Lamination thickness detection system and method for transformer core production line

Abstract

The invention discloses a thickness detection system and method, belongs to the technical field of transformer core lamination thickness detection, and particularly relates to a lamination thickness detection system and method for a transformer core production line, and the system comprises a beat monitoring module, a lamination thickness detection module, an upper computer, and a wireless communication module. According to the invention, the lamination thickness data of each step can be effectively detected in the automatic iron core lamination process. Through high-speed visual processing, a 3D laser sensor and a mobile robot technology, a detection position is rapidly and accurately positioned, the production takt is followed, and a production line is helped to find and process abnormal data in time, so that the production efficiency and the product quality are improved, and an important support is provided for automatic and intelligent development of a production process.

Description

Technical Field

[0001] This invention discloses a thickness detection system and method, belonging to the field of transformer core lamination thickness detection technology, specifically relating to a lamination thickness detection system and method for use in transformer core production lines. Background Technology

[0002] The transformer core is the core component of a transformer, responsible for conducting magnetic flux. To effectively reduce eddy current losses generated in the core under alternating magnetic fields, modern transformer cores are generally made of high-permeability, low-loss silicon steel sheets (or amorphous alloy strips), which are sheared and then stacked layer by layer in a specific manner. The quality of the lamination process, especially the uniformity and consistency of the thickness of individual sheets and the overall stacked thickness, is crucial to the performance of the transformer.

[0003] Therefore, accurate and reliable thickness testing of core laminations is a fundamental and indispensable quality control step to ensure high performance, high efficiency, low noise, long life and good assemblability of transformers.

[0004] Traditional sheet thickness inspection mainly relies on manual sampling measurement (such as micrometers and calipers). This method has significant drawbacks such as low efficiency, small coverage, easy to miss detection, poor consistency, high labor intensity, and difficulty in integrating into high-speed production lines. Especially in large-scale, fast-paced modern production, it has become a bottleneck restricting quality and efficiency. Summary of the Invention

[0005] Purpose of the invention: To provide a lamination thickness detection system and method for transformer core production lines, and to solve the problems mentioned above.

[0006] Technical solution: A lamination thickness detection system and method for transformer core production line, including a cycle monitoring module that monitors the operation status of the automatic lamination robot in real time through a cycle monitoring camera to determine whether it can enter the lamination thickness detection state;

[0007] The stack thickness detection module uses a visual positioning camera, a laser 3D contour camera, and a pressure sensor to measure the stack thickness. The visual positioning camera is positioned at the detection location, the laser 3D contour camera measures the contour and calculates the upper and lower stack thickness using the contour information, and sends the measurement information to the host computer. The pressure sensor performs a pressing operation to prevent the gaps between the silicon steel sheets from being too large, which would affect the measurement accuracy.

[0008] The host computer receives the measurement information output by the stacking thickness detection module and displays it in real time. At the same time, it performs status alarms, reports detection data, and coordinates with the automatic stacking robot and the thickness detection AGV for control.

[0009] The wireless communication module is used for communication between the host computer and other control devices, the stack thickness detection module, and the cycle monitoring module.

[0010] In a further embodiment, the visual positioning camera, laser 3D contour camera, and pressure sensor in the thickness detection module are all mounted on the robotic arm of the thickness detection AGV.

[0011] A lamination thickness detection system for a transformer core production line is implemented, wherein the lamination thickness detection method includes:

[0012] Step 1, Task Assignment: After the automatic stacking robot completes a stacking operation, the host computer sends a stacking thickness detection command to the cycle monitoring module and the thickness detection AGV through the wireless communication module.

[0013] Step 2, Cycle Time Monitoring Status: The cycle time monitoring module monitors the operation status of the automatic stacking machine in real time to determine whether it can enter the stacking thickness detection state;

[0014] Step 3, Stack Thickness Detection Status: Enter the stack thickness detection status, and perform thickness detection through the stack thickness detection module. The system uses a visual positioning camera, a laser 3D contour camera, and a pressure sensor to detect and calculate the upper and lower stack thickness data.

[0015] Step 4: Report stack thickness data: Send the measurement information to the host computer.

[0016] Step 5: Determine whether the thickness information meets the process requirements.

[0017] In a further embodiment, the cycle time monitoring status in step 2 is determined by detecting the pose of the automatic stacker to determine whether the automatic stacker is in operation. The specific steps are as follows:

[0018] Step 21: Acquire pose images of the automatic stacking machine using a beat monitoring camera, perform image processing, and extract pose features of the automatic stacking machine from the images;

[0019] Step 22: Establish a coordinate system and input the pose features of the automatic stacker into the coordinate system to determine the pose features of the automatic stacker;

[0020] Step 23: Determine whether the positional characteristics of the automatic stacking machine meet the requirements.

[0021] In a further embodiment, in step 21, the pose image of the automatic stacking machine is first acquired by a beat monitoring camera, and then the acquired image is subjected to image grayscale conversion, image filtering, threshold segmentation, edge detection and feature extraction to obtain the feature points of the pose.

[0022] In a further embodiment, the specific steps for detecting the stack thickness in step 3 are as follows:

[0023] Step 31: When the automatic stacking robot is in standby mode, the detection program starts and drives the mechanical arm of the thickness detection AGV to move the laser 3D contour camera to the detection position.

[0024] Step 32: Perform visual relocalization using a visual positioning camera to further optimize the detection position;

[0025] Step 33: Drive the pressure sensor to operate and tighten the iron core;

[0026] Step 34: Drive the laser 3D contour camera to acquire contour information, calculate the upper and lower stack thickness data, and send it to the host computer;

[0027] Step 35: The robotic arm moves out of the automatic stacking operation area and waits to enter the next cycle monitoring state.

[0028] In a further embodiment, the specific steps for visual relocalization via the visual positioning camera in step 32 are as follows:

[0029] Step 321: The visual positioning camera acquires images and performs feature extraction. First, a 5×5 Sobel kernel is used to calculate the gradients in the x and y directions of the image. Then, the average gradients in the x and y directions of the image are calculated, and the FAST feature threshold is determined based on this value. At the same time, the image is divided into blocks according to the pre-set grid size. Finally, FAST features are extracted in each image block, and the feature with the highest score in each image block is retained. A mask with a radius of r is set with this feature as the center.

[0030] Step 322: Use IMU data to estimate the motion state information of the current frame of the image, thereby realizing the reprojection of feature points, then constructing a linear photometric model, and using the grayscale information of the previous frame image to correct it.

[0031] Step 323: Calculate the noise level of the image patch containing feature points, and then prioritize the processing of image patches with higher noise. In order to ensure that the number of remaining feature points after denoising meets the requirements of subsequent pose solving, stop denoising and output all features when the number of remaining feature points is lower than the threshold.

[0032] Step 324: Search the image feature library using feature points. When the system identifies a potential loop closure candidate frame, it will further refine the matching of the local features of the query key frame and the candidate frame to confirm the loop closure frame.

[0033] Step 325: Compare the loopback frame with the surface features to determine the error value and perform error correction to obtain the accurate detection position.

[0034] In a further embodiment, in step 34, the laser 3D contour camera is driven to operate, acquire contour information, calculate the upper and lower stack thickness data, and send it to the host computer. The specific steps are as follows:

[0035] Step 341, Data Acquisition: Use a laser 3D contour camera to scan the moving object under test and acquire high-density three-dimensional point cloud data of its surface in the sensor coordinate system;

[0036] Step 342, Data Processing: Preprocess the original point cloud to remove noise and extract the contour lines;

[0037] Step 343, Reference plane calibration: Accurately calculate the distance from the laser 3D contour camera to the stationary platform and establish a reference system for height calculation.

[0038] Step 344: Contour Analysis: Analyze the height information on the contour line of each frame to identify the top edge of the object;

[0039] Step 345, Stack Thickness Calculation: Subtract the height value of the top edge from the height value of the reference plane to obtain the thickness of a single object or the total height of the stacked objects.

[0040] Beneficial effects: This invention can effectively detect the stacking thickness data of each step during the automatic core stacking process. Through high-speed vision processing, 3D laser sensors, and mobile robot technology, it can quickly and accurately locate the detection position, keep up with the production cycle, and help the production line to promptly detect and handle abnormal data, thereby improving production efficiency and product quality, and providing important support for the automation and intelligent development of production processes. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the stack thickness detection system of the present invention.

[0042] Figure 2 This is a flowchart of the laminate thickness detection system of the present invention.

[0043] Figure 3 This is a flowchart illustrating the stack thickness detection method of the present invention.

[0044] Figure 4 This is a flowchart of the beat monitoring status of the sheet thickness detection method of the present invention.

[0045] Figure 5 This is a flowchart of the stack thickness detection state of the sheet thickness detection method of the present invention. Detailed Implementation

[0046] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Example 1:

[0048] A lamination thickness detection system for transformer core production lines, such as Figures 1 to 2 As shown, it includes: a beat monitoring module, which monitors the working status of the automatic stacking robot in real time through a beat monitoring camera and determines whether it can enter the stacking thickness detection state;

[0049] The stack thickness detection module uses a visual positioning camera, a laser 3D contour camera, and a pressure sensor to measure the stack thickness. The visual positioning camera is positioned at the detection location, the laser 3D contour camera measures the contour and calculates the upper and lower stack thickness using the contour information, and sends the measurement information to the host computer. The pressure sensor performs a pressing operation to prevent the gaps between the silicon steel sheets from being too large, which would affect the measurement accuracy.

[0050] The host computer receives the measurement information output by the stacking thickness detection module and displays it in real time. At the same time, it performs status alarms, reports detection data, and coordinates with the automatic stacking robot and the thickness detection AGV for control.

[0051] The wireless communication module is used for communication between the host computer and other control devices, the stack thickness detection module, and the cycle monitoring module.

[0052] In one embodiment, such as Figures 1 to 2 As shown, the visual positioning camera, laser 3D contour camera, and pressure sensor in the thickness detection module are all mounted on the robotic arm of the thickness detection AGV.

[0053] Example 2:

[0054] A method for detecting the lamination thickness in a transformer core production line, such as... Figure 3 As shown, the laminate thickness detection method includes:

[0055] Step 1, Task Assignment: After the automatic stacking robot completes a stacking operation, the host computer sends a stacking thickness detection command to the cycle monitoring module and the thickness detection AGV through the wireless communication module.

[0056] Step 2, Cycle Time Monitoring Status: The cycle time monitoring module monitors the operation status of the automatic stacking machine in real time to determine whether it can enter the stacking thickness detection state;

[0057] Step 3, Stack Thickness Detection Status: Enter the stack thickness detection status, and perform thickness detection through the stack thickness detection module. The system uses a visual positioning camera, a laser 3D contour camera, and a pressure sensor to detect and calculate the upper and lower stack thickness data.

[0058] Step 4: Report stack thickness data: Send the measurement information to the host computer.

[0059] Step 5: Determine whether the thickness information meets the process requirements.

[0060] In one embodiment, such as Figure 4 As shown, in step 2, the cycle time monitoring status is determined by detecting the pose of the automatic stacker to determine whether the automatic stacker is in operation. The specific steps are as follows:

[0061] Step 21: Acquire pose images of the automatic stacking machine using a beat monitoring camera, perform image processing, and extract pose features of the automatic stacking machine from the images;

[0062] Step 22: Establish a coordinate system and input the pose features of the automatic stacker into the coordinate system to determine the pose features of the automatic stacker;

[0063] Step 23: Determine whether the positional characteristics of the automatic stacking machine meet the requirements.

[0064] In one embodiment, in step 21, the pose image of the automatic stacking machine is first acquired by a beat monitoring camera, and then the acquired image is subjected to image grayscale conversion, image filtering, threshold segmentation, edge detection and feature extraction to obtain the feature points of the pose.

[0065] In one embodiment, such as Figure 5 As shown, the specific steps for detecting the stack thickness in step 3 are as follows:

[0066] Step 31: When the automatic stacking robot is in standby mode, the detection program starts and drives the mechanical arm of the thickness detection AGV to move the laser 3D contour camera to the detection position.

[0067] Step 32: Perform visual relocalization using a visual positioning camera to further optimize the detection position;

[0068] Step 33: Drive the pressure sensor to operate and tighten the iron core;

[0069] Step 34: Drive the laser 3D contour camera to acquire contour information, calculate the upper and lower stack thickness data, and send it to the host computer;

[0070] Step 35: The robotic arm moves out of the automatic stacking operation area and waits to enter the next cycle monitoring state.

[0071] In one embodiment, the specific steps for visual relocalization using a visual positioning camera in step 32 are as follows:

[0072] Step 321: The visual positioning camera acquires images and performs feature extraction. First, a 5×5 Sobel kernel is used to calculate the gradients in the x and y directions of the image. Then, the average gradients in the x and y directions of the image are calculated, and the FAST feature threshold is determined based on this value. At the same time, the image is divided into blocks according to the pre-set grid size. Finally, FAST features are extracted in each image block, and the feature with the highest score in each image block is retained. A mask with a radius of r is set with this feature as the center.

[0073] Step 322: Use IMU data to estimate the motion state information of the current frame of the image, thereby realizing the reprojection of feature points, then constructing a linear photometric model, and using the grayscale information of the previous frame image to correct it.

[0074] Step 323: Calculate the noise level of the image patch containing feature points, and then prioritize the processing of image patches with higher noise. In order to ensure that the number of remaining feature points after denoising meets the requirements of subsequent pose solving, stop denoising and output all features when the number of remaining feature points is lower than the threshold.

[0075] Step 324: Search the image feature library using feature points. When the system identifies a potential loop closure candidate frame, it will further refine the matching of the local features of the query key frame and the candidate frame to confirm the loop closure frame.

[0076] Step 325: Compare the loopback frame with the surface features to determine the error value and perform error correction to obtain the accurate detection position.

[0077] In one embodiment, step 34 involves driving a laser 3D contour camera to acquire contour information, calculate the upper and lower stack thickness data, and send it to the host computer. The specific steps are as follows:

[0078] Step 341, Data Acquisition: Use a laser 3D contour camera to scan the moving object under test and acquire high-density three-dimensional point cloud data of its surface in the sensor coordinate system;

[0079] Step 342, Data Processing: Preprocess the original point cloud to remove noise and extract the contour lines;

[0080] Step 343, Reference plane calibration: Accurately calculate the distance from the laser 3D contour camera to the stationary platform and establish a reference system for height calculation.

[0081] Step 344: Contour Analysis: Analyze the height information on the contour line of each frame to identify the top edge of the object;

[0082] Step 345, Stack Thickness Calculation: Subtract the height value of the top edge from the height value of the reference plane to obtain the thickness of a single object or the total height of the stacked objects.

[0083] This invention has autonomous sensing capabilities, which can accurately detect when the core stacking equipment stops operating during material changing, schedule detection sensors to quickly complete post-stack detection, and vacate the work position after the core stacking equipment finishes changing material, thus automating and seamlessly streamlining the production cycle.

[0084] This invention boasts a repeatability accuracy of up to 4µm, helping to adjust production parameters, avoid defective products, and improve product quality.

[0085] This invention is an autonomous mobile device that can serve multiple stacked iron core workbenches with a single device, resulting in high product utilization.

[0086] This invention exhibits strong stability and excellent repeatability, ensuring high detection performance even during long-term operation.

[0087] This invention has a high degree of automation, enabling automated testing without human intervention, thereby improving production efficiency.

[0088] This invention enables data integration with the production line, transmitting detection data to the MES system in real time to provide data support for production management.

[0089] The detection equipment of this invention can provide timely warnings of stacking thickness issues, help production adjust parameters, avoid the generation of defective products, and improve product quality.

[0090] This invention provides a graphical interface to meet the flexible usage needs of system users. It offers a user-friendly interface, allowing users to operate the system intuitively and conveniently, and supports various interactive operations using a mouse or keyboard.

[0091] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A lamination thickness detection system for a transformer core production line, characterized in that, include: The beat monitoring module monitors the operation status of the automatic stacking robot in real time through the beat monitoring camera and determines whether it can enter the stacking thickness detection state. The stack thickness detection module uses a visual positioning camera, a laser 3D contour camera, and a pressure sensor to measure the stack thickness. The visual positioning camera is positioned at the detection location, the laser 3D contour camera measures the contour and calculates the upper and lower stack thickness using the contour information, and sends the measurement information to the host computer. The pressure sensor performs a pressing operation to prevent the gaps between the silicon steel sheets from being too large, which would affect the measurement accuracy. The host computer receives the measurement information output by the stacking thickness detection module and displays it in real time. At the same time, it performs status alarms, reports detection data, and coordinates with the automatic stacking robot and the thickness detection AGV for control. The wireless communication module is used for communication between the host computer and other control devices, the stack thickness detection module, and the cycle monitoring module.

2. The laminate thickness detection system for a transformer core production line according to claim 1, characterized in that, In the thickness detection module, the visual positioning camera, laser 3D contour camera, and pressure sensor are all mounted on the robotic arm of the thickness detection AGV.

3. A method for detecting the lamination thickness in a transformer core production line, implemented using a lamination thickness detection system for a transformer core production line as described in any one of claims 1 to 2, characterized in that... The method for detecting the thickness of the laminated sheets includes: Step 1, Task Assignment: After the automatic stacking robot completes a stacking operation, the host computer sends a stacking thickness detection command to the cycle monitoring module and the thickness detection AGV through the wireless communication module. Step 2, Cycle Time Monitoring Status: The cycle time monitoring module monitors the operation status of the automatic stacking machine in real time to determine whether it can enter the stacking thickness detection state; Step 3, Stack Thickness Detection Status: Enter the stack thickness detection status, and perform thickness detection through the stack thickness detection module. The system uses a visual positioning camera, a laser 3D contour camera, and a pressure sensor to detect and calculate the upper and lower stack thickness data. Step 4: Report stack thickness data: Send the measurement information to the host computer. Step 5: Determine whether the thickness information meets the process requirements.

4. The method for detecting the lamination thickness in a transformer core production line according to claim 3, characterized in that, In step 2, the cycle time monitoring status is determined by detecting the pose of the automatic stacker to determine whether the automatic stacker is in operation. The specific steps are as follows: Step 21: Acquire pose images of the automatic stacking machine using a beat monitoring camera, perform image processing, and extract pose features of the automatic stacking machine from the images; Step 22: Establish a coordinate system and input the pose features of the automatic stacker into the coordinate system to determine the pose features of the automatic stacker; Step 23: Determine whether the positional characteristics of the automatic stacking machine meet the requirements.

5. The method for detecting the lamination thickness in a transformer core production line according to claim 4, characterized in that, In step 21, the pose image of the automatic stacking machine is first acquired by the beat monitoring camera. Then, the acquired image is converted to grayscale, filtered, thresholded, edge detected and feature extracted to obtain the feature points of the pose.

6. The method for detecting the lamination thickness in a transformer core production line according to claim 3, characterized in that, The specific steps for detecting the stack thickness status in step 3 are as follows: Step 31: When the automatic stacking robot is in standby mode, the detection program starts and drives the mechanical arm of the thickness detection AGV to move the laser 3D contour camera to the detection position. Step 32: Perform visual relocalization using a visual positioning camera to further optimize the detection position; Step 33: Drive the pressure sensor to operate and tighten the iron core; Step 34: Drive the laser 3D contour camera to acquire contour information, calculate the upper and lower stack thickness data, and send it to the host computer; Step 35: The robotic arm moves out of the automatic stacking operation area and waits to enter the next cycle monitoring state.

7. A method for detecting the lamination thickness in a transformer core production line according to claim 6, characterized in that, In step 32, the specific steps for visual relocalization using the visual positioning camera are as follows: Step 321: The visual positioning camera acquires images and performs feature extraction. First, a 5×5 Sobel kernel is used to calculate the gradients in the x and y directions of the image. Then, the average gradients in the x and y directions of the image are calculated, and the FAST feature threshold is determined based on this value. At the same time, the image is divided into blocks according to the pre-set grid size. Finally, FAST features are extracted in each image block, and the feature with the highest score in each image block is retained. A mask with a radius of r is set with this feature as the center. Step 322: Use IMU data to estimate the motion state information of the current frame of the image, thereby realizing the reprojection of feature points, then constructing a linear photometric model, and using the grayscale information of the previous frame image to correct it. Step 323: Calculate the noise level of the image patch containing feature points, and then prioritize the processing of image patches with higher noise. In order to ensure that the number of remaining feature points after denoising meets the requirements of subsequent pose solving, stop denoising and output all features when the number of remaining feature points is lower than the threshold. Step 324: Search the image feature library using feature points. When the system identifies a potential loop closure candidate frame, it will further refine the matching of the local features of the query key frame and the candidate frame to confirm the loop closure frame. Step 325: Compare the loopback frame with the surface features to determine the error value and perform error correction to obtain the accurate detection position.

8. A method for detecting the lamination thickness in a transformer core production line according to claim 6, characterized in that, In step 34, the laser 3D contour camera is driven to acquire contour information, calculate the upper and lower stack thickness data, and send it to the host computer. The specific steps are as follows: Step 341, Data Acquisition: Use a laser 3D contour camera to scan the moving object under test and acquire high-density three-dimensional point cloud data of its surface in the sensor coordinate system; Step 342, Data Processing: Preprocess the original point cloud to remove noise and extract the contour lines; Step 343, Reference plane calibration: Accurately calculate the distance from the laser 3D contour camera to the stationary platform and establish a reference system for height calculation. Step 344: Contour Analysis: Analyze the height information on the contour line of each frame to identify the top edge of the object; Step 345, Stack Thickness Calculation: Subtract the height value of the top edge from the height value of the reference plane to obtain the thickness of a single object or the total height of the stacked objects.

Images