System for monitoring settlement and eccentricity of enclosed bus conductor based on machine vision

Abstract

The invention provides a system for monitoring settlement and eccentricity of an enclosed bus conductor based on machine vision. The bus conductor is arranged in an enclosed bus shell; comprising an industrial camera, a visual identifier and an image processing system, the number of the visual identifications is two, and the visual identifications are symmetrically arranged on the outer surface of the bus conductor. When the bus conductor is in an initial state, a connecting line between the two visual identifications is parallel to the axis of the bus conductor; the industrial camera is arranged in the bus shell and is used for respectively acquiring images of the visual identification and transmitting the images to the image processing system; and the image processing system is used for identifying the coordinates of the visual identification according to the image of the visual identification and identifying the relative displacement and the space offset of the bus conductor according to the coordinate change of the visual identification. According to the invention, machine vision and state monitoring of the state of the conductor in the totally-enclosed metal shell can be realized, and the safety and reliability of the key leading-out equipment of the generator are greatly improved.

Description

Technical Field

[0001] The invention belongs to the technical field of electromechanical engineering, and in particular relates to a system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision. Background Art

[0002] The phase-isolated enclosed busbar is a large current lead-out device for the generator in the power station. It is generally connected in a metal shell, and the internal conductor is supported on the shell by supporting insulators. Since the enclosed busbar adopts a metal shell full connection, its internal conductor is partially invisible. At present, the monitoring of large current enclosed busbar equipment uses RTD or infrared temperature measurement on the shell. Due to the electrical insulation and full enclosure of the internal conductor, there is almost no temperature or deformation measurement device. Therefore, once the enclosed busbar is put into operation, the temperature rise of the internal conductor can only be obtained indirectly through the measurement of the shell, and the possible dangerous conditions inside, such as the falling off of the insulator support hardware, the aging and deterioration of the insulator surface, and the partial failure of the load-bearing support structure due to long-term operation and low-frequency vibration, cannot be monitored. Since the closed busbar is the transmission channel for all the electrical energy generated by the generator, it has the characteristics of large current, high heat generation, many welding connections, and many internal supporting insulators. Therefore, under long-term operation conditions, the conductor will inevitably have settlement and concentricity deviation. When the situation accumulates to a certain extent, it is very easy to cause a single-phase short circuit in the internal conductor, or even expand to a serious three-phase short circuit, which seriously threatens the operation and safety of the generator and main transformer.

[0003] The prior art lacks relevant equipment and methods for monitoring the settlement and concentricity deviation of busbar conductors. Summary of the invention

[0004] The purpose of the present invention is to address the deficiencies of the above-mentioned background technology and to provide a system for monitoring the settlement and eccentricity of closed bus conductors based on machine vision, which can realize machine vision and status monitoring of the conductor status inside a fully enclosed metal casing, thereby changing the current situation where there is no long-term monitoring of the conductor inside the closed bus, and greatly improving the safety and reliability of the key lead-out equipment of the generator.

[0005] The technical solution adopted by the present invention is: a system for monitoring the settlement and eccentricity of a closed bus conductor based on machine vision, wherein the bus conductor is arranged inside a closed bus housing; comprising an industrial camera, a visual identification and an image processing system;

[0006] There are two visual marks, which are symmetrically arranged on the outer surface of the bus conductor; when the bus conductor is in the initial state, the line between the two visual marks is parallel to the axis of the bus conductor;

[0007] The industrial camera is arranged inside the busbar housing and is used to respectively acquire images of the visual identification and transmit them to the image processing system;

[0008] The image processing system is used to identify the coordinates of the visual marker according to the image of the visual marker, and to identify the relative displacement and spatial offset of the bus conductor according to the change of the coordinates of the visual marker.

[0009] In the above technical solution, the industrial camera is fixed to the busbar housing via a camera base.

[0010] In the above technical solution, a support insulator for supporting the bus conductor is also provided inside the busbar housing; and the camera base is sleeved on the outside of the support insulator.

[0011] In the above technical solution, a limiting structure is provided on the end of the supporting insulator close to the busbar housing; and the camera base is clamped between the limiting structure and the inner surface of the busbar housing.

[0012] In the above technical solution, the camera base is a reverse arc disk structure made of elastic material, and its concave surface faces the busbar conductor.

[0013] In the above technical solution, when the busbar conductor is in the initial state, the image processing system calibrates the binocular camera and establishes a visual three-dimensional coordinate system, and then determines the initial midpoint coordinates of the line segment connecting the two visual markers based on the images of the two visual markers acquired by the binocular camera;

[0014] The image processing system calculates the current midpoint coordinates of the line segment connecting the two visual markers based on the images of the two visual markers acquired in real time by the binocular camera and subtracts them from the initial midpoint coordinates to obtain the spatial offset of the busbar conductor.

[0015] In the above technical solution, the image processing system calculates the current coordinates of the two visual markers based on the images of the two visual markers acquired in real time by the industrial camera; and calculates the current axis deviation of the bus conductor based on the current coordinates of the two visual markers.

[0016] In the above technical solution, the image processing system calculates the real-time relative displacement of the busbar conductor in the axial direction according to the current axial deviation of the busbar conductor.

[0017] In the above technical solution, the image processing system calculates the real-time unit time variation of the busbar conductor in the axial direction according to the real-time relative displacement of the busbar conductor in the axial direction.

[0018] In the above technical solution, if the image processing system determines that the real-time unit time variation of the bus conductor in the axial direction exceeds the set threshold, it outputs an alarm signal to the display device; the display device issues an alarm prompt upon receiving the alarm signal.

[0019] The beneficial effects of the present invention are as follows: the present invention proposes a method based on machine vision and high-precision identification inside a closed busbar, and obtains the settlement and eccentricity parameters of the conductor relative to the outer shell by identifying the special mark on the conductor through a machine vision system arranged inside the busbar. This method can effectively detect dangerous conditions such as settlement of the busbar conductor, asymmetric support force, and centerline offset eccentricity between the conductor and the outer shell, and obtains the long-term state change trend of the conductor inside the busbar relative to the outer shell through online machine vision identification, providing early warning of dangerous stress conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a structural schematic diagram a of the present invention;

[0021] Figure 2 It is a structural schematic diagram b of the present invention;

[0022] Figure 3 It is a structural schematic diagram c of the present invention;

[0023] Figure 4 Schematic diagram a of the calculation principle diagram of the present invention;

[0024] Figure 5 Schematic diagram b of the calculation principle diagram of the present invention;

[0025] Figure 6 Schematic diagram c of the calculation principle diagram of the present invention.

[0026] Among them, 1-busbar conductor, 2-busbar casing, 3-support insulator, 4-industrial camera, 5-camera base, 6-flange, 7-casing support structure, 8-visual identification, 9-limiting structure. DETAILED DESCRIPTION

[0027] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments to facilitate a clear understanding of the present invention, but they do not constitute a limitation on the present invention.

[0028] like Figure 1-3 As shown, the present invention provides a system for monitoring the settlement and eccentricity of a closed bus conductor based on machine vision, wherein the bus conductor 1 is arranged inside a closed bus housing 2; it includes an industrial camera 4, a visual marker 8 and an image processing system;

[0029] There are two visual marks 8, which are symmetrically arranged on the outer surface of the bus conductor 1; when the bus conductor 1 is in the initial state, the line between the two visual marks 8 is parallel to the axis of the bus conductor 1;

[0030] The industrial camera 4 is arranged inside the busbar housing 2, and is used to respectively acquire images of the visual markers 8 and transmit them to the image processing system;

[0031] The image processing system is used to identify the coordinates of the visual marker 8 according to the image of the visual marker 8 , and to identify the relative displacement and spatial offset of the bus conductor 1 according to the change of the coordinates of the visual marker 8 .

[0032] This specific embodiment takes the status monitoring of the internal conductor of a 20kV, 26kA (700MW class) isolated phase enclosed busbar in a power station as an example. The outer diameter of the enclosed busbar housing 2 of this specific embodiment is 1600mm, and the outer diameter of the conductor is 970mm.

[0033] The busbar housing 2 is a circular metal housing of a fully connected enclosed busbar, providing support points and electromagnetic shielding protection for the internal busbar conductor 1. The busbar housing 2 is fixed by a housing support structure 7.

[0034] The busbar conductor 1 is the current-conducting part inside the closed busbar. It is a circular metal conductor with high voltage (generator voltage 10kV~26kV / √3) and large current (10kA~28kA). It is concentric with the busbar casing 2 and supported on the casing by supporting insulators 3 (3~4) arranged symmetrically along the circumference.

[0035] The supporting insulator 3 is an insulating and supporting element between the busbar housing 2 and the busbar conductor 1, and can be an insulator made of electric porcelain or epoxy cast. It is connected to the busbar housing 2 by a flange 6 base, and is connected (supported) to the busbar conductor 1 by a hardware fitting (such as a hemispherical or mushroom head type).

[0036] The industrial camera 4 uses a composite camera (monocular or multi-eye industrial camera 4) with its own light source, which can be an area array camera or a linear array camera, and the light source and the camera (industrial camera 4) are integrated. The industrial camera 4 is a monocular or multi-eye camera with high resolution and low frame rate, and is a black and white or color imaging camera according to the target of monitoring. This specific embodiment uses a symmetrical arrangement of binocular industrial cameras 4. A high-resolution, high-precision area array camera with a resolution of 1920 pixels × 1080 pixels is used to identify and measure the special markings on the conductor in conjunction with the special light source provided by the equipment.

[0037] The visual mark 8 adopts a special machine vision recognition pattern, and two visual marks 8 are symmetrically arranged above and below, with a spacing of 50cm to 60cm. This pattern enables the machine vision to accurately identify the ± displacement of the conductor in the X-axis and Y-axis directions, thereby identifying and monitoring the settlement and eccentricity of the conductor.

[0038] Specifically, the industrial camera 4 is fixed to the busbar housing 2 via the camera base 5, which can ensure that the industrial camera 4 body does not move, thereby accurately acquiring the image of the busbar displacement. The camera head of the industrial camera 4 passes through the busbar housing 2 and faces the visual mark 8 to effectively capture its changing image.

[0039] The image processing system is equipped with a machine recognition and AI algorithm processing unit for the bus conductor 1, which can perform high-precision recognition of the special identification pattern of the bus conductor 1, obtain the coordinate position of the visual identification 8 after applying the image enhancement, edge and contour pixel-level recognition algorithm, further identify the relative displacement and change trend (change per unit time) of the visual identification 8, and issue an early warning based on the set threshold.

[0040] Specifically, a support insulator 3 for supporting the bus conductor 1 is also provided inside the busbar housing 2; the camera base 5 is sleeved outside the support insulator 3. The present invention rationalizes the layout, fixes the industrial camera 4 without affecting the internal arrangement of the busbar housing 2, and ensures the normal operation of the busbar.

[0041] The industrial camera comprises two groups, which are symmetrically arranged on the camera bases of the supporting insulators located on the left and right sides of the bus conductor 1. The visual markers are distributed along the radial direction of the bus conductor.

[0042] Through two groups of industrial cameras, images of visual markers that are at the same radial position on the bus conductor and are facing the industrial cameras are acquired simultaneously; then, the coordinates of visual markers that are located at two different radial positions on the bus conductor can be identified based on the visual marker images acquired by the two groups of cameras, that is, the image processing system obtains the coordinates of the visual markers based on the visual marker images at the same radial position acquired by the two groups of industrial cameras, thereby balancing and eliminating the recognition errors between multiple spatial visual coordinates and the errors of the cameras themselves.

[0043] Specifically, a limiting structure 9 is provided on the end of the supporting insulator 3 close to the busbar housing 2; and the camera base 5 is clamped between the limiting structure 9 and the inner surface of the busbar housing 2. The present invention clamps the camera base 5 by the limiting structure 9 and the busbar housing 2, and does not need to add a fixing point on the busbar housing 2, which is beneficial to the strength of the busbar housing 2, and effectively provides a deformation space for the camera base 5, thereby ensuring the overall safety of the device.

[0044] Specifically, the camera base 5 is a reverse arc disk structure made of non-metallic elastic material, and its concave surface faces the bus conductor 1. A limiting structure 9 is provided at a certain height position (5 cm height) of the supporting insulator 3. The present invention generates extrusion between the limiting structure 9 and the bus housing 2, and generates sufficient pre-tightening force through its reverse arc disk structure (which can be metal or non-metallic elastic material) to firmly compact and fit the camera base 5 to the inner surface of the housing, thereby ensuring the installation stability of the industrial camera 4.

[0045] like Figure 4-6As shown, when the bus conductor 1 is in the initial state, the image processing system calibrates the binocular camera and establishes a visual three-dimensional coordinate system, and then determines the initial midpoint coordinates O0 (x, y, z) of the line segment connecting the two visual markers 8 (as the initial reference axis) according to the images of the two visual markers 8 acquired by the binocular camera; and then establishes the axis fitting straight line equation Y0 of the bus conductor 1:

[0046] Y0=k 0x X0+a0; (projection of the initial reference axis on the XY plane)

[0047] Y0=k 0Z Z0+b0; (projection of the initial reference axis on the ZY plane)

[0048] O0(x, y, z) = 1 / 2·(X 0上 +X 0下 , Y 0上 +Y 0下 , Z 0上 +Z 0下 )

[0049] (Coordinates of the midpoint of the initial fitting straight line segment)

[0050] Among them, the initial coordinates of the upper mark point are (X 0上 , Y 0上 , Z 0上 ), the initial coordinates of the lower mark point are (X 0下 , Y 0下 , Z 0下 )

[0051] In this specific embodiment, when the initial state of non-settlement or offset occurs, k 0x =k 0Z =0, which is the plumb line (vertical arrangement) state; through camera calibration, it is measured that a0=b0=1000mm; O0(x, y, z)=O0(1000, 1000, 1000) (unit: mm).

[0052] The image processing system calculates the current midpoint coordinates of the line segment connecting the two visual identifiers 8 based on the images of the two visual identifiers 8 acquired in real time by the binocular camera, and subtracts the current midpoint coordinates from the initial midpoint coordinates to obtain the spatial offset of the busbar conductor 1. The present invention uses the midpoint coordinate change of the line segment connecting the two visual identifiers 8 as the spatial offset of the busbar conductor 1, and through the symmetrical binocular camera in space visual identifier acquisition and axis fitting straight line segment midpoint comparison method, it can balance and eliminate the recognition errors between multiple spatial visual coordinates and the camera's own errors, thereby effectively ensuring the recognition accuracy.

[0053] Identify the spatial coordinates of the upper and lower marking points after the busbar conductor 1 sinks or deflects, which are (X 上, Y 上 , Z 上 )=(1530,1450,1550),(X 下 , Y 下 , Z 下 )=(490,460,550), the straight line segment equation Y after fitting offset and the midpoint of the line segment O'(x,y,z)=(1010,955,1050):

[0054] Y=k x '·X+a'; (projection of the fitting axis on the XY plane after offset)

[0055] Y=k Z '·Z+b'; (projection of the fitting axis on the ZY plane after offset)

[0056] O'(x, y, z) = 1 / 2·(X 上 +X 下 , Y 上 +Y 下 , Z 上 +Z 下 )

[0057] (Coordinates of the midpoint of the fitted straight line segment after offset)

[0058] The spatial offset is obtained by comparing O0(x, y, z) and O'(x, y, z):

[0059] △O(x,y,z)=O'(x,y,z)-O0(x,y,z)=(-10,-45,+50)(mm).

[0060] Specifically, the image processing system calculates the current coordinates of the two visual markers 8 based on the images of the two visual markers 8 acquired in real time by the binocular camera, and calculates the current axis deviation of the bus conductor 1 based on the current coordinates of the two visual markers 8. The above method of comparing the three-dimensional spatial coordinate variables of the two line segments fitted by the acquired spatial coordinates and the weighted average midpoint utilizes the symmetrical arrangement characteristics of the binocular camera to balance the recognition error between the symmetrical spatial visual coordinates and the error of the camera itself, thereby effectively ensuring the recognition accuracy and efficiency.

[0061] By the upper and lower marking point spatial coordinates (X 上 , Y 上 , Z 上 )(X 下 , Y 下 , Z 下 ) is substituted into the offset straight line segment equation Y, and k can be solved. x ' and k Z '.

[0062] kx ' describes the degree of displacement of the busbar conductor 1 along the X-axis direction, which reflects the proportion of the change in the Y coordinate when the busbar conductor 1 is displaced laterally (along the X-axis). This proportionality coefficient can be regarded as the axial displacement of the conductor in the X-axis direction.

[0063] k Z ' describes the degree of displacement of the busbar conductor 1 along the Z axis, which reflects the proportion of the change in the Y coordinate when the conductor is longitudinally displaced (along the Z axis). This proportionality coefficient can be regarded as the axis displacement of the busbar conductor 1 in the Z axis direction.

[0064] Specifically, the image processing system calculates the real-time relative displacement of the bus conductor 1 in the axial direction according to the current axial deviation of the bus conductor 1, and utilizes the above-mentioned camera symmetrical arrangement, line segment fitting, and midpoint averaging methods to effectively ensure recognition accuracy and efficiency.

[0065] Preferably, by obtaining k x ', k Z ' and the spatial coordinates of the upper and lower marking points (X 上 , Y 上 , Z 上 )(X 下 , Y 下 , Z 下 ) is substituted into the equation of the offset straight line segment Y, and a' and b' can be solved, that is, the Y value of the projection of the offset busbar axis at any point between the two visual identification points 8 in the XY plane and the ZY plane. By comparing with Y0, the relative displacement of the busbar conductor 1 can be identified.

[0066] Specifically, the image processing system calculates the real-time unit time variation of the bus conductor 1 in the axial direction according to the real-time relative displacement of the bus conductor 1 in the axial direction. If the image processing system determines that the real-time unit time variation of the bus conductor 1 in the axial direction exceeds the set threshold, the image processing system outputs an alarm signal to the display device; the display device issues an alarm prompt after receiving the alarm signal.

[0067] Preferably, the display device can issue a sound prompt or a text prompt.

[0068] The present invention also includes a local power supply and signal control box for providing safe voltage power supply and required signal data acquisition and control for the industrial camera 4, display device and image processing system. A dedicated communication interface is used between the industrial camera 4 and the image processing system of the present invention to realize the transmission of machine vision image data.

[0069] The industrial camera 4 is equipped with an image acquisition card for image acquisition and preprocessing of the industrial camera lens. The data interface determines the transmission bandwidth according to the resolution and frame rate. The USB3.0, CameraLink or GigE interface can be selected in combination with the transmission distance. The USB3.0 interface is selected in this specific embodiment.

[0070] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

Claims

1. A system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision, wherein the busbar conductor is arranged inside a closed busbar housing; characterized in that: Including industrial cameras, visual identification and image processing systems; There are two visual marks, which are symmetrically arranged on the outer surface of the bus conductor; when the bus conductor is in the initial state, the line between the two visual marks is parallel to the axis of the bus conductor; The industrial camera is arranged inside the busbar housing and is used to respectively acquire images of the visual identification and transmit them to the image processing system; The image processing system is used to identify the coordinates of the visual marker according to the image of the visual marker, and to identify the relative displacement and spatial offset of the bus conductor according to the change of the coordinates of the visual marker.

2. A system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 1, characterized in that: The industrial camera is fixed to the busbar housing via a camera base.

3. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 2, characterized in that: A supporting insulator for supporting the bus conductor is also arranged inside the busbar shell; the camera base is sleeved on the outside of the supporting insulator.

4. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 3 is characterized in that: A limiting structure is arranged on the end of the supporting insulator close to the busbar housing; and the camera base is clamped between the limiting structure and the inner surface of the busbar housing.

5. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 4, characterized in that: The camera base is a reverse arc disk structure made of elastic material, and its concave surface faces the busbar conductor.

6. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 4, characterized in that: When the busbar conductor is in the initial state, the image processing system calibrates the binocular camera and establishes a visual three-dimensional coordinate system, and then determines the initial midpoint coordinates of the line segment connecting the two visual markers based on the images of the two visual markers acquired by the binocular camera; The image processing system calculates the current midpoint coordinates of the line segment connecting the two visual markers based on the images of the two visual markers acquired in real time by the binocular camera and subtracts them from the initial midpoint coordinates to obtain the spatial offset of the busbar conductor.

7. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 6, characterized in that: The image processing system calculates the current coordinates of the two visual markers based on the images of the two visual markers acquired in real time by the industrial camera; and calculates the current axis deviation of the bus conductor based on the current coordinates of the two visual markers.

8. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 7, characterized in that: The image processing system calculates the real-time relative displacement of the bus conductor in the axial direction according to the current axial deviation of the bus conductor.

9. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 8, characterized in that: The image processing system calculates the real-time unit time variation of the bus conductor in the axial direction according to the real-time relative displacement of the bus conductor in the axial direction.

10. The system for monitoring the settlement and eccentricity of a closed busbar conductor based on machine vision according to claim 8, characterized in that: If the image processing system determines that the real-time unit time variation of the bus conductor in the axial direction exceeds the set threshold, an alarm signal is output to the display device; the display device issues an alarm prompt upon receiving the alarm signal.

Images