventer base flange at equal intervals, the positions of well mouth flange screw holes are fully covered from multiple visual angles, the robustness of centering judgment is enhanced, and the centering accuracy is improved. And based on a centering judgment algorithm of image center deviation, the position of the wellhead flange and the distribution of the screw holes are accurately detected in combination with a computer vision algorithm, rapid and accurate positioning of the wellhead flange screw holes is achieved, and rapid centering and efficient installation are achieved.

Description

Technical Field

[0001] The present application belongs to the technical field of oil and gas production, and specifically relates to a method and device for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition. Background Art

[0002] Deep and ultra-deep oil and gas resources are the development direction of my country's energy strategy. As the burial depth increases, the pressure of oil and gas storage increases, and the potential danger of on-site operations increases. The role of well control blowout preventers as safety equipment for on-site drilling operations is increasing day by day. As an important safety equipment for drilling operations, well control blowout preventers are usually installed above the wellhead to control the pressure of oil and gas wells and prevent abnormal blowout accidents. The well control blowout preventer group has a complex structure, a large volume, and a variety of combinations. At the same time, the height reaches 5 to 6 meters and the weight exceeds dozens of tons. At present, the well control blowout preventer group and the wellhead device at the drilling site are mainly connected by flanges. The implementation process relies on mechanical devices and manual assistance, and there are the following problems: 1. Low manual centering accuracy: Traditional centering methods rely on manual visual inspection or simple mechanical auxiliary equipment, which can easily lead to insufficient centering accuracy due to operational errors, thus prolonging installation time; 2. Complex wellhead environment: The wellhead is usually accompanied by high temperature, high pressure, oil pollution and complex lighting conditions, which further increases the difficulty of alignment; 3. Flange screw holes are difficult to locate: The distribution characteristics of the bolt holes are difficult to quickly detect and align, and multiple adjustments are required during docking.

[0003] The existing connection and centering methods cause installation positioning deviation, low centering accuracy, and construction difficulties, which affect the improvement of on-site operation efficiency. Therefore, the connection between large well control blowout preventer groups and wellhead devices has become a technical problem for improving on-site engineering operation efficiency and mining safety in oil fields. Summary of the invention

[0004] The purpose of this application is to solve the problems of the prior art and to provide a method and device for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition.

[0005] In order to solve the technical problem, the technical solution of the present application is: a method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition, comprising the following steps: Step 1: Install the camera to capture images of the wellhead flange; Multiple cameras are installed on the blowout preventer base flange. The cameras are coaxially arranged at the screw holes of the blowout preventer base flange, and the optical axis of the camera is perpendicular to the plane where the wellhead flange is located. Each camera obtains image information of the wellhead flange in real time. Step 2: Image preprocessing: grayscale processing, noise filtering and edge enhancement processing are performed on the collected image; Step 3: Visual processing and centering judgment; Using the image preprocessed by the computer vision algorithm processing step, the center of the screw hole of the wellhead flange is located, the distance between the center of the screw hole of the wellhead flange and the center of the camera imaging plane is calculated, and it is determined whether the center of the screw hole of the wellhead flange is aligned with the center of the camera imaging plane; If the center of the screw hole of the wellhead flange photographed by a certain camera is located at the center of the camera imaging plane, it is determined that the centering condition of the camera is met; otherwise, it is not met and deviation information is provided; Step 4: Mechanical adjustment; According to the deviation information, the position of the BOP base flange is adjusted by the stepper motor to move it in the deviation direction, and then steps 2 and 3 are repeated until all cameras meet the centering conditions; When the screw holes of the wellhead flange photographed by multiple cameras all meet the alignment conditions, it is determined that the screw holes of the wellhead flange are completely aligned with the screw holes of the blowout preventer base flange; Step 5: Accuracy verification and alignment completed; Use a laser rangefinder or a high-precision posture sensor to verify the centering effect to ensure that the installation error is within the allowable range. When the screw holes of the blowout preventer base flange are fully aligned with the screw holes of the wellhead flange, the centering is completed and the operator is prompted to perform the flange connection operation.

[0006] Preferably, in step 1, four cameras are installed on the blowout preventer base flange, the four cameras are spaced apart at the screw holes of the blowout preventer base flange, and the connecting lines of the two sets of cameras facing each other are perpendicular to each other.

[0007] Preferably, the camera is calibrated with an internal parameter using a calibration plate to obtain the focal length, principal point coordinates and distortion coefficient parameters of the camera to ensure the imaging accuracy of the camera.

[0008] Preferably, the grayscale processing of the collected image in step 2 is specifically as follows: The collected color image Convert to grayscale image , use the weighted average of the three RGB color channels to calculate the grayscale value of each pixel;

[0009] Where: Represents a pixel in a grayscale image The pixel value of , , Represents color images Medium pixel The pixel values ​​for the red, green, and blue channels.

[0010] Preferably, in step 2, the collected image is subjected to noise filtering, specifically using Gaussian filtering to remove image noise:

[0011]

[0012] Where: Represents the pixel after Gaussian filtering The pixel value of Represents a pixel in a grayscale image The pixel value of Represents the Gaussian kernel, which determines the weight of the pixel; Represents the relative coordinates within the filter; represents the Gaussian kernel standard deviation; represents the Gaussian kernel radius.

[0013] Preferably, the edge enhancement processing of the collected image in step 2 is specifically performed as follows: The edge information of the image is extracted by using the Sobel operator, and then edge enhancement is achieved through the weighted gradient image. The Sobel operator is used to detect horizontal and vertical edges in the image and calculate the gradient of each pixel in the image. Horizontal Sobel operator:

[0014] Vertical Sobel operator:

[0015]

[0016] Represents pixel Gradient in the horizontal direction; Represents pixel Gradient in the vertical direction; Represents the pixel point in the image after Gaussian filtering The pixel value of Represents the convolution operation; Continue to calculate the gradient amplitude :

[0017] The places where the gradient magnitude is greater than 120 are identified as image edges. The gradient magnitude is added to the image after Gaussian filtering, and the weight factor Control the degree of enhancement:

[0018] Where: Represents the pixel point in the image after edge enhancement The pixel value of Represents the pixel point in the calculated gradient image The gradient magnitude.

[0019] Preferably, the computer vision algorithm in step 3 is Hough circle transform, which finds possible circular edge points through edge detection and then calculates the circle edge points in the parameter space. Voting: is the image after edge enhancement. Pixels on , assuming it is a point on the circle, deduce the possible center of the circle based on the equation of the circle and radius ;

[0020] in yes[ ] is used to traverse all points on the circle, and in this way a set of possible coordinates of the center of the circle is obtained. , and store it in the parameter space, in which the point with the most votes corresponds to the most likely center coordinate in the image, which is used to locate the center of the screw hole of the wellhead flange.

[0021] Preferably, the distance between the center of the screw hole of the wellhead flange and the center of the camera imaging plane is calculated in step 3 as follows: Compare the coordinates of the circle centers obtained by Hough circle transform with the coordinates of the center of the camera imaging plane. If the coordinate error is less than the distance error threshold, it is determined that the coordinates of the circle center coincide with the coordinates of the center of the camera imaging plane. The coordinates of the center of the camera imaging plane are:

[0022] Assume that the coordinates of the circle center obtained by Hough circle transformation are , the distance error threshold is set to , and then calculate the Euclidean distance between the two coordinates:

[0023] Then set the following conditions to judge the alignment:

[0024] if Less than the preset distance error threshold , it is considered that the center of the screw hole of the wellhead flange coincides with the center of the camera imaging plane, meeting the centering condition, and then it is determined that the installation operation can be carried out. Greater than or equal to the preset distance error threshold , if the alignment conditions are not met, mechanical adjustment is required.

[0025] Preferably, the well control blowout preventer on-site installation centering device based on stereoscopic vision recognition is used for implementing the well control blowout preventer on-site installation centering method based on stereoscopic vision recognition, and comprises a visual perception module, a visual processing and centering judgment module, a mechanical adjustment module, an accuracy verification module, and a power supply and communication module; The visual perception module includes a plurality of cameras, which are coaxially arranged at the screw holes of the blowout preventer base flange, and the optical axis of the camera is perpendicular to the plane where the wellhead flange is located, and each camera obtains image information of the wellhead flange in real time; The visual processing and centering judgment module includes a high-performance industrial computer equipped with image processing software for preprocessing image information and judging whether the center of the screw hole of the wellhead flange is aligned with the center of the camera imaging plane. If the center of the screw hole of the wellhead flange photographed by a certain camera is located at the center of the camera imaging plane, it is judged that the centering condition of the camera is met; otherwise, it is not met and deviation information is provided; The mechanical adjustment module includes a stepper motor and a linear slide rail, and the stepper motor and the linear slide rail adjust the position of the blowout preventer base flange according to the deviation information; The accuracy verification module includes a laser rangefinder or a high-precision posture sensor, which is installed on the flange of the blowout preventer base to verify the centering effect and ensure that the installation error is within the allowable range; The power supply and communication module is used to provide power support to the visual perception module, the visual processing and centering judgment module, the mechanical adjustment module and the accuracy verification module, and to realize data communication therebetween.

[0026] Preferably, the control accuracy of the mechanical adjustment module is 0.1 mm.

[0027] Compared with the prior art, the advantages of this application are: (1) The present invention proposes a method for on-site installation centering of a well control blowout preventer based on stereoscopic vision recognition. Four cameras are equidistantly distributed around the flange of the blowout preventer base, which fully covers the position of the wellhead flange screw holes from multiple perspectives, thereby enhancing the robustness of the centering judgment. The centering judgment algorithm based on the image center deviation is combined with the computer vision algorithm to accurately detect the position of the wellhead flange and the distribution of the screw holes, thereby realizing fast and accurate positioning of the wellhead flange screw holes, and achieving fast centering and efficient installation. (2) This application determines whether the center of the screw hole of the wellhead flange is aligned with the center of the camera imaging plane by the distance between the center of the screw hole of the wellhead flange and the center of the camera imaging plane. Compared with the existing wellhead docking method, this application focuses on the alignment of the screw holes of the blowout preventer and the wellhead flange, with high centering accuracy, which facilitates the subsequent installation of the wellhead blowout preventer bolts; (3) This application proposes a well control blowout preventer on-site installation centering device based on stereoscopic visual recognition, which innovatively combines the visual perception module with the mechanical adjustment module. Through a stepper motor, hydraulic drive or mechanical mobile platform, the blowout preventer position is adjusted in real time according to the deviation information provided by the visual processing and centering judgment module, thereby realizing automated and intelligent precise centering. (4) The device of the present application adopts a modular design, which is convenient for rapid integration into existing well control blowout preventer equipment. It has strong versatility and adaptability, can adapt to complex wellhead conditions, is not affected by external factors such as manual experience and light conditions, and greatly improves installation efficiency and reliability through automated and intelligent operation processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a schematic diagram of the assembly of the camera, wellhead flange and blowout preventer base flange for this application; Figure 2 This is a schematic diagram of the assembly of the camera on the flange of the blowout preventer base; Figure 3 This is a flow chart of the on-site installation and centering method of a well control blowout preventer based on stereoscopic vision recognition in this application; Figure 4 This is a schematic diagram of the image before grayscale processing in this application; Figure 5 This is a schematic diagram of the image after grayscale processing in this application; Figure 6 It is a schematic diagram of adjustment of an embodiment of the present application.

[0029] Description of reference numerals: 1. Camera, 2. Wellhead flange, 3. BOP base flange. DETAILED DESCRIPTION

[0030] The present application is described in detail below in conjunction with the accompanying drawings and specific embodiments, but the present application is not limited to these embodiments. The present application covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present application. In order to make the public have a thorough understanding of the present application, specific details are described in detail in the following embodiments of the present application, and those skilled in the art can fully understand the present application without the description of these details.

[0031] The BOP (Blowout Preventer) is the core well control equipment in oil and gas drilling operations, mainly used to prevent blowout accidents. It is installed on the top of the wellhead device and quickly closes the wellhead through hydraulic or mechanical control to control abnormal pressure or fluid overflow in the well. The BOP group usually consists of multiple BOP units, including annular BOPs and gate BOPs, as well as a series of auxiliary components. These units are stacked together to form a complete multi-level well control system to meet different well control needs. The working principle of the BOP group is to close the wellhead and direct the high-pressure fluid in the well to control equipment such as throttling pipes, reduce the risk of blowouts, and provide safety for subsequent mud circulation or pressure control.

[0032] There are two main types of BOPs: annular BOPs and ram BOPs. The annular BOP can adapt to various shapes of well tools (such as drill pipes, casings) through its flexible rubber seals, and can even completely close the open wellhead. It is usually used as the first line of defense to quickly respond to abnormal pressure in the well. The ram BOP includes different types such as blind plates, shear rams and tube sheets, which are used to cut off the tubing in the well or completely close the wellhead. The shear ram can cut off the drill pipe or casing and form a seal at the same time to deal with extreme blowout scenarios. Through multi-stage stacking combinations, the BOP can provide redundant well control means to enhance the safety and reliability of the system.

[0033] The installation of the BOP group needs to be connected to the wellhead equipment through a flange. There are multiple threaded holes on the flange, which are aligned with the threaded holes of the wellhead flange to ensure good sealing and mechanical stability. In actual operation, the BOP group needs to be strictly calibrated and maintained to ensure that it can operate normally under extreme conditions such as high pressure, high temperature, and corrosive fluids. At the same time, the BOP group cooperates with the hydraulic control system to achieve rapid response and protect the safety of equipment, personnel and the environment at the critical moment of well kick or blowout. The wide application of this equipment has greatly improved the safety and efficiency of modern drilling operations and is an indispensable and important guarantee equipment for the oil and gas industry.

[0034] Advantages of visual recognition: Computer vision algorithms have the advantages of high precision, automation, and real-time feedback. They have been widely used in industrial scenarios in recent years, and are particularly suitable for high-precision alignment needs.

[0035] The uniqueness of multi-camera vision: The multi-camera vision system can provide multi-view information through the collaborative work of multiple cameras, solve the problem of limited view of a single camera, and significantly improve the centering accuracy and stability.

[0036] Therefore, it is necessary to develop a well control blowout preventer installation method based on multi-eye vision recognition, which can effectively solve the current problems of low installation efficiency and insufficient centering accuracy, and has important engineering value and commercial prospects.

[0037] like Figures 1 to 3 As shown, the present application discloses a method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition, comprising the following steps: Step 1: Install camera 1 to collect images of wellhead flange 2; A plurality of cameras 1 are installed on the blowout preventer base flange 3. The cameras 1 are coaxially arranged at the screw holes of the blowout preventer base flange 3, and the optical axis of the cameras 1 is perpendicular to the plane where the wellhead flange 2 is located. Each camera 1 acquires image information of the wellhead flange 2 in real time. Step 2: Image preprocessing: grayscale processing, noise filtering and edge enhancement processing are performed on the collected image; Step 3: Visual processing and centering judgment; Process the image preprocessed in step 2 by a computer vision algorithm, locate the center of the screw hole of the wellhead flange 2, calculate the distance between the center of the screw hole of the wellhead flange 2 and the center of the camera imaging plane, and determine whether the center of the screw hole of the wellhead flange 2 is aligned with the center of the camera imaging plane; If the center of the screw hole of the wellhead flange 2 photographed by a certain camera 1 is located at the center of the camera imaging plane (image center), it is determined that the centering condition of the camera 1 is satisfied; otherwise, it is not satisfied and deviation information is provided; Step 4: Mechanical adjustment; According to the deviation information, the position of the BOP base flange 3 is adjusted by a stepper motor to move it in the deviation direction, and then steps 2 to 3 are repeated until all cameras 1 meet the centering conditions; When the screw holes of the wellhead flange 2 photographed by the multiple cameras 1 all meet the centering condition, it is determined that the screw holes of the wellhead flange 2 are completely aligned with the screw holes of the blowout preventer base flange 3; Step 5: Accuracy verification and alignment completed; Use a laser rangefinder or a high-precision posture sensor to verify the centering effect to ensure that the installation error is within the allowable range. When the screw holes of the blowout preventer base flange 3 are fully aligned with the screw holes of the wellhead flange 2, the centering is completed and the operator is prompted to perform the flange connection operation.

[0038] Preferably, in step 1, four cameras 1 are installed on the blowout preventer base flange 3, the four cameras 1 are spaced apart at the screw holes of the blowout preventer base flange 3, and the connecting lines of two groups of cameras 1 facing each other are perpendicular to each other.

[0039] Preferably, the camera 1 is calibrated with an internal parameter using a calibration plate to obtain the focal length, principal point coordinates and distortion coefficient parameters of the camera 1 to ensure the imaging accuracy of the camera 1 .

[0040] The camera 1 is an industrial camera. Four industrial cameras are installed equidistantly around the blowout preventer base flange 3. The model is a high-resolution industrial camera of a certain brand. The lens field of view covers the wellhead flange 2 area. The optical axis of the camera 1 is perpendicular to the plane of the wellhead flange 2.

[0041] Preferably, the grayscale processing of the collected image in step 2 is specifically as follows: The collected color image Convert to grayscale image , use the weighted average of the three RGB color channels to calculate the grayscale value of each pixel;

[0042] Where: Represents a pixel in a grayscale image The pixel value of , , Represents color images Medium pixel The pixel values ​​for the red, green, and blue channels.

[0043] like Figure 4 As shown, it is a schematic diagram of the image before grayscale processing of this application; Figure 5 The figure shows the image after grayscale processing in this application.

[0044] Preferably, in step 2, the collected image is subjected to noise filtering, specifically using Gaussian filtering to remove image noise:

[0045]

[0046] Where: Represents the pixel after Gaussian filtering The pixel value of Represents a pixel in a grayscale image The pixel value of Represents the Gaussian kernel, which determines the weight of the pixel; Represents the relative coordinates within the filter; represents the Gaussian kernel standard deviation; represents the Gaussian kernel radius.

[0047] Preferably, the edge enhancement processing of the collected image in step 2 is specifically performed as follows: The edge information of the image is extracted by using the Sobel operator, and then edge enhancement is achieved through the weighted gradient image. The Sobel operator is used to detect horizontal and vertical edges in the image and calculate the gradient of each pixel in the image. Horizontal Sobel operator:

[0048] Vertical Sobel operator:

[0049]

[0050] Represents pixel Gradient in the horizontal direction; Represents pixel Gradient in the vertical direction; Represents the pixel point in the image after Gaussian filtering The pixel value of Represents the convolution operation; Continue to calculate the gradient amplitude :

[0051] The gradient magnitude is greater than 120 and is considered as the image edge. The gradient magnitude is added to the Gaussian filtered image, and the weight factor Control the degree of enhancement:

[0052] Where: Represents the pixel point in the image after edge enhancement The pixel value of Represents the pixel point in the calculated gradient image The gradient magnitude.

[0053] Preferably, the computer vision algorithm in step 3 is Hough circle transform, which finds possible circular edge points through edge detection and then calculates the circle edge points in the parameter space. Voting: is the image after edge enhancement. Pixels on , assuming it is a point on the circle, deduce the possible center of the circle based on the equation of the circle and radius ;

[0054] in yes[ ] is used to traverse all points on the circle, and in this way a set of possible coordinates of the center of the circle is obtained. , and store it in the parameter space, in which the point with the most votes corresponds to the most likely center coordinate in the image, which is used to locate the center of the screw hole of the wellhead flange 2; Preferably, the distance between the center of the screw hole of the wellhead flange 2 and the center of the camera imaging plane is calculated in step 3 as follows: Compare the coordinates of the circle centers obtained by Hough circle transform with the coordinates of the center of the camera imaging plane. If the coordinate error is less than the distance error threshold, it is determined that the coordinates of the circle center coincide with the coordinates of the center of the camera imaging plane. The coordinates of the center of the camera imaging plane are:

[0055] Assume that the coordinates of the circle center obtained by Hough circle transformation are , the distance error threshold is set to , and then calculate the Euclidean distance between the two coordinates:

[0056] Then set the following conditions to judge the alignment:

[0057] if Less than the preset distance error threshold , it is considered that the center of the screw hole of the wellhead flange 2 coincides with the center of the camera imaging plane, which meets the centering condition, and then it is determined that the installation operation can be carried out. Greater than or equal to the preset distance error threshold , if the alignment conditions are not met, mechanical adjustment is required.

[0058] Preferably, the well control blowout preventer on-site installation centering device based on stereoscopic vision recognition is used for implementing the well control blowout preventer on-site installation centering method based on stereoscopic vision recognition, and comprises a visual perception module, a visual processing and centering judgment module, a mechanical adjustment module, an accuracy verification module, and a power supply and communication module; The visual perception module includes a plurality of cameras 1, which are coaxially arranged at the screw holes of the blowout preventer base flange 3, and the optical axis of the camera 1 is perpendicular to the plane where the wellhead flange 2 is located, and each camera 1 obtains image information of the wellhead flange 2 in real time; The visual processing and centering judgment module includes a high-performance industrial computer equipped with image processing software for preprocessing image information and judging whether the center of the screw hole of the wellhead flange 2 is aligned with the center of the camera imaging plane. If the center of the screw hole of the wellhead flange 2 photographed by a certain camera 1 is located at the center of the camera imaging plane, it is judged that the centering condition of the camera 1 is met; otherwise, it is not met and deviation information is provided; The mechanical adjustment module includes a stepper motor and a linear slide rail, and the stepper motor and the linear slide rail adjust the position of the blowout preventer base flange 3 according to the deviation information; The accuracy verification module includes a laser rangefinder or a high-precision posture sensor, which is installed on the blowout preventer base flange 3 to verify the centering effect and ensure that the installation error is within the allowable range; The power supply and communication module is used to provide power support to the visual perception module, the visual processing and centering judgment module, the mechanical adjustment module and the accuracy verification module, and to realize data communication therebetween.

[0059] Preferably, the control accuracy of the mechanical adjustment module is 0.1 mm, and the adjustment range meets the on-site requirements of well control.

[0060] The visual processing and centering judgment module includes a high-performance industrial computer equipped with image processing software (such as MATLAB, OpenCV, etc.) to process the collected screw hole images of the wellhead flange 2.

[0061] Example 1 like Figure 3 As shown, the implementation steps of this application are as follows: (1) Camera installation and calibration: Before installation, the four cameras 1 are calibrated with internal parameters using the calibration board to obtain parameters such as the focal length, principal point coordinates, and distortion coefficient of the camera 1 to ensure the imaging accuracy of the camera.

[0062] (2) Image acquisition, image processing, screw hole boundary extraction, screw hole center fitting: The BOP base flange 3 is preliminarily aligned with the wellhead flange 2, the camera is started, the wellhead flange image is collected, the screw hole features are extracted through the image processing algorithm, and the center position of the screw hole is located.

[0063] (3) Alignment judgment and adjustment: Is the center of the screw hole located in the center of the image? Yes, the alignment condition is met; No, rotate and move the BOP; Calculate the deviation between the center of the screw hole of each camera and the center of the image; if all four cameras meet the centering condition (deviation is less than 0.2mm), the centering is considered to be completed; otherwise, adjust the position of the BOP in the direction of the deviation.

[0064] (4) Installation and accuracy verification: After the alignment is completed, tighten the connecting bolts between the BOP and the wellhead flange. Use a laser rangefinder to verify the alignment accuracy between the BOP flange and the wellhead flange to ensure that the installation error is within the allowable range.

[0065] Example 2 In order to verify the reliability of the method in this application, a centering test was carried out according to the design process to observe the program status and generated results during the centering process. Figure 1 ; From the feedback results, we can see that when the distance error threshold is designed <5mm, Euclidean distance in initial state is 76.87mm; after the first adjustment, due to the Euclidean distance between the two coordinates (40.44mm) is greater than the distance error threshold, so it needs to be adjusted again. When the Euclidean distance between the two coordinates is (1.18mm) less than the distance error threshold The program prompts that the alignment is completed. Thus, the feasibility of the alignment judgment technology of the wellhead flange and the blowout preventer flange of the present application has been verified through experiments.

[0066] This application works as follows: This application uses multi-eye stereo vision recognition technology and Hough transform to detect the center position of the wellhead flange screw hole, judges the current alignment state according to the deviation between the center of the screw hole and the camera imaging plane, and controls the stepper motor according to the deviation information to adjust the position of the blowout preventer until the alignment condition is met, thereby solving the engineering problem of alignment and positioning of the well control blowout preventer group and the wellhead flange, significantly improving the installation accuracy and efficiency, ensuring the complete alignment of the flange threaded holes, and avoiding deviations and equipment damage caused by human errors; at the same time, through automation and intelligent operation, human intervention is reduced, and the safety risks in the installation process are reduced, especially in complex well site environments, and the installation posture can be monitored and adjusted in real time to ensure the sealing and stability of the flange connection and the reliability of the subsequent operation of the blowout preventer group; this method also greatly saves manpower and equipment maintenance costs, shortens the installation time, and is efficient, automated and universal; in addition, this method fully meets the intelligent development needs of modern digital oilfields, provides a new technical solution for the installation and maintenance of oilfield equipment, and has the potential for widespread application in other industrial high-precision alignment scenarios.

[0067] This application completes the precise alignment of the blowout preventer and the wellhead flange. The installation time is greatly reduced compared to the traditional manual alignment method, which significantly improves efficiency. The system adapts to different lighting conditions, has high robustness, and ensures the safety of operators.

[0068] This application takes multi-eye stereo vision technology as the core, combines intelligent recognition and closed-loop control, and realizes high-precision centering positioning of the blowout preventer and the wellhead. This method not only improves the efficiency and safety of well control blowout preventer installation, but also provides a reference for equipment that requires high-precision centering positioning in other industrial scenarios.

[0069] This application uses multi-eye vision collaborative recognition: four cameras are equidistantly distributed at the screw hole positions of the blowout preventer base flange to achieve all-round screw hole alignment monitoring. The camera optical axis is perpendicular to the wellhead flange plane to ensure that the collected images have accurate geometric relationships. The visual processing and alignment judgment module achieves real-time alignment judgment through image feature extraction and position deviation calculation.

[0070] The visual processing and centering judgment algorithm of this application: Based on computer vision technology (such as Hough transform, edge detection, etc.), the screw hole position of the wellhead flange is accurately located; by calculating the deviation between the center of the screw hole and the center of the camera image, the centering condition of each camera perspective is judged one by one; multi-perspective data is integrated to comprehensively judge the overall centering status of the blowout preventer flange and the wellhead flange Mechanical adjustment and feedback control system of this application: The adjustment device (such as a stepper motor, hydraulic drive or mechanical mobile platform) can automatically adjust the position of the blowout preventer based on the deviation information provided by the visual processing and centering judgment module. Through closed-loop feedback control, it is ensured that the adjustment process gradually approaches the centering target.

[0071] This application's high-precision installation and verification method: Use a laser rangefinder or a high-precision posture sensor to verify the alignment effect after installation, ensure that the final installation accuracy meets the engineering requirements, and combine visual and mechanical adjustment technology to ensure that the alignment error is low enough.

[0072] The preferred implementation modes of the present application have been described in detail above, but the present application is not limited to the above implementation modes, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present application.

[0073] Many other changes and modifications can be made without departing from the concept and scope of the present application.It should be understood that the present application is not limited to specific embodiments, and the scope of the present application is defined by the appended claims.

Claims

1. A method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition, characterized in that: The following steps are involved: Step 1: Install a camera (1) to capture an image of the wellhead flange (2); A plurality of cameras (1) are installed on the blowout preventer base flange (3), the cameras (1) are coaxially arranged at the screw holes of the blowout preventer base flange (3), and the optical axes of the cameras (1) are perpendicular to the plane where the wellhead flange (2) is located, and each camera (1) obtains image information of the wellhead flange (2) in real time; Step 2: Image preprocessing: grayscale processing, noise filtering and edge enhancement processing are performed on the collected image; Step 3: Visual processing and centering judgment; Processing the image pre-processed in step 2 by a computer vision algorithm, locating the center of the screw hole of the wellhead flange (2), calculating the distance between the center of the screw hole of the wellhead flange (2) and the center of the camera imaging plane, and determining whether the center of the screw hole of the wellhead flange (2) is aligned with the center of the camera imaging plane; If the center of the screw hole of the wellhead flange (2) photographed by a certain camera (1) is located at the center of the camera imaging plane, it is determined that the centering condition of the camera (1) is satisfied; otherwise, it is not satisfied and deviation information is provided; Step 4: Mechanical adjustment; According to the deviation information, the position of the blowout preventer base flange (3) is adjusted by a stepper motor to move it in the deviation direction, and then steps 2 to 3 are repeated until all cameras (1) meet the centering conditions; When the screw holes of the wellhead flange (2) photographed by the multiple cameras (1) all meet the centering condition, it is determined that the screw holes of the wellhead flange (2) and the screw holes of the blowout preventer base flange (3) are completely aligned; Step 5: Accuracy verification and alignment completed; The alignment effect is verified by using a laser rangefinder or a high-precision posture sensor to ensure that the installation error is within the allowable range. When the screw holes of the blowout preventer base flange (3) are completely aligned with the screw holes of the wellhead flange (2), the alignment is completed and the operator is prompted to perform the flange connection operation.

2. The method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition according to claim 1, characterized in that: In step 1, four cameras (1) are installed on the blowout preventer base flange (3). The four cameras (1) are arranged at intervals at the screw holes of the blowout preventer base flange (3), and the connecting lines of two groups of cameras (1) facing each other are perpendicular to each other.

3. The method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition according to claim 2, characterized in that: The camera (1) is calibrated with an internal parameter using a calibration plate to obtain the focal length, principal point coordinates and distortion coefficient parameters of the camera (1), thereby ensuring the imaging accuracy of the camera (1).

4. The method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition according to claim 1, characterized in that: The grayscale processing of the collected image in step 2 is specifically as follows: The collected color image Convert to grayscale image , use the weighted average of the three RGB color channels to calculate the grayscale value of each pixel; ； Where: Represents a pixel in a grayscale image The pixel value of , , Represents color images Medium pixel The pixel values ​​for the red, green, and blue channels.

5. The method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition according to claim 4, characterized in that: In step 2, the collected image is subjected to noise filtering, specifically using Gaussian filtering to remove image noise: ； ； Where: Represents the pixel after Gaussian filtering The pixel value of Represents a pixel in a grayscale image The pixel value of Represents the Gaussian kernel, which determines the weight of the pixel; Represents the relative coordinates within the filter; represents the Gaussian kernel standard deviation; represents the Gaussian kernel radius.

6. The method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition according to claim 5, characterized in that: The edge enhancement process of the collected image in step 2 is specifically as follows: The edge information of the image is extracted by using the Sobel operator, and then edge enhancement is achieved through the weighted gradient image. The Sobel operator is used to detect horizontal and vertical edges in the image and calculate the gradient of each pixel in the image. Horizontal Sobel operator: ； Vertical Sobel operator: ； ； Represents pixel Gradient in the horizontal direction; Represents pixel Gradient in the vertical direction; Represents the pixel point in the image after Gaussian filtering The pixel value of Represents the convolution operation; Continue to calculate the gradient amplitude : ； The places where the gradient magnitude is greater than 120 are identified as image edges. The gradient magnitude is added to the image after Gaussian filtering, and the weight factor Control the degree of enhancement: ； Where: Represents the pixel point in the image after edge enhancement The pixel value of Represents the pixel point in the calculated gradient image The gradient magnitude.

7. The method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition according to claim 6, characterized in that: The computer vision algorithm in step 3 is the Hough circle transform, which finds possible circular edge points through edge detection and then calculates the circle edge points in the parameter space. Voting: is the image after edge enhancement. Pixels on , assuming it is a point on the circle, deduce the possible center of the circle based on the equation of the circle and radius ; ； in yes[ ] is used to traverse all points on the circle, and in this way a set of possible coordinates of the center of the circle is obtained. and store it in a parameter space, in which the point with the most votes corresponds to the most likely center coordinate of the circle in the image, which is used to locate the center of the screw hole of the wellhead flange (2).

8. The method for on-site installation and centering of a well control blowout preventer based on stereoscopic vision recognition according to claim 7, characterized in that: The distance between the center of the screw hole of the wellhead flange (2) and the center of the camera imaging plane is calculated in step 3 as follows: Compare the coordinates of the circle centers obtained by Hough circle transform with the coordinates of the center of the camera imaging plane. If the coordinate error is less than the distance error threshold, it is determined that the coordinates of the circle center coincide with the coordinates of the center of the camera imaging plane. The coordinates of the center of the camera imaging plane are: ； Assume that the coordinates of the circle center obtained by Hough circle transformation are , the distance error threshold is set to , and then calculate the Euclidean distance between the two coordinates: ； Then set the following conditions to judge the alignment: ； if Less than the preset distance error threshold , then it is considered that the center of the screw hole of the wellhead flange (2) coincides with the center of the camera imaging plane, satisfying the centering condition, and then it is determined that the installation operation can be carried out. Greater than or equal to the preset distance error threshold , if the alignment conditions are not met, mechanical adjustment is required.

9. A well control blowout preventer on-site installation centering device based on stereoscopic vision recognition, characterized in that: Used for the implementation of the on-site installation and centering method of a well control blowout preventer based on stereoscopic visual recognition as described in any one of claims 1 to 8, comprising a visual perception module, a visual processing and centering judgment module, a mechanical adjustment module, an accuracy verification module, and a power supply and communication module; The visual perception module comprises a plurality of cameras (1), the plurality of cameras (1) being coaxially arranged at screw holes of a blowout preventer base flange (3), and the optical axes of the cameras (1) being perpendicular to the plane where the wellhead flange (2) is located, and each camera (1) acquires image information of the wellhead flange (2) in real time; The visual processing and centering judgment module includes a high-performance industrial computer equipped with image processing software for preprocessing image information and judging whether the center of the screw hole of the wellhead flange (2) is aligned with the center of the camera imaging plane. If the center of the screw hole of the wellhead flange (2) photographed by a certain camera (1) is located at the center of the camera imaging plane, it is judged that the centering condition of the camera (1) is satisfied; otherwise, it is not satisfied and deviation information is provided; The mechanical adjustment module comprises a stepper motor and a linear slide rail, and the stepper motor and the linear slide rail adjust the position of the blowout preventer base flange (3) according to the deviation information; The accuracy verification module includes a laser rangefinder or a high-precision posture sensor, which is installed on the blowout preventer base flange (3) to verify the centering effect and ensure that the installation error is within the allowable range; The power supply and communication module is used to provide power support to the visual perception module, the visual processing and centering judgment module, the mechanical adjustment module and the accuracy verification module, and to realize data communication therebetween.

10. The well control blowout preventer on-site installation centering device based on stereoscopic vision recognition according to claim 9, characterized in that: The control accuracy of the mechanical adjustment module is 0.1 mm.

Images