Device and method for detecting tilt of power catwalk loading string based on machine vision

Abstract

The present invention discloses a device and method for detecting the inclination of a power catwalk loaded pipe string based on machine vision, comprising a catwalk PLC, an ultrasonic sensor, a lever detector, a central industrial controller, a central, industrial color camera and an industrial monitoring all-in-one machine. The industrial color camera comprises a control unit and an acquisition unit. The catwalk PLC, the central industrial controller and the control unit are connected via Ethernet to form a local area network. The detection method comprises the following steps: the ultrasonic sensor and the lever detector detect whether a signal is blocked by the pipe string, while the industrial color camera collects and extracts a V-groove ROI area image frame by frame, runs an image analysis algorithm, detects whether the pipe string is completely loaded into the catwalk V-groove, and sends a signal to the industrial monitoring all-in-one machine. The present invention has the characteristics of being able to switch detection means when a sensor fails, highly integrated programs and reducing the efficiency of misjudgment.

Description

Technical Field

[0001] The invention belongs to the technical field of oil drilling and production detection equipment, and relates to a device for detecting the inclination of a power catwalk loading string based on machine vision, and also relates to a method for detecting the inclination of a power catwalk loading string based on machine vision. Background Art

[0002] The automation level of domestic drilling rigs used to extract oil and other mineral resources is increasing. Power catwalks, the automated equipment used to deliver tubing to the drill floor, are also becoming increasingly common. Improving the automation of power catwalks is crucial for improving tubing delivery efficiency and, ultimately, overall geological drilling and production efficiency.

[0003] While the power catwalk has a self-control system that can perform general logic and functional control, it can easily tilt when transporting the tubing to the V-groove, sometimes causing both ends of the tubing to not be fully seated. Existing technologies use sensors to determine whether the tubing is in place in the V-groove, but these sensors often fail or misjudge due to factors such as collision and vibration, necessitating a pause in the process for manual correction, which compromises drilling efficiency. Summary of the Invention

[0004] The purpose of the present invention is to provide a device for detecting the inclination of a power catwalk loading string based on machine vision, which solves the problem in the prior art that sensor failure cannot be detected, thus affecting drilling efficiency.

[0005] Another object of the present invention is to provide a method for detecting the inclination of a power catwalk loading string based on machine vision, which has the characteristic of reducing the efficiency of misjudgment.

[0006] One solution of the present invention is a device for detecting the inclination of a power catwalk loading string based on machine vision, comprising a catwalk PLC disposed inside a catwalk body, one end of the catwalk PLC being connected to an ultrasonic sensor via a signal line, and the other end being connected to a lever detector via a signal line, the catwalk PLC being further connected to a central industrial controller via a signal line, the input end of the central industrial controller being connected to an industrial color camera, and the output end of the central industrial controller being connected to an all-in-one industrial monitoring machine;

[0007] The industrial color camera includes an interconnected control unit and acquisition unit, and the central industrial controller is connected to the control unit signal; the catwalk PLC, central industrial controller and control unit are connected to Ethernet to form a local area network.

[0008] The present invention is also characterized in that the central industrial controller is any one of a programmable controller PLC, an embedded industrial computer or a PXI controller.

[0009] Another solution of the present invention is a method for detecting the inclination of a power catwalk loading string based on machine vision, which specifically comprises the following steps:

[0010] Step 1: Arrange the device for detecting the inclination of the power catwalk loading string in the actual drilling rig structure and initialize the catwalk PLC, industrial color camera and central industrial controller;

[0011] Step 2: The ultrasonic sensor and lever detector detect whether the signal is blocked by the pipe string and send it to the industrial monitoring integrated machine through the catwalk PLC and the central industrial controller in sequence;

[0012] If so, the industrial monitoring machine receives a signal indicating that the pipe is blocked and displays "Sensor judgment: pipe loaded in place";

[0013] If not, the industrial monitoring machine receives a signal that the pipe string is not blocked and displays "Sensor judgment: pipe string is not loaded in place";

[0014] Step 3: The acquisition unit acquires and extracts the catwalk V-groove ROI area image frame by frame and sends it to the control unit;

[0015] Step 4: The control unit runs an image analysis algorithm based on the catwalk V-groove ROI area image and determines whether the tubing string is completely loaded into the catwalk V-groove;

[0016] If so, a signal indicating that the pipe string has been successfully loaded into the catwalk V-groove is sent via the central industrial controller to the industrial monitoring integrated machine, which displays "Machine vision interpretation: pipe string successfully loaded into the catwalk";

[0017] If not, a signal indicating that the pipe string has failed to be loaded into the catwalk V-groove is sent to the industrial monitoring all-in-one machine through the central industrial controller, and the industrial monitoring all-in-one machine displays "Machine vision interpretation: Failed to load the pipe string into the catwalk".

[0018] The present invention is also characterized in that: the device for detecting the inclination of the power catwalk loading string in step 1 is arranged as follows: a central industrial controller and an industrial monitoring integrated machine are placed in the driller's room during actual drilling, an industrial color camera is placed on the drill floor guardrail and aimed at the end face of the catwalk V-groove, an ultrasonic sensor is placed in the middle of the catwalk V-groove, and a lever detector is placed at the end of the catwalk V-groove;

[0019] The industrial color camera initialization in step 1 includes initializing the exposure time, gain, and sampling mode parameters. These parameters can also be modified on the industrial monitoring device. The industrial color camera initialization also includes setting the image's ROI rectangular analysis area to only the area in front of and behind the V-groove of the horizontal catwalk when the catwalk is at its lowest point.

[0020] The present invention is also characterized in that: step 2 is specifically implemented according to the following steps:

[0021] Step 2.1: The catwalk PLC program executes the pipe string kick-in command. The catwalk pipe rack rises, and the pipe string rolls onto the tilting mechanism. Then all the tilting structures on the same side are lifted. After the pipe string rolls over the catwalk support end surface, it rolls into the V-groove. The ultrasonic sensor and lever detector detect whether it is blocked by the closed pipe string.

[0022] Step 2.2: When the ultrasonic sensor and lever detector detect a pipe obstruction signal, the signal is sent to the industrial monitoring machine through the catwalk PLC and the central industrial controller. The industrial monitoring machine receives the pipe obstruction signal and displays "Sensor judgment: pipe loaded in place";

[0023] When the ultrasonic sensor and lever detector do not detect the signal of pipe column obstruction or only detect one of them, the signal is sent to the industrial monitoring machine through the catwalk PLC and the central industrial controller in sequence. The industrial monitoring machine receives the signal that the pipe column is not obstructed and displays "Sensor judgment: pipe column is not loaded in place";

[0024] In step 3, the industrial color camera extracts the front and rear end areas of the catwalk V-groove of the captured image according to the initialized ROI area setting;

[0025] Step 4 is implemented as follows:

[0026] Step 4.1, the control unit performs filtering and noise reduction, histogram equalization, binarization, morphological operation, and edge extraction on the ROI detection area to obtain the contour surface of the catwalk V-groove and the pipe column;

[0027] Step 4.2: The algorithm sets the middle area of ​​the couplings at both ends of the pipe string and selects three detection rectangular areas with a uniform interval length ΔL, which are defined as: the proximal detection point, the mid-end detection point, and the distal detection point. ΔL is approximately equal to 1 / 4 of the pipe string length.

[0028] Calculate the areas S1, S1′, S2, S2′, S3, S3′ of the quadrilaterals formed by the intersection of the rectangular frames of the three detection points and the upper and lower contour frames of the V-shaped groove and the catwalk respectively;

[0029] Step 4.3: Let K1 = S1 / S1′, K2 = S2 / S2′, and K3 = S3 / S3′, where K1, K2, and K3 are the area ratios of the three rectangular frames. If K1, K2, and K3 are all within the reasonable threshold value set by the program, 0.95-1, the program deems that the three detection points S1 = S1′, S2 = S2′, and S3 = S3′ have all passed the detection. A straight line is determined based on two points, and a plane is determined based on three points. The image analysis algorithm of the machine vision determines that the pipe string has been successfully loaded into the catwalk V-groove. A signal indicating that the pipe string has been successfully loaded into the catwalk V-groove is sent to the industrial monitoring integrated machine via the central industrial controller, and the industrial monitoring integrated machine displays "Machine Vision Interpretation: Pipe string successfully loaded into the catwalk."

[0030] Step 4.4: If the three detection points S1=S1′, S2=S2′, and S3=S3′ fail, the program determines that the pipe string has not been successfully loaded into the catwalk V-groove. A signal indicating that the pipe string has failed to be loaded into the catwalk V-groove is sent through the central industrial controller to the industrial monitoring integrated machine, which displays "Machine vision interpretation: pipe string failed to be loaded into the catwalk."

[0031] The beneficial effects of the present invention are:

[0032] 1. This invention uses an industrial color camera, machine vision, and image analysis technology to automatically and contactlessly detect whether the tubing has been successfully loaded onto the power catwalk. This serves as a supplementary verification of the sensor's normal operation. Furthermore, if a sensor fails, the system automatically switches to a reliable backup detection method, avoiding the problem of successful tubing loading but inability to detect, which can lead to process suspension.

[0033] 2. The present invention establishes communication between various devices through signal lines to establish a local area network, and a unified industrial monitoring all-in-one machine monitors and operates the status of the equipment in the local area network, thereby increasing data fusion and system integration. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Schematic diagram of the structure of the device for detecting the inclination of the power catwalk loading string based on machine vision of the present invention;

[0035] Figure 2 This is a schematic diagram of the position of the device for detecting the inclination of the power catwalk loading string based on machine vision in the actual drilling rig structure of the present invention;

[0036] Figure 3 This is a schematic diagram of the specific position of the sensor of the device for detecting the inclination of the power catwalk loading string based on machine vision in the catwalk of the present invention;

[0037] Figure 4 This is an algorithm flow chart of the method for detecting the inclination of a power catwalk loading string based on machine vision of the present invention;

[0038] Figure 5 It is the ROI area taken by the industrial color camera in the method of detecting the inclination of the power catwalk loading string based on machine vision of the present invention;

[0039] Figure 6 This is a schematic diagram of the successful loading of a pipe string into a catwalk V-groove captured by an industrial color camera in the method for detecting the inclination of a pipe string loaded on a power catwalk based on machine vision of the present invention;

[0040] Figure 7 This is a schematic diagram of an industrial color camera capturing an image of a pipe string not being successfully loaded into a catwalk V-groove in a method for detecting the inclination of a pipe string loaded on a power catwalk based on machine vision according to the present invention;

[0041] In the figure, 1. Catwalk PLC, 2. Ultrasonic sensor, 3. Lever detector, 4. Central industrial controller, 5. Industrial color camera, 6. Industrial monitoring all-in-one machine, 501. Control unit, 502. Acquisition unit. DETAILED DESCRIPTION

[0042] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] The present invention provides a device for detecting the inclination of a power catwalk loading string based on machine vision, such as Figure 1 As shown, it includes a catwalk PLC 1, an ultrasonic sensor 2, a lever detector 3, a central industrial controller 4, an industrial color camera 5 and an industrial monitoring all-in-one machine 6 connected via data cables to form a local area network;

[0044] The catwalk PLC1 is arranged inside the catwalk body. One end of the catwalk PLC1 is connected to the ultrasonic sensor 2 through a signal line, and the other end is connected to the lever detector 3 through a signal line. The catwalk PLC is connected to the central industrial controller 4 through a signal line. The input end of the central industrial controller 4 is connected to the industrial color camera 5, and the output end of the central industrial controller 4 is connected to the industrial monitoring all-in-one machine 6; the industrial color camera 5 includes a control unit 501 and an acquisition unit 502, and the central industrial controller 4 is signal-connected to the control unit 501; the catwalk PLC1 is connected to the central industrial controller 4 and the control unit 501 through Ethernet to form a local area network.

[0045] The central industrial controller 4 serves as the integrated controller of the entire system and undertakes the functions of reliable data fusion and processing. It can be any one of a programmable controller PLC, an embedded industrial computer or a PXI controller.

[0046] The industrial monitoring all-in-one machine 6 serves as the system human-machine interface terminal, and can complete remote control and status monitoring of the power cat; the industrial monitoring all-in-one machine 6 can complete the parameter setting function of the industrial color camera 5, set its exposure time, gain, sampling mode (continuous acquisition / single acquisition), and can also complete the parameter setting function of the image algorithm embedded in the industrial color camera 5, such as the geometric setting of the ROI rectangular analysis area, the image grayscale threshold setting, etc.

[0047] The present invention provides a method for detecting the inclination of a power catwalk loading string based on machine vision, such as Figure 4 As shown, please follow the steps below:

[0048] Step 1: Arrange the device for detecting the inclination of the power catwalk loading string in the actual drilling rig structure and initialize the catwalk PLC1, industrial color camera 5 and central industrial controller 4;

[0049] like Figure 2 As shown, the central industrial controller 4 and the industrial monitoring integrated machine 6 are placed in the driller's room, the industrial color camera 5 is placed on the guardrail of the drill floor, the ultrasonic sensor 2 is placed in the middle of the catwalk V-groove, and the lever detector 3 is placed at the end of the catwalk V-groove;

[0050] The initialization of the industrial color camera 5 in step 1 includes initialization of the parameter settings of exposure time, gain, and sampling mode, and also supports modification in the industrial monitoring integrated machine 6. The initialization of the industrial color camera 5 also includes setting the ROI rectangular analysis area of ​​the image to be limited to the area before and after the V-groove of the horizontal catwalk when the power catwalk is at its lowest point on the ground.

[0051] Step 2: The ultrasonic sensor 2 and the lever detector 3 detect whether the signal is blocked by the pipe string and send it to the industrial monitoring integrated machine 6 through the catwalk PLC 1 and the central industrial controller 4 in sequence; if so, the industrial monitoring integrated machine 6 receives the signal of pipe string blocking and displays "Sensor judgment: pipe string loaded in place"; if not, the industrial monitoring integrated machine 6 receives the signal of pipe string not blocking and displays "Sensor judgment: pipe string not loaded in place";

[0052] Step 2.1: The catwalk PLC program executes the pipe string kick-in command. The catwalk pipe rack rises, and the pipe string rolls onto the tilting mechanism. Then, all the tilting structures on the same side are lifted. After the pipe string rolls over the catwalk support end surface, it rolls into the V-groove. The ultrasonic sensor 2 and lever detector 3 detect whether the pipe string is blocking the signal. At this time, regardless of whether a signal is detected, the industrial color camera 5 is triggered to capture an image according to the initialized set parameters. The purpose is to complement and verify each other with the two detection methods.

[0053] Step 2.2: When the ultrasonic sensor 2 and the lever detector 3 detect a pipe obstruction signal, the signal is sent to the industrial monitoring integrated machine 6 through the catwalk PLC 1 and the central industrial controller 4. The industrial monitoring integrated machine 6 receives the pipe obstruction signal and displays "Sensor judgment: pipe loaded in place";

[0054] When the ultrasonic sensor 2 and the lever detector 3 do not detect the signal of pipe column obstruction or only detect one of them, they are sent to the industrial monitoring integrated machine 6 through the catwalk PLC 1 and the central industrial controller 4 in sequence. The industrial monitoring integrated machine 6 receives the signal of pipe column unobstruction and displays "Sensor judgment: pipe column is not loaded in place";

[0055] Step 3, the acquisition unit (502) acquires and extracts the catwalk V-groove ROI area image frame by frame;

[0056] The industrial color camera 5 extracts the front and rear end areas of the catwalk V-groove of the captured image according to the initialized ROI area setting.

[0057] Although the camera captures a wide range, the rectangular analysis area within the image is limited to the V-groove. In conventional implementations, the entire surface of the catwalk's V-groove is yellow, while the outer cylindrical surface is black. This ROI restriction reduces data processing and facilitates subsequent algorithm design by leveraging the significant color contrast.

[0058] Step 4: The control unit (501) runs an image analysis algorithm based on the catwalk V-groove ROI area image and determines whether the pipe string is completely loaded into the catwalk V-groove; the geometric graphics of the ROI area at the algorithm processing level, when viewed by a color camera, are divided into three colors from top to bottom: yellow-black-yellow.

[0059] If so, a signal indicating that the pipe string has been successfully loaded into the catwalk V-groove is sent via the central industrial controller 4 to the industrial monitoring integrated machine 6, which displays "Machine vision interpretation: pipe string successfully loaded into the catwalk".

[0060] If not, a signal indicating that the pipe string has failed to be loaded into the catwalk V-groove is sent through the central industrial controller 4 to the industrial monitoring all-in-one machine 6, which displays "Machine vision interpretation: Failed to load the pipe string into the catwalk".

[0061] Step 4.1, the control unit (501) performs filtering and noise reduction, histogram equalization processing, binarization processing, morphological operation, and edge extraction on the ROI detection area to obtain the contour surface of the catwalk V-groove and the pipe column;

[0062] Step 4.2: The algorithm sets the center area of ​​the couplings at both ends of the string and selects three detection rectangular areas with a uniform interval length ΔL, defined as the near-end detection point, the mid-end detection point, and the far-end detection point. ΔL is approximately equal to 1 / 4 of the string length.

[0063] like Figure 5 As shown, taking the near-end detection point as an example, the intersection points of the detection rectangle frame and the upper contour surface of the V-shaped groove and catwalk are A, B, C, and D, and the areas of quadrilaterals A, B, C, and D are calculated; the intersection points of the lower contour surface are A', B', C', and D', and the area S1' of quadrilaterals A', B', C', and D' is calculated. Similarly, the mid-end detection point corresponds to the areas S2 and S2' of the quadrilaterals; the far-end detection point corresponds to the areas S3 and S3' of the quadrilaterals.

[0064] Step 4.3: Let K1 = S1 / S1′, K2 = S2 / S2′, and K3 = S3 / S3′, where K1, K2, and K3 are the area ratios of the three rectangular borders. If K1, K2, and K3 are all within the program-defined reasonable threshold of 0.95-1, the program deems that the three detection points S1 = S1′, S2 = S2′, and S3 = S3′ have all passed. Based on two points defining a straight line and three points defining a plane, the machine vision image analysis algorithm determines:

[0065] like Figure 6 As shown, the pipe string is successfully loaded into the catwalk V-groove, and a signal indicating the successful loading of the pipe string into the catwalk V-groove is sent to the industrial monitoring integrated machine 6 via the central industrial controller 4, and the industrial monitoring integrated machine 6 displays "Machine vision interpretation: pipe string successfully loaded into the catwalk";

[0066] Step 4.4: If the three test points S1=S1′, S2=S2′, and S3=S3′ fail, Figure 7 As shown, the program determines that the pipe string has not been successfully loaded into the catwalk V-groove, and sends a signal of failure to load the pipe string into the catwalk V-groove through the central industrial controller 4 and then to the industrial monitoring all-in-one machine 6, and the industrial monitoring all-in-one machine 6 displays "Machine vision interpretation: Failed to load the pipe string into the catwalk".

[0067] The advantage of the present invention's machine vision-based device and method for detecting the tilt of a pipe string loaded onto a power catwalk lies in the use of machine vision and image analysis technology to automatically and contactlessly detect whether the pipe string has been successfully transferred to the power catwalk. This system and method achieve a high degree of integration between the detection program and the equipment control program, providing a backup for traditional sensor detection methods, reducing false positives and improving the automation efficiency of the operational process.

Claims

1. A method for detecting the inclination of a power catwalk loading string based on machine vision, characterized in that: The method uses a device for detecting the inclination of a power catwalk loading string based on machine vision, the device comprising a catwalk PLC (1) arranged inside a catwalk body, one end of the catwalk PLC (1) being connected to an ultrasonic sensor (2) via a signal line, and the other end being connected to a lever detector (3) via a signal line, the catwalk PLC (1) being further connected to a central industrial controller (4) via a signal line, the input end of the central industrial controller (4) being connected to an industrial color camera (5), and the output end of the central industrial controller (4) being connected to an industrial monitoring all-in-one machine (6); The industrial color camera (5) includes an interconnected control unit (501) and an acquisition unit (502), and the central industrial controller (4) is connected to the control unit (501) by signal; the catwalk PLC (1), the central industrial controller (4) and the control unit (501) are connected via Ethernet to form a local area network; The specific steps of this method are as follows: Step 1: Arrange the device for detecting the inclination of the power catwalk loading string in the actual drilling rig structure and initialize the program of the catwalk PLC (1), the industrial color camera (5) and the central industrial controller (4); Step 2: The ultrasonic sensor (2) and the lever detector (3) detect whether the signal is blocked by the pipe column and send it to the industrial monitoring integrated machine (6) through the catwalk PLC (1) and the central industrial controller (4) in sequence; If so, the industrial monitoring integrated machine (6) receives the signal of the pipe column being blocked and displays "Sensor judgment: pipe column loaded in place"; If not, the industrial monitoring integrated machine (6) receives a signal indicating that the pipe string is not blocked and displays "Sensor judgment: the pipe string is not loaded in place"; Step 3, the acquisition unit (502) acquires and extracts the catwalk V-groove ROI area image frame by frame and sends it to the control unit (501); Step 4, the control unit (501) runs an image analysis algorithm based on the catwalk V-groove ROI area image, and determines whether the tubing string is completely loaded into the catwalk V-groove; If so, a signal indicating that the pipe string has been successfully loaded into the catwalk V-groove is sent to the industrial monitoring all-in-one machine (6) via the central industrial controller (4), and the industrial monitoring all-in-one machine (6) displays "Machine vision interpretation: pipe string has been successfully loaded into the catwalk"; If not, a signal indicating that the pipe string has failed to be loaded into the catwalk V-groove is sent via the central industrial controller (4) to the industrial monitoring all-in-one machine (6), which displays "Machine vision interpretation: Failed to load the pipe string into the catwalk".

2. The method for detecting the inclination of a power catwalk loading string based on machine vision according to claim 1, characterized in that: The central industrial controller (4) is any one of a programmable controller PLC, an embedded industrial computer or a PXI controller.

3. The method for detecting the inclination of a power catwalk loading string according to claim 1, characterized in that: The device for detecting the inclination of the power catwalk loading string in step 1 is arranged as follows: a central industrial controller (4) and an industrial monitoring all-in-one machine (6) are placed in the driller's room during actual drilling; the industrial color camera (5) is placed on the guardrail of the drilling table and is aimed at the end face of the catwalk V-groove; the ultrasonic sensor (2) is placed in the middle of the catwalk V-groove; and the lever detector (3) is placed at the end of the catwalk V-groove.

4. The method for detecting the inclination of a power catwalk loading string according to claim 3, characterized in that: The initialization of the industrial color camera (5) in step 1 includes initialization of the parameter settings of exposure time, gain, and sampling mode, and also supports modification in the industrial monitoring all-in-one machine (6). The initialization of the industrial color camera (5) also includes setting the ROI rectangular analysis area of ​​the image to be limited to the front and rear areas of the horizontal catwalk V-groove when the power catwalk is at its lowest point on the ground.

5. The method for detecting the inclination of a power catwalk loading string according to claim 1, characterized in that: The step 2 is specifically implemented according to the following steps: Step 2.1, the catwalk PLC program executes the pipe string kick-in command, the catwalk pipe rack rises, the pipe string rolls onto the tilting mechanism, and then all the tilting structures on the same side are lifted. After the pipe string rolls over the catwalk support end face, it rolls into the V-shaped groove. The ultrasonic sensor (2) and the lever detector (3) detect whether it is blocked by the closed pipe string; Step 2.2: When the ultrasonic sensor (2) and the lever detector (3) detect a signal indicating that the pipe is blocked, the signal is sent to the industrial monitoring integrated machine (6) via the catwalk PLC (1) and the central industrial controller (4). The industrial monitoring integrated machine (6) receives the signal indicating that the pipe is blocked and displays "Sensor judgment: pipe is loaded in place". When the ultrasonic sensor (2) and the lever detector (3) do not detect a signal indicating that the pipe is blocked or only detect one of them, the signal is sent to the industrial monitoring integrated machine (6) via the catwalk PLC (1) and the central industrial controller (4) in sequence. The industrial monitoring integrated machine (6) receives the signal indicating that the pipe is not blocked and displays "Sensor judgment: the pipe is not loaded in place".

6. The method for detecting the inclination of a power catwalk loading string according to claim 1, characterized in that: In step 3, the industrial color camera (5) extracts the front and rear end areas of the catwalk V-groove of the captured image according to the initialized ROI area setting.

7. The method for detecting the inclination of a power catwalk loading string according to claim 1, characterized in that: The step 4 is specifically implemented according to the following steps: Step 4.1, the control unit (501) performs filtering and noise reduction, histogram equalization processing, binarization processing, morphological operation, and edge extraction on the ROI detection area to obtain the contour surface of the catwalk V-groove and the pipe column; Step 4.2: The algorithm sets the middle area of ​​the couplings at both ends of the pipe string and selects three detection rectangular areas with a uniform interval length ΔL, which are defined as: the proximal detection point, the mid-end detection point, and the distal detection point. ΔL is equal to 1 / 4 of the pipe string length. Calculate the areas S1, S1′, S2, S2′, S3, S3′ of the quadrilaterals formed by the intersection of the rectangular frames of the three detection points and the upper and lower contour frames of the V-shaped groove and the catwalk respectively; Step 4.3, let K1 = S1 / S1′, K2 = S2 / S2′, K3 = S3 / S3′, where K1, K2, and K3 are the area ratios of the three rectangular frames. If K1, K2, and K3 are all within the reasonable threshold period 0.95-1 set by the program, then the program considers that the three detection points S1 = S1′, S2 = S2′, and S3 = S3′ have all passed the detection. A straight line is determined based on two points, and a plane is determined based on three points. The image analysis algorithm of the machine vision determines that the pipe string is successfully loaded into the catwalk V-groove. The signal of the successful loading of the pipe string into the catwalk V-groove is sent to the industrial monitoring integrated machine (6) through the central industrial controller (4). The industrial monitoring integrated machine (6) displays "Machine vision interpretation: the pipe string is successfully loaded into the catwalk". Step 4.4: If the three detection points S1=S1′, S2=S2′, and S3=S3′ fail to pass, the program determines that the pipe string has not been successfully loaded into the catwalk V-groove, and a signal indicating that the pipe string has failed to be loaded into the catwalk V-groove is sent to the industrial monitoring integrated machine (6) through the central industrial controller (4). The industrial monitoring integrated machine (6) displays "Machine vision interpretation: pipe string failed to be loaded into the catwalk".

Images