Welding inspection method and system

Abstract

A welding inspection method and system. The welding inspection method comprises: when a battery cell (505) on a turntable (404) reaches an inspection station (405), a controller (401) controls a lifting mechanism on the turntable (404) to execute a lifting operation to lift the battery cell (505) to a preset height, and sends an image acquisition signal to an image acquisition system (403); in response to the image acquisition signal, the image acquisition system (403) performs image acquisition on a welding area of the battery cell (505) located at the preset height to obtain an acquired image, and sends the acquired image to an industrial personal computer (402); the industrial personal computer (402) performs welding inspection on the acquired image to obtain a welding inspection result of the battery cell (505), wherein the welding inspection at least comprises inspecting a weld bead area of the battery cell (505) on the acquired image.

Description

Welding detection method and system

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure is based on and claims the priority of Chinese patent application with application number 202410213396.1, application date February 27, 2024, and invention name “Welding Detection Method and System”. The entire content of this Chinese patent application is hereby incorporated into this disclosure as a reference. Technical Field

[0003] The present disclosure relates to the field of battery detection, and in particular to a welding detection method and system. Background Art

[0004] New energy batteries are being used more and more widely in life and industry. For example, new energy vehicles equipped with batteries have been widely used. In addition, batteries are also being increasingly used in energy storage fields.

[0005] In a power battery system, after the cylindrical bare cell is placed in the shell, to ensure the connection between the bare positive and negative electrodes, the positive electrode post, and the steel shell, penetration welding is required between the positive electrode post and the positive current collector plate, and simultaneously between the negative electrode and the negative current collector plate. After welding, the weld seam and the surface where the weld seam is located are inspected for dimensions and defects.

[0006] In the related art, after the battery cells are welded, manual inspection and judgment of weld defects are usually required. This has the problem of high manual inspection costs and the possibility of missed inspections due to visual fatigue.

[0007] Summary of the Invention

[0008] The main purpose of the present disclosure is to provide a welding detection method and system that can improve the accuracy and efficiency of weld detection.

[0009] The technical solution of the embodiment of the present disclosure is implemented as follows:

[0010] In a first aspect, an embodiment of the present disclosure provides a welding detection method, the welding detection method comprising:

[0011] When the battery cell on the turntable reaches the inspection station, the controller controls the lifting mechanism on the turntable to perform a lifting operation, lifts the battery cell to a preset height, and sends an image acquisition signal to the image acquisition system;

[0012] The image acquisition system acquires an image of the welding area of ​​the battery cell at the preset height in response to the image acquisition signal, obtains an acquired image, and sends the acquired image to the industrial computer;

[0013] The industrial computer performs welding detection on the collected image to obtain a welding detection result of the battery cell; wherein the welding detection at least includes detecting a weld area of ​​the battery cell on the collected image.

[0014] According to the above technical means, an automatic welding detection system is used to replace manual visual inspection to perform abnormal detection on the welding information of the welding area of ​​the battery cell, which can not only improve the accuracy and efficiency of weld detection, but also reduce labor costs and time costs; at the same time, the present invention controls the rotation of a turntable that can hold multiple batteries to realize the transfer of battery cells from the loading station to the detection station, making the transfer of battery cells more intelligent, which can improve the efficiency of battery cell production, and transporting battery cells to the welding station by the turntable to ensure the consistency of incoming battery cells; finally, the battery cells are raised to a preset height by a lifting mechanism to ensure that each battery cell is in the same position, avoid the problem of inaccurate image taking position of the image acquisition system, and avoid problems in image extraction caused by fluctuations in the depth of field of the camera of the image acquisition system.

[0015] In the above scheme, a plurality of cups for placing battery cells are provided on the turntable; the welding detection method further includes: when the battery cell is placed on the cup at the loading station, the controller controls the rotation of the turntable to drive the battery cell from the loading station to the detection station; when the battery cell moves to the detection station, the controller sends the acquisition signal to the image acquisition system.

[0016] According to the above technical means, when the battery cell is placed in the support cup at the loading station, the controller drives the support cup to the inspection station, thereby sending an acquisition signal to the visual inspection system. This not only allows the controller to more accurately determine when to trigger the acquisition signal to the image acquisition system, thereby improving the accuracy of the subsequent acquired images; but also by controlling the rotation of the turntable capable of holding multiple batteries, the battery cells can be transferred from the loading station to the inspection station, making the transfer of battery cells more intelligent and improving the efficiency and intelligence of battery cell production. At the same time, the turntable transports the battery cells to the welding station, ensuring the consistency of the incoming battery cells.

[0017] In the above scheme, the industrial computer performs welding detection on the acquired image to obtain the welding detection result of the battery cell, including: the industrial computer performs image segmentation on the acquired image to obtain a weld area and a non-weld area; the industrial computer performs at least one defect detection on at least one of the weld area and the non-weld area through a trained detection model to obtain at least one defect detection result; wherein, the trained model is trained based on sample images with marked defects; the defect detection results include at least welding slag detection results, burst point detection results, breakpoint detection results, weld offset results and size detection results; based on the at least one defect detection result, the welding detection result of the battery cell is obtained.

[0018] Based on the above technical means, the image acquisition system and industrial computer are used to detect defects such as pinholes, explosion points, weld wire deviation, weld slag residue, weld wire size, broken welds, and cold welds in through-hole welding. No manual inspection is required, which not only improves the accuracy and efficiency of weld detection, but also reduces labor and time costs.

[0019] In the above scheme, the industrial computer performs at least one defect detection on at least one of the weld area and the non-weld area through the trained detection model, and obtains at least one defect detection result, including: the industrial computer performs welding slag detection on the non-weld area through the trained detection model to obtain a welding slag detection result; when the welding slag result indicates that welding slag exists in the non-weld area, the industrial computer measures the size of the welding slag to obtain the welding slag size; and compares the welding slag size with a preset welding slag size to obtain the welding slag detection result.

[0020] According to the above technical means, whether there is welding slag in the welding area is detected and the identified welding slag is further judged based on the preset welding slag size, which can improve the accuracy of welding slag identification in the welding area of ​​the battery cell.

[0021] In the above scheme, the industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area through the trained detection model, and obtains at least one defect detection result, including: the industrial computer detects the brightness values ​​of all pixels corresponding to the weld bead area through the trained detection model to obtain the brightness value of each pixel; based on the brightness value of each pixel, the number of hot spot areas in the weld bead area is determined; wherein the value of the pixel in the hot spot area is greater than a preset value, and the brightness values ​​of the pixel in the hot spot area are all greater than the preset brightness; and the hot spot detection result is obtained based on the number of hot spot areas.

[0022] According to the above technical means, by detecting the relatively small pinholes and explosion point areas on the weld, the problem of missed detection and false detection during visual inspection is avoided, and the accuracy of detecting small defects on the weld is improved.

[0023] In the above scheme, the industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area through the trained detection model to obtain at least one defect detection result, including: the industrial computer performs contour extraction on the weld bead area through the trained detection model to obtain the number of closed contours corresponding to the weld bead area; and obtains the breakpoint detection result based on the number of contours.

[0024] According to the above technical means, the breakpoints on the weld are detected by the number of closed contours, which avoids the problem of missed detection and false detection during visual inspection, and improves the accuracy of detecting discontinuous defects in the weld.

[0025] In the above scheme, the industrial computer performs at least one defect detection on at least one of the weld area and the non-weld area through a trained detection model to obtain at least one defect detection result, including: the industrial computer determines the reference point of the battery cell on the acquired image through the trained detection model and the acquired image; performs center point fitting on the weld area to obtain the weld center of the weld area, and determines the offset distance between the weld center and the reference point; compares the offset distance with a preset offset distance to obtain the weld offset result.

[0026] According to the above technical means, the distance value between the weld bead and the reference point can be quickly determined, thereby improving the accuracy and efficiency of offset detection of the weld bead.

[0027] In the above scheme, the industrial computer performs at least one defect detection on at least one of the weld area and the non-weld area through the trained detection model, and obtains at least one defect detection result, including: the industrial computer performs straight line fitting on the weld area through the trained detection model to obtain a straight weld corresponding to the weld area; performs a first measurement on the straight weld to obtain a first measurement size of the straight weld; segments the straight weld along the length direction of the straight weld to obtain multiple segmented areas; performs a second measurement on multiple non-adjacent target segmented areas in the multiple segmented areas to obtain multiple second measurement sizes; and compares the first measurement size with a first preset size interval, and multiple second measurement sizes with a second preset size interval, to obtain a size detection result.

[0028] According to the above technical means, not only can it be quickly determined whether there are dimensional abnormalities in the weld area, but it also reduces the detection workload of the industrial computer and improves the detection efficiency and speed.

[0029] In the above scheme, the welding detection result of the battery cell is obtained based on the at least one defect detection result, including: when there are no abnormalities in the welding slag detection result, the burst point detection result, the breakpoint detection result, the weld offset result and the size detection result, obtaining a welding detection result characterizing that the battery cell is normal; when there is a detection abnormality, obtaining a welding detection result characterizing that the battery cell is abnormal; wherein, the detection abnormality includes at least one of the following: the welding slag detection result is characterized by the welding slag size being greater than the preset welding slag size, the burst point detection result is characterized by the number of burst point areas being greater than a first preset number, the breakpoint detection result is characterized by the number of contours being greater than a second preset number, the weld offset result is characterized by the offset distance being greater than the preset offset distance, and the size detection result is characterized by the first measurement dimension being outside the first preset size interval, and multiple second measurement dimensions being outside the second preset size interval.

[0030] According to the above technical means, by detecting defects such as pinholes, burst points, welding wire deviation, welding slag residue, welding wire size, broken welds, and cold welds, it is determined whether the welding of the battery cell is normal. The welding results can be tested in multiple dimensions and aspects, reducing the risk of false detection and missed detection, and improving detection accuracy.

[0031] In the above scheme, the welding detection method further includes: when the welding detection result indicates that the battery cell is abnormal, the industrial computer sends a marking signal to the controller; and the controller controls the marking mechanism to mark the battery cell with abnormal detection based on the marking signal.

[0032] According to the above technical means, if the welding detection result is abnormal, the industrial computer sends a marking signal to the controller, so that the controller can promptly mark the battery cell. In this way, the relevant marking operation can be executed immediately when the abnormality is detected, so as to identify defective products and issue an alarm, thereby quickly preventing the outflow of defective products and the resulting mass scrapping.

[0033] In the above scheme, the welding detection method also includes: the industrial computer sends an end signal containing the welding detection result to the controller; the controller controls the rotation of the turntable based on the end signal to drive the battery cell from the detection station to the unloading station.

[0034] In a second aspect, an embodiment of the present disclosure provides a welding detection system, which includes: a controller, configured to control a lifting mechanism on a turntable to perform a lifting operation when a battery cell on the turntable reaches a detection station, thereby lifting the battery cell to a preset height and sending an image acquisition signal to an image acquisition system; the image acquisition system, configured to perform image acquisition on a welding area of ​​the battery cell at the preset height in response to the image acquisition signal, obtain an acquired image, and send the acquired image to an industrial computer; the industrial computer, configured to perform welding detection on the acquired image to obtain a welding detection result of the battery cell; wherein the welding detection at least includes detecting the weld area of ​​the battery cell on the acquired image.

[0035] According to the above technical means, an automatic welding detection system is used to replace manual visual inspection to perform abnormal detection on the welding information of the welding area of ​​the battery cell, which can not only improve the accuracy and efficiency of weld detection, but also reduce labor costs and time costs; at the same time, the present invention controls the rotation of a turntable that can hold multiple batteries to realize the transfer of battery cells from the loading station to the inspection station, making the transfer of battery cells more intelligent, which can improve the efficiency of battery cell production, and transporting battery cells to the welding station by the turntable to ensure the consistency of incoming battery cells.

[0036] In the above scheme, the jacking mechanism includes at least a jacking drive, a cam, a jacking transmission, an upper top plate and a limit plate; the controller is also used to send a jacking signal to the jacking drive when the battery cell on the turntable reaches the detection station; the cam is slidably connected to the turntable and connected to the jacking drive; the cam is provided with a slope structure; the jacking transmission is connected to the upper top plate at one end and located on the slope structure at the other end; the upper top plate is used to support the battery cell; the limit plate is provided on the turntable and has the preset height; the jacking drive, one end of the jacking drive is connected to the turntable and the other end is connected to the cam, and is used to drive the cam to slide on the turntable in response to the jacking signal to drive the jacking transmission to move along the first direction until the battery cell contacts the limit plate.

[0037] According to the above technical means, the consistent position of each battery cell is ensured, which avoids the problem of inaccurate image acquisition position of the image acquisition system and the problem of image extraction caused by fluctuation of the depth of field of the camera of the image acquisition system.

[0038] In the above solution, the turntable further includes a plurality of turntable supporting cups; the inner root of the turntable supporting cups is provided with an avoidance groove, and the turntable supporting cups are used to place the battery cells.

[0039] According to the above technical means, it is possible to avoid the accumulation of dust and impurities at the root of the inner side of the turntable cup after long-term use, which may cause battery cell contamination or battery cell jamming in the turntable cup.

[0040] In the above solution, the inspection station includes at least two image acquisition systems; the at least two image acquisition systems are used to respectively respond to the image acquisition signals to acquire images of the welding area of ​​the battery cell to obtain the acquired images.

[0041] In the above scheme, the image acquisition system includes at least a light source, a camera, a locking structure and a slide rail structure; the camera is fixed on the inspection station through the locking structure and the slide rail structure, and is used to capture images of the welding area of ​​the battery cell to obtain the captured image; the light source is fixed on the inspection station through the locking structure and the slide rail structure, and is located between the camera and the turntable, and is used to provide light for the camera; the slide rail structure is used to adjust the positions of the camera and the light source on the inspection station.

[0042] According to the above technical means, the locking structure and the slide rail structure enable the image acquisition system to perform axial adjustment of the camera and light source to be compatible with battery cells of different lengths and sizes, and can be applied to more usage scenarios.

[0043] The above description is only an overview of the technical solution of the present disclosure. In order to more clearly understand the technical means of the present disclosure, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present disclosure more obvious and easy to understand, the specific implementation methods of the present disclosure are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and, together with the specification, serve to illustrate the technical solutions of the present disclosure. Obviously, the drawings described below are only some embodiments of the present disclosure. It is clear that those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0045] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0046] FIG1 is a flow chart of an optional welding detection method provided in an embodiment of the present disclosure;

[0047] FIG2 is a schematic structural diagram of a turntable provided in an embodiment of the present disclosure;

[0048] FIG3 is a second flow chart of an optional welding detection method provided in an embodiment of the present disclosure;

[0049] FIG4 is a structural diagram 1 of a welding detection system provided in an embodiment of the present disclosure;

[0050] FIG5 is a schematic structural diagram of a lifting mechanism provided in an embodiment of the present disclosure;

[0051] FIG6 is a second structural diagram of a welding detection system provided in an embodiment of the present disclosure;

[0052] FIG7 is a schematic diagram of a detection process provided by an embodiment of the present disclosure;

[0053] FIG8 is a schematic diagram of the layout of the inspection stations provided in an embodiment of the present disclosure;

[0054] FIG9 is a schematic diagram of a detection station provided in an embodiment of the present disclosure;

[0055] FIG10 is a schematic diagram of a detection communication process according to an embodiment of the present disclosure;

[0056] FIG11 is a schematic diagram of a pinhole or explosion point on a weld provided by an embodiment of the present disclosure;

[0057] FIG12 is a schematic diagram of welding slag residue provided by an embodiment of the present disclosure;

[0058] FIG13 is a schematic diagram of a weld bead offset provided by an embodiment of the present disclosure;

[0059] FIG. 14 is a schematic diagram of weld bead dimensions according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] To make the purpose, technical solutions and advantages of the embodiments of the present disclosure more clear, the specific technical solutions of the present disclosure will be further described in detail below in conjunction with the drawings in the embodiments of the present disclosure. The following embodiments are used to illustrate the present disclosure, but are not intended to limit the scope of the present disclosure.

[0061] Unless otherwise defined, all technical and scientific terms used in this disclosure have the same meaning as commonly understood by those skilled in the art in the art of this disclosure. The terms used in this disclosure are only for the purpose of describing the present embodiment and are not intended to limit this disclosure.

[0062] In the following description, references to “some embodiments,” “this embodiment,” “this embodiment,” and examples, etc., describe a subset of all possible embodiments, but it can be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict.

[0063] If similar descriptions of "first / second" appear in the application documents, the following explanation is added. In the following description, the terms "first\second\third" are merely used to distinguish similar objects and do not represent a specific order of the objects. It can be understood that "first\second\third" can be interchanged with the specific order or sequence where permitted, so that the embodiment described herein can be implemented in an order other than that illustrated or described herein.

[0064] In the embodiments of the present disclosure, the term "and / or" is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, object A and / or object B may represent three situations: object A exists alone, object A and object B exist at the same time, and object B exists alone.

[0065] At present, the application of new energy batteries in life and industry is becoming more and more extensive. New energy batteries are not only used in energy storage power systems such as hydropower, thermal power, wind power and solar power stations, but are also widely used in electric vehicles such as electric bicycles, electric motorcycles, electric cars, as well as in aerospace and other fields. With the continuous expansion of the application field of power batteries, the market demand is also constantly expanding. In the embodiments of the present disclosure, the battery involved may be a battery cell. A battery cell refers to a basic unit that can realize the mutual conversion of chemical energy and electrical energy, which can be used to make a battery module or battery pack, thereby being used to supply power to electrical devices. A battery cell may be a secondary battery, which refers to a battery cell that can continue to be used by activating the active material by charging after the battery cell is discharged. The battery cell may be a lithium-ion battery, a sodium-ion battery, a sodium-lithium-ion battery, a lithium metal battery, a sodium metal battery, a lithium-sulfur battery, a magnesium-ion battery, a nickel-hydrogen battery, a nickel-cadmium battery, a lead-acid battery, etc., and the embodiments of the present disclosure are not limited to this.

[0066] In the embodiment of the present disclosure, the battery may also be a single physical module including one or more battery cells to provide higher voltage and capacity. When there are multiple battery cells, the multiple battery cells are connected in series, in parallel or in hybrid through a busbar.

[0067] In related technologies, welding the positive and negative electrodes of batteries is a very important step in the battery assembly process. Abnormal welding can easily lead to serious consequences such as battery short circuit and explosion. Among them, after the cylindrical lithium battery cell is placed in the shell, in order to ensure that the positive and negative electrodes of the bare cell and the positive electrode column and the steel shell are overlapped, the back cover of the cylindrical lithium battery cell is closed with the steel shell, and the positive electrode column is welded to the positive electrode collector plate, and the back cover (corresponding to the negative electrode of the cell) is welded to the negative electrode collector plate. During the battery welding process, a weld bead will be left on the battery shell. Due to the problem of unstable welding process, there may be various defects on the weld bead, such as weld bead offset, pinholes, pits, explosion points, protrusions, etc.

[0068] Therefore, battery welds need to be quality inspected to ensure battery performance and safety. In the field of battery weld inspection, manual visual inspection is usually performed, which has high manual inspection costs and is prone to missed inspections due to visual fatigue.

[0069] Based on the above technical problems, the present inventors believe that an inspection station can be set up on the battery production line. After the battery welding is completed, the battery is moved to the inspection station by a turntable, and the welding area of ​​the battery is automatically inspected by the inspection station. This not only improves the inspection efficiency, but also can achieve the accuracy of detection of various post-weld defect items through automatic inspection, thereby reducing labor costs and time costs.

[0070] Based on the above-mentioned inventive concept, an embodiment of the present disclosure provides a welding detection method, which is applied to a welding detection system of a battery production line. The welding detection system includes at least an industrial computer, a controller, and an image acquisition system at a detection station. The controller can be any one of a programmable logic controller (PLC), a single-chip microcomputer, an intermediate computer, and a host computer. The image acquisition system can refer to a two-dimensional camera system arranged at the detection station. The industrial computer can be a host computer. Visual inspection software can be installed in the industrial computer for performing welding inspection on images of batteries captured by the two-dimensional camera system.

[0071] In the welding detection method provided by the embodiment of the present disclosure, when the battery cell on the turntable reaches the detection station, the controller controls the lifting mechanism on the turntable to perform a lifting operation, lifts the battery cell to a preset height, and sends an image acquisition signal to the image acquisition system. In response to the image acquisition signal, the image acquisition system performs image acquisition on the welding area of ​​the battery cell at the preset height to obtain a captured image, and sends the captured image to the industrial computer. The industrial computer performs welding detection on the captured image to obtain the welding detection result of the battery cell.

[0072] In this way, the embodiment of the present disclosure replaces manual visual inspection with an automatic welding detection system to detect abnormalities in the welding information of the welding area of ​​the battery cell, which can not only improve the accuracy and efficiency of weld detection, but also reduce labor costs and time costs; at the same time, the present disclosure controls the rotation of a turntable that can hold multiple batteries to realize the transfer of battery cells from the loading station to the inspection station, making the transfer of battery cells more intelligent and improving the efficiency of battery cell production.

[0073] Batteries inspected using the welding inspection method disclosed in the embodiments of this disclosure can be used, but are not limited to, in electrical devices such as vehicles, ships, or aircraft. A power supply system comprising the battery pack and batteries disclosed herein can be used to mitigate and automatically regulate deterioration in cell expansion, replenish electrolyte consumption, and improve battery performance stability and battery life.

[0074] The battery detected by the welding detection method disclosed in the embodiments of the present disclosure can be used as a power source for an electrical device, and the electrical device can include but is not limited to a mobile phone, a tablet, a laptop computer, an electric toy, an electric tool, a battery-powered vehicle, an electric car, a ship, a spacecraft, etc. Among them, the electric toy can include a fixed or mobile electric toy, such as a game console, an electric car toy, an electric ship toy, and an electric airplane toy, etc. The spacecraft can include an airplane, a rocket, a space shuttle, and a spacecraft, etc.

[0075] The present disclosure provides a welding detection method to improve the accuracy and efficiency of battery welding detection. The technical solution of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0076] Referring to FIG. 1 , FIG. 1 is a flow chart illustrating an optional welding detection method provided in an embodiment of the present disclosure. The method is applied to a welding detection system, which includes a controller, an image acquisition system, and an industrial computer. The welding detection method is implemented through steps S101 to S103:

[0077] Step S101: When the battery cell on the turntable reaches the inspection station, the controller controls the lifting mechanism on the turntable to perform a lifting operation, lifts the battery cell to a preset height, and sends an image acquisition signal to the image acquisition system.

[0078] In the disclosed embodiment, the controller (PLC) is the center of the system, responsible for overall control and coordination. The controller communicates with the industrial computer to receive the detection results of the industrial computer and / or controls the image acquisition system to execute relevant instructions.

[0079] The image acquisition system may refer to a camera system at an inspection station, including components such as a camera, a light source, and a display interface. The display interface may be used to display the captured image captured by the camera, or it may be used to receive the detection results of the industrial computer after the inspection and display them on the display interface. In some embodiments, the display interface may be a human-machine interface (HMI), which allows on-site engineers to adjust parameters of the image acquisition system, such as exposure, to adjust the quality of the captured image. Here, the camera may be a charge coupled device (CCD) camera, or a two-dimensional camera or a three-dimensional camera capable of image acquisition.

[0080] It should be noted that the inspection station can be a station for performing abnormality detection on the welding area of ​​the battery cell. The welding station is before the inspection station, and the welded battery cell reaches the inspection station via a conveyor belt and a turntable. Among them, the abnormality detection of the welding area of ​​the battery cell includes but is not limited to: detecting whether the weld of the battery cell is offset, whether the weld size is abnormal (but not limited to: the length, width and height of the weld), whether there is welding slag in the area outside the weld (here, the area outside the weld can be the area adjacent to the weld within the welding area), and whether there are cracks, protrusions, pinholes, etc. in the weld.

[0081] In the embodiments of the present disclosure, the welding area of ​​a battery cell may refer to the area on the battery casing where the weld bead is located after the battery cell is welded (e.g., during the process of welding the positive and negative electrodes of the battery, or during the welding process after the battery pack is mylared). Here, the welding area of ​​the battery cell includes but is not limited to: the weld bead area on the battery cell, the area outside the weld bead adjacent to the weld bead area, etc. Here, the welding area of ​​the battery cell may be located anywhere on the battery cell, such as: the top cover area of ​​the battery cell, the side of the battery cell, the bottom cover of the battery cell, etc.

[0082] In the embodiments of the present disclosure, the battery cell may refer to a battery cell of any shape, such as a square battery cell, a round battery cell, etc. And the battery cell usually refers to a battery cell (Battery Cell), which is one of the basic units that constitute the battery. The battery cell is the core component of the battery, responsible for storing and releasing electrical energy. The battery cell can be: a lithium-ion battery cell (Li-ion Cell), a lithium polymer battery cell (Li-polymer Cell), a nickel-metal hydride battery cell (NiMH Cell), etc. The embodiments of the present disclosure do not impose any restrictions on the type of battery cell, and the specific selection can be made according to the actual application scenario.

[0083] In the embodiment of the present disclosure, the battery cell is the core component of the battery pack, and a battery pack usually contains multiple battery cells, which are combined together to provide the required power capacity and voltage. Among them, the battery pack refers to a device composed of multiple battery cells, which is intended to store electrical energy and provide power supply. The components of the battery pack include at least: battery cells, battery management system (BMS, Battery Management System), casing, connecting wiring harness, connector and interface, etc. These components work together to combine the battery cells into a fully functional battery pack for various application scenarios. For example, the battery pack can be applied to electric vehicles, energy storage systems, portable electronic devices, solar energy systems, wind energy systems, emergency backup power supplies, power tools or electric bicycles, etc. The embodiment of the present disclosure does not make any restrictions on this, and the specific selection can be made according to the actual application scenario.

[0084] It should be noted that the battery pack can use different types of battery cells, such as lithium-ion batteries, nickel-metal hydride batteries, lithium polymer batteries, etc., depending on the actual application needs and performance requirements.

[0085] In the embodiment of the present disclosure, a turntable is provided on the inspection station, and the battery cells welded at the welding station are transferred to the inspection station via the turntable. When the battery cells on the turntable arrive at the inspection station, the controller controls the lifting mechanism on the turntable to perform a lifting operation, lifts the battery cells to a preset height, and sends an image acquisition signal to the image acquisition system to perform image acquisition.

[0086] Here, the controller uses sensors or other detection equipment installed in fixed locations to detect the position, signal, or other specific characteristics of the battery cell to determine whether the battery cell has reached the inspection station. Once the controller confirms that the battery cell has reached the inspection station, it sends a lifting signal to the lifting mechanism to control the lifting mechanism to perform the lifting operation, raising the battery cell to a preset height, and controlling the image acquisition system to capture the welding area of ​​the battery cell at the preset height.

[0087] It can be understood that the welding detection method provided in the embodiment of the present disclosure ensures that the height of each battery cell during detection is consistent through the jacking mechanism in the turntable, that is, the distance from the image acquisition system to the surface of the battery cell to be tested is the same, avoiding errors in image acquisition due to fluctuations in the depth of field of the camera, resulting in inaccurate detection results.

[0088] Step S102 : The image acquisition system acquires an image of the welding area of ​​the battery cell at the preset height in response to the image acquisition signal to obtain an acquired image, and sends the acquired image to the industrial computer.

[0089] In some embodiments, the inspection station may include at least two image acquisition systems to capture images of two battery cells respectively. In order to ensure the accuracy of the image, multiple images can be captured for one battery cell. The acquisition frequency corresponding to the image capture of the battery cell can be determined according to actual needs, so the number of captured images can be one, two or more.

[0090] In some embodiments, the image acquisition system controls the camera to move along the edge of the battery cell, or the battery cell itself moves under the camera, to obtain complete information about the weld area of ​​the battery cell. In this way, edge scanning to achieve image acquisition can help detect surface anomalies or other problems in the weld area.

[0091] In the embodiment of the present disclosure, the captured image may include all information about the surface features of the welding area of ​​the battery cell, such as shape, size, color, etc., so that subsequent image processing and analysis can detect whether there are defects in the welding area. Exemplarily, the captured image may include the following: the shape and size of the weld in the welding area: the shape, diameter, length, width and other dimensional information of the weld; the quality of the weld surface, including flatness, uniformity and possible defects or damage, such as: burst points, pinholes, protrusions or depressions, etc.; welding slag information outside the weld in the welding area: information such as the shape, position and size of the welding slag; color information of the welding area, such as the surface color of the welding area. In actual application scenarios, the captured image may also include other display content, which is not limited in the embodiment of the present disclosure.

[0092] In some embodiments, after the image acquisition system acquires the acquired image, the acquired image is sent to the industrial computer, and the industrial computer detects the welding area of ​​the battery cell based on the acquired image.

[0093] Step S103: The industrial computer performs welding detection on the collected image to obtain a welding detection result of the battery cell; wherein the welding detection at least includes detecting the weld area of ​​the battery cell on the collected image.

[0094] In the disclosed embodiment, the battery cell on the captured image includes at least a weld area and a non-weld area, and the industrial computer performs abnormality detection on the welding area of ​​the captured image, including but not limited to: whether the weld is offset, whether the weld size is abnormal (including but not limited to: the length of the weld, the width of the weld, the height of the weld), whether there is welding slag in the area outside the weld (here, the area outside the weld can be a non-weld area adjacent to the weld within the welding area), whether there are burst points, protrusions, pinholes, etc. in the weld.

[0095] In the disclosed embodiment, visual inspection software may be installed in the industrial computer. After receiving the captured image, the industrial computer detects the welding area of ​​the battery cell on the captured image through the visual inspection software. Here, the visual inspection software may be based on a trained model to detect the captured image. The trained inspection model may be trained by using sample images with defects marked and supervised learning. The sample image may be an image marked with information such as welding slag information, weld burst point information, weld breakpoint information, weld offset information, and weld size anomaly. The inspection model inspects the sample image to obtain a sample inspection result, compares the sample inspection result with the annotation information of the sample image, determines the similarity between the sample inspection result and the annotation information, determines the loss of the inspection model based on the similarity, and corrects the parameters of the inspection model through the loss to obtain a trained model, such as adjusting the weight value so that the loss of the welding slag information output by the trained model converges.

[0096] In some embodiments, after obtaining the welding detection results of the battery cells, the industrial computer can send the welding detection results to the controller. Based on the received welding detection results, the controller continues to transport the battery cells to the unloading station, and at the unloading station, determines whether to transport the battery cells to the abnormal marking station or the next station corresponding to normal operation, such as the battery cell grouping station, based on the welding detection results.

[0097] In the disclosed embodiments, welding inspection results can be expressed in text, such as: normal weld bead size inspection, abnormal weld bead size inspection, welding slag present in the welding area of ​​the battery cell, no welding slag present in the welding area of ​​the battery cell, offset of the weld bead of the battery cell, no offset of the weld bead of the battery cell, and welding abnormality of the weld bead of the battery cell. The welding inspection results can be displayed on the display interface of the industrial computer.

[0098] The embodiment of the present disclosure replaces manual visual inspection with an automatic welding detection system to perform abnormal detection on the welding information of the welding area of ​​the battery cell, which can not only improve the accuracy and efficiency of weld detection, but also reduce labor costs and time costs; at the same time, the present disclosure controls the rotation of a turntable that can hold multiple batteries to realize the transfer of battery cells from the loading station to the inspection station, making the transfer of battery cells more intelligent, which can improve the efficiency of battery cell production, and transporting battery cells to the welding station by the turntable to ensure the consistency of incoming battery cells.

[0099] In some embodiments of the present disclosure, a plurality of cups for placing battery cells are provided on the turntable. FIG2 is a schematic structural diagram of the turntable provided in an embodiment of the present disclosure. As shown in FIG2 , the turntable is provided with eight cups, namely, the first cup 1 to the eighth cup 8, for holding battery cells. Currently, the first cup 1 and the second cup 2 on the turntable can be located at a loading station, which includes two loading ports, such as the first loading port 201 and the second loading port 202. After welding, the battery cells are transported to the first loading port 201 and the second loading port 202, and the battery cells in the first loading port 201 and the second loading port 202 are placed in the first cup 1 and the second cup 2 by a robot. The third and fourth cups 3 and 4 are located at an inspection station, which includes at least two inspection ports, such as the first inspection port 203 and the second inspection port 204. The image acquisition system is located at the first inspection port 203 and the second inspection port 204 to capture images of the battery cells of the third and fourth cups 3 and 4. The sixth cup 6 can be located at a blanking station, which can include at least one blanking port 205. When the cup on the turntable rotates to the blanking port 205, the grasping robot corresponding to the blanking port 205 grasps the battery cell of the sixth cup 6 and transports it to the next station. Among them, the fifth, seventh, and eighth cups 5, 7, and 8 can be cups located at other operating stations, for example, stations for processing abnormal battery cells.

[0100] Here, when the turntable rotates and the battery cells on the first cup 1 and the second cup 2 are moved to the first inspection port 203 and the second inspection port 204, the seventh cup 7 and the eighth cup 8 are located at the first loading port 201 and the second loading port 202. At this time, the first loading port 201 and the second loading port 202 can place the battery cells on the seventh cup 7 and the eighth cup 8, and the transportation is circulated in this way, thereby improving the inspection efficiency.

[0101] Based on the turntable with multiple cups shown in FIG2 , the welding detection method provided by the embodiment of the present disclosure may further include step S1 and step S2:

[0102] Step S1: When the battery cell is placed on the support cup at the loading station, the controller controls the turntable to rotate to drive the battery cell from the loading station to the inspection station.

[0103] In some embodiments, the loading station can be connected to the welding station. After welding, the battery cells are transported to the loading station via a conveyor belt. A robot with a gripper can place the battery cells in a cup at the loading station. A controller controls the rotation of the turntable on which the cup is located to drive the cup to the inspection station. The loading station can be adjacent to the inspection station or remote from it, and this is not limited in the present disclosure.

[0104] It should be noted that multiple cups are arranged on the turntable, which can be used to hold several battery cells. The battery cells held in the cups can be used for operations such as liquid injection, welding, or welding detection. At the same time, the multiple cups arranged on the turntable have corresponding shapes and sizes that match the battery cells or batteries. Here, the number of cups on the turntable can be 8, 10, etc. Here, the number of cups can be determined according to actual production needs; there is no limitation on this in the embodiments of the present disclosure.

[0105] Step S2: When the battery cell moves to the detection station, the controller sends the acquisition signal to the image acquisition system.

[0106] In an embodiment of the present disclosure, when the cup with the battery cell is transported to the inspection station, the sensor of the inspection station detects that the battery cell is in place and sends a position signal to the controller, so that the controller sends an acquisition signal to the image acquisition system.

[0107] Through the above steps, when the battery cell is placed in the support cup at the loading station, the controller drives the support cup to the inspection station, thereby sending a collection signal to the visual inspection system. This not only allows the controller to more accurately determine when to trigger the collection signal to the image acquisition system, thereby improving the accuracy of the subsequent captured images; but also by controlling the rotation of the turntable capable of holding multiple batteries, the battery cells can be transferred from the loading station to the inspection station, making the transfer of battery cells more intelligent and improving the efficiency and intelligence of battery cell production. At the same time, the turntable transports the battery cells to the welding station, ensuring the consistency of the incoming battery cells.

[0108] In the disclosed embodiment, the industrial computer can perform welding detection on the collected image through a trained detection model to obtain the welding detection result of the battery cell. The detection model can be trained by using sample images with defects marked and supervised learning. The sample image can be an image marked with information such as welding slag information, weld burst point information, weld breakpoint information, weld offset information, and weld size anomaly. The detection model detects the sample image to obtain a sample detection result, compares the sample detection result with the labeled information of the sample image, determines the similarity between the sample detection result and the labeled information, determines the loss of the detection model based on the similarity, and corrects the parameters of the detection model through the loss to obtain a trained model, such as adjusting the weight value so that the loss of the welding slag information output by the trained model converges.

[0109] In some embodiments, FIG3 is a second flow chart of an optional welding detection method provided by an embodiment of the present disclosure. As shown in FIG3 , step S103 can be implemented through steps S301 to S303:

[0110] Step S301: The industrial computer performs image segmentation on the acquired image to obtain weld bead areas and non-weld bead areas.

[0111] In the disclosed embodiments, image segmentation is a natural step in coarse-to-fine reasoning. The goal of segmentation is to label each pixel in an image with a corresponding class. That is, image segmentation is to assign each pixel in an image to a class.

[0112] Here, after image segmentation, the weld bead area and the non-weld bead area are obtained, wherein the weld bead area and the non-weld bead area are adjacent to each other, the weld bead area refers to the welded area on the battery cell, and the non-weld bead area refers to the area on the battery cell where no welding is performed.

[0113] Here, the industrial computer divides the welded area and non-welded area of ​​the captured image so that the range of subsequent defect detection can be located in a more precise area, thereby obtaining more accurate measurement results.

[0114] Step S302: The industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area using a trained detection model to obtain at least one defect detection result; wherein the trained model is trained based on sample images with marked defects; and the defect detection result includes at least a slag detection result, a burst point detection result, a breakpoint detection result, a weld bead offset result, and a size detection result.

[0115] In some embodiments, the industrial computer can detect whether the weld bead area is offset, whether the battery cell's weld bead dimensions are abnormal (including but not limited to: whether the weld bead length, width, and height are abnormal), and whether the battery cell's weld bead has defects such as cracks, protrusions, and pinholes. It can also detect whether non-weld bead areas contain weld slag. Therefore, defect detection results include at least weld slag detection results, cracks detection results, breakpoint detection results, weld bead offset results, and size detection results.

[0116] The industrial computer can detect the weld area and the non-weld area synchronously or asynchronously, and the embodiments of the present disclosure do not impose any limitation on this.

[0117] Step S303: Obtain the welding detection result of the battery cell based on the at least one defect detection result.

[0118] In some embodiments, when there are no abnormalities in the welding slag detection results, the explosion point detection results, the breakpoint detection results, the weld bead offset results, and the size detection results, the welding detection result of the battery cell is normal.

[0119] In some embodiments, when at least one of the welding slag detection result, the explosion point detection result, the breakpoint detection result, the weld bead offset result, and the size detection result is abnormal, the welding detection result of the battery cell is abnormal.

[0120] The disclosed embodiment uses an image acquisition system and an industrial computer to detect defects such as pinholes, burst points, weld line deviation, weld slag residue, weld line size, broken welds, and cold welds in through-the-wire welding without the need for manual inspection. This not only improves the accuracy and efficiency of weld inspection, but also reduces labor and time costs.

[0121] In some embodiments, welding slag has a significant impact on welding quality, such as: welding slag will reduce the mechanical properties of the weld. Due to the presence of welding slag, the strength of the weld will be affected, thereby affecting the strength and stability of the entire welded structure; welding slag will reduce the sealing performance of the weld joint. Welding slag will enlarge the gaps in the weld joint, thereby affecting the sealing performance of the weld joint and posing a potential safety hazard to the welded structure. Therefore, corresponding measures need to be taken to reduce the generation of welding slag and identify welding slag information to ensure the stability and reliability of welding quality. Therefore, defect detection can be welding slag detection in non-weld areas, that is, step S302 can be implemented through steps S10 to S12:

[0122] Step S10: The industrial computer performs welding slag detection on the non-weld area by using the trained detection model to obtain a welding slag detection result.

[0123] In some embodiments, the industrial computer uses a trained detection model to perform image processing on the non-weld bead area to obtain a grayscale image corresponding to the non-weld bead area. Pixels on the grayscale image with grayscale values ​​less than a preset value (e.g., 50), i.e., black pixels, are identified as welding slag, thereby obtaining a welding slag detection result indicating that welding slag is present in the non-weld bead area. However, the identified welding slag is not necessarily welding slag that affects the welding detection results of the battery cell. If the identified welding slag is small and insufficient to affect the welding quality of the battery cell, it can also indicate that welding slag is not present in the area outside the weld bead within the welding area of ​​the battery cell.

[0124] In some embodiments, if the grayscale values ​​of all pixels on the grayscale image are less than 200, a slag detection result indicating that the non-weld bead region does not have slag is obtained.

[0125] Therefore, after the industrial computer determines the welding slag based on the grayscale image, it also needs to determine whether the size of the welding slag will affect the welding quality.

[0126] Step S11: When the welding slag result indicates that welding slag exists in the non-weld area, the industrial computer measures the size of the welding slag to obtain the welding slag size.

[0127] In some embodiments, when welding slag exists in the non-weld area, the industrial computer determines the size of each welding slag. The size can be measured by the number of pixels or based on the length of the welding slag in a certain direction.

[0128] Step S12: Compare the welding slag size with a preset welding slag size to obtain the welding slag detection result.

[0129] In some embodiments, if judged by quantity, the preset slag size can be the number of pixel points corresponding to the slag that affects the welding quality. For example, the preset slag size is 200 pixels. When the number of pixels of any slag is greater than 200, a slag detection result indicating that there is slag in the non-weld area is obtained.

[0130] In some embodiments, if the length of the weld slag in a certain direction is used for judgment, the preset weld slag size can be the longest dimension of the weld slag that affects the welding quality, for example, 1 millimeter (mm). When any weld slag is longer than 1 mm in one direction, a weld slag detection result indicating that there is weld slag in the non-weld area is obtained.

[0131] The embodiment of the present disclosure detects whether there is welding slag in the welding area and further judges the identified welding slag based on a preset welding slag size, which can improve the accuracy of welding slag identification in the welding area of ​​the battery cell.

[0132] In some embodiments, when welding in the weld area, due to the welding process, the weld may be penetrated, resulting in abnormalities such as pinholes and cracks. Therefore, defect detection can also be crack point detection in the weld area. Therefore, step S302 can also be implemented through steps S13 to S15:

[0133] Step S13: The industrial computer detects the brightness values ​​of all pixels corresponding to the weld area using the trained detection model to obtain the brightness value of each pixel.

[0134] In some embodiments, pinholes and hot spots on the weld bead appear as bright spots in a grayscale image. Therefore, the brightness of each pixel is detected in the grayscale image to obtain the brightness value of each pixel, determine the hot spot area, and then determine whether there is a hot spot in the weld bead area based on the size of the hot spot area.

[0135] Step S14: determining the number of hotspot areas in the weld area based on the brightness value of each pixel point; wherein the value of the pixel points in the hotspot area is greater than a preset value, and the brightness values ​​of the pixel points in the hotspot area are all greater than the preset brightness.

[0136] In some embodiments, based on the brightness value of each pixel point, it is determined whether there is an area in which the brightness values ​​of all pixels are greater than a preset brightness (for example, 200), and the number of pixel values ​​is greater than a preset value (for example, 100), and the bright spot area is determined as a hot spot area.

[0137] Step S15: Obtain the burst point detection result based on the number of burst point areas.

[0138] In some embodiments, if the number of burst point areas is greater than or equal to 1, it is obtained that there are burst point areas in the weld area; if the number of burst point areas is equal to 0, it means that there are no burst point areas in the weld area, that is, there are no defects such as pinholes and burst points in the weld area.

[0139] The disclosed embodiment detects relatively small pinholes and explosion point areas on the weld, thereby avoiding the problems of missed detection and false detection during visual inspection, and improving the accuracy of detecting small defects on the weld.

[0140] In some embodiments, during penetration welding, due to solder replacement or other reasons, the weld may have breakpoints, disconnected welds, and cold welds, i.e., the weld is discontinuous, which may also cause low welding quality. Therefore, defect detection can also be performed by detecting breakpoints in the weld area. Therefore, step S302 can also be implemented through steps S16 to S17:

[0141] Step S16: The industrial computer extracts the contour of the weld area using the trained detection model to obtain the number of closed contours corresponding to the weld area.

[0142] In some embodiments, the industrial computer extracts the contour of the weld bead region based on the trained detection model to obtain a closed contour of the weld bead region. If there is a breakpoint, there will be at least two closed contours.

[0143] Step S17: Obtain the breakpoint detection result based on the number of contours.

[0144] Here, whether the weld bead region has breakpoints can be determined based on the number of closed contours. For example, if only one closed contour is identified in the weld bead region, the weld bead is continuous and has no breakpoints. If only multiple closed contours are identified in the weld bead region, the weld bead is discontinuous and has breakpoints. The number of breakpoints is the number of closed contours minus one.

[0145] The embodiment of the present disclosure detects breakpoints on the weld by the number of closed contours, thereby avoiding the problem of missed detection or false detection during visual inspection, and improving the accuracy of detecting discontinuous defects in the weld.

[0146] In some embodiments, when the battery cell is a cylindrical battery cell, the weld bead is generally a circular weld bead. If the solder deforms during welding, the weld bead may be offset. Therefore, defect detection may also be offset detection of the weld bead area. Therefore, step S302 may also be implemented through steps S18 to S20:

[0147] Step S18: The industrial computer determines the reference points of the battery cells on the collected image using the trained detection model and the collected image.

[0148] In some embodiments, the reference point of a battery cell can refer to a reference point for determining whether a weld has shifted. For example, the reference point is the center of the weld when the weld has not shifted. Here, the positive electrode of the battery cell itself has an injection hole located at the center of the positive electrode top cover of the battery cell. When the welding area of ​​the battery cell is the positive electrode welding area of ​​the battery cell, the reference point can refer to the injection hole of the battery cell. The negative electrode of the battery cell does not have an injection hole. Therefore, when the welding area of ​​the battery cell is the negative electrode welding area of ​​the battery cell, it is necessary to fit the center of the negative electrode on the captured image as the reference point of the negative electrode to determine whether the weld has shifted.

[0149] Step S19: performing center point fitting on the weld area to obtain the weld center of the weld area, and determining the offset distance between the weld center and the reference point.

[0150] In some embodiments, after determining the reference point, the center point of the weld bead region is fitted to obtain the weld bead center of the weld bead region. Based on the weld bead center and the reference point, an offset distance between the weld bead center and the reference point is calculated. The offset distance may be the straight-line distance between the weld bead center and the reference point.

[0151] Step S20: Compare the offset distance with a preset offset distance to obtain the weld bead offset result.

[0152] In some embodiments, a preset offset distance is used to represent an offset error. For example, if the straight-line distance between the weld center and the reference point is less than 1 mm, the weld area is not offset. By comparing the offset distance with the preset offset distance, a weld offset result can be obtained to determine whether the weld area is offset.

[0153] In some embodiments, the weld bead region may be annular. Determining whether the weld bead region has shifted may also involve setting a circular region within a predetermined area of ​​the battery cell. If the weld bead does not extend beyond the circular region, it indicates that the weld bead has not shifted. If the weld bead extends beyond the circular region, it indicates that the weld bead has shifted. This method may also be used to determine whether the weld bead has shifted if the weld bead has other shapes.

[0154] The disclosed embodiments can quickly determine the distance between a weld bead and a reference point, thereby improving the accuracy and efficiency of weld bead offset detection.

[0155] In some embodiments, due to the temperature during welding and other reasons, the length and width of the weld may not meet the requirements. Therefore, defect detection can also be a size detection of the weld area. Therefore, step S302 can also be implemented through steps S21 to S25:

[0156] Step S21: The industrial computer performs straight line fitting on the weld area using the trained detection model to obtain a straight weld corresponding to the weld area.

[0157] In the disclosed embodiment, the industrial computer can perform straight line fitting on the weld area and geometric transformation on the circular weld using the trained detection model to obtain a straight weld corresponding to the weld area.

[0158] Step S22: perform a first measurement on the straight weld bead to obtain a first measured dimension of the straight weld bead.

[0159] Here, the first measurement refers to measuring the length of the weld bead area, and the first measured dimension refers to the length of the weld bead area.

[0160] Step S23: segmenting the straight weld along the length direction of the straight weld to obtain a plurality of segmented areas.

[0161] In some embodiments, measuring the width of each point along the length of the linear weld bead would be time-consuming. Therefore, to reduce the computational complexity of the industrial computer, the disclosed embodiments can segment the linear weld bead along its length to obtain multiple segmented regions. The segmented regions can have the same or different lengths along their lengths.

[0162] Step S24 : performing a second measurement on a plurality of non-adjacent target segmented regions among the plurality of segmented regions to obtain a plurality of second measured sizes.

[0163] The embodiment of the present disclosure can detect the widths of multiple segmented areas at intervals, so as to reduce the detection workload of the industrial computer and improve detection efficiency.

[0164] Here, the number of segmented regions among the multiple target segmented regions may be the same or different. For example, the first of the multiple segmented regions may be used as the first target segmented region, the fifth segmented region after three segmented regions may be used as the second target segmented region, and the eleventh segmented region after five segmented regions may be used as the third target segmented region. Finally, the second measurement size corresponding to each of the multiple target segmented regions is obtained. The second measurement size refers to the width of each target segmented region.

[0165] Step S25 : Compare the first measured size with the first preset size interval, and compare the plurality of second measured sizes with the second preset size interval to obtain size detection results.

[0166] In some embodiments, a first preset size interval is used to represent the length of the weld bead, and a weld bead length within this interval is considered normal. A second preset size interval is used to represent the width of the weld bead, and a width of each target segmented region within this interval is considered normal. If both the first and second measured dimensions are normal, a size detection result indicating the weld bead region dimensions are normal is obtained.

[0167] The disclosed embodiment can not only quickly determine whether there is dimensional abnormality in the weld area, but also reduce the detection workload of the industrial computer and improve the detection efficiency and rate.

[0168] Based on the above embodiment, step S303 can be implemented through steps S3031 and S3032:

[0169] Step S3031: Obtain a welding detection result indicating that the battery cell is normal when the welding slag detection result, the burst point detection result, the breakpoint detection result, the weld bead offset result, and the size detection result are all normal.

[0170] In an embodiment of the present disclosure, if the welding slag detection result satisfies the conditions that the welding slag size is smaller than the preset welding slag size, the burst point detection result is characterized as the number of burst point areas is less than a first preset number (for example, 1), the breakpoint detection result is characterized as the number of contours is less than or equal to a second preset number (for example, 1), the weld offset result is characterized as the offset distance is greater than the preset offset distance, and the size detection result is characterized as the first measured dimension is within the first preset size interval, and multiple second measured dimensions are within the second preset size interval, a welding detection result characterizing that the battery cell is normal is obtained.

[0171] Step S3032: In the event of a detection abnormality, a welding detection result characterizing the abnormality of the battery cell is obtained; wherein the detection abnormality includes at least one of the following: the welding slag detection result is characterized by the welding slag size being larger than a preset welding slag size, the burst point detection result is characterized by the number of burst point areas being larger than a first preset number, the breakpoint detection result is characterized by the number of contours being larger than a second preset number, the weld offset result is characterized by the offset distance being larger than the preset offset distance, and the size detection result is characterized by the first measurement size being outside a first preset size interval, and multiple second measurement sizes being outside a second preset size interval.

[0172] In the embodiment of the present disclosure, if there is at least one abnormality among the welding slag detection result, the explosion point detection result, the breakpoint detection result, the weld bead offset result and the size detection result, it indicates that the welding of the battery cell is abnormal.

[0173] The disclosed embodiment determines whether the welding of the battery cell is normal by detecting defects such as pinholes, burst points, welding wire deviation, welding slag residue, welding wire size, broken welds, and cold welds. It can detect the welding results in multiple dimensions and aspects, reducing the risk of false detection and missed detection, and improving detection accuracy.

[0174] In some embodiments, if the welding detection result is normal, the battery cell is cut normally. If it is abnormal, the battery cell needs to be marked as abnormal. Therefore, the welding detection method provided in the embodiment of the present disclosure may further include steps S40 to S41:

[0175] Step S40: When the welding detection result indicates that the battery cell is abnormal, the industrial computer sends a marking signal to the controller.

[0176] In some embodiments, the industrial computer may send a marking signal carrying abnormality information to the controller when the welding detection result is abnormal.

[0177] In the embodiments of the present disclosure, detection abnormalities include but are not limited to the multiple detection abnormalities mentioned above, such as: the presence of welding slag in the welding area of ​​the battery cell, the offset of the weld bead of the battery cell, the abnormal size of the weld bead of the battery cell, and the presence of welding abnormalities in the weld bead of the battery cell.

[0178] Step S41: The controller controls the marking mechanism to mark the battery cell detected to be abnormal based on the marking signal.

[0179] In the embodiment of the present disclosure, the controller can control the marking mechanism set on the detection station based on the received marking signal, such as: a marking structure carrying a marking pen to mark the battery cell as abnormal, or control a marking mechanism to attach an abnormal label to the battery cell.

[0180] In the disclosed embodiment, if the welding test result is abnormal, the industrial computer sends a marking signal to the controller, causing the controller to promptly mark the battery cell. This allows the relevant marking operation to be executed promptly when an abnormality is detected, thereby identifying defective products and issuing an alarm, thereby quickly preventing the outflow of defective products and the resulting mass scrapping.

[0181] To implement the welding detection method provided in the embodiment of the present disclosure, the embodiment of the present disclosure further provides a welding detection system. FIG4 is a structural schematic diagram of the welding detection system provided in the embodiment of the present disclosure. As shown in FIG4 , the welding detection system includes: a controller 401, an industrial computer 402, and an image acquisition system 403.

[0182] Among them, the controller 401 is used to control the jacking mechanism on the turntable 404 to perform a jacking operation when the battery cell on the turntable 404 reaches the detection station 405 (such as the position of the battery cell in Figure 4), so as to lift the battery cell to a preset height and send an image acquisition signal to the image acquisition system 403; the image acquisition system 403 is used to respond to the image acquisition signal, acquire an image of the welding area of ​​the battery cell located at the preset height, obtain a captured image, and send the captured image to the industrial computer 402; the industrial computer 402 is used to perform welding detection on the captured image to obtain the welding detection result of the battery cell.

[0183] In some embodiments, the industrial computer 402 is also used to send an end signal containing the welding detection result to the controller 401. Based on the end signal, the controller 401 controls the rotation of the turntable 404 to drive the battery cell from the detection station 405 to the unloading station 406.

[0184] In some embodiments, Figure 5 is a structural schematic diagram of the jacking mechanism provided in an embodiment of the present disclosure. As shown in Figure 5, the jacking mechanism includes at least a jacking drive member (not shown in Figure 5), a cam 501, a jacking transmission member 502, an upper top plate 503 and a limit plate 504.

[0185] In some embodiments, the controller 401 is also used to send a lifting signal to the lifting drive member when the battery cell 505 on the turntable 404 reaches the inspection station; the cam 501 is slidably connected to the turntable and connected to the lifting drive member, wherein the cam 501 is provided with a slope structure; the lifting transmission member 502 is connected to the upper top plate 503 at one end, and the other end is located on the slope structure of the cam 501; the upper top plate 503 is used to support the battery cell 505; the limit plate 504 is provided on the turntable 404 and has a preset height; one end of the lifting drive member is connected to the turntable 404, and the other end is connected to the cam 501, and is used to drive the cam 501 to slide on the turntable 404 in response to the lifting signal to drive the lifting transmission member 502 to move along the first direction until the battery cell 505 contacts the limit plate 504.

[0186] 5 , the turntable 404 further includes a plurality of turntable support cups 506. A relief groove 5061 (shown in the dotted box in FIG5 ) is provided at the inner base of the turntable support cups 506. The turntable support cups 506 are used to accommodate the battery cells 505. The relief grooves prevent dust and impurities from accumulating at the inner base of the turntable support cups 506 after prolonged use, thereby contaminating the battery cells 505 or causing the battery cells 505 to become stuck in the turntable support cups 506.

[0187] In some embodiments, each battery cell 505 is pushed up by the lifting transmission part 502, and the battery cell 505 is driven by the cam 501 to push the battery cell 505 up until the battery cell 505 is tightly fitted with the limit plate 504. The battery cell 505 does not need to be separated from the turntable support cup 506. The turntable support cup 506 can ensure that each battery cell 505 to be tested is in the same position, avoiding the problem of inaccurate image taking position of the image acquisition system, and also avoiding problems in image extraction due to fluctuations in the depth of field of the camera of the image acquisition system.

[0188] In some embodiments, Figure 6 is a second structural schematic diagram of the welding detection system provided in an embodiment of the present disclosure. As shown in Figure 6, the detection station in the welding detection system includes at least two image acquisition systems 601; at least two image acquisition systems 601 are used to respectively respond to image acquisition signals to capture images of the welding areas of the battery cells to obtain captured images.

[0189] In some embodiments, the image acquisition system 601 includes at least a light source 6011 , a camera 6012 , a locking structure 6013 , and a slide rail structure 6014 .

[0190] Among them, the camera 6012 is fixed on the inspection station through the locking structure 6013 and the slide rail structure 6014, and is used to capture images of the welding area of ​​the battery cell to obtain a captured image; the light source 6011 is fixed on the inspection station through the locking structure 6013 and the slide rail structure 6014, and is located between the camera 6012 and the turntable, and is used to provide light for the camera 6012; the slide rail structure 6014 is used to adjust the positions of the camera 6012 and the light source 6011 on the inspection station.

[0191] In some embodiments, the light source 6011 may include an axially arranged dome light source and a coaxial light source to achieve better lighting, making the captured image clearer.

[0192] In the embodiment of the present disclosure, the locking structure and the slide rail structure enable the image acquisition system to perform axial adjustment of the camera and light source to be compatible with battery cells of different lengths and sizes, and can be applied to more usage scenarios.

[0193] In the embodiment of the present disclosure, in order to prevent the light source from causing damage to human eyes and the camera from being contaminated, a protective cover may be added to the outside of the camera.

[0194] Next, an application of a welding detection method in a practical scenario is provided.

[0195] In order to solve the problem of unstable camera imaging in related technologies and the problem that the consistency of incoming battery cells cannot be guaranteed during battery cell inspection, manual inspection of welding quality is required.

[0196] In the embodiment of the present disclosure, multiple groups of battery cells are in place at the same time at a single station, and the camera is controlled by module motion to take pictures for welding testing; a pair of limit plates (i.e., upper plate and limit plate) are added to each battery cell at the detection station, and the battery cell is pushed up to the limit reference plate (i.e., limit plate) by a cam and an elastic lifting mechanism (i.e., lifting transmission member), ensuring that the detection height of each battery cell (i.e., the distance from the camera to the surface to be tested) is consistent, avoiding image extraction errors (NG, NO) caused by fluctuations in the camera's depth of field GOOD) problem; improve equipment efficiency by coordinating the cup with the turntable; make an avoidance groove at the root of the inner side of the cup to avoid the accumulation of dust and impurities at the root of the inner side of the cup after long-term use, which may cause battery cell contamination or battery cell jamming in the cup; use a 2D line scan camera (i.e. camera) in conjunction with a coaxial light source and a dome light source to detect defects such as penetrating welding pinholes, burst points, welding wire offset, welding slag residue, welding wire size, broken welds, and cold welds; provide post-penetration welding size detection and defect detection logic; the main equipment PLC communicates with the visual software installed on the industrial computer through the network, the main equipment informs the CCD system (i.e. image acquisition system) of the current battery cell QR code information, and informs the CCD system of the arrival of the battery cell, the PLC gives the camera light source trigger signal respectively, and the camera takes pictures in 2D mode. Using the processing algorithm, the characteristics of the detection area can be more intuitively displayed, and the detection data can be obtained in real time. By connecting the visual software to the camera, the detection efficiency and detection rate can be improved. The deep learning algorithm is used for defect judgment. The deep learning algorithm performs defect detection and judgment based on the located region of interest (ROI), and finally displays the display results on the CCD system interface.

[0197] The disclosed embodiment can solve the problems of low detection accuracy and detection efficiency with fewer visual systems. The battery cell is pushed up to the limit reference plate through the cam and elastic lifting mechanism to ensure that the detection height of each battery cell (the distance from the CCD to the surface to be measured) is consistent, avoiding the image extraction NG problem caused by the fluctuation of the camera depth of field; the root of the inner side of the cup is treated with an avoidance groove to avoid the accumulation of dust and impurities at the root of the inner side of the cup after long-term use, which may cause battery cell contamination or battery cell jamming in the cup. It can realize the detection of defects such as penetrating welding pinholes, burst points, welding wire offset, welding slag residue, welding wire size, broken welds, and cold welds.

[0198] FIG7 is a schematic diagram of a detection process provided by an embodiment of the present disclosure. As shown in FIG7 , the detection process of the battery cell penetration welding can be implemented through steps S701 to S705:

[0199] Step S701: The CCD system is ready.

[0200] In the embodiment of the present disclosure, after welding is completed in the cylindrical battery cell penetration welding equipment (such as the positive electrode penetration welding machine and the negative electrode penetration welding machine), the battery cell flows to the visual inspection station along with the fixture. After the battery cell flows to the inspection starting position, the CCD system is ready.

[0201] Step S702: The PLC controls the light source of the CCD system to light up.

[0202] At this time, after the PLC detects that the battery cell is in place, it controls the light source of the CCD system to turn on.

[0203] Step S703: The camera of the CCD system takes a picture.

[0204] After the light source of the CCD system is turned on, the camera of the CCD system takes pictures of the battery cell to obtain a captured image.

[0205] Step S704: the visual software performs algorithm processing.

[0206] Here, the CCD system sends the captured image to the industrial computer, and the visual software on the industrial computer performs algorithm processing and recognition on the captured image to obtain the detection result.

[0207] Here, the algorithm can be implemented based on a pre-trained model. By annotating offline data to obtain sample data, the model is trained based on the sample data to obtain detection results. The detection results are compared with the annotated data of the sample data to determine the model loss. Based on the loss, the model parameters are optimized until the loss converges to obtain a trained model. The trained model is used to infer images collected by the CCD system to obtain detection results.

[0208] Step S705: Get the detection result.

[0209] In some embodiments, if the test is qualified (OK), the inspection station releases the battery cell and proceeds to the next station; if the test is unqualified (NG), the inspection station marks the battery cell as NG.

[0210] Here, the industrial computer can upload the test results to the PLC and host computer according to the TCP protocol, and the host computer will then feed the test results back to the manufacturing execution system (MES).

[0211] FIG8 is a schematic diagram of the layout of the detection stations provided in an embodiment of the present disclosure, and FIG9 is a schematic diagram of the detection stations provided in an embodiment of the present disclosure. Based on FIG8 and FIG9 , an embodiment of the present disclosure further provides a detection logic.

[0212] As shown in Figure 8, there are two supporting cups on the turntable corresponding to the first inspection station 801 and the second inspection station 802, two corresponding to the first welding station 803 and the second welding station 804, and two corresponding to the first unloading station 805 and the second unloading station 806. The battery cells are transferred to different stations through the rotation of the turntable.

[0213] The disclosed embodiment can drive two battery cells from the first welding station 803 and the second welding station 804 to the first inspection station 801 and the second inspection station 802 by a turntable to detect the two incoming battery cells. As shown in Figure 9, the detection system of each group of detection stations includes a dome light source 901, a coaxial light source 902, a lens 903, a 2D camera 904 and other auxiliary equipment (such as an industrial computer, a display, a light source controller, an uninterruptible power supply (UPS) to avoid sudden power outages, a communication harness, etc.). The battery cells are transferred to the post-weld inspection station by the turntable, and each inspection station detects one battery cell; before detection, the battery cells are pushed to the limit reference plate 907 by the turntable cam 905 (i.e., cam) and the elastic top mechanism 906, and then the weld size and defect detection of the welding area of ​​the battery cells are performed by the detection camera.

[0214] In some embodiments, in order to prevent the light source from causing damage to the human eye and contamination of the camera, a protective cover can be added to the outside of the camera; in order to be compatible with different battery cell lengths and sizes and facilitate the debugging of the camera and light source, the fixing plates of the camera and light source are respectively added with a slide rail structure and a spiral locking structure to facilitate the axial position adjustment of the camera and light source.

[0215] Each cell is lifted up by a resilient lifting mechanism 906 driven by a turntable cam 905 until it is tightly fitted against a limit reference plate 907. The cell does not need to be removed from the support cup 908 (i.e., the turntable support cup). The support cup 908 ensures that each cell under test is in the same position, preventing the CCD inspection system from taking inaccurate images. After the cell is lifted up and tightened, the CCD inspection system inspects the weld seam and the surface to be tested where the weld seam is located.

[0216] In addition, the inner root of the support cup 908 is treated with an avoidance groove to prevent dust and impurities from accumulating at the inner root of the support cup after long-term use, which may cause battery cell contamination or battery cell jamming in the support cup.

[0217] In the embodiment of the present disclosure, the main device PLC communicates with the visual software over the network. The main device informs the CCD system of the QR code information of the current battery cell and informs the CCD that the battery cell has arrived. The PLC gives the camera a light source trigger signal respectively, the light source turns on, and the camera takes pictures in 2D mode. When the image of the first battery cell is completed, the image is sent to the visual software. Using the processing algorithm, the characteristics of the detection area can be intuitively displayed, and the detection data can be obtained in real time.

[0218] FIG10 is a schematic diagram of a detection communication process provided by an embodiment of the present disclosure. As shown in FIG10 , the detection communication process may be implemented through steps S111 to S114:

[0219] Step S111: When the turntable is in place, a trigger signal is sent to the CCD system.

[0220] Here, when the turntable is in place, a trigger signal is sent to the CCD system, which is used to trigger the photo.

[0221] Step S112: PLC triggers the camera to take a photo.

[0222] Here, if the camera fails to take a photo, the PLC will give a trigger signal again or set the battery cell as unqualified. If the photo is successful, it will be sent to the industrial computer for inspection.

[0223] Step S113: The PLC sends a result request signal to the industrial computer.

[0224] Here, the PLC sends a result request signal to the industrial computer to obtain the test results of the industrial computer. If the waiting time for the result request signal times out, the PLC will alarm, press the reset button, and set the battery cell to unqualified. If the result request signal is successful, step S114 is executed. Here, if the PLC times out receiving the reset signal, it will also alarm and set the battery cell to unqualified.

[0225] Step S114: The industrial computer sends the detection result to the PLC.

[0226] In some embodiments, if the result request signal is successful, the industrial computer sends the detection result to the PLC.

[0227] In some embodiments, there are multiple inspection stations, and the communication logic of each inspection station is the same.

[0228] This disclosed embodiment connects to a camera via visual software. To improve inspection efficiency and detection accuracy, deep learning is used for defect detection. The 2D images captured by the camera are fed into a deep learning algorithm for identification and detection. The algorithm then performs defect detection based on the located ROI. Finally, the results are displayed on the camera interface.

[0229] FIG11 is a schematic diagram of pinholes or explosion points on a weld provided by an embodiment of the present disclosure. As shown in FIG11 , 110 is a first weld area identified by the visual software, and 111 is a pinhole or explosion point on the weld area identified by the visual software.

[0230] Figure 12 is a schematic diagram of the welding slag residue provided by an embodiment of the present disclosure. As shown in Figure 12, 120 is the second weld bead area identified by the visual software, 121 is the non-weld bead area identified by the visual software, and 122 is the welding slag residue on the non-weld bead area identified by the visual software.

[0231] Figure 13 is a schematic diagram of weld offset provided in an embodiment of the present disclosure. As shown in Figure 13, 131 is the third weld area identified by the visual software; 132 is the weld contour extracted from the weld area; 133 is the center coordinates corresponding to the weld contour; 134 is the comparison coordinates of the injection hole of the battery cell on the captured image; 135 is a schematic diagram of positionally comparing the center coordinates with the comparison coordinates, and determining whether the weld is offset based on the offset between the center coordinates and the comparison coordinates.

[0232] Figure 14 is a schematic diagram of the weld bead size provided in an embodiment of the present disclosure. As shown in Figure 14, 141 can be used for the weld bead contour; 142 is the contour to be measured that forms a straight line, that is, the weld bead contour is unfolded to obtain the corresponding straight weld bead, which can be used to calculate or measure the length of the contour; 143 is a schematic diagram of dividing the contour to be measured into equal pixels; here, the width value corresponding to each pixel after the equal pixel division can be calculated and measured.

[0233] It should be understood that “one embodiment” or “an embodiment” mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present disclosure. Therefore, “in one embodiment” or “in an embodiment” appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in the various embodiments of the present disclosure, the size of the serial numbers of the above-mentioned steps / processes does not mean the order of execution, and the execution order of each step / process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure. The serial numbers of the embodiments of the present disclosure are for description only and do not represent the advantages and disadvantages of the embodiments.

[0234] The present disclosure uses descriptions of directions or positional relationships indicated by “upper”, “lower”, “top”, “bottom”, “front”, “back”, “inside” and “outside”, etc., which are only for the convenience of describing the present disclosure, and do not indicate or imply that the device referred to must have a specific direction, be constructed and operate in a specific direction. Therefore, it should not be understood as limiting the scope of protection of the present disclosure.

[0235] It should also be noted that, in the description of this disclosure, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood broadly. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to direct connections or indirect connections through an intermediary. Those skilled in the art will understand the specific meanings of these terms in this disclosure depending on the specific circumstances.

[0236] It should be noted that, in this disclosure, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0237] In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0238] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, the functional units in the various embodiments of the present disclosure may all be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0239] The above are only implementation methods of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any changes or replacements that can be easily conceived by any technician familiar with this technical field within the technical scope disclosed in this disclosure should be covered by the protection scope of the present disclosure.

Claims

1. A welding detection method, comprising: When the battery cell on the turntable reaches the inspection station, the controller controls the lifting mechanism on the turntable to perform a lifting operation, lifts the battery cell to a preset height, and sends an image acquisition signal to the image acquisition system; The image acquisition system acquires an image of the welding area of ​​the battery cell at the preset height in response to the image acquisition signal, obtains an acquired image, and sends the acquired image to the industrial computer; The industrial computer performs welding detection on the collected image to obtain a welding detection result of the battery cell; wherein the welding detection at least includes detecting a weld area of ​​the battery cell on the collected image.

2. The welding detection method according to claim 1, wherein: The turntable is provided with a plurality of cups for placing battery cells; the welding detection method further comprises: When the battery cell is placed on the support cup at the loading station, the controller controls the turntable to rotate so as to drive the battery cell from the loading station to the inspection station; When the battery cell moves to the detection station, the controller sends the acquisition signal to the image acquisition system.

3. The welding detection method according to claim 1 or 2, wherein: The industrial computer performs welding detection on the collected image to obtain welding detection results of the battery cell, including: The industrial computer performs image segmentation on the collected image to obtain weld bead areas and non-weld bead areas; The industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area using a trained detection model to obtain at least one defect detection result; wherein the trained model is trained based on sample images with defects annotated; and the defect detection result includes at least a slag detection result, a burst point detection result, a breakpoint detection result, a weld bead offset result, and a size detection result; The welding detection result of the battery cell is obtained based on the at least one defect detection result.

4. The welding detection method according to claim 3, wherein: The industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area using a trained detection model to obtain at least one defect detection result, including: The industrial computer performs welding slag detection on the non-weld area using the trained detection model to obtain a welding slag detection result; When the welding slag result indicates that welding slag exists in the non-weld area, the industrial computer measures the size of the welding slag to obtain the welding slag size; The welding slag size is compared with a preset welding slag size to obtain the welding slag detection result.

5. The welding detection method according to claim 3 or 4, wherein: The industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area using a trained detection model to obtain at least one defect detection result, including: The industrial computer detects the brightness values ​​of all pixels corresponding to the weld area using the trained detection model to obtain the brightness value of each pixel; Determining the number of hotspot areas in the weld bead area based on the brightness value of each pixel point; wherein the value of the pixel points in the hotspot area is greater than a preset value, and the brightness values ​​of the pixel points in the hotspot area are all greater than the preset brightness; The burst point detection result is obtained based on the number of burst point areas.

6. The welding detection method according to any one of claims 3 to 5, wherein: The industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area using a trained detection model to obtain at least one defect detection result, including: The industrial computer extracts the contour of the weld bead area using the trained detection model to obtain the number of closed contours corresponding to the weld bead area; Based on the number of contours, the breakpoint detection result is obtained.

7. The welding detection method according to any one of claims 3 to 6, wherein: The industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area using a trained detection model to obtain at least one defect detection result, including: The industrial computer determines the reference point of the battery cell on the collected image through the trained detection model and the collected image; Performing center point fitting on the weld area to obtain a weld center of the weld area, and determining an offset distance between the weld center and the reference point; The offset distance is compared with a preset offset distance to obtain the weld bead offset result.

8. The welding detection method according to any one of claims 3 to 7, wherein: The industrial computer performs at least one defect detection on at least one of the weld bead area and the non-weld bead area using a trained detection model to obtain at least one defect detection result, including: The industrial computer performs straight line fitting on the weld area using a trained detection model to obtain a straight weld corresponding to the weld area; Performing a first measurement on the straight weld bead to obtain a first measured dimension of the straight weld bead; Segmenting the straight weld bead along its length to obtain a plurality of segmented areas; Performing a second measurement on a plurality of non-adjacent target segmented regions among the plurality of segmented regions to obtain a plurality of second measured sizes; The first measured size is compared with a first preset size interval, and the plurality of second measured sizes are compared with a second preset size interval to obtain size detection results.

9. The welding detection method according to any one of claims 4 to 8, wherein: The obtaining the welding detection result of the battery cell based on the at least one defect detection result includes: When the welding slag detection result, the burst point detection result, the breakpoint detection result, the weld bead offset result, and the size detection result are all normal, obtaining a welding detection result indicating that the battery cell is normal; In the presence of a detection abnormality, a welding detection result characterizing the abnormality of the battery cell is obtained; wherein the detection abnormality includes at least one of the following: the welding slag detection result is characterized by the welding slag size being larger than a preset welding slag size, the burst point detection result is characterized by the number of burst point areas being larger than a first preset number, the breakpoint detection result is characterized by the number of contours being larger than a second preset number, the weld bead offset result is characterized by the offset distance being larger than the preset offset distance, and the size detection result is characterized by the first measurement dimension being outside a first preset size interval, and multiple second measurement dimensions being outside a second preset size interval.

10. The welding detection method according to any one of claims 1 to 9, wherein: The welding detection method further comprises: When the welding detection result indicates that the battery cell is abnormal, the industrial computer sends a marking signal to the controller; The controller controls the marking mechanism to mark the battery cell in which abnormality is detected based on the marking signal.

11. The welding detection method according to any one of claims 1 to 10, wherein: The welding detection method further comprises: The industrial computer sends an end signal including the welding detection result to the controller; The controller controls the rotation of the turntable based on the end signal to drive the battery cell to move from the inspection station to the unloading station.

12. A welding detection system, comprising: a controller, configured to control a lifting mechanism on the turntable to perform a lifting operation when the battery cell on the turntable reaches the inspection station, lift the battery cell to a preset height, and send an image acquisition signal to an image acquisition system; The image acquisition system is configured to acquire an image of the welding area of ​​the battery cell at the preset height in response to the image acquisition signal, obtain an acquired image, and send the acquired image to the industrial computer; The industrial computer is used to perform welding detection on the collected image to obtain the welding detection result of the battery cell; wherein the welding detection at least includes detecting the weld area of ​​the battery cell on the collected image.

13. The welding detection system according to claim 12, wherein: The lifting mechanism at least includes a lifting drive member, a cam, a lifting transmission member, an upper lift plate and a limit plate; The controller is further configured to send a lifting signal to the lifting drive member when the battery cells on the turntable reach the detection station; The cam is slidably connected to the turntable and connected to the lifting drive member; the cam is provided with an inclined surface structure; The lifting transmission member has one end connected to the upper top plate and the other end located on the inclined structure; The upper plate is used to support the battery cell; The limiting plate is provided on the turntable and has the preset height; The lifting drive member, one end of which is connected to the turntable, and the other end is connected to the cam, is used to drive the cam to slide on the turntable in response to the lifting signal, so as to drive the lifting transmission member to move along the first direction until the battery cell contacts the limit plate.

14. The welding detection system according to any one of claims 12 or 13, wherein: The turntable also includes a plurality of turntable cups; The turntable support cup is provided with an avoidance groove at the inner root portion, and the turntable support cup is used for placing the battery core.

15. The welding detection system according to any one of claims 12 to 13, wherein: The inspection station includes at least two image acquisition systems; The at least two image acquisition systems are used to respectively acquire images of the welding area of ​​the battery cell in response to the image acquisition signals to obtain the acquired images.

16. The welding detection system according to any one of claims 12 to 15, wherein: The image acquisition system at least includes a light source, a camera, a locking structure and a slide rail structure; The camera is fixed to the inspection station via the locking structure and the slide rail structure, and is used to capture an image of the welding area of ​​the battery cell to obtain the captured image; The light source is fixed to the inspection station through the locking structure and the slide rail structure, is located between the camera and the turntable, and is used to provide light for the camera; The slide rail structure is used to adjust the positions of the camera and the light source on the detection station.

Images