Rapid detection device for appearance defects of battery piece
By combining dual high-speed cameras and multispectral light sources, along with servo motor drives and deep learning models, the problem of insufficient front and side defect recognition capabilities of battery cell inspection equipment has been solved, achieving efficient and accurate automated inspection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- DONGGUAN MOORE INTELLIGENT TECHNOLOGY CO LTD
- Filing Date
- 2025-06-05
- Publication Date
- 2026-05-12
AI Technical Summary
Existing solar cell inspection equipment cannot effectively identify defects on the front and sides of solar cells. The single light source configuration leads to unstable image quality. The image processing algorithm is outdated, the level of intelligence is low, the inspection efficiency is low, and it relies on manual intervention.
It employs dual high-speed cameras with a visible light LED array and a short-wave infrared light source, combined with a multispectral filter switching mechanism, a servo motor-driven conveyor belt and a pneumatic sorting valve, and utilizes FPGA real-time processing and deep learning models for image processing and automatic sorting.
It enables multi-dimensional inspection of the front and sides of the solar cells, improving inspection efficiency, reducing the false negative rate and labor costs, and ensuring the accuracy and automation of the inspection.
Smart Images

Figure CN224222039U_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The utility model relates to battery detection technical field, concretely is a battery piece appearance defect rapid detection device. BACKGROUND
[0002] With the scale development of battery industry, higher requirements are put forward to the speed, precision and automation degree of battery piece appearance defect detection. In the production process of battery piece, appearance defect detection is the key link to ensure product quality. The traditional manual detection mode depends on visual observation, and there are problems such as low detection efficiency, strong subjectivity, high omission rate and unable to adapt to high-speed assembly line production. Although the existing automatic detection equipment introduces visual detection technology, the following deficiencies exist generally: first, the detection mechanism is only for detecting the front of battery piece, and the recognition ability for side defects (such as hidden crack, delamination, etc.) is insufficient;Second, the light source configuration is single, and it is difficult to consider the imaging needs of different defect types, resulting in unstable image quality;Third, the image processing algorithm is backward, cannot accurately identify complex defects, and the sorting process relies on manual intervention, and the intelligent degree is low. UTILITY MODEL CONTENT
[0003] In order to overcome the deficiencies of the prior art, the utility model provides a battery piece appearance defect rapid detection device, which can effectively solve the problems put forward in the background art.
[0004] The utility model solves the technical problems thereof by adopting the technical scheme of:
[0005] A battery piece appearance defect rapid detection device, comprising a rack, a conveying mechanism arranged on the rack, a visual detection mechanism located above the conveying mechanism and a control system electrically connected with the visual detection mechanism, the conveying mechanism comprising a conveying belt driven by a servo motor and limiting baffles arranged on both sides of the conveying belt, the conveying belt surface is uniformly distributed with anti-skid protrusions, and the conveying belt is provided below the visual detection mechanism with a pneumatic sorting valve;
[0006] The visual detection mechanism comprises a first high-speed camera and a second high-speed camera arranged in sequence along the transmission direction of battery piece, the first high-speed camera is used for collecting battery piece front image, and the second high-speed camera is used for collecting battery piece side image;
[0007] The both sides of the conveying belt are provided with light source modules, the light source modules comprise visible light LED array and shortwave infrared light source, the visible light LED array is used in cooperation with the first high-speed camera, and the shortwave infrared light source is used in cooperation with the second high-speed camera.
[0008] As a further description of the above technical solution, the rack is further provided with a code scanning module located at the entrance end of the conveying belt, and the code scanning module is electrically connected with the control system.
[0009] As a further description of the above technical solution, the visual detection mechanism further comprises a multi-spectrum filter switching mechanism arranged between the first high-speed camera and the second high-speed camera, and the multi-spectrum filter switching mechanism comprises a driving motor and at least three filters of different wavelengths.
[0010] As a further description of the above technical solution, the light source module further comprises a diffuse reflection plate arranged between the visible light LED array and the short-wave infrared light source, and the surface of the diffuse reflection plate is sprayed with a high-reflectivity coating.
[0011] As a further description of the above technical solution, the control system comprises an FPGA real-time processing unit, an edge computing module with a built-in deep learning model, an alarm module connected with the edge computing module, and a sorting control module, the FPGA real-time processing unit is used for pre-processing images collected by the first high-speed camera and the second high-speed camera, the edge computing module is used for defect recognition based on the pre-processed images, the alarm module is used for issuing sound and light alarms when defects are detected, and the sorting control module is used for controlling the pneumatic sorting valve.
[0012] As a further description of the above technical solution, the limiting baffle on both sides of the conveying belt is an elastic baffle with adjustable spacing, and a pressure sensor is arranged on the inner side of the limiting baffle, and the pressure sensor is electrically connected with the control system.
[0013] Compared with the prior art, the utility model has the beneficial effects that:
[0014] The battery piece appearance defect rapid detection device has at least one of the following beneficial effects in the process of use:
[0015] The double high-speed cameras cooperate with the visible light LED array and the short-wave infrared light source to perform multi-dimensional detection of front regular defects (such as scratches and cracks) and side hidden defects (such as hidden cracks and delamination), the detection efficiency is more prominent in combination with the multi-spectrum filter switching mechanism: the servo motor drives the conveying belt to stably transmit, the high-speed camera cooperates with the FPGA real-time pre-processing and the edge computing module, and the single detection time consumption is less. The elastic limiting baffle automatically adjusts the spacing through the pressure sensor, and the code scanning module realizes intelligent tracing of detection data and production batches; the diffuse reflection plate optimizes the uniformity of illumination, and in combination with the anti-skid conveying belt design, the transmission stability and image quality are ensured. The deep learning model identifies defects in real time and triggers sound and light alarms, the pneumatic sorting valve has a fast response time, realizes automatic rejection of defective pieces, greatly improves the detection efficiency compared with traditional manual detection, and greatly reduces the missed detection rate and labor cost.BRIEF DESCRIPTION OF DRAWINGS BRIEF DESCRIPTION OF DRAWINGS
[0016] Fig. 1 It is whole structure schematic view of the battery piece appearance defect fast detection device of the utility model;
[0017] Fig. 2 It is first side structure schematic view of the battery piece appearance defect fast detection device of the utility model;
[0018] Fig. 3 It is second side structure schematic view of the battery piece appearance defect fast detection device of the utility model.
[0019] Reference numerals in the drawing:
[0020] 1, rack;101, control system;102, pneumatic sorting valve;2, conveying mechanism;201, code scanning module;202, servo motor;203, limit baffle;204, light source module;205, multi-spectral filter switching mechanism;206, conveying belt;3, visual inspection mechanism;301, first high-speed camera;302, diffuse reflection plate;303, second high-speed camera. DETAILED DESCRIPTION
[0021] The technical solutions in the embodiments of the utility model will be clearly and completely described below with reference to the drawings in the embodiments of the utility model. Obviously, the described embodiments are only part of the embodiments of the utility model, rather than all the embodiments. Based on the embodiments in the utility model, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the utility model.
[0022] As Figs. 1-3 shown, the utility model provides a kind of battery piece appearance defect fast detection device, including rack 1, the conveying mechanism 2 being arranged on the rack 1, visual inspection mechanism 3 located above the conveying mechanism 2 and the control system 101 being electrically connected with visual inspection mechanism 3, the conveying mechanism 2 includes the conveying belt 206 driven by servo motor 202 and the limit baffle 203 being arranged on the both sides of the conveying belt 206, the conveying belt 206 surface is uniformly distributed with anti-skid boss, the conveying belt 206 below corresponding visual inspection mechanism 3 is equipped with pneumatic sorting valve 102.
[0023] The working process of the battery piece appearance defect rapid detection device of the embodiment mainly includes battery piece transmission, image acquisition, image processing and analysis, defect sorting and information management, etc. In the whole process, the conveying mechanism 2 is responsible for stable transmission of the battery piece, the visual detection mechanism 3 acquires the front and side images of the battery piece, the control system 101 processes and analyzes the images, and controls the alarm and sorting actions according to the analysis results, and the code scanning module 201 realizes the identification and traceability of the battery piece information.
[0024] The servo motor 202 serves as a power source, drives the conveying belt 206 to transmit the battery piece at a stable speed by precisely controlling the rotating speed and direction. The limiting baffle 203 on both sides of the conveying belt 206 plays a role in limiting the lateral movement of the battery piece, ensuring that the battery piece moves along the predetermined transmission direction. Since the limiting baffle 203 is an elastic baffle with adjustable spacing and is provided with a pressure sensor on the inner side, the pressure sensor monitors the pressure between the battery piece and the baffle in real time. When the size of the battery piece changes, the control system 101 can adjust the spacing of the limiting baffle 203 according to the feedback signal of the pressure sensor, so that the device can adapt to the transmission of battery pieces of different sizes, ensure that the battery piece maintains a stable position during transmission, and avoid affecting the subsequent image acquisition due to position deviation.
[0025] The anti-skid protrusions uniformly distributed on the surface of the conveying belt 206 increase the friction between the conveying belt 206 and the battery piece, preventing the battery piece from slipping during transmission and ensuring that the battery piece can be accurately moved to the visual detection mechanism 3 below for detection. When the visual detection mechanism 3 detects a defective battery piece, the control system 101 sends a command to the sorting control module, which in turn controls the pneumatic sorting valve 102 located below the conveying belt 206 corresponding to the visual detection mechanism 3 to act, sorting the defective battery piece out of the conveying belt 206, realizing the automatic rejection of the defective battery piece.
[0026] The conveying mechanism 2 drives the conveying belt 206 with the servo motor 202, which can realize high-speed and stable transmission of the battery piece, and cooperate with the continuous shooting of the high-speed camera to complete the detection of a large number of battery pieces in a short time, meeting the detection needs of large-scale production.
[0027] The visual detection mechanism 3 includes a first high-speed camera 301 and a second high-speed camera 303 arranged in sequence along the transmission direction of the battery piece, the first high-speed camera 301 is used for acquiring the front image of the battery piece, and the second high-speed camera 303 is used for acquiring the side image of the battery piece.
[0028] The visual detection mechanism 3 sequentially arranges the first high-speed camera 301 and the second high-speed camera 303 along the transmission direction of the battery piece, which are respectively used for acquiring the front and side images of the battery piece. Under the cooperation of the light source module 204, clear images of the battery piece can be obtained.
[0029] The light source module 204 is arranged on both sides of the conveying belt 206, and includes a visible light LED array and a short-wave infrared light source. The visible light LED array is used in cooperation with the first high-speed camera 301, and the short-wave infrared light source is used in cooperation with the second high-speed camera 303.
[0030] During the conveying of the battery piece, the first high-speed camera 301 and the second high-speed camera 303 continuously capture the front and side images of the battery piece at a high speed in real time, and transmit the image data to the control system 101.
[0031] The light source module 204 includes a visible light LED array and a short-wave infrared light source, and a diffuse reflection plate 302 is arranged between the two light sources. The surface of the diffuse reflection plate 302 is sprayed with a high-reflectivity coating. The visible light emitted by the visible light LED array is used in cooperation with the first high-speed camera 301 to capture the front image of the battery piece, and is suitable for detecting the general appearance defects on the front of the battery piece, such as surface scratches, stains, cracks, etc. The diffuse reflection plate 302 makes the light emitted by the visible light LED array more uniformly irradiate on the front of the battery piece, reduces the reflection and shadow, and improves the quality of the front image.
[0032] Further, the rack 1 is also provided with a code scanning module 201 located at the entrance end of the conveying belt 206, and the code scanning module 201 is electrically connected with the control system 101.
[0033] The code scanning module 201 arranged on the rack 1 is located at the entrance end of the conveying belt 206 and is electrically connected with the control system 101. The code scanning module 201 is used to scan the two-dimensional code or bar code of the battery piece, obtain the related information of the battery piece such as production batch, model, etc., and transmit the information to the control system 101. The control system 101 associates these information with the detection results, which is convenient for subsequent quality tracing and management. The anti-skid protrusions of the conveying belt 206 prevent the battery piece from sliding, and the limiting baffle 203 and the pressure sensor ensure the stable conveying of the battery piece, avoiding the detection error caused by the position deviation of the battery piece, and improving the reliability of the detection process.
[0034] Further, the visual detection mechanism 3 further includes a multi-spectral filter switching mechanism 205 arranged between the first high-speed camera 301 and the second high-speed camera 303, and the multi-spectral filter switching mechanism 205 includes a driving motor and at least three filters of different wavelengths.
[0035] The short-wave infrared light emitted by the short-wave infrared light source cooperates with the second high-speed camera 303 to collect the image of the side of the battery piece. The short-wave infrared light has a certain penetration ability and can detect some defects on the side of the battery piece that are difficult to detect by the naked eye, such as side cracks and delamination. The multi-spectral filter switching mechanism 205 is arranged between the first high-speed camera 301 and the second high-speed camera 303 and includes a driving motor and at least three filters of different wavelengths. According to different detection requirements, the driving motor drives the filter switching to enable the camera to collect images at different wavelengths, further improving the detection capability of various defects.
[0036] Further, the light source module 204 further comprises a diffuse reflection plate 302 arranged between the visible light LED array and the short-wave infrared light source, and the surface of the diffuse reflection plate 302 is sprayed with a high reflectivity coating.
[0037] The diffuse reflection plate 302 in the light source module 204 uniformly distributes the light, reduces the influence of uneven illumination on the detection results, and ensures that high-quality images can be obtained under different illumination conditions.
[0038] The visual detection mechanism 3 adopts the first high-speed camera 301 and the second high-speed camera 303 to collect the front and side images of the battery piece, respectively, and combines the visible light LED array and the short-wave infrared light source to comprehensively and clearly obtain the appearance information of the battery piece. The visible light LED array is suitable for detecting regular defects on the front, the short-wave infrared light source can detect hidden defects on the side, and the multi-spectral filter switching mechanism 205 can switch the filter according to different detection requirements to collect images at different wavelengths, thereby improving the detection capability of various defects.
[0039] Further, the control system 101 comprises an FPGA real-time processing unit, an edge computing module with a built-in deep learning model, an alarm module connected with the edge computing module, and a sorting control module. The FPGA real-time processing unit is used for pre-processing the images collected by the first high-speed camera 301 and the second high-speed camera 303, the edge computing module is used for defect recognition based on the pre-processed images, the alarm module is used for issuing an audible and visual alarm when a defect is detected, and the sorting control module is used for controlling the pneumatic sorting valve 102.
[0040] The control system 101 is the core of the entire detection device and comprises an FPGA real-time processing unit, an edge computing module with a built-in deep learning model, an alarm module, and a sorting control module.
[0041] After the FPGA real-time processing unit receives the image data collected by the first high-speed camera 301 and the second high-speed camera 303, it immediately performs preprocessing on the image, including image noise reduction, contrast adjustment, edge detection, and other operations, to remove noise and interference information from the image and enhance the features of the image, providing high-quality image data for subsequent defect identification.
[0042] The edge computing module is built-in with a deep learning model that has been trained with a large number of battery sheet defect images and can perform in-depth analysis and defect identification on the preprocessed image. Through deep learning algorithms, the edge computing module can accurately identify various appearance defects of the battery sheet, such as scratches, cracks, and stains on the front surface, and hidden cracks and delamination on the side surface, and judge the type and severity of the defects.
[0043] When the edge computing module detects defects in the battery sheet, it sends a signal to the alarm module, which immediately issues an audible and visual alarm to alert the operator. At the same time, the sorting control module controls the pneumatic sorting valve 102 to act according to the instructions of the edge computing module, sorting out the defective battery sheet from the conveyor belt 206, achieving automatic rejection of defective battery sheets.
[0044] The edge computing module in the control system 101 is built-in with a deep learning model that has been trained with a large amount of data and can accurately identify various types of defects, reducing the occurrence of missed and false detections and improving the accuracy of defect detection. The pneumatic sorting valve 102 and the alarm module can timely reject defective battery sheets and issue an alarm, ensuring that defective battery sheets do not flow into the next process and ensuring the reliability of product quality.
[0045] Further, the limiting baffle 203 on both sides of the conveyor belt 206 is an elastic baffle with adjustable spacing, and a pressure sensor is arranged on the inner side of the limiting baffle 203, and the pressure sensor is electrically connected with the control system 101. The limiting baffle 203 on both sides of the conveyor belt 206 is an elastic baffle with adjustable spacing, which cooperates with the pressure sensor to automatically adjust the spacing according to the size of the battery sheet, adapt to the detection of battery sheets of different specifications, and improve the universality of the device.
[0046] It is apparent for a person skilled in the art that the present application is not restricted to the details of the above exemplary embodiments, but that it can be implemented in other concrete forms without departing from the spirit or the essential characteristics of the present application. Therefore, the embodiments should be considered as exemplary only, and not limiting, the scope of the present application being defined by the appended claims rather than the above description, and all changes coming within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims should not be construed as limiting the claims concerned.
Claims
1. A rapid inspection device for appearance defects of battery cells, comprising a frame, a conveying mechanism mounted on the frame, a visual inspection mechanism located above the conveying mechanism, and a control system electrically connected to the visual inspection mechanism, characterized in that, The conveying mechanism includes a servo motor driven conveyor belt and limiting baffles set on both sides of the conveyor belt. Anti-slip protrusions are evenly distributed on the surface of the conveyor belt, and a pneumatic sorting valve is provided below the visual inspection mechanism corresponding to the conveyor belt. The visual inspection mechanism includes a first high-speed camera and a second high-speed camera arranged sequentially along the battery cell transmission direction. The first high-speed camera is used to capture images of the front of the battery cell, and the second high-speed camera is used to capture images of the side of the battery cell. Light source modules are provided on both sides of the conveyor belt. The light source modules include a visible light LED array and a short-wave infrared light source. The visible light LED array is used in conjunction with the first high-speed camera, and the short-wave infrared light source is used in conjunction with the second high-speed camera.
2. The rapid detection device for appearance defects of battery cells according to claim 1, characterized in that: The frame is also equipped with a barcode scanning module located at the inlet end of the conveyor belt, and the barcode scanning module is electrically connected to the control system.
3. The rapid detection device for appearance defects of battery cells according to claim 1, characterized in that: The visual inspection mechanism also includes a multispectral filter switching mechanism disposed between the first high-speed camera and the second high-speed camera. The multispectral filter switching mechanism includes a drive motor and at least three filters of different wavelengths.
4. The rapid detection device for appearance defects of battery cells according to claim 1, characterized in that: The light source module also includes a diffuse reflector plate disposed between the visible light LED array and the short-wave infrared light source, and the surface of the diffuse reflector plate is coated with a high reflectivity coating.
5. The rapid detection device for appearance defects of battery cells according to claim 1, characterized in that: The control system includes an FPGA real-time processing unit, an edge computing module with a built-in deep learning model, an alarm module connected to the edge computing module, and a sorting control module. The FPGA real-time processing unit is used to preprocess the images captured by the first high-speed camera and the second high-speed camera. The edge computing module is used to identify defects based on the preprocessed images. The alarm module is used to issue an audible and visual alarm when a defect is detected. The sorting control module is used to control the pneumatic sorting valve.
6. The rapid detection device for appearance defects of battery cells according to claim 1, characterized in that: The limiting baffles on both sides of the conveyor belt are elastic baffles with adjustable spacing. A pressure sensor is provided on the inner side of the limiting baffle, and the pressure sensor is electrically connected to the control system.