Rapid detection method for number of turns of pancake coil of transformer
Through the image detection method, the identification model is used to identify the handover wire, shielded wire or insulating cardboard in the cake coil, and the number of turns is automatically determined, which solves the problem of low detection efficiency in the prior art and realizes efficient and accurate number of turns detection.
Patent Information
- Application Number
- CN202510153573.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-12
- Publication Date
- 2025-05-30
AI Technical Summary
At this stage, the transformer cake coil turns detection efficiency is low, especially the fractional turns and special wire cakes cannot be accurately judged, and manual secondary verification is required.
The image detection method is adopted, and the number of turns of the cake coil is automatically determined through the pre-trained recognition model using the handover wire, shielded wire or insulated cardboard in the image to be detected as identification features.
No manual review is required, which improves the efficiency and accuracy of turns detection, and can accurately identify all cake coils including fraction turns and special wire cakes.
Smart Images
Figure CN120070382A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of transformer manufacturing, and particularly relates to a method for quickly detecting the number of turns of a transformer disk-type coil. Background Art
[0002] The detection of the number of turns of a transformer coil is an important link in transformer manufacturing and maintenance, which is directly related to the operating efficiency and safety of the transformer. For a continuous transformer disk-type coil, each disk has n turns of working conductors, and in some special disks, there are also various situations such as filling insulation paperboards for filling the radial direction, inner diameter side insulation paperboards for lifting the inner diameter of the coil, and shielding wires. For this type of disk-type coil, at present, generally after each disk is wound, it is manually rechecked, or the corresponding number of turns is determined by referring to the machine tool turn counter (recording the number of turns wound by the machine tool).
[0003] However, the current inspection methods for the number of turns of disk-type coils have the following disadvantages: 1) The manual rechecking efficiency is low; 2) Since the machine tool turn counter records the number of turns of the machine tool rotation, for fractional turns (the situation where the circumference is less than one full turn) and special disks, it is impossible to accurately determine the number of turns, and manual secondary verification is still required. Summary of the Invention
[0004] The embodiments of the present invention provide a method for quickly detecting the number of turns of a transformer disk-type coil to solve the problem of low efficiency of the current detection method for the number of turns of transformer disk-type coils.
[0005] In a first aspect, the embodiments of the present invention provide a method for quickly detecting the number of turns of a transformer disk-type coil, including:
[0006] Obtaining a to-be-detected image of a target disk-type coil;
[0007] Inputting the to-be-detected image into a pre-trained recognition model to obtain the number of turns of the target disk-type coil;
[0008] Wherein, the recognition model is used to use the intersection line of the target disk-type coil in the to-be-detected image as an identification feature to determine the number of turns of the target disk-type coil based on the intersection line.
[0009] In a possible implementation manner, the recognition model is further used to use the shielding wire of the target disk-type coil in the to-be-detected image as an identification feature to determine the number of turns of the target disk-type coil based on the shielding wire.
[0010] In a possible implementation manner, the recognition model is further used to use the insulation paperboard of the target disk-type coil in the to-be-detected image as an identification feature to determine the number of turns of the target disk-type coil based on the insulation paperboard.
[0011] In a possible implementation, inputting the image to be detected into a pre-trained recognition model to obtain the number of turns of the target pancake coil includes:
[0012] Cut the image to be detected into a first image with a preset width, and input the first image into the pre-trained recognition model to obtain the number of turns of the target pancake coil.
[0013] In a possible implementation, after inputting the image to be detected into the pre-trained recognition model to obtain the number of turns of the target pancake coil, it further includes:
[0014] Display the number of turns of the target pancake coil.
[0015] In a possible implementation, displaying the number of turns of the target pancake coil includes:
[0016] Restore the first image to the image to be detected, and based on the number of turns of the target pancake coil, mark the corresponding recognition features and the number of turns on the image to be detected.
[0017] In a possible implementation, the pre-trained recognition model is a neural network model.
[0018] In a second aspect, an embodiment of the present invention provides a fast detection device for the number of turns of a transformer pancake coil, including:
[0019] An acquisition module, configured to acquire an image to be detected of the target pancake coil;
[0020] A recognition module, configured to input the image to be detected into a pre-trained recognition model to obtain the number of turns of the target pancake coil;
[0021] Wherein, the recognition model is configured to use the intersection line of the target pancake coil in the image to be detected as a recognition feature, and based on the intersection line, determine the number of turns of the target pancake coil.
[0022] In a third aspect, an embodiment of the present invention provides a system, including a camera, a development board, and a display terminal; wherein, the development board includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, it implements the steps of the method described in the first aspect or any possible implementation manner of the first aspect above.
[0023] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the steps of the method described in the first aspect or any possible implementation manner of the first aspect above.
[0024] An embodiment of the present invention provides a method for quickly detecting the number of turns of a transformer disk-type coil. By means of image detection and using an identification model for identification, manual recheck is not required, improving the identification efficiency. In addition, considering that in the current disk-type coils, there are cross-connection lines at adjacent positions of each turn of wire, therefore, using the cross-connection lines as identification features can ensure the accuracy of quickly detecting the number of turns of the transformer disk-type coil. Description of the Drawings
[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of related technologies. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.
[0026] Figures 1A - 1B is a schematic diagram of a disk-type coil;
[0027] Figures 2A - 2B is a cross-sectional view of a disk-type coil;
[0028] Figure 3 is a flowchart of the implementation of the method for quickly detecting the number of turns of a transformer disk-type coil provided by an embodiment of the present invention;
[0029] Figure 4 is the image to be detected provided by an embodiment of the present invention;
[0030] Figure 5 is the first image provided by an embodiment of the present invention;
[0031] Figures 6A - 6B is a schematic diagram of the identification features provided by an embodiment of the present invention;
[0032] Figure 7 is a schematic diagram of the display result provided by an embodiment of the present invention;
[0033] Figure 8 is a schematic structural diagram of the device for quickly detecting the number of turns of a transformer disk-type coil provided by an embodiment of the present invention.
[0034] Figure 9 is a schematic structural diagram of the system for quickly detecting the number of turns of a transformer disk-type coil provided by an embodiment of the present invention; Detailed Embodiments
[0035] In the following description, specific details such as specific system architectures and technologies are presented for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, those skilled in the art should understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from obscuring the description of the present invention.
[0036] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will be described through specific embodiments in conjunction with the accompanying drawings.
[0037] Figures 1A - 1B is a schematic diagram of a pancake coil, as shown in Figure 1, where Figure 1A is a three-dimensional structural diagram of the pancake coil, Figure 1B is a schematic diagram of the number of internal turns of the wire in the pancake coil. Based on Figures 1A - 1B it can be seen that the pancake coil is formed by connecting several turns in the form of a ring or a semi-ring, and the turns are arranged in parallel, and the laying of the wire is relatively tight.
[0038] Figures 2A - 2B is a cross-sectional view of the pancake coil; where Figure 2A is a characteristic of the number of turns of the wire in a common pancake coil, where the vertical line is the working wire. Since there are arcs at the corners of each turn of the wire, there are intersection lines at adjacent positions of each turn of the wire, that is, Figure 2A the horizontal lines corresponding to each turn of the wire in Figure 2B Based on this, for a common pancake coil, the intersection line can be used as an identification feature to determine the number of turns of the wire in the entire pancake coil.
[0039] is a characteristic of the number of turns of the wire in a special pancake coil. For a special pancake coil, in addition to the intersection line, it also includes a shield wire and an insulating cardboard. Therefore, for a special pancake coil, in addition to using the intersection line as an identification line, the shield wire and the insulating cardboard can also be used as identification features to identify the number of turns of the wire in the entire pancake coil. Figure 3 Based on the above analysis, the method provided by the embodiments of the present invention can be as shown in Figure 3 The following will describe the method provided by the embodiments of the present invention in conjunction with
[0040] Figure 3 is a flowchart of the implementation of the method for quickly detecting the number of turns of a transformer pancake coil provided by the embodiments of the present invention. As shown in Figure 3 the method may include:
[0041] Step 110: Obtain a to-be-detected image of the target pancake coil.
[0042] Figure 4 is the to-be-detected image provided by the embodiments of the present invention, as shown inFigure 4 As shown, a camera can be used to photograph the pancake coil during the production process to collect the corresponding image to be detected.
[0043] Step 120: Input the image to be detected into a pre-trained recognition model to obtain the number of turns of the target pancake coil.
[0044] In this embodiment, the pre-trained recognition model can be a neural network model, specifically, it can be a YOLO series model, such as YOLOv5, YOLOv6, YOLOv7, etc. Exemplarily, the pre-trained recognition model can be trained in the following way:
[0045] Collect a large number of images of pancake coil wires, where the recognition features are labeled in the images. Based on these images, a turn number dataset is obtained. Model training is performed based on the turn number dataset to obtain the pre-trained recognition model.
[0046] In this embodiment, before inputting the image to be detected into the pre-trained recognition model, considering that most of the images to be detected collected are images of the large area of the wire cake of the pancake coil, in order to improve the recognition accuracy, on the basis of retaining the original height, the image to be detected can be first sheared into a first image with a preset width. The first image is as Figure 5 shown. Among them, the preset width can be 9mm, 10mm, 11mm, etc.
[0047] Input the first image into the pre-trained recognition model, and the pre-trained recognition model recognizes the intersection line in the first image, that is Figure 6A the features within the box in, and based on the recognition result of the intersection line, determine the number of turns of the target pancake coil.
[0048] Correspondingly, in some alternative embodiments, for special pancake coils that have shielded wires and insulating cardboard, then, for these special pancake coils, the shielded wires or insulating cardboard can also be used as recognition features for recognition, that is, the features within the box as Figure 6B shown, to obtain the number of turns of its corresponding coil.
[0049] In an alternative embodiment, to ensure the recognition accuracy, for pancake coils with two or more of the intersection line, shielded wire, and insulating cardboard, the corresponding number of turns of the coil can be obtained respectively based on their corresponding recognition features. When the number of turns of the coil obtained from different recognition features is different, re-recognition can be performed to obtain accurate results.
[0050] Alternatively, when the number of turns of the coil obtained from different identification features is different, an alarm can be issued to remind the staff that there is an abnormal identification situation, so that the identifier can control the corresponding device to re-acquire the image to be detected of the target disk coil and re-identify it to obtain an accurate result.
[0051] Optionally, in some embodiments, after obtaining the number of turns of the target disk coil, it may further include:
[0052] Displaying the number of turns of the target disk coil.
[0053] In this embodiment, since the image to be detected is sheared during the identification to obtain the first image, the recognition model itself recognizes the first image. For the convenience of the operator's observation, when displaying, the first image can be restored to the image to be detected, and then the corresponding identification features can be marked on the image to be detected, and the number of turns of the disk coil recognized according to the first image can also be marked.
[0054] Correspondingly, Figure 7 is a schematic diagram of the display result provided by the embodiment of the present invention. As Figure 7 shown, Figure 7 taking the intersection line as the identification feature as an example, the display result is obtained. The recognized intersection line is marked, and the number of turns of the recognized coil is marked in the upper left corner. Among them, Figure 7 the number of turns of the coil recognized in is 16. Therefore, the number of turns of the coil marked in the upper left corner is 16.
[0055] In summary, in the embodiment of the present invention, the intersection line, shield wire or insulating cardboard in the disk coil is used as the identification mark, which can ensure the accuracy of the rapid detection of the number of turns of the transformer disk coil. The obtained image to be detected is input into the recognition model for recognition by the recognition model, without manual recheck, improving the recognition efficiency. In order to improve the accuracy during the recognition process, the image to be detected is also sheared, which can reduce the interference items during the recognition process. Finally, the recognition result is displayed in the image to be detected restored based on the first image for the convenience of the operator to view.
[0056] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
[0057] The following is the device embodiment of the present invention. For the details not described in detail, reference can be made to the corresponding method embodiment above.
[0058] Figure 8The structural schematic diagram of the rapid detection device for the number of turns of the transformer disc coil provided by the embodiment of the present invention is shown. For the sake of convenience of description, only the parts related to the embodiment of the present invention are shown and are described in detail as follows:
[0059] As Figure 8 shown, the rapid detection device 8 for the number of turns of the transformer disc coil includes:
[0060] An acquisition module 81, configured to acquire a to-be-detected image of the target disc coil;
[0061] An identification module 82, configured to input the to-be-detected image into a pre-trained identification model to obtain the number of turns of the target disc coil;
[0062] Among them, the identification model is configured to use the intersection line of the target disc coil in the to-be-detected image as an identification feature to determine the number of turns of the target disc coil based on the intersection line.
[0063] In a possible implementation manner, the identification model is further configured to use the shield wire of the target disc coil in the to-be-detected image as an identification feature to determine the number of turns of the target disc coil based on the shield wire.
[0064] In a possible implementation manner, the identification model is further configured to use the insulating cardboard of the target disc coil in the to-be-detected image as an identification feature to determine the number of turns of the target disc coil based on the insulating cardboard.
[0065] In a possible implementation manner, the identification module 82 is specifically configured to:
[0066] Cut the to-be-detected image into a first image with a preset width, and input the first image into a pre-trained identification model to obtain the number of turns of the target disc coil.
[0067] In a possible implementation manner, the rapid detection device 8 for the number of turns of the transformer disc coil further includes a display module 83;
[0068] The display module 83 is configured to display the number of turns of the target disc coil.
[0069] In a possible implementation manner, the display module 83 is specifically configured to:
[0070] Restore the first image to the to-be-detected image, and mark the corresponding identification features and the number of turns on the to-be-detected image based on the number of turns of the target disc coil.
[0071] In a possible implementation manner, the pre-trained identification model is a neural network model.
[0072] The following is the system embodiment of the present invention. For the details not described in detail, reference may be made to the corresponding method embodiment above.
[0073] Figure 9 It is a schematic structural diagram of a fast detection system for the number of turns of a transformer disc coil provided by an embodiment of the present invention; for the convenience of description, only the parts related to the embodiments of the present invention are shown and are described in detail as follows:
[0074] In this embodiment, the system may include a camera, a development board, and a display terminal; wherein, the development board includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the corresponding method provided by the embodiments of the present invention are implemented.
[0075] In this embodiment, the system may include a development board. The development board may be a Jetson Xavier development board. An identification model is built in it, and the identification model may be a coil turn number feature model based on YOLOv5x, and this model is trained with a local coil turn number feature data set.
[0076] The system may further include a display terminal and a camera. Among them, the display terminal and the camera are respectively connected to the development board. The camera is used to collect an image of the corresponding target disc coil on the coil winding machine tool and send the image to the development board. Based on the identification model in the development board, the identification features in the image are identified to obtain the corresponding identification result, and the identification result is displayed through the display terminal, so that the operator can adjust the production process of the disc coil based on the identification result.
[0077] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.
[0078] Those of ordinary skill in the art can realize that the templates, units, and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.
[0079] If the module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described embodiments of the rapid detection method for the number of turns of each transformer disk-type coil can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium, etc.
[0080] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A method for quickly detecting the number of turns of a transformer pancake coil, characterized in that: include: Acquire the image of the target pancake coil to be detected; Inputting the image to be detected into a pre-trained recognition model to obtain the number of turns of the target pancake coil; Wherein, the recognition model is used to use the intersection line of the target pancake coil in the image to be detected as a recognition feature, so as to determine the number of turns of the target pancake coil based on the intersection line.
2. The method for quickly detecting the number of turns of a transformer pancake coil according to claim 1, characterized in that: The recognition model is further used to use the shielding line of the target pancake coil in the image to be detected as a recognition feature, so as to determine the number of turns of the target pancake coil based on the shielding line.
3. The method for quickly detecting the number of turns of a transformer pancake coil according to claim 1, characterized in that: The recognition model is further used to use the insulating paperboard of the target pancake coil in the image to be detected as a recognition feature, so as to determine the number of turns of the target pancake coil based on the insulating paperboard.
4. The method for quickly detecting the number of turns of a transformer pancake coil according to claim 1, characterized in that: The step of inputting the image to be detected into a pre-trained recognition model to obtain the number of turns of the target pancake coil includes: The image to be detected is cut into a first image of a preset width, and the first image is input into a pre-trained recognition model to obtain the number of turns of the target pancake coil.
5. The method for quickly detecting the number of turns of a transformer pancake coil according to claim 4, characterized in that: After inputting the image to be detected into a pre-trained recognition model to obtain the number of turns of the target pancake coil, the method further includes: The number of turns of the target pancake coil is displayed.
6. The method for quickly detecting the number of turns of a transformer pancake coil according to claim 5, characterized in that: The displaying of the number of turns of the target pancake coil comprises: The first image is restored to the image to be detected, and based on the number of turns of the target pancake coil, corresponding identification features and the number of turns are marked on the image to be detected.
7. The method for quickly detecting the number of turns of a transformer pancake coil according to claim 1, characterized in that: The pre-trained recognition model is a neural network model.
8. A rapid detection device for the number of turns of a transformer pancake coil, characterized in that: include: An acquisition module, used for acquiring an image to be detected of the target pancake coil; A recognition module, used for inputting the image to be detected into a pre-trained recognition model to obtain the number of turns of the target pancake coil; Wherein, the recognition model is used to use the intersection line of the target pancake coil in the image to be detected as a recognition feature, so as to determine the number of turns of the target pancake coil based on the intersection line.
9. A system comprising a camera, a development board and a display terminal; the development board comprises a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method as claimed in any one of claims 1 to 7 are implemented.