Welding testing system and defect detection method

Abstract

A welding testing system and a defect detection method. The welding testing system comprises a welding mechanism and a first testing mechanism (1); the welding mechanism is used for welding a casing and a top cover of a battery cell (01) together; the first testing mechanism (1) is arranged downstream of the welding mechanism; the first testing mechanism (1) comprises a support assembly (11), a planar imaging assembly (12), and a processing assembly; the planar imaging assembly (12) is mounted to the support assembly (11); the planar imaging assembly (12) is used for capturing a planar image at a welding pass; and the processing assembly is used for acquiring the planar image and determining, on the basis of the planar image, whether the welding pass is standard-compliant. The planar imaging assembly (12) comprises imaging units (121); each imaging unit (121) comprises a camera (1211) and a reflective mirror (123); the camera (1211) and the testing position of the first testing mechanism (1) are located on the same side of the reflective mirror (123); a lens of the camera (1211) faces the reflective mirror (123); and when the battery cell (01) is located at the testing position, the camera (1211) captures an image of the welding pass reflected by the reflective mirror (123).

Description

Welding detection system and defect detection method

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This disclosure is based on and claims the priority of Chinese patent application with application number 202410001433.2, application date January 2, 2024, and invention name “A welding detection system and defect detection method”. 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 technology, and in particular to a welding detection system and a defect detection method. 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] During the assembly of the battery cells, the shell and the top cover need to be welded. In order to ensure the quality of the battery cells, the weld between the shell and the top cover needs to be inspected.

[0006] In related technologies, a 3D image of the weld is often captured, and the weld area of ​​the 3D image is analyzed. When the weld surface shows height information that is higher or lower than a reference position and outside the detection specifications, it can be determined that the weld has defects.

[0007] However, the above method may miss weld defects, resulting in low accuracy of weld detection.

[0008] Summary of the Invention

[0009] In view of this, the present disclosure provides a welding inspection system and a defect detection method, which can improve the accuracy of weld bead inspection and reduce the risk of misjudging the weld bead.

[0010] The present disclosure is achieved through the following technical solutions.

[0011] The first aspect of the present disclosure provides a welding detection system. The welding detection system includes a welding mechanism and a first detection mechanism. The welding mechanism is used to weld the shell and the top cover of the battery cell together. The first detection mechanism is arranged downstream of the welding mechanism and is used to detect the weld between the shell and the top cover after welding. The first detection mechanism includes a bracket assembly, a planar imaging assembly and a processing assembly. The planar imaging assembly is mounted to the bracket assembly. The planar imaging assembly is used to capture a planar image at the weld; the processing assembly is used to obtain the planar image and determine whether the weld is qualified based on the planar image. The planar imaging assembly includes an imaging unit. The imaging unit includes a camera and a reflector. The detection position of the camera and the first detection mechanism is located on the same side of the reflector. The lens of the camera faces the reflector. When the battery cell is located at the detection position, the camera captures the reflected image of the weld through the reflector.

[0012] In the welding inspection system provided in the embodiments of the present disclosure, a planar imaging component can be used to capture a planar image of the weld bead. Based on this, a processing component can be used to analyze the planar image of the weld bead to determine whether the weld bead is qualified. Compared to the related art method of using height information in a 3D image to determine whether the weld bead is qualified, the embodiments of the present disclosure can use the grayscale characteristics of different areas of the planar image for analysis, which can effectively determine defects in the weld bead, especially welds that have been rolled. Therefore, the welding inspection system provided in the embodiments of the present disclosure can improve the accuracy of weld bead inspection and reduce the risk of misjudging the weld bead. Moreover, by using a reflector to reflect the weld bead and arranging the camera and the detection position on the same side of the reflector, the space occupied by the planar imaging component can be reduced, making it easier to set up the first detection mechanism.

[0013] In a possible implementation of the present disclosure, the imaging unit further includes a fill light component, and the fill light component is used to fill light toward the weld bead.

[0014] In a possible implementation of the present disclosure, the welding detection system further includes a conveying mechanism, which is used to convey the battery cells to the detection position and to convey the detected battery cells out of the detection position.

[0015] In a possible implementation of the present disclosure, the fill light component includes a first light-emitting component, and the imaging unit further includes a first driving component, the fixed end of the first driving component is fixedly connected to the bracket assembly, and the output end of the first driving component is connected to the first light-emitting component; the first driving component is used to drive the first light-emitting component to switch between a fill light position and an avoidance position; when the first light-emitting component is in the fill light position, the first light-emitting component extends into the conveying path of the battery cell to fill light the weld; when the first light-emitting component is in the avoidance position, the first light-emitting component is retracted from the conveying path of the battery cell to avoid the conveyance of the battery cell.

[0016] In a possible implementation of the present disclosure, the conveying mechanism includes a second driving member, a conveying member and a first position detection module, the second driving member is used to drive the conveying member to operate, the conveying member is used to convey the battery cell, and the first position detection module is used to detect the position of the battery cell; the first position detection module and the second driving member are electrically connected to the processing component; the processing component is also used to control the operation of the imaging unit and the second driving member based on the position of the battery cell.

[0017] In a possible implementation of the present disclosure, when the battery cell is located at the detection position, the processing component controls the first driving component to drive the first light-emitting component to the fill-light position and controls the second driving component to stop running, and the processing component controls the fill-light component to fill-light the weld bead and controls the camera component to photograph the weld bead; after the planar imaging component completes photographing the weld bead, the processing component controls the first driving component to drive the first light-emitting component to the avoidance position, and the processing component controls the second driving component to continue running.

[0018] In a possible implementation of the present disclosure, the bracket assembly includes a vertical support column arranged along the vertical direction, and an upper mounting frame and a lower mounting frame arranged on the vertical support column, the upper mounting frame is equipped with a planar imaging assembly, and the lower mounting frame is equipped with a conveying mechanism.

[0019] In a possible implementation of the present disclosure, the upper mounting frame includes a first mounting frame and a camera mounting frame; the first mounting frame is fixedly connected to the vertical support column, and the camera mounting frame is fixedly connected to the first mounting frame; the first driving member is installed to the first mounting frame, and the first driving member is used to drive the first light-emitting member to move in the vertical direction; the reflector is fixedly connected to the lower end of the first mounting frame; and the conveying mechanism is used to convey the battery cell to move in the horizontal direction.

[0020] In a possible implementation of the present disclosure, the upper mounting bracket further includes a second mounting bracket, and the second mounting bracket is fixedly connected to the lower end of the camera mounting bracket;

[0021] The fill light also includes a second light-emitting component and a third light-emitting component. The second light-emitting component extends in the horizontal direction and is installed to the lower end of the second mounting frame. There are two third light-emitting components. The two third light-emitting components extend in the vertical direction and are installed at intervals below the second light-emitting component. The space between the two third light-emitting components and the second light-emitting component forms a conveying path for the battery cell.

[0022] In a possible implementation of the present disclosure, the first light-emitting element, the second light-emitting element, and the third light-emitting element all have a planar light-emitting surface.

[0023] In a possible implementation of the present disclosure, a first adjustment structure is provided between the first driving member and the first mounting bracket, and the first driving member can adjust its position in the vertical direction through the first adjustment structure; and / or a second adjustment structure is provided between the second mounting bracket and the camera mounting bracket, and the second mounting bracket can adjust its position in the vertical direction through the second adjustment structure.

[0024] In a possible implementation of the present disclosure, a third adjustment structure is provided between the camera element and the camera element mounting bracket, and the camera element can adjust its position along the direction of the camera element lens through the third adjustment structure; and / or, a fourth adjustment structure is provided between the reflector and the first mounting bracket, and the reflector can adjust its position along the vertical direction through the fourth adjustment structure.

[0025] In a possible implementation of the present disclosure, a fifth adjustment structure is provided between the third light-emitting element and the second light-emitting element, and the position of the third light-emitting element can be adjusted along the horizontal direction through the fifth adjustment structure.

[0026] In a possible implementation of the present disclosure, the planar imaging assembly includes an imaging module, each imaging module includes two imaging units, and the two imaging units are symmetrically arranged along a first direction, corresponding to the welds at both ends of the battery cell along the first direction, respectively. The first direction refers to the extension direction of the conveying path of the battery cell.

[0027] In a possible implementation of the present disclosure, the planar imaging assembly includes at least two imaging modules, which are arranged along a second direction for simultaneously detecting welds of at least two battery cells; wherein the second direction is perpendicular to the first direction.

[0028] In a possible implementation of the present disclosure, the imaging units located on the same side along the first direction in all the imaging modules share the first light-emitting element and the second light-emitting element.

[0029] In a possible implementation of the present disclosure, the welding detection system further includes a posture adjustment mechanism, which is arranged between the welding mechanism and the first detection mechanism, and is used to adjust the posture of the battery cell to the posture required by the first detection mechanism for battery cell detection.

[0030] In a possible implementation of the present disclosure, the detection mechanism also includes a first identity recognition component, which is used to obtain identification information of the battery cell; the first identity recognition component is electrically connected to the processing component, and the processing component is also used to match the identification information of the battery cell with the detection results of the weld.

[0031] In a possible implementation of the present disclosure, the welding detection system also includes a second detection mechanism, which is arranged downstream of the first detection mechanism, and the second detection mechanism is provided with a waste discharge mechanism; the second detection mechanism includes a second identity recognition component, and the second identity recognition component is electrically connected to the processing component; before the second detection mechanism detects the battery cell, the second identity recognition component confirms the identification information of the battery cell; when the weld corresponding to the identification information of the battery cell fails the inspection, the processing component controls the second detection mechanism not to detect the battery cell, and the battery cell is discharged through the waste discharge mechanism.

[0032] A second aspect of the present disclosure provides a defect detection method applied to a welding detection system, the welding detection system comprising: a welding mechanism and a first detection mechanism;

[0033] The first detection mechanism is arranged downstream of the welding mechanism, wherein the first detection mechanism includes a bracket assembly, a plane imaging assembly and a processing assembly, the plane imaging assembly is mounted on the bracket assembly, the plane imaging assembly includes an imaging unit, the imaging unit includes a camera and a reflector, the camera and the detection position of the first detection mechanism are located on the same side of the reflector, and the lens of the camera faces the reflector; the processing assembly includes: a visual host computer and a controller;

[0034] The method comprises:

[0035] When the battery cell reaches the first detection mechanism, the controller sends a position signal to the visual host computer based on the position information of the battery cell obtained by the first position detection module;

[0036] The visual host computer controls the lighting of the fill light of the camera in the first detection mechanism and controls the camera to take pictures to obtain a plane image of the weld, and performs defect detection on the plane image to determine the detection result of the weld.

[0037] In the above solution, the welding detection system further includes: a posture adjustment mechanism; the first detection mechanism includes: a first identity recognition component;

[0038] The method further comprises:

[0039] The battery cell is transferred to the posture adjustment mechanism through the conveying mechanism, and after being flipped into position by the posture adjustment mechanism, control information is sent to the visual host computer;

[0040] transporting the battery cell to the first detection mechanism through a conveying mechanism;

[0041] The visual host computer controls the first identity recognition component to scan the code to obtain the identification information of the battery cell.

[0042] It can be understood that the battery cell is transferred to the posture adjustment mechanism through the conveying mechanism, and after being flipped into place by the posture adjustment mechanism, the control information is sent to the visual host computer; the visual host computer controls the first identity recognition component to scan the code and obtain the identification information of the battery cell, so as to facilitate the subsequent update of the identification information of the battery cell.

[0043] In the above solution, the welding detection system includes: production control equipment;

[0044] The method further comprises:

[0045] The visual host computer displays the detection result of the battery cell and sends the detection result of the battery cell to the production control device;

[0046] The production control device reads identification information of the battery cell;

[0047] Based on the test result of the battery cell and the identification information of the battery cell, the identification information of the battery cell is updated to obtain a product identification code with the test result.

[0048] It can be understood that the visual host computer displays the test results of the battery cell and sends the test results of the battery cell to the production control equipment, and the production control equipment reads the identification information of the battery cell; based on the test results of the battery cell and the identification information of the battery cell, the identification information of the battery cell is updated to obtain a product identification code with the test results, which can facilitate the subsequent second testing agency to scan the code to obtain the test results.

[0049] In the above solution, the welding detection system includes: a second detection mechanism; the second detection mechanism is located behind the first detection mechanism on the conveying mechanism;

[0050] The method further comprises:

[0051] When the battery cell is transferred to the second detection mechanism by the conveying mechanism, the second detection mechanism determines the treatment of the battery cell according to the detection result of the battery cell.

[0052] It can be understood that when the battery cells are transferred to the second detection mechanism through the conveying mechanism, the second detection mechanism determines the treatment of the battery cells based on the detection results of the battery cells, and can effectively detect the missed battery cells to prevent the NG battery cells from continuing to flow and causing safety risks.

[0053] In the above solution, when the battery cell is transferred to the second detection mechanism by the conveying mechanism, the second detection mechanism determines the treatment of the battery cell according to the detection result of the battery cell, including:

[0054] When the battery cell is transferred to the second detection mechanism by the conveying mechanism, the second detection mechanism scans the product identification code of the battery cell to obtain an abnormal result of the battery cell or a normal result of the battery cell;

[0055] According to the abnormal result of the battery cell, the battery cell is discharged from the waste discharge mechanism of the helium detector; or

[0056] According to the normal result of the battery cell, a helium test is performed on the battery cell.

[0057] It can be understood that the second detection mechanism scans the product identification code of the battery cell to obtain the abnormal result of the battery cell or the normal result of the battery cell; according to the abnormal result of the battery cell, the battery cell is discharged from the waste discharge mechanism of the helium detection machine; or, according to the normal result of the battery cell, the battery cell is subjected to helium detection, which can effectively detect the missed battery cells and prevent the NG battery cells from continuing to flow and causing safety risks.

[0058] In the above solution, the method further includes:

[0059] When the battery cell reaches the first detection mechanism, the controller triggers the first driving member to extend to prevent the battery cell from sliding; the first driving member is located on the conveying mechanism;

[0060] When the battery cell does not reach the first detection mechanism, the controller triggers the first driving member to contract to make the battery cell slide.

[0061] It is understandable that when the battery cell reaches the first detection mechanism, the controller triggers the first driving member to extend to prevent the battery cell from sliding, thereby fixing the battery cell, facilitating subsequent acquisition of the battery cell image, and improving the image quality of the battery cell image.

[0062] In the above solution, the first through-beam sensor in the first position detection module is arranged near the position adjustment mechanism; the second through-beam sensor is arranged in the middle of the detection station;

[0063] The method further comprises:

[0064] When the battery cell just reaches the first detection mechanism, the first through-beam sensor obtains first sub-sensor information;

[0065] The battery cell continues to move, and when the battery cell completely reaches the first detection mechanism, the second beam sensor obtains second sub-sensor information; both the first sub-sensor information and the second sub-sensor information are position information of the battery cell.

[0066] It can be understood that when the battery cell just reaches the first detection mechanism, the first corresponding sensor obtains the first sub-sensor information; the battery cell continues to move, and when the battery cell completely reaches the first detection mechanism, the second corresponding sensor obtains the second sub-sensor information, which can determine that the battery cell is in place, facilitating the subsequent acquisition of the battery cell image.

[0067] In the above solution, the performing defect detection on the planar image and determining the detection result of the weld bead includes:

[0068] The visual host computer performs image preprocessing on the plane image using the preset defect detection model to obtain at least one defect information;

[0069] performing defect information fusion on the at least one defect information to obtain fused information;

[0070] Defect detection is performed based on the fusion information to determine the detection result of the weld.

[0071] It can be understood that, through the preset defect detection model, the planar image of the battery cell is preprocessed to obtain at least one defect information; the at least one defect information is fused to obtain fused information; based on the fused information, defect detection is performed to determine the detection result of the weld. Since the preset defect detection model is trained in advance, the accuracy of defect detection can be improved by detecting the planar image of the battery cell.

[0072] In the above solution, the method further includes:

[0073] The visual host computer obtains planar images of multiple sample weld beads;

[0074] performing image preprocessing and defect annotation on the planar images of the plurality of sample weld beads respectively to obtain sample defect information corresponding to each of the planar images of the plurality of sample weld beads;

[0075] The initial defect detection model is trained using the sample defect information corresponding to each of the planar images of the plurality of sample weld beads to determine a preset defect detection model.

[0076] It can be understood that, by obtaining planar images of multiple sample welds; performing image preprocessing and defect annotation on the planar images of multiple sample welds respectively, and obtaining sample defect information corresponding to each of the planar images of the multiple sample welds; and training the initial defect detection model through the sample defect information corresponding to each of the planar images of the multiple sample welds, and determining the preset defect detection model, the accuracy of detection by the preset defect detection model can be improved.

[0077] In the above solution, the initial defect detection model is trained using the sample defect information corresponding to each of the planar images of the plurality of sample welds to determine the preset defect detection model, including:

[0078] Determining a mapping relationship between defects and inspection specifications based on sample defect information corresponding to each of the planar images of the plurality of sample weld beads;

[0079] Defect learning of the labeled image based on the mapping relationship between the defects and the inspection specifications;

[0080] The initial defect detection model is trained through defect learning of the image and sample defect information corresponding to each of the planar images of the plurality of sample weld beads to determine the preset defect detection model.

[0081] It can be understood that the mapping relationship between defects and detection specifications is determined through the sample defect information corresponding to each planar image of multiple sample welds; based on the mapping relationship between defects and detection specifications, the defects of the labeled images are learned; the initial defect detection model is trained through the defect learning of the images and the sample defect information corresponding to each planar image of multiple sample welds, and the preset defect detection model is determined. The accuracy of the detection of the preset defect detection model can be improved by model training through the mapping relationship between defects and detection specifications and the defect learning of the labeled images.

[0082] A third aspect of the present disclosure provides a welding detection system, the welding detection system comprising: a welding mechanism and a first detection mechanism;

[0083] The first detection mechanism is arranged downstream of the welding mechanism, wherein the first detection mechanism includes a bracket assembly, a plane imaging assembly and a processing assembly, the plane imaging assembly is mounted on the bracket assembly, the plane imaging assembly includes an imaging unit, the imaging unit includes a camera and a reflector, the camera and the detection position of the first detection mechanism are located on the same side of the reflector, and the lens of the camera faces the reflector; the processing assembly includes: a visual host computer and a controller;

[0084] A welding mechanism for welding the battery shell and the top cover of the battery cell together;

[0085] a controller, configured to send a position signal to the visual host computer when the battery cell reaches the first detection mechanism based on the position information of the battery cell obtained by the first position detection module;

[0086] The visual host computer is used to control the lighting of the fill light of the camera in the first detection mechanism and control the camera to take pictures in response to the in-position signal, and to perform defect detection on the plane image of the weld of the battery cell to determine the detection result of the weld.

[0087] The defect detection method provided by the disclosed embodiment mainly involves the following steps: when a battery cell reaches the first detection mechanism, the controller sends a position signal to the visual host computer based on the position information of the battery cell obtained by the first position detection module; the visual host computer controls the lighting of the fill light of the camera element in the first detection mechanism and controls the camera element to take a photo, obtains a planar image of the weld bead, and performs defect detection on the planar image to determine the weld bead detection result. In this process, by controlling the lighting of the fill light and the camera element to take photos, a more effective battery cell image can be obtained, which can effectively detect missed battery cells, prevent NG battery cells from continuing to flow and causing safety risks, and improve the accuracy of defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present disclosure. The same reference numerals are used throughout the drawings to denote the same components. In the drawings:

[0089] FIG1 is a schematic diagram of the arrangement of some mechanisms of a welding inspection system provided by an embodiment of the present disclosure (showing a protective cover);

[0090] FIG2 is a schematic diagram of the arrangement of part of the mechanism of a welding detection system provided by an embodiment of the present disclosure (protective cover is omitted);

[0091] FIG3 is a schematic diagram of the overall structure of a first detection mechanism of a welding detection system provided by an embodiment of the present disclosure;

[0092] FIG4 is a partial structural diagram of a first detection mechanism of a welding detection system provided by an embodiment of the present disclosure;

[0093] FIG5 is a schematic diagram of a conveying mechanism of a welding detection system provided in an embodiment of the present disclosure;

[0094] FIG6 is a top view of a first detection mechanism of a welding detection system provided by an embodiment of the present disclosure;

[0095] FIG7 is a schematic diagram of a partial structure of a first detection mechanism of a welding detection system provided by an embodiment of the present disclosure;

[0096] FIG8 is a schematic diagram of a partial structure of a first detection mechanism of a welding detection system provided by an embodiment of the present disclosure;

[0097] FIG9 is a schematic diagram showing the arrangement of an imaging unit of a first detection mechanism of a welding detection system provided in an embodiment of the present disclosure.

[0098] FIG10A is a schematic diagram of a missed kill product provided by an embodiment of the present disclosure;

[0099] FIG10B is a schematic diagram of a height map rendering effect of a battery cell provided by an embodiment of the present disclosure;

[0100] FIG10C is a schematic diagram of a grayscale image effect of a battery cell provided by an embodiment of the present disclosure;

[0101] FIG11 is a first optional flow chart of a defect detection method provided by an embodiment of the present disclosure;

[0102] FIG12 is a schematic diagram of an optional flow chart of a model training and application method provided in an embodiment of the present disclosure;

[0103] FIG13A is a schematic diagram of loading an ROI frame in a defect detection method provided by an embodiment of the present disclosure;

[0104] FIG13B is a schematic diagram of a positioning weld bead body according to a defect detection method provided by an embodiment of the present disclosure;

[0105] FIG13C is a schematic diagram of defect determination in a defect detection method provided by an embodiment of the present disclosure;

[0106] FIG14A is a first schematic diagram of a detection result of a defect detection method provided by an embodiment of the present disclosure;

[0107] FIG14B is a second schematic diagram of detection results of a defect detection method provided by an embodiment of the present disclosure;

[0108] FIG15 is a second optional flow chart of a defect detection method provided in an embodiment of the present disclosure.

[0109] Description of Reference Numerals

[0110] 1-first detection mechanism; 11-bracket assembly; 111-vertical support column; 112-upper mounting frame; 1121-first mounting frame; 11211-first adjustment structure; 11212-fourth adjustment structure; 1122-camera mounting frame; 11221-first extension wall; 11222-second extension wall; 11223-third adjustment structure; 1123-second mounting frame; 11231-second adjustment structure; 113-lower mounting frame; 12-planar imaging assembly; 121- Imaging unit; 1211-camera element; 1212-fill light element; 12121-first light-emitting element; 12122-second light-emitting element; 121221-fifth adjustment structure; 12123-third light-emitting element; 122-first driving element; 123-reflector; 13-first identity recognition component; 2-conveyance mechanism; 21-second driving element; 22-conveyance member; 3-posture adjustment mechanism; 31-flipping mechanism; 4-second detection mechanism; 41-second identity recognition component; 01-battery cell. DETAILED DESCRIPTION

[0111] The following embodiments of the technical solution of the present disclosure are described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present disclosure and are therefore only examples and are not intended to limit the scope of protection of the present disclosure.

[0112] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which the present disclosure belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure; the terms "including" and "having" and any variations thereof in the specification of the present disclosure and the above-mentioned drawings are intended to cover non-exclusive inclusions.

[0113] In the description of the embodiments of the present disclosure, technical terms such as "first," "second," and "third" are used solely to distinguish different objects and should not be understood to indicate or imply relative importance or to implicitly specify the quantity, specific order, or primary and secondary relationship of the technical features indicated. In the description of the embodiments of the present disclosure, "plurality" means more than two, unless otherwise specifically defined.

[0114] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present disclosure. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0115] In the description of the embodiments of the present disclosure, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0116] In the description of the embodiments of the present disclosure, the orientations or positional relationships indicated by technical terms such as "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", and "circumferential" are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the embodiments of the present disclosure and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed, operated or used in a specific orientation. Therefore, they should not be understood as limiting the embodiments of the present disclosure.

[0117] In the description of the embodiments of the present disclosure, unless otherwise expressly specified or limited, technical terms such as "mounted," "connected," "connect," and "fixed" should be understood broadly. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection, or electrical connection; direct connection or indirect connection through an intermediate medium; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of these terms in the embodiments of the present disclosure based on specific circumstances.

[0118] In the description of the embodiments of the present disclosure, unless otherwise clearly specified and limited, the technical term "contact" should be understood in a broad sense, and can be direct contact, contact through an intermediate medium layer, contact with essentially no interaction force between the two contacting parties, or contact with interaction force between the two contacting parties.

[0119] Hereinafter, the present disclosure will be described in detail.

[0120] Currently, new energy batteries are increasingly being used in everyday life and industry. They are not only used in energy storage systems such as hydropower, thermal power, wind power, and solar power plants, but are also widely used in electric vehicles like electric bicycles, electric motorcycles, and electric cars, as well as in a variety of fields such as aerospace. As the application of power batteries continues to expand, market demand is also growing.

[0121] In many application scenarios, multiple battery cells can be arranged and combined to form a battery pack for use, which can increase the capacity of the battery pack. Generally speaking, for the convenience of description, the battery in the embodiments of the present disclosure may generally refer to a battery pack or a battery module.

[0122] It should be noted that the batteries in the embodiments of the present disclosure can be used, but are not limited to, in power storage systems, vehicles, ships, aircraft, and other electrical devices. Furthermore, a battery pack or battery module is formed by placing multiple battery cells in a sealed enclosure, which provides more reliable dust and water resistance and can therefore be used in harsh, humid, or even submerged environments.

[0123] The embodiments of the present disclosure provide an electrical device including the above-mentioned battery or battery pack for providing electrical energy. The electrical device may be, 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. The electric toy may 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 may include an airplane, a rocket, a space shuttle, and a spacecraft, etc.

[0124] In the following embodiments, for the convenience of description, a vehicle is used as an example of an electrical device according to an embodiment of the present disclosure.

[0125] The vehicle can be a fuel-powered vehicle, a gas-powered vehicle, or a new energy vehicle. The latter can be a pure electric vehicle, a hybrid vehicle, or an extended-range vehicle. The vehicle is equipped with a battery, which can be located at the bottom, front, or rear of the vehicle. The battery can be used to power the vehicle, for example, as a power source for operation. The vehicle may also include a controller and a motor. The controller controls the battery to power the motor, for example, for starting the vehicle, navigation, and operating power requirements during driving.

[0126] In some embodiments of the present disclosure, the battery can serve not only as the operating power source of the vehicle, but also as the driving power source of the vehicle, replacing or partially replacing fuel or natural gas to provide driving power for the vehicle.

[0127] The battery includes a battery bottom plate, a battery cover, a battery vertical plate and at least one battery cell. The battery cover is covered on the battery bottom plate, thereby forming a battery cell accommodating space between the battery bottom plate and the battery cover.

[0128] In a battery, there can be multiple battery cells, which can be connected in series, in parallel, or in a hybrid connection. A hybrid connection refers to a combination of series and parallel connections. Multiple battery cells can be directly connected in series, in parallel, or in a hybrid connection, and then the entire battery cell structure can be placed in the storage space formed by the battery base and the battery cover. Of course, the battery can also be formed by first connecting multiple battery cells in series, in parallel, or in a hybrid connection to form a battery module, and then connecting multiple battery modules in series, in parallel, or in a hybrid connection to form a whole, and then accommodated in the storage space formed by the battery base and the battery cover. The battery can also include other structures. For example, the battery can also include a busbar component for achieving electrical connection between multiple battery cells.

[0129] In the embodiment of the present disclosure, the battery cell may be a secondary battery. A secondary battery refers to a battery cell that can be continuously used by activating active materials by charging after the battery cell is discharged.

[0130] The battery cells may be lithium-ion batteries, sodium-ion batteries, sodium-lithium-ion batteries, lithium metal batteries, sodium metal batteries, lithium-sulfur batteries, magnesium-ion batteries, nickel-hydrogen batteries, nickel-cadmium batteries, lead-acid batteries, etc., which are not limited in the embodiments of the present disclosure.

[0131] Although not shown, a battery cell generally includes an electrode assembly. The electrode assembly includes a positive electrode, a negative electrode, and a separator. During the charge and discharge process of the battery cell, active ions (such as lithium ions) are inserted and removed between the positive and negative electrodes. The separator is placed between the positive and negative electrodes to prevent short circuits between the positive and negative electrodes while allowing active ions to pass through.

[0132] In some embodiments, the electrode assembly is provided with tabs that can conduct current from the electrode assembly. The tabs include a positive tab and a negative tab.

[0133] In some embodiments, a battery cell may include a housing. The housing is used to encapsulate components such as the electrode assembly and the electrolyte. The housing may be a steel housing, an aluminum housing, a plastic housing (e.g., polypropylene), a composite metal housing (e.g., a copper-aluminum composite housing), or an aluminum-plastic film.

[0134] As an example, the battery cell can be a cylindrical battery cell, a prismatic battery cell, a soft-pack battery cell or a battery cell of other shapes. The prismatic battery cell includes a square-shell battery cell, a blade-shaped battery cell, a polygonal battery, such as a hexagonal battery, etc. There is no special limitation in the present disclosure.

[0135] During the assembly of battery cells, the shell and top cover need to be welded. To ensure the quality of the battery cells, the weld between the shell and top cover needs to be inspected. Related technologies often capture 3D images of the welds and analyze the weld areas in these 3D images. If the weld surface shows height information that is above or below a reference position and outside the inspection specifications, a weld defect is identified.

[0136] For example, in some related technologies, a 3D line scan camera can be used to scan the weld to obtain a weld image with height information. By analyzing the above weld image, it is possible to detect whether there are defects on the weld surface that exceed the specification height (such as pinholes, burst points, etc.).

[0137] However, in some battery production lines, a rolling mechanism is often provided after the welding mechanism. The rolling mechanism can be used to roll the weld between the shell and the top cover. At this time, the welded weld will be flattened by the roller. This causes the pinholes, burst points and other defects of the original weld to lose height information. The above-mentioned method of judging whether the weld has defects based on the height information that appears above or below the reference position and outside the detection specifications on the weld surface may result in inaccurate detection of the weld. In addition, due to the limitations of the imaging principle of the 3D camera, the grayscale features of the 3D imaging of the weld taken by the 3D camera are not obvious. All of the above situations will lead to misjudgment of weld defects.

[0138] In view of this, embodiments of the present disclosure provide a welding inspection system and a defect detection method.

[0139] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to FIG. 1 to FIG. 15 .

[0140] Specifically, referring to Figures 1, 2, and 3, the welding inspection system provided in the embodiment of the present disclosure includes a welding mechanism and a first inspection mechanism 1. The welding mechanism is used to weld the shell and top cover of a battery cell 01 together. The first inspection mechanism 1 is arranged downstream of the welding mechanism and is used to inspect the weld between the shell and top cover after welding. The first inspection mechanism 1 includes a bracket assembly 11, a planar imaging assembly 12, and a processing assembly. The planar imaging assembly 12 is mounted to the bracket assembly 11 and is used to capture a planar image of the weld. The processing assembly is used to obtain the planar image and determine whether the weld is qualified based on the planar image.

[0141] It is understandable that, in addition to the above-mentioned welding mechanism and first detection mechanism 1, the welding detection system may also include other mechanisms.

[0142] In the embodiment of the present disclosure, a welding robot can be used as the welding mechanism, and the welding robot can complete the welding of the shell and the top cover according to a predetermined process.

[0143] It should be noted that, in order to vividly illustrate the welding detection system provided by the embodiment of the present disclosure, a square-shell battery cell is taken as an example for description in the specification and the accompanying drawings.

[0144] It is understandable that the housing is provided with a receiving cavity open to one side, and the battery cell can be placed in the receiving cavity. After the top cover is placed in the opening according to the assembly relationship, the welding mechanism can be used to complete the welding between the housing and the top cover.

[0145] In addition, the first detection mechanism 1 is arranged downstream of the welding mechanism, which means that the first detection mechanism 1 is arranged downstream of the welding mechanism according to the flow direction of the battery cell 01 in the welding detection system. The first detection mechanism 1 can detect the weld between the shell and the top cover after welding.

[0146] It should be noted that, for the convenience of description, the weld bead between the shell and the top cover after welding is collectively referred to as a weld bead.

[0147] Furthermore, the bracket assembly 11 can provide a mounting base for the planar imaging assembly 12 and the like. In the disclosed embodiment, the specific structural form of the bracket assembly 11 is not limited. For example, referring to FIG. 3 , a plurality of columnar structures can be assembled to form the bracket assembly 11, and then the planar imaging assembly 12 can be mounted to the bracket assembly 11.

[0148] In addition, for the processing component, the processing component can be installed on the bracket component 11, or the processing component can be independently set and installed at other locations, which is not limited in the embodiments of the present disclosure.

[0149] Furthermore, the planar imaging component 12 is used to capture a planar image of the weld. The position of the planar imaging component 12 can be adaptively set according to the position of the battery cell 01 during weld detection, so that the planar imaging component 12 can capture a planar image of the weld.

[0150] It should be noted that the plane image here can also be considered as a 2D image, which is relative to the 3D image. It can be considered that the plane image does not include the height information of the aforementioned 3D image.

[0151] For example, a black and white area array camera can be used to set the planar imaging component 12 in the embodiment of the present disclosure. More specifically, in the embodiment of the present disclosure, a 12-megapixel black and white area array camera can be used in conjunction with a 16 mm fixed-focus lens to set the planar imaging component 12 in the embodiment of the present disclosure.

[0152] Furthermore, the first detection mechanism 1 of the welding detection system provided in the embodiment of the present disclosure also includes a processing component, and the plane imaging component 12 is electrically connected to the processing component. The processing component can obtain the plane image of the weld taken by the plane imaging component 12 and determine whether the weld is qualified based on the plane image.

[0153] For example, for a planar image of a weld, the grayscale value at the defect will be significantly different from that at other locations on the weld. Therefore, the weld can be inspected based on the grayscale values ​​of different parts of the planar image and the shapes corresponding to the different grayscale values ​​to determine whether the weld is qualified.

[0154] It should be noted that the type of processing component is not limited in the embodiment of the present disclosure. For example, a programmable logic controller (PLC) can be used to set the processing component, and other devices with processing functions can also be used to set the processing component, such as an industrial computer. In addition, in the embodiment of the present disclosure, a distributed control system can also be used to perform distributed processing on the plane imaging component 12 and the above-mentioned plane image. Specifically, a PLC can be set for the plane imaging component 12, and an industrial computer can be set for the plane image. For this purpose, the industrial computer can also be referred to as a host computer. In this way, the strong processing power of the industrial computer can be used to process the plane image of the weld captured by the plane imaging component 12, and the industrial computer can be used to transmit a control signal to the above-mentioned PLC to control the plane imaging component 12.

[0155] Through the above-described configuration, the welding inspection system provided by the embodiment of the present disclosure can utilize the planar imaging component 12 to capture a planar image of the weld bead. Based on this, the processing component can analyze the planar image of the weld bead to determine whether the weld bead is qualified. Compared to the related art method of determining whether a weld bead is qualified by utilizing height information in a 3D image, the present embodiment utilizes the grayscale characteristics of different regions of the planar image for analysis, effectively determining defects in the weld bead, particularly weld bead that has been subjected to rolling. Therefore, the welding inspection system provided by the embodiment of the present disclosure can improve the accuracy of weld bead detection and reduce the risk of misjudgment of weld beads.

[0156] On this basis, referring to Figure 7, in the embodiment of the present disclosure, the planar imaging component 12 includes an imaging unit 121, and the imaging unit 121 includes a camera 1211 and a fill light 1212. The camera 1211 is used to capture a planar image at the weld, and the fill light 1212 is used to fill light toward the weld.

[0157] It should be noted that in the embodiment of the present disclosure, the specific structural type of the camera 1211 is not limited. For example, referring to the above description, a 12-megapixel black and white area array camera can be used in conjunction with a 16 mm fixed-focus lens as the camera 1211 in the embodiment of the present disclosure.

[0158] Furthermore, the fill light component 1212 can be used to fill light toward the weld bead, and a flash light or a plane light source can be used to fill light the weld bead.

[0159] It can be understood that in the embodiment of the present disclosure, the orientation of the fill light component 1212 can be adaptively adjusted so that the fill light component 1212 can fill light on the weld.

[0160] Through the above arrangement, by using the fill light component 1212 to fill light toward the weld, it is possible to further improve the distinction between the defective portion of the weld and other portions, and further improve the accuracy of defect detection on the weld.

[0161] On this basis, referring to Figures 1, 2, and 3, the welding inspection system provided in the embodiment of the present disclosure further includes a conveying mechanism 2, which is used to convey the battery cell 01 to the inspection position of the first inspection mechanism 1 and to remove the inspected battery cell 01 from the inspection position. When the battery cell 01 is at the inspection position, the camera 1211 can capture a planar image of the weld bead.

[0162] For example, in the embodiment of the present disclosure, a conveyor belt can be used as the conveying mechanism 2. Specifically, the conveyor belt can be set between the welding mechanism and the first detection mechanism 1, and the conveyor belt can be extended to the downstream of the first detection mechanism 1. In this way, the battery cell 01 can move from the welding mechanism to the first detection mechanism 1 and continue to move downstream of the first detection mechanism 1 as the conveyor belt moves.

[0163] It can be understood that by controlling the conveying mechanism 2 , the battery cell 01 can be paused at the welding position of the welding mechanism, and at the detection position of the first detection mechanism 1 for detecting the weld bead.

[0164] It should be noted that the detection position here refers to the position at which the plane imaging component 12 can capture a plane image of the weld bead.

[0165] Specifically, the lens orientation of the camera 1211 of the planar imaging assembly 12 and the orientation of the fill light 1212 can be adaptively set so that when the battery cell 01 is located at the detection position, the lens and the fill light 1212 face the weld.

[0166] For example, referring to Figures 3 and 5, in some embodiments of the present disclosure, the conveying mechanism 2 can be set as a conveyor belt, and the conveyor belt can be set below the planar imaging component 12. When the battery cell 01 is suspended at a position below the planar imaging component 12 using the conveyor belt, the planar imaging component 12 can capture a planar image of the weld.

[0167] Through the above arrangement, in the welding detection system provided by the embodiment of the present disclosure, the conveying mechanism 2 is used to convey the battery cell 01, which can reduce manual operations and improve the intelligence level of the welding detection system.

[0168] On this basis, referring to FIG7 , in the embodiment of the present disclosure, the fill light component 1212 includes a first light-emitting component 12121, and the imaging unit 121 further includes a first driving component 122. The fixed end of the first driving component 122 is fixedly connected to the bracket assembly 11, and the output end of the first driving component 122 is connected to the first light-emitting component 12121. The first driving component 122 is used to drive the first light-emitting component 12121 to switch between a fill light position and a avoidance position. When the first light-emitting component 12121 is in the fill light position, the first light-emitting component 12121 extends into the conveying path of the battery cell 01 to fill light the weld bead; when the first light-emitting component 12121 is in the avoidance position, the first light-emitting component 12121 is retracted from the conveying path of the battery cell 01 to avoid the conveyance of the battery cell 01.

[0169] 7 , it can be understood that, in the embodiment of the present disclosure, the first driving member 122 can be used to drive the first light-emitting member 12121 to move, thereby enabling the first light-emitting member 12121 to switch between the fill light position and the avoidance position.

[0170] It should be noted that in the embodiments of the present disclosure, the specific structural form of the first driving member 122 is not limited. The first driving member 122 can be configured as a driving member with a power source of various types, such as a cylinder, a hydraulic rod, or a motor. For example, with reference to FIG. 7 , in some embodiments of the present disclosure, the first driving member 122 can be configured as a cylinder.

[0171] In addition, the movement direction of the first driving member 122 driving the first light-emitting member 12121 can be determined according to the conveying direction of the battery cell 01 conveyed by the conveying mechanism 2.

[0172] For example, referring to FIG1 , FIG2 and FIG3 , the description is made by taking the conveying mechanism 2 conveying the battery cell 01 in the horizontal direction and detecting the short side of a square battery as an example.

[0173] In this case, the conveying path of the battery cell 01 can be considered to be the direction extending along the length of the battery cell 01. With reference to Figure 7 , the direction in which the first driving member 122 drives the first light-emitting member 12121 can be set to a vertical direction. Specifically, the first driving member 122 can be positioned above the conveying path of the battery cell 01 and configured to drive the first light-emitting member 12121 in the vertical direction.

[0174] In the process of the battery cell 01 moving from the position where the welding mechanism is located to the detection position, the first driving member 122 drives the first light-emitting member 12121 to move in the vertical direction to above the conveying path of the battery cell 01, thereby avoiding the conveying of the battery cell 01. In this regard, the avoidance position of the first light-emitting member 12121 refers to the position where the first light-emitting member 12121 is located above the conveying path of the battery cell 01, as shown in Figure 7.

[0175] After battery cell 01 moves to the detection position, first driver 122 drives first light-emitting element 12121 to move vertically into the conveying path of battery cell 01, so that first light-emitting element 12121 and battery cell 01 are positioned opposite each other to the side of battery cell 01. This allows the first light-emitting frame to be directly oriented toward battery cell 01, facilitating fill-lighting of the weld bead. The fill-lighting position of first light-emitting element 12121 refers to the position where first light-emitting element 12121 is located within the conveying path of battery cell 01 and opposite to battery cell 01, as shown in FIG9 .

[0176] Through the above-mentioned setting, in the welding detection system provided in the embodiment of the present disclosure, by setting the first driving member 122 to drive the first light-emitting member 12121 to move so as to switch between the avoidance position and the fill light position, the first light-emitting member 12121 can be prevented from affecting the transportation of the battery cell 01. At the same time, the volume of the first detection mechanism 1 can be set to be smaller, thereby improving the adaptability of the first detection mechanism 1 in the embodiment of the present disclosure.

[0177] On this basis, with reference to FIG5 , in some embodiments of the present disclosure, the conveying mechanism 2 includes a second driving member 21, a conveying member 22, and a first position detection module. The second driving member 21 is used to drive the conveying member 22 to operate, and the conveying member 22 is used to convey the battery cell 01. The first position detection module is used to detect the position of the battery cell 01. The first position detection module and the second driving member 21 are electrically connected to a processing assembly. The processing assembly is also used to control the operation of the imaging unit 121 and the second driving member 21 based on the position of the battery cell 01.

[0178] It is understood that in the embodiment of the present disclosure, the conveying member 22 is used to directly convey the battery cells 01. For example, the conveying member 22 can be set as a belt. The second driving member 21 is used to drive the conveying member 22 to operate. The second driving member 21 can be set as a motor.

[0179] In addition, in the embodiments of the present disclosure, the first position detection module is used to detect the position of the battery cell 01. The first detection module can be configured as a contact position sensor or a non-contact position sensor, which is not limited in the embodiments of the present disclosure. For example, in some embodiments of the present disclosure, the first position detection module can be configured as a through-beam photoelectric sensor.

[0180] Thus, in the embodiment of the present disclosure, by electrically connecting the driving member and the first position detection module to the processing assembly, the processing assembly can be used to control the operation of the imaging unit 121 and the second driving member 21 based on the position of the battery cell 01 .

[0181] For example, the first position detection module can be set to correspond to the detection position of battery cell 01. When battery cell 01 moves to the detection position, it can trigger a signal from the first position detection module. After receiving this signal, the processing component issues a command to stop the second drive member 21. At this time, battery cell 01 is suspended at the detection position. At the same time, the processing component also issues a command to the planar imaging unit 121 to photograph the weld bead. After the planar imaging unit 121 completes the photographing of the weld bead, the processing component issues a command to resume the operation of the second drive member 21, thereby continuing to transport the battery cell 01.

[0182] Through the above settings, in the battery cell 01 provided in the embodiment of the present disclosure, the operation of the plane imaging component 12 and the conveying mechanism 2 is controlled based on the position of the battery cell 01 by the processing component, which can further improve the intelligence of the welding detection system provided in the embodiment of the present disclosure.

[0183] On this basis, in some embodiments of the present disclosure, when the battery cell 01 is located in the detection position, the processing component controls the first driving component 122 to drive the first light-emitting component 12121 to the fill light position and controls the second driving component 21 to stop running, and the processing component controls the fill light component 1212 to fill light the weld and controls the camera component 1211 to photograph the weld; after the planar imaging component 12 completes photographing the weld, the processing component controls the first driving component 122 to drive the first light-emitting component 12121 to the avoidance position, and the processing component controls the second driving component 21 to continue running.

[0184] It can be understood that in the embodiment of the present disclosure, by electrically connecting the first driving member 122 to the processing component, the operation of the first driving member 122 can also be controlled by using the processing component.

[0185] Specifically, when the processing component determines, based on the first position detection module, that battery cell 01 is at the detection position, it issues a command to control the first driver 122 to drive the first light-emitting element 12121 to the fill-light position, and issues a command to control the second driver 21 to stop operating. Furthermore, the processing component issues a control command to control the fill-light element 1212 to fill-light the weld bead and to control the camera 1211 to photograph the weld bead.

[0186] After the planar imaging component 12 completes photographing the weld, the processing component issues an instruction to control the first driving component 122 to drive the first light-emitting component 12121 to move to the avoidance position, and the processing component issues an instruction to control the second driving component 21 to continue operating, thereby continuing to transport the battery cell 01.

[0187] Through the above-mentioned settings, in the welding detection system provided in the embodiment of the present disclosure, the processing component can be used to interlock the operations among the first driving member 122, the second driving member 21, the camera member 1211 and the fill light member 1212, thereby further improving the intelligence level of the welding detection system in the embodiment of the present disclosure.

[0188] In addition, referring to Figures 8 and 9, in some embodiments of the present disclosure, the planar imaging assembly 12 also includes a reflector 123, the camera 1211 and the detection position are located on the same side of the reflector 123, and the lens of the camera 1211 faces the reflector 123, for capturing the reflected image of the weld through the reflector 123.

[0189] It should be noted that in the embodiment of the present disclosure, the weld bead can be reflected and imaged using the reflector 123. By using the camera 1211 to capture the reflected image, a planar image of the weld bead can also be obtained.

[0190] In addition, in the embodiment of the present disclosure, the size of the reflector 123 is not limited. For example, the size of the reflector 123 can be set according to the length of the weld to be detected. The size of the reflector 123 can be set smaller so that the reflector 123 can meet the reflective imaging of the weld. In addition, the size of the reflector 123 can also be set larger, and the embodiment of the present disclosure is not limited to this.

[0191] It can be understood that, according to the reflective imaging principle of the reflector 123, the camera 1211 and the detection position need to be set on the same side of the reflector 123, and by adaptively setting the orientation of the lens of the camera 1211, the camera 1211 can be used to capture the reflected image of the weld through the reflector 123.

[0192] Through the above-mentioned arrangement, in the welding detection system provided in the embodiment of the present disclosure, the reflector 123 is used to reflectively image the weld bead, and the camera 1211 and the detection position are arranged on the same side of the reflector 123, which can reduce the space occupied by the planar imaging component 12 and facilitate the arrangement of the first detection mechanism 1 provided in the embodiment of the present disclosure.

[0193] On this basis, referring to Figure 3, in some embodiments of the present disclosure, the bracket assembly 11 includes a vertical support column 111 arranged along the vertical direction, and an upper mounting frame 112 and a lower mounting frame 113 arranged on the vertical support column 111, the upper mounting frame 112 is equipped with a planar imaging assembly 12, and the lower mounting frame 113 is equipped with a conveying mechanism 2.

[0194] For example, referring to FIG. 3 , in some embodiments of the present disclosure, the overall structure of the bracket assembly 11 may be set to a cube shape, with four pillars extending in a vertical direction to form a vertical support column 111 .

[0195] At the same time, a bracket-like structure can be installed on the upper end of the vertical support column 111 to form an upper mounting frame 112 , and a bracket-like structure can be installed on the vertical support column 111 at a position below the upper mounting frame 112 to form a lower mounting frame 113 .

[0196] In this way, the upper mounting frame 112 can be used to install the planar imaging component 12, and the lower mounting frame 113 can be used to install the conveying mechanism 2.

[0197] Through the above configuration, in the welding inspection system provided in the embodiment of the present disclosure, by configuring the support assembly 11 to include a vertical support column 111, an upper mounting frame 112, and a lower mounting frame 113, it is possible to facilitate assembly of the support assembly 11. Simultaneously, it is also possible to facilitate installation of the planar imaging assembly 12 and the conveying mechanism 2.

[0198] On this basis, referring to Figures 4 and 6 , in some embodiments of the present disclosure, the upper mounting frame 112 includes a first mounting frame 1121 and a camera mounting frame 1122. The first mounting frame 1121 is fixedly connected to the vertical support column 111, and the camera mounting frame 1122 is fixedly connected to the first mounting frame 1121. A first driving member 122 is mounted to the first mounting frame 1121 and is used to drive the first light-emitting member 12121 to move in a vertical direction. The reflective mirror 123 is fixedly connected to the lower end of the first mounting frame 1121. The conveying mechanism 2 is used to convey the battery cell 01 in a horizontal direction.

[0199] It should be noted that, referring to FIG. 4 , in the embodiment of the present disclosure, the main structure of the first mounting frame 1121 may be configured as a plate-like structure extending in the vertical direction, and the first mounting frame 1121 may be fixedly connected to the vertical support column 111 through pillars.

[0200] 4 and 7 , the camera mounting frame 1122 can also be configured as a right-angle structure. Specifically, referring to FIG7 , the camera mounting frame 1122 includes a first extension wall 11221 extending horizontally and a second extension wall 11222 extending vertically. One end of the first extension wall 11221 is fixedly connected to the first mounting frame 1121, and the second extension wall 11222 is connected to the other end of the first extension wall 11221.

[0201] 8 , the reflector 123 may be fixedly connected to the lower end of the first mounting bracket 1121 , and the camera 1211 may be fixedly mounted at the angle between the first extension wall 11221 and the second extension wall 11222 , so that the lens of the camera 1211 faces the reflector 123 .

[0202] Furthermore, the fixed end of the first driving member 122 can be connected to the first mounting bracket 1121 by a fastener, and by adjusting the output direction of the output shaft of the first driving member 122, the first driving member 122 can be used to drive the first light-emitting member 12121 to move in the vertical direction.

[0203] 8 , when the first driving member 122 drives the first light emitting member 12121 to move vertically downward, the weld can be illuminated; and when the first driving member 122 drives the first light emitting member 12121 to move vertically upward, the transportation of the battery cell 01 can be avoided.

[0204] Through the above arrangement, in the battery cell 01 provided in the embodiment of the present disclosure, the structure of the upper mounting frame 112 can be simplified and conveniently arranged, and the installation of the first driving member 122 , the first light emitting member 12121 and the reflector 123 can be facilitated.

[0205] On this basis, referring to Figures 7 and 8, in some embodiments of the present disclosure, the upper mounting frame 112 further includes a second mounting frame 1123, which is fixedly connected to the lower end of the camera mounting frame 1122. The fill light 1212 further includes a second light-emitting member 12122 and a third light-emitting member 12123. The second light-emitting member 12122 extends horizontally and is mounted to the lower end of the second mounting frame 1123. There are two third light-emitting members 12123, which extend vertically and are installed below the second light-emitting member 12122 at intervals. The space between the two third light-emitting members 12123 and the second light-emitting member 12122 forms a conveying path for the battery cells 01.

[0206] For example, referring to FIG8 , in the embodiment of the present disclosure, the second mounting bracket 1123 can be fixedly connected to the lower end of the second extension wall 11222. Furthermore, a mounting plate extending horizontally can be provided at the lower end of the second mounting bracket 1123, so that the second light-emitting element 12122 can be mounted to the lower surface of the mounting plate.

[0207] Furthermore, two third light-emitting components 12123 may be mounted below the second light-emitting component 12122 via fasteners.

[0208] In this way, the space between the two third light-emitting elements 12123 forms a gap, which can be used as a conveying path for the battery cell 01. At the same time, the space between the two light-emitting elements also forms a detection position for the battery cell 01.

[0209] 8 and 9 , in some embodiments of the present disclosure, when the first light-emitting element 12121 moves upward and is located at the avoidance position, the battery cell 01 may be transported to a position between the two third light-emitting elements 12123 .

[0210] When the battery cell 01 is located at the detection position, the first light-emitting component 12121 moves downward to be located at the detection position, and the first light-emitting component 12121, the second light-emitting component 12122 and the third light-emitting component 12123 can be used to fully surround the weld.

[0211] Through the above-mentioned arrangement, in the welding detection system provided in the embodiment of the present disclosure, the first light-emitting component 12121, the second light-emitting component 12122 and the third light-emitting component 12123 can be used to form a fully surrounded lighting for the weld, so that the weld can be evenly brightened without over-exposure, thereby further improving the accuracy of weld detection.

[0212] On this basis, referring to FIG. 9 , in the embodiment of the present disclosure, the first light-emitting element 12121 , the second light-emitting element 12122 and the third light-emitting element 12123 all have planar light-emitting surfaces.

[0213] It should be noted that, relative to a point light source, in the embodiment of the present disclosure, by arranging the first light-emitting component 12121, the second light-emitting component 12122 and the third light-emitting component 12123 to have a planar light-emitting surface, it is also equivalent to arranging the first light-emitting component 12121, the second light-emitting component 12122 and the third light-emitting component 12123 to be a surface light source.

[0214] Through the above arrangement, the uniformity of the lighting of the weld by the first light-emitting component 12121, the second light-emitting component 12122 and the third light-emitting component 12123 can be further improved, and the accuracy of the weld detection can be further improved.

[0215] On this basis, referring to Figure 7, a first adjustment structure 11211 is provided between the first driving member 122 and the first mounting bracket 1121, and the first driving member 122 can adjust its position in the vertical direction through the first adjustment structure 11211; and / or, a second adjustment structure 11231 is provided between the second mounting bracket 1123 and the camera mounting bracket 1122, and the second mounting bracket 1123 can adjust its position in the vertical direction through the second adjustment structure 11231.

[0216] For example, referring to Figure 7, in the embodiment of the present disclosure, the first driving member 122 can be fixedly connected to the first mounting frame 1121 by fasteners. To this end, the corresponding fastener connection holes on the first driving member 122 or the first mounting frame 1121 can be set as long strip holes. In this way, the first adjustment structure 11211 can be formed using the above-mentioned long strip holes.

[0217] In addition, referring to Figure 7, in the embodiment of the present disclosure, the second mounting bracket 1123 can be fixedly connected to the camera mounting bracket 1122 by fasteners. For this, the corresponding fastener connection holes on the second mounting bracket and the camera mounting bracket 1122 can be set as long strip holes. In this way, the second adjustment structure 11231 can be formed by using the above-mentioned long strip holes.

[0218] It should be noted that the first adjustment structure 11211 and the second adjustment structure 11231 may also be set to other forms with reference to the above-mentioned setting of the long strip holes, and the embodiments of the present disclosure are not limited to this.

[0219] Through the above-mentioned settings, in the embodiment of the present disclosure, the relative position of the first driving member 122 relative to the first mounting bracket 1121 can be adjusted, and the relative position of the second mounting bracket 1123 relative to the camera mounting bracket 1122 can be adjusted, thereby improving the scope of application of the first detection mechanism 1 provided in the embodiment of the present disclosure and enabling the first detection mechanism 1 provided in the embodiment of the present disclosure to be applicable to different models of battery cells 01.

[0220] In addition, referring to Figure 7, in some embodiments of the present disclosure, a third adjustment structure 11223 is provided between the camera element 1211 and the camera element mounting bracket 1122, and the camera element 1211 can adjust its position along the direction of the lens of the camera element 1211 through the third adjustment structure 11223; and / or, a fourth adjustment structure 11212 is provided between the reflector 123 and the first mounting bracket 1121, and the reflector 123 can adjust its position along the vertical direction through the fourth adjustment structure 11212.

[0221] It should be noted that, referring to Figure 7, in the embodiment of the present disclosure, the camera 1211 can be fixedly connected to the camera mounting bracket 1122 by fasteners. To this end, the corresponding fastener connection holes on the camera 1211 and the camera mounting bracket 1122 can be set as long strip holes. In this way, the above-mentioned long strip holes can be used to form a third adjustment structure 11223.

[0222] In addition, referring to Figure 8, in the embodiment of the present disclosure, the reflector 123 can be fixedly connected to the first mounting frame 1121 by fasteners. To this end, the corresponding fastener connection holes on the reflector 123 and the first mounting frame 1121 can be set as long strip holes. In this way, the fourth adjustment structure 11212 can be formed using the above-mentioned long strip holes.

[0223] Through the above arrangement, in the disclosed embodiment, the relative position of the camera 1211 with respect to the camera mounting bracket 1122 can be adjusted, thereby adapting the camera 1211 to lenses of different focal lengths. Furthermore, in the disclosed embodiment, the relative position of the reflective mirror 123 and the first mounting bracket 1121 can also be adjusted, thereby adapting the reflective mirror 123 to battery cells 01 of different models and specifications.

[0224] In addition, referring to FIG. 7 , in some embodiments of the present disclosure, a fifth adjustment structure 121221 is provided between the third light-emitting member 12123 and the second light-emitting member 12122 , and the position of the third light-emitting member 12123 can be adjusted in the horizontal direction through the fifth adjustment structure 121221 .

[0225] 7 , in some embodiments of the present disclosure, the third light-emitting element 12123 can be fixedly connected to the second light-emitting element 12122 via fasteners. To this end, a plurality of fastener holes can be provided on the third light-emitting element 12123, and the third light-emitting element 12123 can be connected to different positions on the second light-emitting element 12122 by adapting to different fastener holes.

[0226] Through the above-mentioned setting, in the welding detection system provided in the embodiment of the present disclosure, the position of the third light-emitting component 12123 can be adjusted in the horizontal direction, so that the distance between the two third light-emitting components 12123 can be adjusted, thereby further making the first detection mechanism 1 provided in the embodiment of the present disclosure suitable for battery cells 01 of different models and specifications.

[0227] In addition, referring to Figures 7 and 9, in the welding detection system provided in the embodiment of the present disclosure, the planar imaging component 12 includes an imaging module, each imaging module includes two imaging units 121, and the two imaging units 121 are symmetrically arranged along a first direction, corresponding to the welds at both ends of the battery cell 01 along the first direction, respectively. The first direction refers to the extension direction of the conveying path of the battery cell 01.

[0228] Exemplarily, referring to FIG. 9 , for a square battery cell, the shape of the weld between the top cover and the shell is rectangular, including welds on two opposite long sides and welds on two opposite short sides.

[0229] 9 , the conveying path of the battery cell 01 may be set parallel to the length direction of the battery cell 01. In this regard, the two imaging units 121 of the imaging module may be set with respect to two opposing welds of the battery cell 01.

[0230] In this way, the two imaging units 121 of the imaging module can be used to simultaneously detect two opposite welds of a battery cell 01, thereby improving the detection efficiency of the welds.

[0231] On this basis, in some embodiments of the present disclosure, the planar imaging assembly 12 includes at least two imaging modules, which are arranged along a second direction for simultaneously detecting welds of at least two battery cells 01, wherein the second direction is perpendicular to the first direction.

[0232] It should be noted that in the embodiment of the present disclosure, the number of imaging modules is not limited. The number of vehicle modules can be set to two, or the number of imaging modules can be set to more than two.

[0233] For example, for a square battery, the first direction may be set to the length direction of the battery cell 01, and the second direction refers to the length direction of the battery cell 01. It is understood that at least two battery cells 01 may be arranged on the conveying mechanism 2 corresponding to the imaging module.

[0234] It should be noted that the number of battery cells 01 on the conveying mechanism 2 can be set to be the same as the number of imaging modules, and the battery cells 01 and the imaging modules can be arranged in a one-to-one correspondence.

[0235] Through the above arrangement, the first detection mechanism 1 in the embodiment of the present disclosure can be used to detect at least two battery cells 01 at the same time, which can further improve the detection efficiency of the weld bead.

[0236] On this basis, referring to FIG. 7 and FIG. 9 , in some embodiments of the present disclosure, the imaging units 121 located on the same side along the first direction in all imaging modules share the first light-emitting element 12121 and the second light-emitting element 12122 .

[0237] It should be noted that in the embodiment of the present disclosure, the dimensions of the first light-emitting element 12121 and the second light-emitting element 12122 along the second direction can be set to be larger, so that the first light-emitting element 12121 and the second light-emitting element 12122 can cover the imaging units 121 located on the same side of all imaging modules.

[0238] Through the above configuration, the number of light-emitting components can be reduced, making it easier to control the first light-emitting component 12121.

[0239] In addition, in the embodiment of the present disclosure, the welding detection system also includes a posture adjustment mechanism 3, which is arranged between the welding mechanism and the first detection mechanism 1. The posture adjustment mechanism 3 is used to adjust the posture of the battery cell 01 to the posture required by the first detection mechanism 1 to detect the battery cell 01.

[0240] It is understood that in the disclosed embodiment, the posture of battery cell 01 when the welding mechanism welds the battery cell 01 may be different from the posture of battery cell 01 when the first detection mechanism 1 detects the battery cell 01. For example, when the welding mechanism welds the battery cell 01, the battery cell 01 may be in a sideways posture; while when the first detection mechanism 1 detects the battery cell 01, the battery cell 01 may be in an upright posture.

[0241] For this purpose, a posture adjustment mechanism 3 may be provided between the welding mechanism and the first detection mechanism 1 to adjust the posture of the battery cell 01 .

[0242] 1 and 2 , in some embodiments of the present disclosure, the posture adjustment mechanism 3 may be configured as a flipping mechanism 31 . Thus, the flipping mechanism 31 may be used to flip the battery cell 01 , facilitating the first detection mechanism 1 to detect the battery cell 01 .

[0243] It should be noted that the posture adjustment mechanism 3 can be configured to perform gripping adjustment on the battery cell 01 , or can be configured to perform clamping adjustment on the battery cell 01 , and the embodiments of the present disclosure are not limited thereto.

[0244] Through the above configuration, in the welding detection system provided in the embodiment of the present disclosure, the posture adjustment mechanism 3 is provided to adjust the posture of the battery cell 01 , so as to facilitate the first detection mechanism 1 to detect the weld of the battery cell 01 .

[0245] In addition, referring to Figures 1, 2, 3 and 4, in some embodiments of the present disclosure, the detection mechanism also includes a first identity recognition component 13, which is used to obtain identification information of the battery cell 01; the first identity recognition component 13 is electrically connected to the processing component, and the processing component is also used to match the identification information of the battery cell 01 with the detection results of the weld.

[0246] It is understood that the identification information of battery cell 01 is used to represent the identity information of battery cell 01, and each battery cell 01 has different identification information. It should be noted that in the embodiments of the present disclosure, the expression method of the identification information of battery cell 01 is not limited. For example, the identification information of battery cell 01 can be expressed as a digital string. In addition, the identification information of battery cell 01 can also be expressed as a QR code or barcode.

[0247] For expressing the identification information of the battery cell 01 using a QR code or a barcode, the first identity recognition component 13 can be set using a barcode scanner.

[0248] By electrically connecting the first identification component 13 to the processing component, the processing component can be used to record the identification information of the battery cell 01.

[0249] It is understood that the processing component can also be used to record the inspection results of the weld bead of the battery cell 01. In this way, the processing component can be used to match the identification information of the battery cell 01 with the inspection results of the weld bead of the battery cell 01 to determine whether the inspection structure of the weld bead corresponding to the battery cell 01 is qualified.

[0250] Through the above-mentioned settings, in the welding detection system provided in the embodiment of the present disclosure, by setting the first identity recognition component 13, the identification information of the battery cell 01 can be identified, and the detection results of the weld of the battery cell 01 can be matched with the identification information of the battery cell 01, so as to facilitate the recording of the detection results of the weld of the battery cell 01.

[0251] On this basis, with reference to Figures 1 and 2 , in some embodiments of the present disclosure, the welding inspection system further includes a second inspection mechanism 4 , which is disposed downstream of the first inspection mechanism 1 and is equipped with a waste discharge mechanism. The second inspection mechanism 4 includes a second identification component 41 , which is electrically connected to a processing component. Before the second inspection mechanism 4 inspects the battery cell 01 , the second identification component 41 verifies the identification information of the battery cell 01 . If the weld bead corresponding to the identification information of the battery cell 01 fails inspection, the processing component controls the second inspection mechanism 4 not to inspect the battery cell 01 , and the battery cell 01 is discharged through the waste discharge mechanism.

[0252] It should be noted that, in the embodiment of the present disclosure, the type of the second detection mechanism 4 is not limited. For example, the second detection mechanism 4 can be set as a helium detection mechanism, which can detect the sealing of the battery cell 01.

[0253] It should be noted that, when the identification information of the battery cell 01 is expressed using a QR code or a barcode, the second identity recognition component 41 can be set using a barcode scanner.

[0254] In this way, the second identity recognition component 41 confirms the identification information of the battery cell 01. When the result corresponding to the identification information of the battery cell 01 is that the inspection result of the weld by the first inspection mechanism 1 is unqualified, the processing component controls the second inspection mechanism 4 not to inspect the battery cell 01, and the battery cell 01 is discharged through the waste discharge mechanism; when the result corresponding to the identification information of the battery cell 01 is that the inspection result of the weld by the first inspection mechanism 1 is qualified, the processing component controls the second inspection mechanism 4 to perform normal inspection on the battery cell 01.

[0255] It should be noted that, in the embodiment of the present disclosure, the specific structural form of the waste discharge mechanism is not limited, and the waste discharge mechanism can be set as a discharge groove on the second detection mechanism 4.

[0256] Through the above arrangement, in the embodiment of the present disclosure, when the weld bead inspection result of the battery cell 01 is unqualified, the battery cell 01 is discharged through the waste discharge mechanism, which can improve the inspection efficiency of the second inspection mechanism 4 .

[0257] The existing technology uses a 3D line scan camera to obtain an image with height information by decoding the reflected laser light from the detection area. The 3D image is then used to retrieve the weld area. When the weld surface shows height information that is higher or lower than the benchmark and outside the detection specifications, it is determined to be a defect. At the same time, the algorithm combines the defect morphology to identify the defect type. Due to the limitations of the imaging principle, the grayscale image presented by the 3D camera does not have obvious defect features, which can easily lead to the omission of battery cells, that is, the inability to detect defective battery cells. The omission of battery cells is shown in Figure 10A. The use of a 3D line scan camera to decode the reflected laser light from the detection area obtains an image with height information, as shown in Figure 10B. The grayscale image presented by the 3D camera is shown in Figure 10C.

[0258] The present disclosure provides a defect detection method. FIG11 is a first optional flow chart of a defect detection method provided by the present disclosure. As shown in FIG11 , the defect detection method has the following steps:

[0259] S101: When a battery cell reaches a first detection mechanism, a controller sends a position signal to a visual host computer based on position information of the battery cell obtained by a first position detection module.

[0260] In some embodiments of the present disclosure, through-beam sensors are provided on both sides of the belt corresponding to the detection station on the conveying mechanism 2. The first through-beam sensor is provided near the posture adjustment mechanism 3; the second through-beam sensor is provided in the middle of the detection station.

[0261] In some embodiments of the present disclosure, when the battery cell 01 moves to the first detection mechanism 1 , the position information of the battery cell 01 is acquired through a corresponding sensor.

[0262] In some embodiments of the present disclosure, after the battery cell 01 moves to the posture adjustment mechanism 3 and the posture adjustment mechanism 3 flips the battery cell 01, the battery cell 01 continues to move. When the battery cell 01 moves to the first detection mechanism 1, the controller sends a positioning signal to the visual host computer based on the position information of the battery cell 01 obtained by the through-beam sensor of the first detection mechanism 1, and obtains sensor information about the battery cell 01 through the through-beam sensor.

[0263] In some embodiments of the present disclosure, when the battery cell 01 just reaches the first detection mechanism 1, the first corresponding sensor obtains the first sub-sensor information; the battery cell 01 continues to move, and when the battery cell 01 completely reaches the first detection mechanism 1, the second corresponding sensor obtains the second sub-sensor information; the first sub-sensor information and the second sub-sensor information both belong to the position information of the battery cell 01.

[0264] S102, the visual host computer controls the fill light of the camera in the first detection mechanism to light up and controls the camera to take pictures, obtains a plane image of the weld bead, and performs defect detection on the plane image to determine the detection result of the weld bead.

[0265] In some embodiments of the present disclosure, the camera 1211 is a 2D camera 1211 .

[0266] In some embodiments of the present disclosure, after obtaining the position information of the battery cell 01, it indicates that the battery cell 01 has reached the shooting position. The visual host computer controls the lighting of the fill light 1212 of the camera 1211 in the first detection mechanism 1 and controls the camera 1211 to take pictures to obtain a planar image of the weld.

[0267] In some embodiments of the present disclosure, the visual host computer performs defect detection on the plane image to determine the detection result of the battery cell 01.

[0268] In some embodiments of the present disclosure, the visual host computer can perform defect detection on the plane image by using a preset defect detection model to determine the detection result of the weld.

[0269] It should be noted that the plane image of the battery cell 01 is a plane image of the edge of the battery cell 01 , which can show the surface features of the battery cell 01 .

[0270] It can be understood that according to the sensor information, the controller sends an in-position signal to the host computer, and the host computer controls the lighting of the fill light 1212 and the camera 1211 to take pictures, so as to obtain a planar image of the battery cell 01. Since the planar image of the battery cell 01 is obtained by combining the lighting of the fill light 1212 and the camera 1211, the features of the planar image are more comprehensive and clearer.

[0271] In some embodiments of the present disclosure, the visual host computer performs image preprocessing on the plane image through a preset defect detection model to obtain at least one defect information; performs defect information fusion on the at least one defect information to obtain fusion information; performs defect detection based on the fusion information to determine the detection result of the weld.

[0272] In some embodiments of the present disclosure, the preset defect detection model is based on an artificial intelligence detection algorithm. The artificial intelligence detection algorithm can use a neural network, which is a research hotspot in the field of artificial intelligence since the 1980s. It abstracts the human brain neural network from the perspective of information processing, establishes a simple model, and forms different networks according to different connection methods. In engineering and academia, it is often simply referred to as a neural network or a quasi-neural network. A neural network is a computational model composed of a large number of nodes (or neurons) connected to each other. Each node represents a specific output function, called an activation function. The connection between each two nodes represents a weighted value for the signal passing through the connection, called a weight, which is equivalent to the memory of the artificial neural network. The output of the network varies depending on the network connection method, weight value and activation function. The network itself is usually an approximation of a certain algorithm or function in nature, or it may be an expression of a logical strategy.

[0273] In some embodiments of the present disclosure, the preset defect detection model is obtained by annotating a large number of defect images of various types and then training the initial defect detection model.

[0274] It should be noted that the defect detection mainly involves post-weld defect detection of the short side of the planar image of the battery cell 01 .

[0275] It is understandable that when the battery cell 01 reaches the first detection mechanism 1, the controller sends a position signal to the visual host computer based on the position information of the battery cell 01 obtained by the first position detection module of the first detection mechanism 1; the visual host computer controls the lighting of the fill light 1212 of the camera 1211 in the first detection mechanism 1 and controls the camera 1211 to take a photo, obtains a planar image of the weld, and performs defect detection on the planar image to determine the detection result of the weld. In this process, by controlling the lighting of the fill light 1212 and the camera 1211 to take a photo, a better planar image of the battery cell 01 can be obtained, and the missed battery cell 01 can be effectively detected, preventing the continued flow of NG battery cells and causing safety risks, thereby improving the accuracy of defect detection.

[0276] In some embodiments of the present disclosure, the first detection mechanism 1 includes: a first identity recognition component 13; the defect detection method also includes: transmitting the battery cell 01 to the posture adjustment mechanism 3 through the conveying mechanism 2, and after the posture adjustment mechanism 3 is flipped into place, sending control information to the visual host computer; the visual host computer controls the first identity recognition component 13 to scan the code to obtain the identification information of the battery cell 01.

[0277] In some embodiments of the present disclosure, the battery cell 01 is transferred to the posture adjustment mechanism 3 by the conveying mechanism 2. After the posture adjustment mechanism 3 is flipped into place, the controller sends control information to the visual host computer so that the visual host computer controls the first identity recognition component 13 to scan the code and obtain the identification information of the battery cell 01.

[0278] It can be understood that the battery cell 01 is transmitted to the posture adjustment mechanism 3 through the conveying mechanism 2, and after being flipped into place by the posture adjustment mechanism 3, the control information is sent to the visual host computer; the visual host computer controls the first identity recognition component 13 to scan the code and obtain the identification information of the battery cell 01, so as to facilitate the subsequent update of the identification information of the battery cell 01.

[0279] In some embodiments of the present disclosure, the defect detection system includes: a production control device; the defect detection method also includes: a visual host computer displays the detection result of the battery cell 01 and sends the detection result of the battery cell 01 to the production control device; the production control device reads the identification information of the battery cell 01; based on the detection result of the battery cell 01 and the identification information of the battery cell 01, the identification information of the battery cell 01 is updated to obtain a product identification code with the detection result.

[0280] It can be understood that the visual host computer displays the test results of the battery cell 01, and can visualize the test results. At the same time, the test results of the battery cell 01 are sent to the production control equipment; the production control equipment reads the identification information of the battery cell 01; based on the test results of the battery cell 01 and the identification information of the battery cell 01, the identification information of the battery cell 01 is updated to obtain a product identification code with the test results, which is convenient for the subsequent second testing agency 4 to scan the code to obtain the test results.

[0281] Before executing S101, S103 is also executed as follows:

[0282] S103 , the battery cell is transferred to the posture adjustment mechanism via the conveying mechanism. After the battery cell is flipped into position by the posture adjustment mechanism, the battery cell is transferred to the first detection mechanism via the conveying mechanism.

[0283] In some embodiments of the present disclosure, the first detection mechanism 1 includes a first identity recognition component 13 .

[0284] In some embodiments of the present disclosure, the battery cell 01 is transferred to the posture adjustment mechanism 3 by the conveying mechanism 2 . After being flipped into position by the posture adjustment mechanism 3 , the battery cell 01 is transferred to the first detection mechanism 1 by the conveying mechanism 2 .

[0285] In some embodiments of the present disclosure, the battery cell 01 is transferred to the posture adjustment mechanism 3 by the conveying mechanism 2. After being flipped into place by the posture adjustment mechanism 3, the first identity recognition component 13 is controlled by the controller and the host computer to scan the code to obtain the identification code of the battery cell 01.

[0286] In some embodiments of the present disclosure, the battery cell 01 is transferred to the posture adjustment mechanism 3 by the conveying mechanism 2. After being flipped into place by the posture adjustment mechanism 3, control information is sent to the visual host computer. The visual host computer controls the first identity recognition component 13 to scan the code and obtain the identification information of the battery cell 01.

[0287] In some embodiments of the present disclosure, each battery cell 01 has a corresponding identification code for marking the battery cell 01 .

[0288] In some embodiments of the present disclosure, the defect detection system includes: a second detection mechanism 4 ; the second detection mechanism 4 is located behind the first detection mechanism 1 on the conveying mechanism 2 .

[0289] In some embodiments of the present disclosure, S104 is further executed in S102 as follows:

[0290] S104 : When the battery cell is transferred to the second testing mechanism via the conveying mechanism, the second testing mechanism determines a treatment of the battery cell according to a test result of the battery cell.

[0291] In some embodiments of the present disclosure, when the battery cell 01 is transferred to the second detection mechanism 4 by the conveying mechanism 2, the second detection mechanism 4 scans the product identification code of the battery cell 01 to obtain an abnormal result of the battery cell 01 or a normal result of the battery cell 01; based on the abnormal result of the battery cell 01, the battery cell 01 is discharged from the waste discharge mechanism of the helium detection machine; or, based on the normal result of the battery cell 01, the battery cell 01 is subjected to helium detection.

[0292] In some embodiments of the present disclosure, when the battery cell 01 is transferred to the second detection mechanism 4 by the conveying mechanism 2, the second detection mechanism 4 scans the product identification code of the battery cell 01, obtains the abnormal detection result of the battery cell 01, and discharges the abnormal battery cell 01 from the helium detection machine waste discharge mechanism.

[0293] In some embodiments of the present disclosure, when the battery cell 01 is transferred to the second detection mechanism 4 by the conveying mechanism 2, the second detection mechanism 4 scans the product identification code of the battery cell 01, obtains the normal detection result of the battery cell 01, and performs helium inspection on the normal battery cell 01.

[0294] It can be understood that the second detection mechanism 4 scans the product identification code of the battery cell 01 to obtain the abnormal result of the battery cell 01 or the normal result of the battery cell 01; according to the abnormal result of the battery cell 01, the battery cell 01 is discharged from the waste discharge mechanism of the helium inspection machine; or, according to the normal result of the battery cell 01, the battery cell 01 is subjected to helium inspection, which can effectively detect the missed battery cell 01 and prevent the NG battery cell from continuing to flow and causing safety risks.

[0295] In some embodiments of the present disclosure, the defect detection method also includes: when the battery cell 01 reaches the first detection mechanism 1, the controller triggers the first driving member 122 to extend to prevent the battery cell 01 from sliding; the first driving member 122 is located on the conveying mechanism 2; when the battery cell 01 does not reach the first detection mechanism 1, the controller triggers the first driving member 122 to contract to allow the battery cell 01 to slide.

[0296] In some embodiments of the present disclosure, when the battery cell 01 has not reached the first detection mechanism 1, the controller triggers the first driving member 122 to contract to allow the battery cell 01 to slide. When the battery cell 01 reaches the first detection mechanism 1, the controller triggers the first driving member 122 to extend to prevent the battery cell 01 from sliding, thereby fixing the position of the battery cell 01.

[0297] It is understandable that when the battery cell 01 reaches the first detection mechanism 1, the controller triggers the first driving member 122 to extend, blocking the battery cell 01 from sliding, thereby fixing the battery cell 01, facilitating subsequent acquisition of the battery cell 01 image, and improving the image quality of the battery cell 01 image.

[0298] In the embodiment of the present disclosure, after S102 , the battery cell image may also be saved to the production control device.

[0299] In some embodiments of the present disclosure, saving the battery cell 01 image to a production control device can facilitate subsequent extraction of the battery cell 01 image for data analysis.

[0300] In the embodiment of the present disclosure, before S102, S105-S107 are further executed as follows:

[0301] S105. The visual host computer obtains planar images of multiple sample weld beads.

[0302] In some embodiments of the present disclosure, a visual host computer obtains planar images of multiple sample welds, which contain different types of defects, and each type of defect has at least one planar image of a sample battery cell 01.

[0303] It should be noted that the planar images of multiple sample welds were collected and acquired during the treadmill operation.

[0304] S106 , performing image preprocessing and defect marking on the planar images of the plurality of sample weld beads respectively, to obtain sample defect information corresponding to each of the planar images of the plurality of sample weld beads.

[0305] In some embodiments of the present disclosure, the planar images of multiple sample welds are first preprocessed to obtain planar images of multiple preprocessed sample welds 1, and then the planar images of the multiple preprocessed sample welds are defect-labeled to obtain sample defect information corresponding to each of the planar images of the multiple sample welds.

[0306] In some embodiments of the present disclosure, defect annotation of the pre-processed planar images of the plurality of sample weld beads is performed by an algorithm engineer.

[0307] It should be noted that the sample defect information corresponding to each of the planar images of the plurality of sample weld beads is, that is, the defect type corresponding to each of the planar images of the plurality of sample weld beads.

[0308] S107 , training an initial defect detection model using sample defect information corresponding to the planar images of a plurality of sample weld beads to determine a preset defect detection model.

[0309] In some embodiments of the present disclosure, the mapping relationship between defects and detection specifications is determined through the defect information corresponding to each planar image of multiple sample welds; based on the mapping relationship between defects and detection specifications, defect learning of the labeled image is performed; and the initial defect detection model is trained through the defect learning of the image and the sample defect information corresponding to each planar image of multiple sample welds to determine a preset defect detection model.

[0310] In some embodiments of the present disclosure, the defect types and detection specifications can be matched one-to-one through the defect information corresponding to each of the planar images of multiple sample welds, and the mapping relationship between defects and detection specifications can be determined; based on the mapping relationship between defects and detection specifications, the defects of the labeled images are learned; through the defect learning of the images and the defect information corresponding to each of the planar images of multiple sample welds, the initial defect detection model is trained to determine the preset defect detection model.

[0311] In some embodiments of the present disclosure, during the process of training the initial defect detection model and determining the preset defect detection model, if the output value of the initial defect detection model is greater than the preset threshold, the model is output to obtain the preset defect detection model; if the output value of the initial defect detection model is not greater than the preset threshold, training is continued until the output value of the initial defect detection model is greater than the preset threshold, then training is stopped, the model is output, and the preset defect detection model is obtained.

[0312] For example, short-side post-weld defect detection is accomplished through AI-based detection algorithm identification and classification. First, a large number of defect images of various types are collected during machine operation. Algorithm engineers annotate each image with the defect type and then provide the images to the algorithm model for training. After training, a 2D detection model for short-side post-rolling is obtained, which is the preset defect detection model. As shown in Figure 12, during the training process to determine the preset defect detection model, S2, image preprocessing, is first performed; then S3, defect annotation / extraction, is performed. S4, mapping of defects to inspection specifications, and S5, defect learning of annotated images are performed to determine the post-rolling 2D defect learning model. During the defect detection process, S1, image input, S2, image preprocessing, S3, defect annotation / extraction, S7, defect information fusion, and S8, defect determination are performed.

[0313] It should be noted that the post-roll 2D defect learning model is a preset defect detection model; after defect labeling / extraction, defects 1, 2, 3, and 4 are obtained. During training, S3 performs defect labeling, and during defect detection, S3 performs defect extraction. During actual detection, the camera 1211 captures the image and transmits it back to the host computer software via a network cable. The software then transmits the image to the algorithm artificial intelligence model. After the image is input, simple image preprocessing (such as filtering) is performed, and an ROI box (such as a region of interest) is loaded. The weld body is located within the ROI box based on the difference in grayscale values. Finally, the artificial intelligence detection algorithm determines the presence and type of defects in the weld body area based on the mapping relationship between defects and specifications and defect characteristics. The loading of the ROI box is shown in Figure 13A, the positioning of the weld body based on the difference in grayscale values ​​is shown in Figure 13B, and the defect determination by the artificial intelligence detection algorithm is shown in Figure 13C. In addition, the plane image is inspected according to the preset defect detection model, and the inspection results are shown in Figures 14A and 14B.

[0314] It can be understood that, by obtaining a plurality of planar images of historical battery cells 01; performing image preprocessing and defect annotation on the planar images of the plurality of historical battery cells 01 respectively, and obtaining defect information corresponding to each of the planar images of the plurality of historical battery cells 01; and training the initial defect detection model through the defect information corresponding to each of the planar images of the plurality of historical battery cells 01, and determining the preset defect detection model, the accuracy of detection by the preset defect detection model can be improved.

[0315] In some embodiments of the present disclosure, FIG15 is a second optional flow chart of a defect detection method provided by the present disclosure. As shown in FIG15 , the defect detection method is as follows: a battery cell 01 moves from the top cover welding outlet to the posture adjustment mechanism 3. Once the flip mechanism is in position, the controller 5 and the visual host computer 6 control the first identification component 13 to scan the code. The scanned code result is stored locally, and the visual host computer 6 and the production control device 7 read the scanned code result. After the visual host computer receives the scanned code result, it executes S9 to determine whether the product code carries an OK or NG information. If OK, the product enters the helium inspection; if NG, the product is directly discharged from the NG slot of the helium inspection machine. The controller 5 and the visual host computer 6 control the first identification component 13 to scan the code, which is triggered by the posture adjustment mechanism 3 being in position. After the scan is complete, the visual host computer 6 retrieves the code from the first identification component 13. After the visual host computer 6 completes the code retrieval, it receives the arrival signal from the controller 5 via the EIP protocol and controls the camera 1211 to take a photo. After the photo is taken, the photo result is sent to the controller 5.

[0316] It should be noted that the helium inspection machine's NG bin, also known as the waste removal mechanism, scans the barcode. Battery cell 01 moves to the inspection station. Two sets of through-beam sensors, one located near the flipping station and the other in the middle, on either side of conveyor mechanism 2 sense the signal and transmit it to the controller. Once battery cell 01 is in position, the controller triggers the first actuator 122 to extend. Simultaneously, the controller sends a position-in-place signal to the visual host computer, which illuminates fill light 1212 and triggers camera 1211 to take a photo, which is then transmitted to the industrial computer. After the photo is taken, the host computer responds with a signal to the controller that it has finished. First actuator 122 retracts, and the controller triggers conveyor mechanism 2 to start moving battery cell 01 to the next station. During this process, the image feedback algorithm completes the inspection and the results are uploaded to the production control equipment, which interacts with the visual host computer. If the battery cell 01 is OK, helium inspection proceeds as usual. The product code carries the OK or NG information, and the helium inspection machine scans the barcode to determine whether to inspect or remove it. If it is NG, it is directly removed from the helium inspection machine's NG bin.

[0317] It can be understood that by controlling the lighting of the fill light 1212 and the camera 1211 to take pictures, a better image of the battery cell 01 can be obtained, and the missed battery cell 01 can be effectively detected to prevent the NG battery cell from continuing to flow and causing safety risks, thereby improving the accuracy of defect detection.

[0318] In order to implement the defect detection method provided by the present disclosure, a welding detection system is provided. As shown in FIG2 , the welding detection system 10 includes: a conveying mechanism 2 , a posture adjustment mechanism 3 and a first detection mechanism 1 .

[0319] Conveying mechanism 2, used for transporting battery cells 01;

[0320] A posture adjustment mechanism 3, used to flip the battery cell 01 on the conveying mechanism 2;

[0321] A first position detection module is provided on both sides of the belt of the conveying mechanism 2 corresponding to the first detection mechanism 1, and is used to sense whether the battery cell 01 is approaching and / or reaching the first detection mechanism 1;

[0322] The first detection mechanism 1 is equipped with a processing component, a fill light 1212 and a camera 1211, and is used to obtain a planar image of the weld of the battery cell 01; the processing component includes a controller and a visual host computer, and the controller is used to send an in-position signal to the visual host computer based on the position information of the battery cell 01 obtained by the first position detection module; the visual host computer is used to control the fill light 1212 of the camera 1211 in the first detection mechanism 1 to light up and control the camera 1211 to take pictures in response to the in-position signal, as well as perform defect detection on the planar image of the weld to determine the detection result of the weld.

[0323] In some embodiments of the present disclosure, the first detection mechanism 1 includes:

[0324] A first identity recognition component 13 is used to scan the battery cell 01 to obtain identification information of the battery cell 01;

[0325] The visual host computer is further configured to control the first identity recognition component 13 to scan a code in response to control information sent by the posture adjustment mechanism 3 after the battery cell 01 is flipped into position.

[0326] In some embodiments of the present disclosure, the welding detection system 10 further includes:

[0327] A production control device, configured to read the identification information of the battery cell 01 and update the identification information of the battery cell 01 based on the test result of the battery cell 01 and the identification information of the battery cell 01 to obtain a product identification code having the test result;

[0328] The visual host computer is further used to display the detection result of the battery cell 01 and send the detection result of the battery cell 01 to the production control device.

[0329] In some embodiments of the present disclosure, the welding detection system 10 further includes:

[0330] The second detection mechanism 4 is located behind the first detection mechanism 1 on the conveying mechanism 2 and is used to determine the treatment of the battery cell 01 according to the detection result of the battery cell 01.

[0331] In some embodiments of the present disclosure, the second detection mechanism 4 is further used to scan the product identification code of the battery cell 01 to obtain an abnormal result of the battery cell 01 or a normal result of the battery cell 01;

[0332] The second detection mechanism 4 is further configured to discharge the battery cell 01 from the helium detection machine waste discharge mechanism according to an abnormal result of the battery cell 01 ; or to perform a helium inspection on the battery cell 01 according to a normal result of the battery cell 01 .

[0333] In some embodiments of the present disclosure, the welding detection system 10 further includes: a first driving member 122 , the first driving member 122 being located on the conveying mechanism 2 ;

[0334] The controller is further configured to trigger the first driving member 122 to extend to prevent the battery cell 01 from sliding when it senses that the battery cell 01 has reached the first detection mechanism 1; or to trigger the first driving member 122 to retract to allow the battery cell 01 to slide when it senses that the battery cell 01 has not reached the first detection mechanism 1.

[0335] In some embodiments of the present disclosure, the first position detection module includes: a first through-beam sensor and a second through-beam sensor; wherein the first through-beam sensor is arranged at a position close to the posture adjustment mechanism 3, and the second through-beam sensor is arranged at a middle position of the detection station;

[0336] The first beam sensor is configured to obtain first sub-sensor information when sensing that the battery cell 01 is close to the first detection mechanism 1;

[0337] The second beam sensor is used to obtain second sub-sensor information when sensing that the battery cell 01 has completely reached the first detection mechanism 1; the first sub-sensor information and the second sub-sensor information both belong to the location information of the battery elevator.

[0338] In some embodiments of the present disclosure, the visual host computer is further used to perform image preprocessing on the plane image through a preset defect detection model to obtain at least one defect information; perform defect information fusion on the at least one defect information to obtain fusion information; perform defect detection based on the fusion information to determine the detection result of the weld.

[0339] In some embodiments of the present disclosure, the visual host computer is further used to obtain planar images of multiple sample weld beads;

[0340] performing image preprocessing and defect annotation on the planar images of the plurality of sample weld beads respectively to obtain sample defect information corresponding to each of the planar images of the plurality of sample weld beads;

[0341] The initial defect detection model is trained using the sample defect information corresponding to each of the planar images of the plurality of sample weld beads to determine the preset defect detection model.

[0342] In some embodiments of the present disclosure, the production control device is further configured to determine a mapping relationship between defects and inspection specifications using sample defect information corresponding to each of the planar images of the plurality of sample weld beads;

[0343] Defect learning of the labeled image based on the mapping relationship between the defects and the inspection specifications;

[0344] The initial defect detection model is trained through defect learning of the image and sample defect information corresponding to each of the planar images of the plurality of sample weld beads to determine the preset defect detection model.

[0345] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present disclosure, and they should all be included in the scope of the present disclosure. In particular, as long as there is no structural conflict, the various technical features mentioned in the various embodiments can be combined in any way. The present disclosure is not limited to the specific embodiments disclosed herein, but includes all technical solutions that fall within the scope of the present disclosure.

Claims

1. A welding detection system, comprising: A welding mechanism for welding the battery case and the top cover of a battery cell together; A first detection mechanism disposed downstream of the welding mechanism for detecting the weld bead between the welded battery case and the top cover; Wherein, the first detection mechanism includes a bracket assembly, a planar imaging assembly, and a processing assembly. The planar imaging assembly is mounted to the bracket assembly. The planar imaging assembly is used to capture a planar image of the weld bead. The processing assembly is used to obtain the planar image and determine whether the weld bead is qualified based on the planar image; The planar imaging assembly includes an imaging unit. The imaging unit includes a camera and a reflector. The detection position of the camera and the first detection mechanism is on the same side of the reflector. The lens of the camera faces the reflector. When the battery cell is at the detection position, the camera captures the reflected image of the weld bead through the reflector.

2. The welding detection system according to claim 1, wherein, The imaging unit further includes a light supplementing member for supplementing light towards the weld bead.

3. The welding detection system according to claim 2, wherein, The welding detection system further includes a conveying mechanism for conveying the battery cell to the detection position and for sending out the detected battery cell from the detection position.

4. The welding detection system according to claim 3, wherein, The light supplementing member includes a first light emitting member. The imaging unit further includes a first driving member. The fixed end of the first driving member is fixedly connected to the bracket assembly. The output end of the first driving member is connected to the first light emitting member; The first driving member is used to drive the first light emitting member to switch between a light supplementing position and an avoidance position; When the first light emitting member is at the light supplementing position, the first light emitting member extends into the conveying path of the battery cell to supplement light to the weld bead; When the first light emitting member is at the avoidance position, the first light emitting member retracts from the conveying path of the battery cell to form an avoidance for the conveying of the battery cell.

5. The welding detection system according to claim 4, wherein, The conveying mechanism includes a second driving member, a conveying member, and a first position detection module. The second driving member is used to drive the conveying member to operate. The conveying member is used to convey the battery cell. The first position detection module is used to detect the position of the battery cell. The first position detection module and the second driving member are electrically connected to the processing assembly; The processing assembly is further used to control the operation of the imaging unit and the second driving member based on the position of the battery cell.

6. The welding detection system according to claim 5, wherein, When the battery cell is at the detection position, the processing assembly controls the first driving member to drive the first light emitting member to the light supplementing position, controls the second driving member to stop operating, controls the light supplementing member to supplement light to the weld bead, and controls the camera to capture the weld bead; After the planar imaging assembly finishes capturing the weld bead, the processing assembly controls the first driving member to drive the first light emitting member to the avoidance position, and controls the second driving member to continue operating.

7. The welding detection system according to any one of claims 4 to 6, wherein, The bracket assembly includes a vertical support column arranged in the vertical direction, and an upper mounting frame and a lower mounting frame provided on the vertical support column. The planar imaging assembly is mounted on the upper mounting frame, and the conveying mechanism is mounted on the lower mounting frame.

8. The welding detection system according to claim 7, wherein, The upper mounting frame includes a first mounting frame and a camera mounting frame; the first mounting frame is fixedly connected to the vertical support column, and the camera mounting frame is fixedly connected to the first mounting frame; the first driving member is mounted on the first mounting frame, and the first driving member is used to drive the first light-emitting member to move in the vertical direction; the reflector is fixedly connected to the lower end of the first mounting frame; The conveying mechanism is used to convey the battery cell to move in the horizontal direction.

9. The welding detection system according to claim 8, wherein, The upper mounting frame further includes a second mounting frame, and the second mounting frame is fixedly connected to the lower end of the camera mounting frame; The light supplementing member further includes a second light-emitting member and a third light-emitting member. The second light-emitting member extends in the horizontal direction and is mounted on the lower end of the second mounting frame; The number of the third light-emitting members is two. The two third light-emitting members extend in the vertical direction and are spaced apart and mounted below the second light-emitting member; the space between the two third light-emitting members and the second light-emitting member forms the conveying path of the battery cell.

10. The welding detection system according to claim 9, wherein, The first light-emitting member, the second light-emitting member, and the third light-emitting member all have a planar light-emitting surface.

11. The welding detection system according to claim 9 or 10, wherein, A first adjusting structure is provided between the first driving member and the first mounting frame, and the first driving member can adjust its position in the vertical direction through the first adjusting structure; and / or, a second adjusting structure is provided between the second mounting frame and the camera mounting frame, and the second mounting frame can adjust its position in the vertical direction through the second adjusting structure.

12. The welding detection system according to any one of claims 9 to 11, wherein, A third adjusting structure is provided between the camera and the camera mounting frame, and the camera can adjust its position in the direction of the lens of the camera through the third adjusting structure; and / or, a fourth adjusting structure is provided between the reflector and the first mounting frame, and the reflector can adjust its position in the vertical direction through the fourth adjusting structure.

13. The welding detection system according to any one of claims 9 to 12, wherein, A fifth adjusting structure is provided between the third light-emitting member and the second light-emitting member, and the third light-emitting member can adjust its position in the horizontal direction through the fifth adjusting structure.

14. The welding detection system according to any one of claims 9 to 13, wherein, The planar imaging assembly includes an imaging module. Each imaging module includes two imaging units. The two imaging units are symmetrically arranged in a first direction and respectively correspond to the weld seams at both ends of the battery cell in the first direction. The first direction refers to the extending direction of the conveying path of the battery cell.

15. The welding detection system according to claim 14, wherein, The planar imaging assembly includes at least two imaging modules. The imaging modules are arranged in a second direction for simultaneously detecting the weld seams of at least two battery cells; wherein, the second direction is perpendicular to the first direction.

16. The welding detection system according to claim 15, wherein, The imaging units on the same side in the first direction among all the imaging modules share the first light-emitting member and the second light-emitting member.

17. The welding detection system according to any one of claims 3 to 16, wherein, The welding detection system further includes an attitude adjustment mechanism, which is arranged between the welding mechanism and the first detection mechanism, and is used to adjust the attitude of the battery cell to the attitude required for the first detection mechanism to detect the battery cell.

18. The welding detection system according to any one of claims 1 to 17, wherein, The detection mechanism further includes a first identity recognition component, which is used to obtain the identification information of the battery cell; the first identity recognition component is electrically connected to the processing component, and the processing component is further used to match the identification information of the battery cell with the detection result of the weld bead.

19. The welding detection system according to claim 18, wherein, The welding detection system further includes a second detection mechanism, which is arranged downstream of the first detection mechanism, and the second detection mechanism is provided with a waste discharging mechanism; the second detection mechanism includes a second identity recognition component, and the second identity recognition component is electrically connected to the processing component; Before the second detection mechanism detects the battery cell, the second identity recognition component confirms the identification information of the battery cell; In the case that the weld bead corresponding to the identification information of the battery cell is detected as unqualified, the processing component controls the second detection mechanism not to detect the battery cell, and the battery cell is discharged through the waste discharging mechanism.

20. A defect detection method, applied to a welding detection system, the welding detection system comprising: Welding mechanism and first detection mechanism; The first detection mechanism is arranged downstream of the welding mechanism. Among them, the first detection mechanism includes a bracket component, a planar imaging component and a processing component. The planar imaging component is installed on the bracket component. The planar imaging component includes an imaging unit, and the imaging unit includes a camera and a reflector. The camera and the detection position of the first detection mechanism are on the same side of the reflector, and the lens of the camera faces the reflector; the processing component includes: a vision host computer and a controller; The method includes: When the battery cell reaches the first detection mechanism, the controller sends a in-place signal to the vision host computer according to the position information of the battery cell obtained by the first position detection module; The vision host computer controls the lighting of the supplementary light component of the camera in the first detection mechanism and controls the camera to take a picture, obtains a planar image of the weld bead, and performs defect detection on the planar image to determine the detection result of the weld bead.

21. The method according to claim 20, wherein The welding detection system further includes: an attitude adjustment mechanism; the first detection mechanism includes: a first identity recognition component; The method further includes: The battery cell is transported to the attitude adjustment mechanism through a conveying mechanism. After being flipped in place by the attitude adjustment mechanism, control information is sent to the vision host computer; The battery cell is transported to the first detection mechanism through a conveying mechanism; The vision host computer controls the first identity recognition component to scan the code to obtain the identification information of the battery cell.

22. The method according to claim 20 or 21, wherein The welding detection system includes: production control equipment; The method further includes: The vision host computer displays the detection result of the battery cell and sends the detection result of the battery cell to the production control equipment; The production control equipment reads the identification information of the battery cell; Based on the detection result of the battery cell and the identification information of the battery cell, update the identification information of the battery cell to obtain a product identification code with the detection result.

23. The method according to claim 20 or 21, wherein The welding detection system includes: a second detection mechanism; the second detection mechanism is located behind the first detection mechanism on the conveying mechanism; The method further includes: When the battery cell is transmitted to the second detection mechanism through the conveying mechanism, the second detection mechanism determines the processing of the battery cell according to the detection result of the battery cell.

24. The method according to claim 23, wherein When the battery cell is transmitted to the second detection mechanism through the conveying mechanism, the second detection mechanism determines the processing of the battery cell according to the detection result of the battery cell, including: When the battery cell is transmitted to the second detection mechanism through the conveying mechanism, the second detection mechanism scans the product identification code of the battery cell to obtain an abnormal result or a normal result of the battery cell; According to the abnormal result of the battery cell, discharge the battery cell from the waste discharge mechanism of the helium leak detector; or, According to the normal result of the battery cell, perform helium leak detection on the battery cell.

25. The method according to any one of claims 20 to 24, wherein, The method further includes: When the battery cell reaches the first detection mechanism, the controller triggers the first driving member to extend to block the sliding of the battery cell; the first driving member is located on the conveying mechanism; When the battery cell does not reach the first detection mechanism, the controller triggers the first driving member to contract to enable the battery cell to slide.

26. The method according to any one of claims 20 to 25, wherein The first pair of photoelectric sensors in the first position detection module are arranged at a position close to the attitude adjustment mechanism; the second pair of photoelectric sensors are arranged at the middle position of the detection station; The method further includes: When the battery cell just reaches the first detection mechanism, the first pair of photoelectric sensors obtain first sub-sensor information; The battery cell continues to move. When the battery cell completely reaches the first detection mechanism, the second pair of photoelectric sensors obtain second sub-sensor information; both the first sub-sensor information and the second sub-sensor information belong to the position information of the battery cell.

27. The method according to any one of claims 20 to 26, wherein, Performing defect detection on the planar image to determine the detection result of the weld bead, including: The vision host computer performs image preprocessing on the planar image through a preset defect detection model to obtain at least one defect information; Performing defect information fusion on the at least one defect information to obtain fusion information; Performing defect detection according to the fusion information to determine the detection result of the weld bead.

28. The method according to any one of claims 20 to 27, wherein The method further includes: The vision host computer obtains planar images of multiple sample weld beads; Performing image preprocessing and defect annotation on the planar images of the multiple sample weld beads respectively to obtain sample defect information corresponding to each of the planar images of the multiple sample weld beads; Training an initial defect detection model through the sample defect information corresponding to each of the planar images of the multiple sample weld beads to determine a preset defect detection model.

29. The method according to claim 28, wherein, Training the initial defect detection model with the sample defect information corresponding to the planar images of the multiple sample weld beads to determine the preset defect detection model includes: Determining the mapping relationship between defects and detection specifications based on the sample defect information corresponding to the planar images of the multiple sample weld beads; Labeling the defect learning of the images based on the mapping relationship between defects and detection specifications; Training the initial defect detection model with the defect learning of the images and the sample defect information corresponding to the planar images of the multiple sample weld beads to determine the preset defect detection model.

30. A welding detection system, the welding detection system comprising: Welding mechanism and first detection mechanism; The first detection mechanism is arranged downstream of the welding mechanism. The first detection mechanism includes a bracket assembly, a planar imaging assembly and a processing assembly. The planar imaging assembly is mounted to the bracket assembly. The planar imaging assembly includes an imaging unit. The imaging unit includes a camera and a reflector. The detection position of the camera and the first detection mechanism is on the same side of the reflector. The lens of the camera faces the reflector. The processing assembly includes: a vision host computer and a controller; The welding mechanism is used to weld the battery case and the top cover of the battery cell together; The controller is used to send a in-place signal to the vision host computer according to the position information of the battery cell obtained by the first position detection module when the battery cell arrives at the first detection mechanism; The vision host computer is used to control the lighting of the fill light of the camera in the first detection mechanism and control the camera to take pictures in response to the in-place signal, and perform defect detection on the planar image of the weld bead to determine the detection result of the weld bead.

Images