An on-line monitoring device for a close contact checker

Abstract

The present application relates to the field of turnout close inspection parameter detection, and discloses a close inspection online monitoring device, comprising: an image acquisition device located in the close inspection device, used for real-time acquisition of image signals of the target to be measured in the close inspection device; a fixed support located in the close inspection device, used for fixing the image acquisition device; a power transmission line for powering the image acquisition device and real-time transmission of the image signals of the target to be measured; an industrial computer platform for acquiring the image signals of the target to be measured and inputting into the pre-trained target monitoring neural network model, and outputting the monitoring data of the target to be measured. The above device can simply and conveniently realize real-time monitoring of the target in the close inspection device through the interaction of the image acquisition device, the fixed support, the power transmission line and the industrial computer platform, has high detection accuracy, simple structure and low cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of turnout tightness inspector parameter detection, in particular to a tightness inspector online monitoring device. BACKGROUND

[0002] With the development of railways, especially the development of high-speed railway transportation, the safety requirements for transportation are becoming higher and higher. The tightness inspector is used to check the tightness and repulsion state between the turnout traction points. The tightness inspector is used as a redundant device for turnout indication and is connected in series in the turnout indication system, which can improve the monitoring reliability of the turnout accident and the monitoring of the turnout fault. However, with the frequent operation of the turnout, due to the influence of the roadbed, the steel rail and other conditions, the tightness inspector will appear various fault phenomena, and many of them are unexplained faults, which are difficult to eliminate at the moment and seriously affect the train operation efficiency.

[0003] The existing tightness inspector can only check the tightness and repulsion state through on-site inspection by maintenance personnel. This operation and maintenance mode is low in efficiency, high in maintenance cost, and difficult to meet the demand of railway operation and maintenance system for improving safety guarantee ability and reducing operation cost.

[0004] Therefore, how to simply and conveniently realize real-time data monitoring of the tightness inspector is a technical problem to be solved by the person skilled in the art. SUMMARY

[0005] Therefore, the purpose of the present application is to provide a tightness inspector online monitoring device which can simply and conveniently monitor the target inside the tightness inspector in real time, has high detection accuracy, simple structure and low cost. The specific scheme is as follows:

[0006] A tightness inspector online monitoring device comprises:

[0007] An image acquisition device is located in the tightness inspector and is used to acquire image signals of a target to be measured in the tightness inspector in real time.

[0008] A fixing support is located in the tightness inspector and is used to fix the image acquisition device.

[0009] A power supply transmission line is used to supply power to the image acquisition device and to transmit the image signals of the target to be measured in real time.

[0010] An industrial computer platform is used to acquire the image signals of the target to be measured and input them into a pre-trained target monitoring neural network model to output monitoring data of the target to be measured.

[0011] Preferably, in the above-mentioned tightness inspector online monitoring device provided by the embodiment of the present application, the image acquisition device comprises a four-in-one camera and a monocular camera.

[0012] The quad camera is arranged at a set distance above the dynamic round contact point and the static spring sheet contact point.

[0013] The monocular camera is arranged on the periphery of the notch contact point composed of a rod groove and a side pulley.

[0014] Preferably, in the above-mentioned close contact inspector online monitoring device provided by the embodiment of the application, the quad camera is configured with a light source.

[0015] Preferably, in the above-mentioned close contact inspector online monitoring device provided by the embodiment of the application, the fixed support includes a reversible Z-shaped bracket shaft, a first component arranged on the Z-shaped bracket shaft for embedding the quad camera, a hinge structure linked to the bottom end of the Z-shaped bracket shaft, a second component connected with the hinge structure and attached to the bottom of the close contact inspector, and a support frame arranged on the second component for fixing the monocular camera.

[0016] Preferably, in the above-mentioned close contact inspector online monitoring device provided by the embodiment of the application, the first component includes a transparent cover plate for carrying the quad camera, a groove for supplying wiring, and a through slot for adjusting the position of the transparent cover plate.

[0017] Preferably, in the above-mentioned close contact inspector online monitoring device provided by the embodiment of the application, the second component has a flat through hole at one end away from the Z-shaped bracket shaft.

[0018] The bolt inside the close contact inspector fixes the fixed support inside the close contact inspector through the flat through hole.

[0019] Preferably, in the above-mentioned close contact inspector online monitoring device provided by the embodiment of the application, the second component has a horizontal shaft at one end close to the Z-shaped bracket shaft; the horizontal shaft has a hole for fixing the monocular camera.

[0020] Preferably, in the above-mentioned close contact inspector online monitoring device provided by the embodiment of the application, it further includes:

[0021] The switch is used to transmit the image signal of the target to be tested transmitted by the power transmission line in real time to the server.

[0022] The server is used to transmit the received image signal of the target to be tested to the industrial computer platform.

[0023] Preferably, in the above-mentioned close contact inspector online monitoring device provided by the embodiment of the application, the industrial computer platform is specifically used to build and train YOLOv7-g nA Conv neural network model; input the acquired image signal of the target to be measured into the trained YOLOv7-g n In the Conv neural network model, the monitoring data of the target to be measured is output.

[0024] Preferably, in the above-mentioned online monitoring device for the close inspection device provided by the embodiment of the present application, further comprising:

[0025] An early warning device is configured to determine whether the close inspection device fails according to the monitoring data, and if so, an early warning prompt is performed.

[0026] As can be seen from the above technical solution, the online monitoring device for the close inspection device provided by the present application comprises: an image acquisition device located in the close inspection device, configured to acquire image signals of a target to be measured in the close inspection device in real time; a fixing support located in the close inspection device, configured to fix the image acquisition device; a power supply transmission line configured to supply power to the image acquisition device and transmit the image signals of the target to be measured in real time; and an industrial computer platform configured to acquire the image signals of the target to be measured and input the image signals into a pre-trained target monitoring neural network model to output monitoring data of the target to be measured.

[0027] The online monitoring device for the close inspection device provided by the present application can use machine vision instead of manual inspection, and can simply and conveniently monitor the target in the close inspection device in real time, has high detection accuracy, simple structure, and low cost, and solves the problem that the existing close inspection device can only be checked by manual on-site inspection, which is low in efficiency and high in cost. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the embodiment or related art description will be briefly introduced. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creative labor.

[0029] Figure 1 The structural schematic diagram of the online monitoring device for the close inspection device provided by the embodiment of the present application;

[0030] Figure 2 The structural schematic diagram of the image acquisition device provided by the embodiment of the present application;

[0031] Figure 3 The structural schematic diagram of the fixing support provided by the embodiment of the present application;

[0032] Figure 4A page display schematic diagram of the industrial computer platform provided by the embodiment of the present application is shown. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0034] The present application provides an online monitoring device for a close contact exaaminer, as shown in the drawings, comprising: Figure 1 An image acquisition device 1 is located in the close contact exaaminer and is used to acquire image signals of a target to be measured in the close contact exaaminer in real time.

[0035] A fixing support 2 is located in the close contact exaaminer and is used to fix the image acquisition device 1.

[0036] A power supply transmission line 3 is used to supply power to the image acquisition device 1 and to transmit the image signals of the target to be measured in real time.

[0037] An industrial computer platform 4 is used to acquire the image signals of the target to be measured and input them into a pre-trained target monitoring neural network model to output monitoring data of the target to be measured.

[0038] In the above online monitoring device for the close contact exaaminer provided by the embodiment of the present application, through the interaction of the image acquisition device, the fixing support, the power supply transmission line and the industrial computer platform, machine vision can be used to replace manual patrol, and the target inside the close contact exaaminer can be simply and conveniently monitored in real time, with high detection accuracy, simple structure, low cost and the problem of low efficiency and high cost of the existing close contact exaaminer which can only be checked by manual on-site inspection is solved.

[0039] It should be noted that the target to be measured can be a side gap composed of a dynamic static contact point group, a rod groove and a side pulley.

[0040] Further, in the specific implementation, in the above online monitoring device for the close contact exaaminer provided by the embodiment of the present application, the image acquisition device 1 can include a four-in-one camera and a monocular camera.

[0041] The four-in-one camera can be arranged at a set distance above the dynamic round contact point and the static spring sheet contact point.

[0042] The monocular camera can be arranged at the periphery of the gap joint composed of the rod groove and the side pulley.

[0043]

[0044] ​Specifically, the quad camera and the monocular camera can be responsible for real-time shooting of the motion state inside the close inspection device, especially shooting images of the dynamic circle contact point, the static spring contact point and the gap node. Preferably, the quad camera can be fixed at 17mm above the dynamic circle contact point and the static spring contact point and embedded in the fixed support 2. The monocular camera can be installed at 25mm in front of the gap composed of the rod groove and the side pulley, facilitating data collection. That is, the fixed support 2 is responsible for fixing the quad camera at 17mm above the dynamic circle contact point and the static spring contact point to be tested and fixing the monocular camera at 25mm in front of the gap to be tested.

[0045] It should be pointed out that the quad camera and the monocular camera can acquire data at the same time, use a network transmission channel together, perform long-distance data transmission, and receive data by the industrial computer platform 4, realizing cross-area monitoring from the monitoring room to the field track turnout.

[0046] In specific implementation, as shown in Figure 2 , the quad camera can be configured with a light source. That is, the quad camera of the application has a built-in illumination light source and does not need to add a redundant light supplement device. When the light source is triggered and the camera records data, it can be as close as possible to the working period of the close inspection device.

[0047] As shown in Figure 2 , the four cameras of the quad camera are integrated on one board, and a single line can realize data transmission of the quad camera, and the size fully meets the limited space inside the close inspection device which is narrow and dark. For example, the length A of the rectangular structure composed of the four cameras can be 50mm, the width B can be 30mm, the length C of the on-board can be 70mm, the width D can be 50mm, the total thickness E of the camera and the on-board can be 7.2mm, the thickness of the on-board can be 1mm, the aperture of the mounting positioning hole O can be 2mm, and the margin can be 1mm.

[0048] In specific implementation, in the above-mentioned close inspection device online monitoring device provided by the embodiment of the application, as shown in Figure 3 , the fixed support 2 can include a reversible Z-shaped support shaft 21, a first component 22 arranged on the Z-shaped support shaft 21 for embedding the quad camera, a hinge structure 23 linked at the bottom end of the Z-shaped support shaft 21, a second component 24 connected with the hinge structure 23 and abutting the bottom of the close inspection device, and a support frame 25 arranged on the second component 24 for fixing the monocular camera.

[0049] The fixed support 2 of the application uses the Z-shaped support shaft 21 to drive the first component 22 to turn over together. When the close inspection device body is maintained, the monitoring device of the application can be removed by only turning over the hinge structure 23, which is very convenient.

[0050] In practical applications, the fixed support 2 is made of insulating composite material, and the internal circuit of the camera and the close inspection device is embedded in the support to prevent electrical errors. Further, the fixed support 2 can be made of light and transparent material, and can be designed as a flip cover type, which is convenient to flip and does not affect the maintenance of the close inspection device body, and the detection is not affected by environmental factors, and the reliability is high.

[0051] In specific implementation, in the above-mentioned close inspection device online monitoring device provided by the embodiment of the application, the Z-shaped support shaft 21 carries the four-in-one camera, the tail end links the rotating shaft of the second component 24, the lower arm has a through slot, and the bolt can be screwed to prevent shaking in the unflipped state, effectively avoiding the normal operation of the close inspection device itself.

[0052] In specific implementation, in the above-mentioned close inspection device online monitoring device provided by the embodiment of the application, the first component 22 can include a transparent cover plate for carrying the four-in-one camera, a groove for supplying the wire, and a through slot for adjusting the position of the transparent cover plate. Specifically, the first component 22 has a transparent cover plate package, is fixed by a bolt, has a groove for supplying the wire, and has a through slot on the left and right sides to fix the component on the Z-shaped support shaft 21 below and adjust the position forward and backward.

[0053] In specific implementation, in the above-mentioned close inspection device online monitoring device provided by the embodiment of the application, the second component 24 can have a flat through hole at the end away from the Z-shaped support shaft; and the bolt inside the close inspection device fixes the fixed support 2 in the close inspection device through the flat through hole.

[0054] Preferably, two hexagonal bolts inside the close inspection device can be used to fix the fixed support 2 through the flat through hole of the second component 24.

[0055] In specific implementation, in the above-mentioned close inspection device online monitoring device provided by the embodiment of the application, the second component 24 has a horizontal shaft at the end close to the Z-shaped support shaft 21; the horizontal shaft has a hole for fixing the monocular camera.

[0056] Preferably, the horizontal shaft at the rear end of the second component 24 is cut, does not block the normal operation of the side gap of the close inspection device, and the hole in the horizontal shaft is used to fix the monocular camera for monitoring the target of the side gap.

[0057] In specific implementation, in the above-mentioned close inspection device online monitoring device provided by the embodiment of the application, the above-mentioned close inspection device online monitoring device can further include:

[0058] The switch is used to transmit the image signal of the to-be-tested target transmitted by the power supply transmission line 3 to the server in real time;

[0059] The server is used to transmit the received image signal of the to-be-tested target to the industrial computer platform 4.

[0060] Specifically, after the image acquisition device 1 acquires the image signal of the target to be detected, the image signal can be transmitted to the switch station through the twisted pair line, the image signal is received by the indoor switch station and accessed into the server, and the industrial computer platform 4 connects the server to call the corresponding image signal.

[0061] In the specific implementation, in the above-mentioned online monitoring device of the close contact inspector provided by the embodiment of the application, the industrial computer platform 4 is specifically used for constructing and training the YOLOv7-g n Conv neural network model, the image signal of the target to be detected is input into the trained YOLOv7-g n Conv neural network model, and the monitoring data of the target to be detected is output.

[0062] It should be noted that the industrial computer platform 4 can first receive a training set, construct a YOLOv7 neural network model, then add a g n Conv module in the YOLOv7 neural network model to obtain a YOLOv7-g n Conv neural network model, and then train the YOLOv7-g n Conv neural network model using the training set, and use the trained YOLOv7-g n Conv neural network model to monitor the state of the close contact inspector. The YOLOv7-g n Conv neural network model is deployed on the industrial computer platform, which solves a series of problems in the traditional monitoring method of the close contact inspector circuit on-off, such as high labor intensity, low work efficiency, limited maintenance area, low safety factor and the like.

[0063] In the specific implementation, in the above-mentioned online monitoring device of the close contact inspector provided by the embodiment of the application, the industrial computer platform 4 is specifically used for constructing and training the YOLOv7-g

[0064] The warning device is used for judging whether the close contact inspector fails according to the monitoring data, and if so, a warning prompt is given.

[0065] As shown in Figure 4 The software interface realizes the functions of real-time detection, detection result saving and detection result display: by clicking real-time detection, the windows corresponding to different targets of the software interface respectively perform image detection on the side gap and the two ends of the front surface of the same close contact inspector (to represent the "close contact" and "repel" states), the detection results are displayed in text below, whether the close contact inspector fails is judged, and the results are marked in the picture and displayed in the window; the picture saving path is set, the images are sequentially saved to the specified position; the picture reading path is selected, and the previously saved detection results are sequentially displayed in the window by clicking the picture reading button.

[0066] The above-mentioned close inspection device online monitoring device provided by the application has low detection delay in real-time transmission, does not occupy extra transmission channels, has functions of real-time detection, trigger detection, fault alarm, data storage review, etc., realizes cross-region monitoring, and solves the maintenance problem of the close inspection device at low cost and high efficiency.

[0067] The various embodiments are described in a progressive manner in the specification, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between the various embodiments can be referred to each other.

[0068] Those skilled in the art will further appreciate that the individual units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate the interchangeability of hardware and software, various illustrative components, blocks, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the particular application and design constraints imposed on the overall system. Skilled persons can implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the application.

[0069] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can be located in random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0070] Finally, it should be noted that the terms such as first and second, etc., used in the text are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0071] The online monitoring device of the close inspection device provided by the present application is described in detail above, the principle and implementation mode of the present application are described by applying specific examples in this paper, and the above example description is only used to help understand the method of the present application and its core idea; at the same time, for the general technical personnel in the art, according to the idea of the present application, the specific implementation mode and application range will be changed, and the above is not understood as the limitation of the present application.

Claims

1. A close-up inspector on-line monitoring device, characterized by, The utility model relates to a kind of image acquisition device, located in close inspection ware, for real-time acquisition the image signal of target to be measured in the close inspection ware;The target to be measured is the side gap of dynamic static contact point group, rod groove and side pulley;The image acquisition device includes four-in-one camera and monocular camera;The four-in-one camera is set at the set distance above dynamic circle contact point and static spring sheet contact point;The monocular camera is set at the periphery of gap joint consisting of rod groove and side pulley;The four-in-one camera and the monocular camera are used to shoot the motion state inside the close inspection ware; Fixed support, located in the close inspection ware, for fixing the image acquisition device, the four-in-one camera is fixed at the set distance above the measured dynamic circle contact point and the measured static spring sheet contact point, and the monocular camera is fixed in front of the measured gap;The fixed support includes reversible Z-shaped support shaft, first component for embedding the four-in-one camera is set on the Z-shaped support shaft, hinge structure linked at the bottom of the Z-shaped support shaft, second component connected with the hinge structure and attached to the bottom of the close inspection ware, and support frame for fixing the monocular camera is set on the second component;The fixed support uses the Z-shaped support shaft to drive the first component to turn over, when the body of the close inspection ware is overhauled, only the hinge structure needs to be turned over;The Z-shaped support shaft carries the four-in-one camera on the upper surface, the tail end links the rotating shaft of the second component, the lower arm has a penetration slot, and the bolt is not imprisoned in the non-overturned state;The first component includes transparent cover plate for carrying the four-in-one camera, groove for supplying wiring, and penetration slot for adjusting the position of the transparent cover plate;The second component has a flat type through hole at the end away from the Z-shaped support shaft;The bolt inside the close inspection ware fixes the fixed support inside the close inspection ware through the flat type through hole;The second component has horizontal shaft at the end close to the Z-shaped support shaft;The horizontal shaft has a punch hole for fixing the monocular camera; Power supply transmission line for powering the image acquisition device and transmitting the image signal of the target to be measured in real time; Industrial computer platform for acquiring the image signal of the target to be measured and inputting into the pre-trained target monitoring neural network model, and outputting the monitoring data of the target to be measured. The four-in-one camera is configured with a light source.

2. The conforming inspector online monitoring device of claim 1, wherein, Further comprising:

3. The conforming inspector online monitoring device of claim 1, wherein, Switch for transmitting the image signal of the target to be measured transmitted by the power supply transmission line in real time to the server; The server is used for transmitting the received image signal of the target to be measured to the industrial computer platform. Further comprising:

4. The conforming inspector online monitoring device of claim 1, wherein, The industrial computer platform is specifically used for constructing and training YOLOv7-g n Conv neural network model; input the obtained image signal of the to-be-measured target into the trained YOLOv7-g n Conv neural network model, and output the monitoring data of the to-be-measured target.

5. The conforming inspector online monitoring device of claim 1, wherein, Early warning device for judging whether the close inspection ware fails according to the monitoring data; If yes, prewarning is prompted. ​

Images