Weld bead detection method and system

Abstract

The invention discloses a weld bead detection method and system. The weld bead detection system comprises an upper computer, a rotating tower, a two-dimensional camera and a three-dimensional camera, wherein the rotating tower transmits a plurality of battery cells; the two-dimensional camera is mounted at a first detection station, and is rotated to the first detection station through the rotating tower, and a two-dimensional image of a welding area of the battery cell is collected when the two-dimensional camera rotates to the first detection station; the three-dimensional camera is mounted on a second detection station, and a rotating motor is arranged on the second detection station; the upper computer rotates through the rotating tower, and when the upper computer rotates to a second detection station, the three-dimensional camera collects a three-dimensional image of a welding area of the battery cell by controlling rotation of the rotating motor; when the turret rotates to a second detection station, a welding bead of a two-dimensional positive electrode welding area in the two-dimensional image is determined for anomaly detection, and a two-dimensional detection result is obtained; performing anomaly detection on a welding bead of a three-dimensional positive electrode welding area in the three-dimensional image to obtain a three-dimensional detection result; and based on the two-dimensional detection result and the three-dimensional detection result, obtaining a welding detection result of the battery cell.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese patent application number 202410175988.9 and application date February 7, 2024. The entire content of the above-mentioned Chinese patent application is hereby incorporated into this application by reference. Technical Field

[0003] The present application relates to the field of battery detection, and in particular to a method and system for detecting welds. Background Art

[0004] Welding the positive and negative electrodes of batteries is a crucial step in the battery assembly process. Conventional technology requires welding the positive electrode to the connecting wires. After the battery cells are welded, manual inspection and assessment of weld defects are often required. This results in high manual inspection costs and the risk of missed inspections due to visual fatigue. Summary of the Invention

[0005] The main purpose of this application is to provide a weld detection method and system that can improve the accuracy and efficiency of weld detection.

[0006] The technical solution of the embodiment of the present application is implemented as follows:

[0007] In a first aspect, an embodiment of the present application provides a weld detection system, which includes: a host computer, a turret, the turret transmits multiple battery cells, a two-dimensional camera, and a three-dimensional camera, wherein:

[0008] The two-dimensional camera is installed at the first inspection station and is configured to rotate through the turret to collect a two-dimensional image of the welding area of the battery cell when the battery cell reaches the first inspection station;

[0009] The three-dimensional camera is installed at the second inspection station, and a rotating motor is provided on the second inspection station; the upper computer is configured to rotate the turret after the two-dimensional camera has completed the collection, and when the battery cell reaches the second inspection station, the three-dimensional camera collects a three-dimensional image of the welding area by controlling the rotation of the rotating motor; and at the same time as arriving at the second inspection station, the two-dimensional positive electrode welding area in the two-dimensional image is determined, and an abnormality detection is performed on the two-dimensional positive electrode welding area to obtain a two-dimensional detection result; and the three-dimensional positive electrode welding area in the three-dimensional image is determined; an abnormality detection is performed on the three-dimensional positive electrode welding area to obtain a three-dimensional detection result; and based on the two-dimensional detection result and the three-dimensional detection result, the welding detection result of the battery cell is obtained.

[0010] According to the above technical means, the battery cells are transported to the first inspection station and the second inspection station in sequence, and the camera can sequentially capture images of the welding area of the battery cells to obtain two-dimensional images and three-dimensional images. The host computer determines the two-dimensional positive electrode welding area in the two-dimensional image and the three-dimensional positive electrode welding area in the three-dimensional image based on the two-dimensional image and the three-dimensional image, respectively. The host computer then performs welding information anomaly detection on the two-dimensional positive electrode welding area and the three-dimensional positive electrode welding area, respectively, to obtain two-dimensional detection results and three-dimensional detection results. The host computer then obtains the welding detection results of the battery cells based on the two-dimensional detection results and the three-dimensional detection results. In this way, the host computer performs welding anomaly detection on the battery cell images, which can replace the manual visual inspection in the related art and perform anomaly detection on the welding information of the positive electrode welding area of the battery cell. This not only improves the accuracy and efficiency of weld detection, but also reduces labor costs and time costs.

[0011] In a second aspect, an embodiment of the present application provides a method for detecting a weld bead, the method comprising:

[0012] When the battery cell arrives at the first inspection station, a two-dimensional image of the welding area of the battery cell is collected;

[0013] When the battery cell arrives at the second inspection station, a three-dimensional image of the welding area of the battery cell is collected, and a two-dimensional positive electrode welding area in the two-dimensional image is simultaneously determined, and an abnormality detection is performed on the weld bead of the two-dimensional positive electrode welding area to obtain a two-dimensional detection result;

[0014] Determine a three-dimensional positive electrode welding area in a three-dimensional image; perform abnormality detection on the three-dimensional positive electrode welding area to obtain a three-dimensional detection result; and obtain a welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result.

[0015] Based on the above technical means, a host computer can perform welding anomaly detection on battery cell images, replacing manual visual inspection in related technologies. This allows for anomaly detection of welding information in the battery cell's weld area, improving the accuracy and efficiency of weld detection while reducing labor and time costs. Furthermore, based on the two-dimensional image and the three-dimensional image, the host computer determines the two-dimensional positive electrode welding area in the two-dimensional image and the three-dimensional positive electrode welding area in the three-dimensional image, respectively. This allows subsequent anomaly detection to be targeted to a more precise area, resulting in more accurate two-dimensional and three-dimensional detection results.

[0016] In the above solution, acquiring a two-dimensional image of the welding area of the battery cell includes:

[0017] When the battery cell arrives at the first inspection station, an image of the welding area of the battery cell is captured based on the received first acquisition signal to obtain a two-dimensional image.

[0018] According to the above technical means, based on the two-dimensional camera set on the first inspection station, the image of the battery cell arriving at the first inspection station is captured, and a more comprehensive and accurate two-dimensional image of the welding area information can be obtained. By controlling the rotation of the turntable, the battery cell can be transported, making the transportation of the battery cell more intelligent, thereby improving the efficiency and intelligence of the battery cell production.

[0019] In the above solution, collecting a three-dimensional image of the welding area of the battery cell includes:

[0020] When the battery cell reaches the second inspection station, an image of the welding area is captured based on the received second acquisition signal to obtain a three-dimensional image.

[0021] In the above solution, based on the received second acquisition signal, an image of the welding area is acquired to obtain a three-dimensional image, including:

[0022] In response to the second acquisition signal, controlling the battery cell to rotate and generate a pulse signal;

[0023] Based on the pulse signal, the welding area is imaged to obtain a three-dimensional image.

[0024] According to the above technical means, a 3D camera installed at the second inspection station captures images of battery cells arriving at the second inspection station, providing a more comprehensive and accurate 3D image of the welding area. Furthermore, by controlling the rotary motor installed at the second inspection station to drive the battery cells and encoder to rotate, the battery cells rotate, providing a basis for diversified image acquisition of the columnar battery cells, thereby improving the accuracy and efficiency of anomaly detection in the subsequent welding area of the battery cells.

[0025] In the above scheme, the detection method further includes:

[0026] After the three-dimensional image acquisition is completed, an end signal is sent; the end signal is used to control the transfer of the battery cell to the unloading station to realize the battery cell unloading.

[0027] According to the above technical means, after the three-dimensional camera finishes imaging the welding area of the battery cell, the first tray cup holding the battery cell is controlled to move to the unloading station accordingly, which not only realizes the automated imaging of the welding area of the battery cell, but also provides a basis for the efficient production of subsequent battery cells.

[0028] In the above scheme, abnormality detection is performed on the weld bead of the two-dimensional positive electrode welding area to obtain a two-dimensional detection result, and abnormality detection is performed on the weld bead of the three-dimensional positive electrode welding area to obtain a three-dimensional detection result, including:

[0029] The host computer performs semantic segmentation on the two-dimensional image and the three-dimensional image respectively to obtain the two-dimensional welding area and the third-dimensional welding area; or,

[0030] The host computer performs semantic segmentation on the two-dimensional image to obtain a two-dimensional welding area, and then maps the two-dimensional welding area to a three-dimensional image to obtain a three-dimensional welding area; or,

[0031] The host computer performs semantic segmentation on the three-dimensional image to obtain the three-dimensional welding area, and then maps the three-dimensional welding area to the two-dimensional image to obtain the two-dimensional welding area.

[0032] According to the above technical means, a two-dimensional welding area can be obtained from a two-dimensional image, and a three-dimensional welding area can be obtained from a three-dimensional image in a variety of ways, which can make the methods of obtaining the two-dimensional welding area and the three-dimensional welding area rich and varied, thereby improving the convenience of implementing the solution.

[0033] In the above solution, the two-dimensional positive electrode welding area includes the two-dimensional weld bead outer area, and the three-dimensional positive electrode welding area includes the three-dimensional weld bead outer area;

[0034] Perform abnormality detection on the weld bead in the two-dimensional positive electrode welding area to obtain a two-dimensional detection result, and perform abnormality detection on the weld bead in the three-dimensional positive electrode welding area to obtain a three-dimensional detection result, including:

[0035] Using the trained model, slag detection is performed on the two-dimensional and three-dimensional areas outside the weld bead, respectively, to obtain two-dimensional and three-dimensional slag results. The trained model is trained based on two-dimensional and three-dimensional sample images with labeled slag information.

[0036] When the two-dimensional welding slag result indicates that the welding slag to be identified exists in a first area outside the two-dimensional weld bead, and the three-dimensional welding slag result indicates that the welding slag to be identified exists in a second area, determining a two-dimensional detection result and a three-dimensional detection result based on the welding slag to be identified, the two-dimensional image, and the three-dimensional image;

[0037] The second region is a region obtained by mapping the first region to the three-dimensional image based on a mapping relationship; the mapping relationship is a corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for registration.

[0038] According to the above technical means, by simultaneously performing welding slag identification and result judgment in the two-dimensional weld bead outer area and the three-dimensional weld bead outer area, the accuracy of determining the two-dimensional detection results and the three-dimensional detection results of the welding slag detection can be improved.

[0039] In the above solution, based on the welding slag to be identified, the two-dimensional image, and the three-dimensional image, determining the two-dimensional detection result and the three-dimensional detection result includes:

[0040] Comparing a first size of the welding slag to be identified in the two-dimensional image with a first preset welding slag size to obtain a two-dimensional detection result, and comparing a second size of the welding slag to be identified in the three-dimensional image with a second preset welding slag size to obtain a three-dimensional detection result;

[0041] Based on the two-dimensional and three-dimensional inspection results, the welding inspection results of the battery cell are obtained, including:

[0042] When the two-dimensional detection result indicates that the first size is larger than the first preset slag size, and / or the three-dimensional detection result indicates that the second size is larger than the second preset slag size, the welding detection result of the battery cell is determined to be that slag exists in the welding area of the battery cell.

[0043] According to the above technical means, on the basis of obtaining the results of the welding slag identified in the two-dimensional images and three-dimensional images of different dimensions, the welding slag to be identified is further judged based on the preset welding slag size, which can improve the accuracy of welding slag identification in the welding area of the battery cell.

[0044] In the above solution, the first size includes: a first length and a first width, the first preset slag size includes: a preset slag length and a preset slag width, the second size includes: a first height, and the second preset slag size includes: a preset slag height;

[0045] Detection methods also include:

[0046] When the first length is greater than a preset slag length, and / or the first width is greater than a preset slag width, determining that the first size is greater than a first preset slag size;

[0047] In a case where the first height is greater than a preset slag height, the second size is determined to be greater than a second preset slag size.

[0048] According to the above technical means, through specific size comparison, a first size of the welding slag to be identified in the two-dimensional image is further determined, and a comparison result is obtained between the first preset welding slag size, and a second size of the welding slag to be identified in the three-dimensional image is further determined, and a comparison result is obtained between the second preset welding slag size. This can make the comparison result between the relevant sizes (first size, second size) and the preset welding slag sizes (first preset welding slag size, second preset welding slag size) more accurate.

[0049] In the above solution, the two-dimensional positive electrode welding area includes a two-dimensional weld bead area, and the three-dimensional positive electrode welding area includes a three-dimensional weld bead area; performing abnormality detection on the weld bead in the two-dimensional positive electrode welding area to obtain a two-dimensional detection result, and performing abnormality detection on the weld bead in the three-dimensional positive electrode welding area to obtain a three-dimensional detection result, including:

[0050] Extracting weld bead contours from the two-dimensional weld bead area and the three-dimensional weld bead area respectively to obtain two-dimensional arc contours and three-dimensional arc contours;

[0051] Anomaly detection is performed on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain two-dimensional detection results and three-dimensional detection results.

[0052] According to the above technical means, not only can a parameter basis be provided for checking the offset and size of subsequent welds, but also the accuracy of abnormal detection of related welds can be improved by taking the two detection results (two-dimensional detection result and three-dimensional detection result).

[0053] In the above solution, the welding area is the positive electrode welding area of the battery cell;

[0054] Perform anomaly detection on the two-dimensional arc profile and the three-dimensional arc profile respectively, and obtain two-dimensional detection results and three-dimensional detection results, including:

[0055] Determining a first distance between a center of the two-dimensional arc-shaped profile and a center of the liquid injection hole of the battery cell, and determining a second distance between a center of the three-dimensional arc-shaped profile and a center of the liquid injection hole;

[0056] Comparing the first distance with a preset offset distance to obtain a two-dimensional detection result, and comparing the second distance with the preset offset distance to obtain a three-dimensional detection result;

[0057] Based on the two-dimensional and three-dimensional inspection results, the welding inspection results of the battery cell are obtained, including:

[0058] When the two-dimensional detection result indicates that the first distance is greater than the preset offset distance, and the three-dimensional detection result indicates that the second distance is greater than the preset offset distance, it is determined that the welding detection result of the battery cell is that the weld of the battery cell is offset.

[0059] Using this technical approach, the host computer determines that the weld bead of a battery cell is offset if the distance between the 2D arc contour and the battery cell's injection hole in the 2D image is greater than a preset offset distance, and if the distance between the 3D arc contour and the battery cell's injection hole in the 3D image is also greater than the preset offset distance. This, by taking the sum of these two detection results, improves the accuracy of weld bead offset detection.

[0060] In the above solution, determining a first distance between the center of the two-dimensional arc profile and the center of the liquid injection hole of the battery cell, and determining a second distance between the three-dimensional arc profile and the center of the liquid injection hole include:

[0061] Performing centerline fitting processing on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a two-dimensional centerline of the two-dimensional arc profile and a three-dimensional centerline of the three-dimensional arc profile;

[0062] Determine the coordinates of a first circle center corresponding to the two-dimensional center line in the two-dimensional image and the coordinates of a second circle center corresponding to the three-dimensional center line in the three-dimensional image;

[0063] Determine the distance between the first circle center coordinate and the first comparison coordinate as the first distance, and determine the distance between the second circle center coordinate and the second comparison coordinate as the second distance;

[0064] The first comparison coordinates are the coordinates of the center of the injection hole in the two-dimensional image, and the second comparison coordinates are the coordinates of the center of the injection hole in the three-dimensional image.

[0065] According to the above technical means, the distance value used for subsequent weld bead offset identification can be determined more conveniently, thereby improving the accuracy and efficiency of subsequent weld bead offset detection.

[0066] In the above scheme, anomaly detection is performed on the two-dimensional arc profile and the three-dimensional arc profile respectively, and the two-dimensional detection results and the three-dimensional detection results are obtained, including:

[0067] determining a two-dimensional size of the two-dimensional arc profile and a three-dimensional size of the three-dimensional arc profile;

[0068] Comparing the two-dimensional size with a first preset weld bead size range to obtain a two-dimensional detection result, and comparing the three-dimensional size with a second preset weld bead size range to obtain a three-dimensional detection result;

[0069] Based on the two-dimensional and three-dimensional inspection results, the welding inspection results of the battery cell are obtained, including:

[0070] When the two-dimensional detection result indicates that the two-dimensional dimension is outside the first preset weld size range, and / or the three-dimensional dimension result indicates that the three-dimensional dimension is outside the second preset weld size range, the host computer determines that the welding detection result of the battery cell is that the weld size of the battery cell is abnormal.

[0071] The above technical approach determines whether the battery cell's weld bead size is abnormal based on the comparison between the two-dimensional dimensions of the two-dimensional arc profile and a first preset weld bead size range, and the comparison between the three-dimensional dimensions of the three-dimensional arc profile and a second preset weld bead size range. This simultaneous identification of weld bead size anomalies for both the two-dimensional and three-dimensional arc profiles improves the accuracy of identifying weld bead size anomalies.

[0072] In the above scheme, determining the two-dimensional size of the two-dimensional arc profile and the three-dimensional size of the three-dimensional arc profile includes:

[0073] Performing geometric transformation on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a first contour to be measured and a second contour to be measured that form a straight line;

[0074] Determine the two-dimensional size corresponding to the first contour to be measured and the three-dimensional size corresponding to the second contour to be measured.

[0075] According to the above technical means, the corresponding dimensions of the two-dimensional arc profile and the three-dimensional arc profile can be determined more accurately, thereby providing a more accurate parameter reference for subsequent determination of whether there is any abnormality in the weld size of the battery cell.

[0076] In the above solution, the two-dimensional dimensions include: the second length, the maximum width, and the minimum width, and the three-dimensional dimensions include: the maximum height and the minimum height;

[0077] Detection methods also include:

[0078] determining a second length corresponding to the first contour to be measured;

[0079] Performing equal pixel segmentation on the first contour to be measured, and determining the width of each pixel obtained by the segmentation to obtain a width set;

[0080] Performing equal pixel segmentation on the second contour to be measured, and determining the height of each pixel obtained by the segmentation to obtain a thickness set;

[0081] Determine the maximum width and minimum width in the width set, and the maximum height and minimum height in the thickness set.

[0082] According to the above technical means, by providing the specific dimensional information contained in the two-dimensional and three-dimensional dimensions, it is possible to provide a more accurate parameter basis for subsequent size-based identification of whether the first and second contours to be measured have dimensional anomalies, thereby improving the accuracy of subsequent identification of whether the weld bead dimensions of the battery cell have abnormalities.

[0083] In the above solution, the first preset weld bead size range includes: a weld bead length range and a weld bead width range, and the second preset weld bead size includes: a weld bead height range;

[0084] Detection methods also include:

[0085] When the second length is outside the weld bead length range, and / or the maximum width is outside the weld bead width range, and / or the minimum width is outside the weld bead width range, determining that the two-dimensional dimension is outside the first preset weld bead size range;

[0086] When the maximum height is outside the weld bead height range, and / or the minimum height is outside the weld bead height range, it is determined that the three-dimensional size is outside the second preset weld bead size range.

[0087] The above technical approach provides comparison results between different parameters to determine whether the two-dimensional dimension falls outside a first preset weld bead dimension range, and whether the three-dimensional dimension falls outside a second preset weld bead dimension range. This makes the dimension comparison logic in determining weld bead dimension anomalies clearer and more accurate.

[0088] In the above solution, the two-dimensional positive electrode welding area includes a two-dimensional weld bead area, and the three-dimensional positive electrode welding area includes a three-dimensional weld bead area; performing abnormality detection on the weld bead in the two-dimensional positive electrode welding area to obtain a two-dimensional detection result, and performing abnormality detection on the weld bead in the three-dimensional positive electrode welding area to obtain a three-dimensional detection result, including:

[0089] Respectively performing pixel value recognition on pixel points in the two-dimensional positive electrode weld bead region and the three-dimensional positive electrode weld bead region to obtain a first recognition result and a second recognition result;

[0090] When the first recognition result indicates that a preset brightness region exists in the third region of the two-dimensional positive electrode weld bead region, and the second recognition result indicates that a preset brightness region exists in the fourth region of the three-dimensional positive electrode weld bead region, determining a two-dimensional detection result and a three-dimensional detection result based on the preset brightness regions, the two-dimensional image, and the three-dimensional image;

[0091] Among them, the pixel value of the pixel point located in the preset brightness area is less than the preset pixel value; the fourth area is the area obtained by mapping the third area to the three-dimensional image based on the mapping relationship; the mapping relationship is the corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for alignment.

[0092] According to the above technical means, by simultaneously performing abnormality identification and result judgment in the two-dimensional weld bead area and the three-dimensional weld bead area, the accuracy of determining the two-dimensional detection results and the three-dimensional detection results of the welding abnormality detection can be improved.

[0093] In the above solution, determining the two-dimensional detection result and the three-dimensional detection result based on the preset brightness area, the two-dimensional image, and the three-dimensional image includes:

[0094] Comparing the size of the preset brightness area in the two-dimensional image with the first preset interval size to obtain a two-dimensional detection result, and comparing the size of the preset brightness area in the three-dimensional image with the second preset interval size to obtain a three-dimensional detection result;

[0095] Based on the two-dimensional and three-dimensional inspection results, the welding inspection results of the battery cell are obtained, including:

[0096] When the two-dimensional detection result indicates that the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size, and / or the three-dimensional detection result indicates that the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size, it is determined that the welding detection result of the battery cell is that there is a welding abnormality in the weld of the battery cell.

[0097] According to the above technical means, on the basis of obtaining the welding results identified in two-dimensional images and three-dimensional images of different dimensions, the weld bead of the battery cell is further judged based on the preset interval size, which can improve the accuracy of identifying the welding results of the battery cell.

[0098] In the above solution, the size of the preset brightness area in the two-dimensional image includes: a third length and a third width, the size of the preset brightness area in the three-dimensional image includes: a third height, the first preset interval size includes: a preset brightness area length and a preset brightness area width, and the second preset interval size includes: a preset brightness area height;

[0099] Detection methods also include:

[0100] When the third length is greater than the length of the preset brightness area, and / or the third width is greater than the width of the preset brightness area, determining that the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size;

[0101] When the third height is greater than the height of the preset brightness area, it is determined that the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size.

[0102] According to the above technical means, through specific size comparison, the comparison results between the relevant sizes (the size of the preset brightness area in the two-dimensional image, the size of the preset brightness area in the three-dimensional image) and the preset interval sizes (the first preset interval size, the second preset interval size) can be made more accurate.

[0103] In the above scheme, the detection method further includes:

[0104] When the weld bead detection result is characterized as a detection abnormality, a marking signal is generated; wherein the detection abnormality includes at least one of the following: the presence of welding slag in the welding area of the battery cell, the presence of offset in the weld bead of the battery cell, the presence of abnormal weld bead size in the battery cell, and the presence of welding abnormality in the weld bead of the battery cell;

[0105] The marking signal is used to mark the battery cell.

[0106] According to the above technical means, relevant marking operations can be performed in a timely manner when an abnormality is detected, so as to mark and alarm defective products, thereby quickly preventing the outflow of defective products and causing large-scale scrapping.

[0107] In the above scheme, the detection method further includes:

[0108] When the number of received abnormal detections reaches a preset number, a shutdown signal is generated; the shutdown signal is used to indicate stopping work and / or detecting the cause of the abnormality.

[0109] According to the above technical means, the weld detection system can automatically perform logic protection when relevant abnormal detection occurs multiple times.

[0110] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] The accompanying drawings herein are incorporated into and constitute a part of this specification. These drawings illustrate embodiments consistent with the present application and, together with the specification, serve to illustrate the technical solutions of the present application. Obviously, the drawings described below are merely some embodiments of the present application. Those skilled in the art can, without inventive effort, derive other drawings from these drawings.

[0112] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0113] Figure 1 A schematic diagram of a process for detecting a weld bead provided in an embodiment of the present application Figure 1 ;

[0114] Figure 2 A schematic diagram of a process for detecting a weld bead provided in an embodiment of the present application Figure 2 ;

[0115] Figure 3 A schematic diagram of a two-dimensional camera and a battery cell at a first inspection station provided in an embodiment of the present application;

[0116] Figure 4 A schematic diagram of a process for detecting a weld bead provided in an embodiment of the present application Figure 3 ;

[0117] Figure 5 A schematic diagram of a three-dimensional camera and a battery cell at a second inspection station provided in an embodiment of the present application;

[0118] Figure 6 A schematic diagram of a turntable provided with a plurality of cups for holding electric cores provided in an embodiment of the present application;

[0119] Figure 7 A schematic diagram of a process for detecting a weld bead provided in an embodiment of the present application Figure 4 ;

[0120] Figure 8 A schematic diagram of detecting weld deviation of a battery cell provided in an embodiment of the present application;

[0121] Figure 9 A schematic diagram of determining the length and width of a weld provided in an embodiment of the present application;

[0122] Figure 10 A schematic diagram of the detection process corresponding to the weld detection method provided in an embodiment of the present application;

[0123] Figure 11 A schematic diagram of the structure of a weld detection system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0124] To make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the specific technical solutions of the present application will be further described in detail below in conjunction with the drawings in the embodiments of the present application. The following embodiments are used to illustrate the present application but are not intended to limit the scope of the present application.

[0125] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used in this application are only for the purpose of describing this embodiment and are not intended to limit this application.

[0126] In the following description, references to “some embodiments,” “this embodiment,” “this embodiment,” and examples, etc., describe a subset of all possible embodiments, but it can be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict.

[0127] If similar descriptions of "first / second" appear in the application documents, the following explanation is added. In the following description, the terms "first\second\third" are merely used to distinguish similar objects and do not represent a specific order of the objects. It can be understood that "first\second\third" can be interchanged with the specific order or sequence where permitted, so that the embodiment described herein can be implemented in an order other than that illustrated or described herein.

[0128] In the embodiment of the present application, the term "and / or" is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, object A and / or object B may represent three situations: object A exists alone, object A and object B exist at the same time, and object B exists alone.

[0129] At present, the application of new energy batteries in life and industry is becoming more and more extensive. New energy batteries are not only used in energy storage power supply systems such as hydropower, thermal power, wind power and solar power stations, but are also widely used in electric vehicles such as electric bicycles, electric motorcycles, electric cars, as well as in aerospace and other fields. With the continuous expansion of the application field of power batteries, the market demand is also constantly expanding. In the embodiments of the present application, the battery involved may be a battery cell. A battery cell refers to a basic unit that can realize the mutual conversion of chemical energy and electrical energy, which can be used to make a battery module or battery pack, thereby being used to supply power to electrical devices. A battery cell may be a secondary battery, which refers to a battery cell that can continue to be used by activating the active material by charging after the battery cell is discharged. The battery cell may be a lithium-ion battery, a sodium-ion battery, a sodium-lithium-ion battery, a lithium metal battery, a sodium metal battery, a lithium-sulfur battery, a magnesium-ion battery, a nickel-hydrogen battery, a nickel-cadmium battery, a lead-acid battery, etc., and the embodiments of the present application are not limited to this.

[0130] In the embodiment of the present application, the battery may also be a single physical module including one or more battery cells to provide higher voltage and capacity. When there are multiple battery cells, the multiple battery cells are connected in series, in parallel or in hybrid via a busbar.

[0131] In related technologies, welding the positive and negative poles of batteries is a very important step in the battery assembly process, because welding anomalies can easily lead to serious consequences such as battery short circuit and explosion. When welding the positive pole, you need to first weld the positive pole of the battery to the connecting wire, and then weld the connecting wire to the positive pole of the next battery, and so on, until all the positive poles of the batteries are connected together; similarly, when welding the negative pole, you need to first weld the connecting wire to the negative pole of the last battery, and then weld the connecting wire to the negative pole of the previous battery, and so on, until all the negative poles of the batteries are connected together; here, in the battery welding process, special attention must be paid to the polarity of the battery to ensure that the positive and negative poles of the battery are connected correctly.

[0132] In practice, the battery welding process leaves weld beads on the battery casing. Furthermore, due to instability in the welding process, these weld beads may contain various defects, such as weld bead offset, pinholes, pits, cracks, and protrusions. Therefore, battery weld bead quality inspection is necessary to ensure battery performance and safety. However, manual visual inspection is typically performed, which is costly and prone to missed inspections due to visual fatigue.

[0133] Based on the above technical problems, the embodiments of the present application provide a method for detecting weld bead to improve the accuracy and efficiency of battery weld bead detection. The technical solution of the present application will be described in detail below with reference to the accompanying drawings.

[0134] See also Figure 1 , Figure 1 A schematic diagram of a process for detecting a weld bead provided in an embodiment of the present application Figure 1 The method is applied to a weld detection system, which includes a controller, a host computer, and a camera. The method includes S101 to S104:

[0135] S101: When a cylindrical battery cell arrives at a first inspection station, a camera collects an image of a welding area of the cylindrical battery cell based on a first collection signal received from a controller to obtain a two-dimensional image.

[0136] In the embodiments of the present application, the weld inspection system includes a controller, a host computer, and a camera. The controller is the core of the system, responsible for overall control and coordination. It communicates with the host computer to control the visual inspection system deployed on the host computer to execute relevant instructions and / or obtain relevant parameters sent by the visual inspection system.

[0137] In the embodiment of the present application, the controller may be a programmable logic controller (PLC).

[0138] It should be noted that the first inspection station can be a station for performing abnormal detection on the welding area of the battery cell, wherein the abnormal detection on the welding area of the battery cell includes but is not limited to: detecting whether the weld of the battery cell is offset, whether the weld size is abnormal (but not limited to: the length of the weld, the width of the weld, the height of the weld), whether there is welding slag in the area outside the weld (here, the area outside the weld can be the area adjacent to the weld within the welding area), and whether there are burst points, protrusions, pinholes, etc. in the weld.

[0139] In the embodiments of the present application, the welding area of the battery cell may refer to the area on the battery casing where the weld bead is located after the battery cell is welded (in the process of welding the positive and negative electrodes of the battery, the welding process after the battery pack is mylared, etc.); the welding area of the battery cell includes but is not limited to: the weld bead area on the battery cell, the area outside the weld bead adjacent to the weld bead area, etc. The welding area of the battery cell may be located anywhere on the battery cell, such as: the top cover area of the battery cell, the side of the battery cell, the bottom cover of the battery cell, etc.

[0140] In the embodiments of the present application, a cell may refer to a round cell. A cell generally refers to a battery cell, which is one of the basic units that make up a battery. A cell is the core component of a battery and is responsible for storing and releasing electrical energy.

[0141] In the embodiment of the present application, the battery cell may be a lithium-ion battery cell (Li-ion Cell), a lithium-polymer battery cell (Li-polymer Cell), a nickel-metal hydride battery cell (NiMH Cell), etc. The embodiment of the present application does not impose any restrictions on the type of battery cell, and the specific type can be selected according to the actual application scenario.

[0142] In the embodiment of the present application, the battery cell is the core component of the battery pack, and a battery pack usually contains multiple battery cells, which are combined together to provide the required power capacity and voltage. Among them, the battery pack refers to a device composed of multiple battery cells, which is intended to store electrical energy and provide power supply. The components of the battery pack include at least: battery cells, battery management system (BMS), casing, connecting wiring harness, connectors and interfaces, etc. These components work together to combine the battery cells into a fully functional battery pack for various application scenarios. For example, the battery pack can be applied to electric vehicles, energy storage systems, portable electronic devices, solar energy systems, wind energy systems, emergency backup power supplies, power tools or electric bicycles, etc. The embodiment of the present application does not impose any restrictions on this, and the specific selection can be made according to the actual application scenario.

[0143] It should be noted that the battery pack can use different types of battery cells, such as lithium-ion batteries, nickel-metal hydride batteries, lithium polymer batteries, etc., depending on the actual application needs and performance requirements.

[0144] In an embodiment of the present application, after the battery cell transport mechanism transports the battery cell to the first inspection station, the battery cell transport mechanism transmits an arrival signal to the controller. Simultaneously, the controller transmits a first acquisition signal to the camera based on the arrival signal, causing the camera to capture an image of the battery cell arriving at the first inspection station to obtain a two-dimensional image. The frequency at which the camera captures images of the battery cell can be determined based on actual needs, and the number of two-dimensional images can be one, two, or more.

[0145] In an embodiment of the present application, when the battery cell arrives at the first inspection station, an image of the welding area of the battery cell is captured based on the received first acquisition signal to obtain a two-dimensional image.

[0146] In an embodiment of the present application, first, the controller uses a sensor or other detection equipment (system) to detect whether the battery cell has arrived at the first detection station; wherein the detection information includes but is not limited to: including the position, signal or other specific characteristics of the detected battery cell. Then, once the controller confirms that the battery cell is in place (reached the first detection station), the controller sends a first acquisition signal to activate the camera to start; here, it includes activating related mechanical or electronic components to put the camera in a startup state. Further, the camera in the startup state can capture images of the welding area of the battery cell to obtain a two-dimensional image. Here, the camera captures images of the welding area of the battery cell, which can be: the camera moves along the edge of the battery cell or the battery cell itself moves under the camera to obtain complete information about the welding area of the battery cell. In this way, edge scanning to achieve image acquisition helps to detect surface anomalies or other problems in the welding area.

[0147] It is understood that through the above steps, the weld bead detection system obtains a two-dimensional image of the weld area of the battery cell. This image will be used for subsequent welding anomaly detection or other quality inspection steps. This helps ensure that the surface of the weld area of the battery cell is free of defects, thereby improving product quality and production efficiency.

[0148] S102: When the cylindrical battery cell reaches the second inspection station, the camera captures an image of the welding area based on a second acquisition signal sent by the controller to obtain a three-dimensional image.

[0149] In an embodiment of the present application, the second inspection station is different from the first inspection station. The difference may be reflected in the different positions of the first inspection station and the second inspection station in the weld inspection system; or the relevant acquisition structure deployed at the first inspection station is different from the relevant acquisition structure deployed at the second inspection station, etc.

[0150] In an embodiment of the present application, the second inspection station can be adjacent to the first inspection station or far away from it; wherein, the first inspection station and the second inspection station can both be stations for collecting relevant images of the welding area of the battery cell to realize abnormal detection of the welding area.

[0151] In an embodiment of the present application, image acquisition modules of different dimensions can be deployed at the first inspection station and the second inspection station, so that based on the image acquisition modules of different dimensions, the welding area of the battery cell arriving at a certain station (the first inspection station or the second inspection station) can be imaged to obtain two-dimensional and three-dimensional images.

[0152] In an embodiment of the present application, a two-dimensional camera can be deployed at the first inspection station, and a three-dimensional camera can be deployed at the second inspection station. When the battery cell arrives at the first inspection station, the two-dimensional camera captures an image of the welding area of the battery cell based on the received first acquisition signal to obtain a two-dimensional image; when the battery cell arrives at the second inspection station, the three-dimensional camera captures an image of the welding area of the battery cell based on the second acquisition signal to obtain a three-dimensional image.

[0153] In an embodiment of the present application, when the battery cell reaches the second inspection station, an image of the welding area is captured based on the received second acquisition signal to obtain a three-dimensional image.

[0154] It should be noted that the two-dimensional welding area may be a two-dimensional positive electrode welding area, and the three-dimensional welding area may be a three-dimensional positive electrode welding area.

[0155] In an embodiment of the present application, the host computer can control different camera structures to capture images of the welding area of the battery cell to obtain multiple first initial images and multiple second initial images, and then screen the multiple first initial images and the multiple second initial images based on image quality to obtain the two-dimensional image and the three-dimensional image.

[0156] In the embodiment of the present application, both the two-dimensional image and the three-dimensional image may contain information about the surface features of the weld area of the battery cell, such as shape, size, color, etc. During the weld bead inspection process, the goal of both the two-dimensional image and the three-dimensional image is to capture the surface details of the weld area of the battery cell so that subsequent image processing and analysis can detect whether there are defects in the weld area. For example, both the two-dimensional image and the three-dimensional image may include the following:

[0157] 1) Shape and size of the weld in the welding area: 2D and 3D images usually contain dimensional information about the weld, such as shape, diameter, length, and width.

[0158] 2) Surface quality of the weld in the welding area: Two-dimensional and three-dimensional images can reflect the quality of the weld surface, including flatness, uniformity, and possible defects or damage, such as: burst points, pinholes, protrusions or depressions, etc.

[0159] 3) Weld slag information outside the weld bead within the welding area: 2D and 3D images can display information such as the shape, position, and size of the weld slag within a preset area adjacent to the weld bead.

[0160] 4) Color information of the welding area: When both the 2D image and the 3D image are color images, the surface color of the welding area can be displayed.

[0161] It should be noted that the display content of the two-dimensional image and the three-dimensional image listed above is only an example. In actual application scenarios, other display content may also be included, and the embodiments of the present application do not impose any limitation on this.

[0162] In an embodiment of the present application, the camera may capture an image of the welding area of the battery cell arriving at the first inspection station based on the first acquisition signal, and after obtaining the two-dimensional image, send an acquisition end signal to the controller, so that the controller transports the battery cell to the second inspection station based on the received acquisition end signal; here, the controller continues to transport the battery cell to the second inspection station, and the controller may control the mechanical gripper to grab the battery cell to the second inspection station, or control the battery cell transport mechanism carrying the battery cell to transport the battery cell to the second inspection station.

[0163] It is understood that through the above steps, the weld detection system obtains a three-dimensional image of the weld area of the battery cell. This image will be used for subsequent weld anomaly detection or other quality inspection steps. This helps ensure that the surface of the weld area of the battery cell is free of defects, thereby improving product quality and production efficiency.

[0164] S103 : The host computer determines, based on the two-dimensional image and the three-dimensional image, a circular arc-shaped two-dimensional welding area in the two-dimensional image and a circular arc-shaped three-dimensional welding area in the three-dimensional image.

[0165] In the embodiment of the present application, the host computer may identify the weld bead area or the area outside the weld bead in the two-dimensional image and the three-dimensional image respectively to obtain the corresponding two-dimensional welding area and three-dimensional welding area.

[0166] In the embodiments of the present application, the two-dimensional weld area and the three-dimensional weld area can be the weld bead area identified in the two-dimensional image and the three-dimensional image, respectively, or can be the area outside the weld bead identified in the two-dimensional image and the three-dimensional image, respectively. Here, the weld bead area and the area outside the weld bead are both within the welding area of the battery cell; the weld bead area and the area outside the weld bead are adjacent, and the area outside the weld bead can be located on either side of the weld bead.

[0167] In the embodiment of the present application, the shapes and sizes corresponding to the two-dimensional welding area and the three-dimensional welding area can be determined according to actual needs, and the embodiment of the present application does not impose any limitations on this.

[0168] Here, the upper computer in S103 determines the two-dimensional welding area in the two-dimensional image and determines the three-dimensional welding area in the three-dimensional image, which can be achieved by any of the following three methods:

[0169] Method 1: The host computer performs semantic segmentation on the two-dimensional image and the three-dimensional image respectively to obtain the two-dimensional welding area and the three-dimensional welding area.

[0170] Method 2: The host computer performs semantic segmentation on the two-dimensional image to obtain the two-dimensional welding area, and then maps the two-dimensional welding area to the three-dimensional image to obtain the three-dimensional welding area.

[0171] Method 3: The host computer performs semantic segmentation on the 3D image to obtain the 3D welding area, and then maps the 3D welding area to the 2D image to obtain the 2D welding area.

[0172] In an embodiment of the present application, the host computer can synchronously perform semantic segmentation on the two-dimensional image and the three-dimensional image to obtain the two-dimensional welding area and the three-dimensional welding area, or asynchronously perform semantic segmentation on the two-dimensional image and the three-dimensional image to obtain the two-dimensional welding area and the three-dimensional welding area.

[0173] It should be noted that semantic segmentation is a natural step from coarse-to-fine reasoning. Its segmentation goal is to label each pixel in the image with a corresponding class. In other words, semantic segmentation is to classify each pixel in the image into one of the classes.

[0174] It can be understood that through the above steps, a two-dimensional welding area can be obtained from a two-dimensional image in a variety of ways, and a three-dimensional welding area can be obtained from a three-dimensional image, which can make the methods of obtaining the two-dimensional welding area and the three-dimensional welding area rich and varied, thereby improving the convenience of implementing the solution.

[0175] S104: The host computer performs abnormality detection on the two-dimensional welding area and the three-dimensional welding area respectively to obtain two-dimensional detection results and three-dimensional detection results; and obtains the welding detection result of the battery cell based on the two-dimensional detection results and the three-dimensional detection results.

[0176] In an embodiment of the present application, the host computer performs abnormality detection on the two-dimensional welding area to obtain a two-dimensional detection result, and performs abnormality detection on the three-dimensional welding area to obtain a three-dimensional detection result; correspondingly, the host computer simultaneously analyzes the two-dimensional detection result and the three-dimensional detection result to obtain the welding detection result of the battery cell.

[0177] In an embodiment of the present application, the host computer performs abnormality detection on the two-dimensional welding area and the three-dimensional welding area, including but not limited to: whether the weld is offset, whether the weld size is abnormal (including but not limited to: the length of the weld, the width of the weld, the height of the weld), whether there is welding slag in the area outside the weld (here, the area outside the weld can be the relevant area adjacent to the weld in the welding area), whether there are burst points, protrusions, pinholes, etc. in the weld.

[0178] In an embodiment of the present application, the host computer can AND the two-dimensional inspection results with the three-dimensional inspection results to obtain the welding inspection results of the battery cell. Here, the host computer can feed the obtained welding inspection results back to the controller, so that the controller can control the battery cell to execute the next process based on the welding inspection results. If the welding inspection result is abnormal, the abnormality is marked and the cause of the abnormality is identified. If the welding inspection result is normal, the battery cell is controlled to execute the next preset process.

[0179] In an embodiment of the present application, after obtaining the welding test results of the battery cell, the host computer can send the welding test results to the controller. Based on the received welding test results, the controller continues to transport the battery cell to the unloading station. At the unloading station, the controller determines whether to transport the battery cell to the abnormality marking station or the next station corresponding to normal operation based on the welding test results. After the three-dimensional image acquisition is completed, an end signal is sent, which is used to control the transportation of the battery cell to the unloading station to complete the battery cell unloading.

[0180] In an embodiment of the present application, the host computer can also collect statistics on the welding inspection results of the battery cells that arrive at the first inspection station and the second inspection station, and when the statistical number reaches a preset value, send a stop signal to the controller to stop the controller, or cause the controller to control the industrial computer to stop working, or cause the controller to control the battery cell transport mechanism to stop transporting the battery cells. The stop signal is used to indicate the cessation of work and / or the cause of the detection abnormality.

[0181] It should be noted that the welding inspection results can be expressed in words, such as: the weld bead size inspection is normal, the weld bead size inspection is abnormal, there is welding slag in the welding area of the battery cell, there is no welding slag in the welding area of the battery cell, there is an offset in the weld bead of the battery cell, there is no offset in the weld bead of the battery cell, and there is a welding abnormality in the weld bead of the battery cell, etc.

[0182] In an embodiment of the present application, when both the two-dimensional welding area and the three-dimensional welding area are weld bead areas, correspondingly, abnormality detection may refer to detecting whether the weld bead area of the battery cell is offset, detecting whether the weld bead size of the battery cell is abnormal (including but not limited to: whether the length of the weld bead is abnormal, whether the width of the weld bead is abnormal, whether the height of the weld bead is abnormal), and detecting whether there are burst points, protrusions, pinholes, etc. in the weld bead of the battery cell; when both the two-dimensional welding area and the three-dimensional welding area are outside the weld bead area, correspondingly, abnormality detection may refer to detecting whether there is welding slag in the outside weld bead area of the battery cell.

[0183] It should be noted that the two-dimensional detection result is the result obtained by performing abnormality detection on the two-dimensional welding area, and the three-dimensional detection result is the result obtained by performing abnormality detection on the three-dimensional welding area; here, the same abnormality detection is performed on both, and it can be synchronous detection or asynchronous detection, and the embodiments of the present application do not impose any restrictions on this.

[0184] It can be understood that through the above steps, the host computer identifies the two-dimensional welding area and the three-dimensional welding area from the welding areas of the battery cells in the two-dimensional image and the three-dimensional image respectively, so that the range of subsequent abnormality detection is positioned to a more accurate area, thereby obtaining more accurate two-dimensional detection results and three-dimensional detection results.

[0185] Exemplarily, when the battery cell arrives at the first inspection station, a two-dimensional image of the welding area of the battery cell is collected; when the battery cell arrives at the second inspection station, a three-dimensional image of the welding area of the battery cell is collected, and the two-dimensional positive electrode welding area in the two-dimensional image is determined at the same time, and the weld of the two-dimensional positive electrode welding area is detected for abnormalities to obtain a two-dimensional detection result; the three-dimensional positive electrode welding area in the three-dimensional image is determined; the weld of the three-dimensional positive electrode welding area is detected for abnormalities to obtain a three-dimensional detection result; based on the two-dimensional detection result and the three-dimensional detection result, the welding detection result of the battery cell is obtained.

[0186] It is understandable that the transported battery cells arrive at the first inspection station and the second inspection station in sequence, and the camera receives the first acquisition signal and the second acquisition signal accordingly to perform image acquisition on the welding area of the battery cell to obtain a two-dimensional image and a three-dimensional image; the host computer determines the arc-shaped two-dimensional welding area in the two-dimensional image and the arc-shaped three-dimensional welding area in the three-dimensional image based on the two-dimensional image and the three-dimensional image respectively; the host computer performs anomaly detection on the two-dimensional welding area and the three-dimensional welding area respectively to obtain a two-dimensional detection result and a three-dimensional detection result, and then the host computer obtains the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result. In this way, the host computer performs welding anomaly detection on the image of the cylindrical battery cell, which can replace the manual visual inspection in the related art and perform anomaly detection on the welding information of the welding area of the battery cell, which can not only improve the accuracy and efficiency of weld detection, but also reduce labor costs and time costs.

[0187] In some embodiments of the present application, the weld detection system further includes a turntable carrying a plurality of cups for holding batteries. When the battery cell arrives at the first inspection station in S101, the two-dimensional camera receives the first acquisition signal sent by the controller, which may include the following steps A1 and A2:

[0188] A1. When the battery cell is placed on the first supporting cup at the loading station, the controller controls the turntable to rotate to drive the first supporting cup to operate.

[0189] In the embodiment of the present application, the loading station is the first station corresponding to the abnormal identification of welding information by the controller after the relevant welding of the battery cell is completed; wherein, the loading station can be adjacent to the first detection station or can be far away from it, and the embodiment of the present application does not impose any restrictions on this.

[0190] It should be noted that the turntable can carry multiple cups, which can be used to hold multiple battery cells (batteries). The battery cells held in the cups can perform related operations, such as liquid injection, welding, or welding inspection. At the same time, the multiple cups arranged on the turntable are configured with corresponding shapes and sizes to match the battery cells or batteries.

[0191] In the embodiment of the present application, the number of the support cups on the turntable can be 8, 10, etc. Here, the number of the support cups can be determined according to actual production needs; there is no limitation on this in the embodiment of the present application.

[0192] In an embodiment of the present application, after the battery cell completes the relevant welding process, the controller controls the mechanical gripper to grab the battery cell to the loading station, and when the battery cell is placed on the first supporting cup of the loading station, the controller controls the turntable where the supporting cup is located to rotate, so as to drive the first supporting cup to the first inspection station; wherein, the turntable can rotate at a preset speed (for example: self-rotation).

[0193] It should be noted that each of the multiple battery-holding cups arranged on the turntable has the opportunity to be transferred to the loading station at a certain time. Meanwhile, when the first cup is at the loading station, the other cups on the turntable can be correspondingly located at the first inspection station, the second inspection station, and so on. As the turret rotates, the battery cells enter the loading station, move through different stations, and are unloaded at the unloading station; wherein, between the loading station and the unloading station, there is at least the first inspection station and the second inspection station.

[0194] It should be noted that the turntable can be understood as a turret for transferring battery cells, and the battery cells can be placed through a supporting cup, etc.

[0195] A2. When the first supporting cup moves to the first inspection station, the controller sends a first acquisition signal to the two-dimensional camera provided on the first inspection station.

[0196] In an embodiment of the present application, when the first cup is transferred to the first inspection station, the functional module deployed on the turntable sends a signal to the controller, so that the controller sends the first acquisition signal to the two-dimensional camera.

[0197] It is understood that through the above steps, when the battery cell is placed in the first support cup at the loading station, the controller drives the first support cup to the first inspection station, thereby sending the first acquisition signal to the two-dimensional camera. This enables the weld detection system to more accurately determine when to trigger the first acquisition signal to the camera, thereby improving the accuracy of the subsequent two-dimensional image acquisition. Moreover, by controlling the rotation of the turntable capable of holding multiple batteries, the battery cells are transferred from the loading station to the first inspection station, making the transfer of the battery cells more intelligent and improving the efficiency and intelligence of battery cell production.

[0198] Continuing from the above description, a two-dimensional camera is set on the first detection station, see Figure 2 In S101, when the battery cell arrives at the first inspection station, the camera obtains a two-dimensional image based on the first acquisition signal sent by the controller, which may include S201 and S202:

[0199] S201 : When the first tray is transferred to the first inspection station, the controller sends a first acquisition signal to the two-dimensional camera.

[0200] In an embodiment of the present application, the controller detects that the first cup is transferred to the first inspection station. There may be a detection module on the first inspection station, which sends a corresponding detection signal to the controller when it detects that the first cup containing the battery cell is transferred to the corresponding position, so that the controller sends a first acquisition signal to the two-dimensional camera.

[0201] S202: The two-dimensional camera captures an image of the welding area based on the first acquisition signal to obtain a two-dimensional image.

[0202] In an embodiment of the present application, the two-dimensional camera can be located directly above the first inspection station or to the side of the first inspection station; the number of two-dimensional cameras can be determined according to actual production needs, such as two; correspondingly, two two-dimensional cameras can be deployed at diagonal positions of the first inspection station and above the first inspection station.

[0203] In an embodiment of the present application, if the welding area of the battery cell is at the top cover position of the battery cell, then the two-dimensional camera can be located directly above the first inspection station and at a preset height from the battery cell; if the welding area of the battery cell is on the side of the battery cell (side A or side B), then the two-dimensional camera can be located above the side of the first inspection station and at a preset distance from the battery cell.

[0204] Here, see Figure 3 From top to bottom, they are: 2D camera 301, coaxial light source mechanism 302, dome lighting mechanism 303, and battery cell 304 arriving at the first inspection station. A fixed distance is set between 2D camera 301, coaxial light source mechanism 302, dome lighting mechanism 303, and battery cell 304 arriving at the first inspection station. This distance can be determined based on actual production requirements.

[0205] In an embodiment of the present application, the camera field of view of the two-dimensional camera is concentrated on the welding area of the battery cell. In this way, the two-dimensional camera can be used to detect the welding information of the weld bead during the welding process of the battery cell, and the corresponding two-dimensional image collected can be used to check the uniformity, integrity and other quality issues of the weld bead in the welding area. The two-dimensional image can also be used to detect welding slag in the welding area, that is, if welding slag is generated during the welding process, the two-dimensional image can help the host computer detect the existence, position and size of the welding slag. The two-dimensional image can also be used to help the host computer measure the shape and size of the weld bead to ensure that it meets the specifications.

[0206] It is understood that through the above steps, when the first support cup containing the battery cell is transferred to the first inspection station, the controller sends a first acquisition signal to the two-dimensional camera, and the two-dimensional camera at the first inspection station captures an image of the welding area of the battery cell to obtain a two-dimensional image. In this way, based on the two-dimensional camera set at the first inspection station, the image of the battery cell arriving at the first inspection station is captured, and a more comprehensive and accurate two-dimensional image of the welding area information can be obtained. By controlling the rotation of the turntable, the battery cell can be transferred, making the transfer of the battery cell more intelligent, thereby improving the efficiency and intelligence of battery cell production.

[0207] It should be noted that the two-dimensional camera captures images of the welding area based on the first acquisition signal. After obtaining the two-dimensional image, or performing preliminary verification on the two-dimensional image, if the two-dimensional image meets certain image detection conditions, an acquisition end signal is sent to the controller to enable the controller to continue transporting the battery cell to the second inspection station.

[0208] Continuing from the above description, a 3D camera is installed on the second inspection station, see Figure 4In S102, when the cylindrical battery cell arrives at the second inspection station, the camera obtains a three-dimensional image based on the second acquisition signal, which may include S301 to S304:

[0209] S301: When the cylindrical battery cell arrives at the second inspection station, the controller sends the second acquisition signal to the rotating motor.

[0210] In an embodiment of the present application, the controller detects that the first cup is transferred to the second inspection station. There may be a detection module on the second inspection station, which sends a corresponding detection signal to the controller when it detects that the first cup containing the battery cell is transferred to the corresponding position, so that the controller sends a second acquisition signal to the rotating motor based on the received detection signal.

[0211] S302: The host computer controls the rotary motor to drive the first support cup, the cylindrical battery cell in the first support cup, and the encoder to rotate in response to the second acquisition signal.

[0212] S303, the encoder generates and sends a pulse signal to the 3D camera during the rotation process;

[0213] In an embodiment of the present application, after receiving the second acquisition signal, the rotating motor can be controlled to rotate so that the three-dimensional camera set on the second detection station can collect the three-dimensional image, that is, the first support cup (which is essentially a battery cell) is driven to rotate in response to the second acquisition signal, and the encoder set on the second detection station is rotated, thereby controlling the rotating encoder to generate a certain pulse signal to the three-dimensional camera, and then the three-dimensional camera can collect the three-dimensional image in response to the pulse signal.

[0214] It should be noted that a rotary motor is a rotating electromagnetic machine that operates on the principle of electromagnetic induction and is used to achieve the mutual conversion of mechanical energy and electrical energy. Here, during the rotation process, the rotary motor can drive the first cup carrying the battery cell contained therein to rotate, and the encoder to rotate.

[0215] Here, see Figure 5 , from top to bottom: 3D camera 501, cylindrical battery cell 502 arriving at the second inspection station. Here, a fixed distance is set between the 3D camera 501 and the cylindrical battery cell 502 arriving at the second inspection station, and the corresponding distance can be determined according to actual production requirements. Figure 5 As shown, the cylindrical battery core 502 arriving at the second inspection station rotates under the control of the rotating motor.

[0216] In the embodiment of the present application, the encoder generates a pulse signal during rotation and sends the pulse signal to the 3D camera to start the 3D camera to capture images of the battery cell. The pulse signal is a discrete signal with a certain periodicity.

[0217] In an embodiment of the present application, the three-dimensional camera can also be located directly above the second inspection station or to the side of the second inspection station; the number of three-dimensional cameras can be determined according to actual production needs, such as two. Correspondingly, two three-dimensional cameras can be deployed at the diagonal position of the second inspection station and above the second inspection station.

[0218] In an embodiment of the present application, if the welding area of the battery cell is at the top cover position of the battery cell, then the three-dimensional camera can be located directly above the second inspection station and at a preset height from the cylindrical battery cell; if the welding area of the cylindrical battery cell is on the side of the cylindrical battery cell (side A or side B), then the three-dimensional camera can be located above the side of the second inspection station and at a preset distance from the cylindrical battery cell.

[0219] S304: The three-dimensional camera collects images of the positive electrode welding area based on the pulse signal to obtain a three-dimensional image.

[0220] In an embodiment of the present application, a three-dimensional camera captures an image of the battery cell in a rotating state based on the pulse signal to obtain a three-dimensional image.

[0221] In an embodiment of the present application, the camera field of view of the three-dimensional camera is concentrated on the welding area of the battery cell. In this way, the three-dimensional camera can be used to detect the welding information of the weld bead during the welding process of the battery cell, and the corresponding three-dimensional image can be used to check the uniformity, integrity and other quality issues of the weld bead in the positive electrode welding area. The three-dimensional image can also be used to detect welding slag in the welding area, that is, if welding slag is generated during the welding process, the three-dimensional image can help the host computer detect the existence, position and size of the welding slag. The three-dimensional image can also be used to help the host computer measure the shape and size of the weld bead to ensure that it meets the specifications.

[0222] In the embodiments of this application, a two-dimensional camera is referred to as a 2D inspection camera, and a three-dimensional camera is referred to as a 3D inspection camera. A 2D inspection camera is the most common type of camera, used to capture flat images with two dimensions: length and width. A 2D inspection camera is used to obtain surface information about static or planar scenes. In manufacturing environments, 2D inspection cameras are commonly used to detect defects on flat surfaces, identify object outlines, and read labels.

[0223] In an embodiment of the present application, a 2D detection camera may be used to capture the weld area of the battery cell for subsequent image processing and analysis. A 3D detection camera captures images with length, width, and depth (height) information. A 3D detection camera is capable of obtaining the three-dimensional shape of a target object, which is very useful for surfaces of different heights or objects of irregular shapes. In manufacturing and quality inspection, 3D detection cameras are widely used to detect three-dimensional shapes, measure the height of objects, detect volumes, and the like. In the detection of welds, a 3D detection camera may be used to obtain three-dimensional information of the welding area of the battery cell in order to more accurately analyze abnormal information in the welding area.

[0224] It should be noted that 2D inspection cameras and 3D inspection cameras can be used individually or in combination, depending on the application requirements and the complexity of the inspection.

[0225] It should be noted that the two-dimensional camera is installed at the first inspection station and is configured to rotate through the turret, and when the battery cell reaches the first inspection station, collect a two-dimensional image of the welding area of the battery cell; the three-dimensional camera is installed at the second inspection station, and a rotating motor is provided on the second inspection station; the host computer is configured to rotate the turret after the two-dimensional camera has collected the image, and when the battery cell reaches the second inspection station, the host computer controls the rotation of the rotating motor to enable the three-dimensional camera to collect a three-dimensional image of the welding area of the battery cell; and at the same time as arriving at the second inspection station, determine the two-dimensional positive electrode welding area in the two-dimensional image, perform abnormality detection on the weld of the two-dimensional positive electrode welding area, and obtain a two-dimensional detection result; and determine the three-dimensional positive electrode welding area in the three-dimensional image; perform abnormality detection on the weld of the three-dimensional positive electrode welding area to obtain a three-dimensional detection result; and obtain the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result.

[0226] It is understood that through the above steps, when the cylindrical battery cell reaches the second inspection station, the controller sends a second acquisition signal to the rotating motor; in response to the second acquisition signal, the rotating motor drives the first support cup, the cylindrical battery cell in the first support cup, and the encoder to rotate; during the rotation process, the encoder generates and sends a pulse signal to the three-dimensional camera; based on the pulse signal, the three-dimensional camera captures an image of the welding area to obtain a three-dimensional image. In this way, based on the three-dimensional camera set at the second inspection station, the battery cell arriving at the second inspection station is imaged, and a more comprehensive and accurate three-dimensional image of the welding area can be obtained. Moreover, by controlling the rotating motor set at the second inspection station to drive the first support cup and encoder to rotate, the rotation of the battery cell is achieved, providing a basis for diversified image acquisition of the battery cell, thereby improving the accuracy and efficiency of abnormality detection in the subsequent welding area of the battery cell.

[0227] In the embodiment of the present application, after obtaining the two-dimensional image and the three-dimensional image, the host computer sends an end signal to the controller, so that the controller controls the turntable to rotate to transfer the first tray cup containing the discharge core to the unloading station. That is, the weld bead detection method provided in the embodiment of the present application can also perform the following B1 and B2:

[0228] B1. After the 3D camera completes the image of the welding area, the host computer sends an end signal to the controller.

[0229] In an embodiment of the present application, after the host computer detects that the 3D camera has acquired a 3D image, or performs preliminary image screening on the 3D image and obtains a 3D image that meets the preset image screening conditions, it sends an end signal to the controller.

[0230] B2. Based on the end signal, the controller controls the rotation of the turntable to transfer the first support cup to the unloading station to realize battery cell unloading.

[0231] In the embodiment of the present application, the unloading station may be a station where a controller controls a related mechanical gripper to grab the battery cell and move it to the next process.

[0232] In an embodiment of the present application, the transportation order of the first tray containing the battery cells can be: loading station, first inspection station, second inspection station and unloading station; among them, the loading station, first inspection station, second inspection station and unloading station can be adjacent to each other in sequence, or there can be a certain interval between them.

[0233] Among them, see Figure 6 The turntable is equipped with eight cups for holding cells, namely cups 1 to 8. Cup 601 is cup 1 at the loading station, cup 602 is cup 2 at the first inspection station, cup 603 is cup 3 at the second inspection station, and cup 604 is cup 6 at the unloading station. Cups 4 and 5 can be cups at other operating stations.

[0234] In an embodiment of the present application, the controller controls the rotation of the turntable to transfer the first support cup to the unloading station. Based on the welding detection result of the battery cell obtained by the upper computer, the battery cell can be unloaded to the next process accordingly. For example, when the welding detection result of the battery cell is an abnormal detection, the battery cell is transferred from the unloading station to the abnormal marking station; or, when the welding detection result of the battery cell is normal, the battery cell is transferred from the unloading station to the next normal operation process.

[0235] It is understood that through the above steps, after the 3D camera completes image acquisition of the welding area, the host computer sends an end signal to the controller, causing the controller to control the rotation of the turntable to drive the first support cup to be transported to the unloading station. In this way, after the host computer completes imaging of the welding area of the battery cell, it correspondingly controls the first support cup holding the battery cell to move to the unloading station, which not only enables automated imaging of the welding area of the battery cell, but also lays a foundation for the efficient production of subsequent battery cells.

[0236] In some embodiments of the present application, the two-dimensional welding area includes a two-dimensional weld bead outer area, and the three-dimensional welding area includes a three-dimensional weld bead outer area; Figure 7 As shown, the "host computer performs abnormality detection on the two-dimensional welding area and the three-dimensional welding area respectively to obtain a two-dimensional detection result and a three-dimensional detection result" in S104 may include S401 and S402:

[0237] S401. The host computer uses the trained model to perform welding slag detection on the two-dimensional weld bead outer area and the three-dimensional weld bead outer area respectively, and obtains the two-dimensional welding slag result and the three-dimensional welding slag result.

[0238] The trained model is obtained by training based on two-dimensional sample images and three-dimensional sample images with labeled welding slag information; the dimensions of the two-dimensional sample images and the three-dimensional sample images are different.

[0239] In an embodiment of the present application, the trained model can be trained by supervised learning, that is, trained based on two-dimensional sample images and three-dimensional sample images with labeled welding slag information. Here, the two-dimensional sample images and three-dimensional sample images of different dimensions can each include multiple first sample sub-images and multiple second sample sub-images. The loss is determined by comparing the similarity between the labeled welding slag information and the identified welding slag information, and the parameters of the model to be trained, such as the weight value, are adjusted based on the loss, so that the loss of the welding slag information output by the trained model converges.

[0240] In the embodiment of the present application, welding slag information is marked inside the two-dimensional sample image and the three-dimensional sample image, which may include but is not limited to: the size of the welding slag, the position of the welding slag, the color of the welding slag, etc.

[0241] It should be noted that welding is achieved by heating and applying pressure to metal to melt it, and then solidifying it by cooling. During the welding process, welding wire or welding rod is usually used as a filler material. These fillers contain some metal alloys and flux. After the filler material is heated and melted, a layer of copper shavings-like substance will form on its surface, which is the welding slag.

[0242] Weld slag has a significant impact on welding quality. 1. Weld slag reduces the mechanical properties of the weld. The presence of slag can affect the weld strength, thereby impacting the strength and stability of the entire welded structure. 2. Weld slag can reduce the sealing performance of weld joints. Weld slag can increase gaps within weld joints, thereby impacting the sealing performance and posing a potential safety hazard to the welded structure. Therefore, appropriate measures are needed to reduce the generation of weld slag and identify weld slag information to ensure the stability and reliability of welding quality.

[0243] In an embodiment of the present application, the host computer uses a trained model to perform welding slag detection on the area outside the two-dimensional weld bead to obtain a two-dimensional welding slag result, and performs welding slag detection on the area outside the three-dimensional weld bead to obtain a three-dimensional welding slag result; wherein, the two-dimensional welding slag result and the three-dimensional welding slag result can be respectively expressed as: no welding slag, the presence of welding slag (and can further carry information such as the position of the welding slag and the size of the welding slag).

[0244] S402. When the two-dimensional welding slag result indicates that there is welding slag to be identified in the first area outside the two-dimensional weld bead, and the three-dimensional welding slag result indicates that there is welding slag to be identified in the second area, the host computer determines the two-dimensional detection result and the three-dimensional detection result based on the welding slag to be identified, the two-dimensional image, and the three-dimensional image.

[0245] The second region is a region obtained by mapping the first region to the three-dimensional image based on a mapping relationship; the mapping relationship is a corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for registration.

[0246] In the embodiment of the present application, if the host computer detects the presence of unidentified weld slag in both the two-dimensional and three-dimensional weld bead areas, the size of the unidentified weld slag can be further identified to obtain two-dimensional and three-dimensional detection results. Here, further identifying the size of the unidentified weld slag means that the unidentified weld slag is considered to be substantial weld slag that affects the welding results of the battery cell only if its size meets certain conditions.

[0247] In embodiments of the present application, a mapping relationship between the two-dimensional image and the three-dimensional image can be further determined, namely, a mapping relationship between the two-dimensional weld bead area and the three-dimensional weld bead area. Here, the two images can be imported into the same coordinate system for registration to obtain a corresponding mapping relationship. The registration of the two-dimensional image and the three-dimensional image into the same coordinate system can be based on a reference object, such as a cell contour point or location that exists in both the two-dimensional and three-dimensional images.

[0248] Here, the same coordinate system can be a two-dimensional plane coordinate system or a three-dimensional coordinate system.

[0249] In an embodiment of the present application, when the upper computer detects that the two-dimensional welding slag result indicates that there is welding slag to be identified in the first area outside the two-dimensional weld bead, and the three-dimensional welding slag result indicates that there is also welding slag to be identified in the second area outside the three-dimensional weld bead, the upper computer can perform size detection on the welding slag to be identified to determine whether the welding slag indicates that there is welding slag that affects the welding results in the area outside the weld bead.

[0250] It should be noted that the welding slag to be identified is not necessarily the welding slag that affects the welding detection results of the battery cell; that is, if the size of the welding slag to be identified is small and not enough to affect the welding quality of the battery cell, it can be said that there is no welding slag in the area outside the weld within the welding area of the battery cell.

[0251] It can be understood that through the above steps, first, the host computer uses the trained model to perform welding slag detection on the two-dimensional and three-dimensional weld bead areas, obtaining two-dimensional and three-dimensional welding slag results. Then, if both the two-dimensional and three-dimensional welding slag results indicate the presence of welding slag to be identified in the relevant area, the two-dimensional and three-dimensional detection results can be further determined based on the welding slag to be identified, the two-dimensional image, and the three-dimensional image. In this way, by simultaneously performing welding slag identification and result judgment on the two-dimensional and three-dimensional weld bead areas, the accuracy of determining the two-dimensional and three-dimensional detection results for welding slag detection can be improved.

[0252] In some embodiments of the present application, the upper computer in S402 determines the two-dimensional detection result and the three-dimensional detection result based on the welding slag to be identified, the two-dimensional image, and the three-dimensional image, which may include C1:

[0253] C1. The host computer compares the first size of the welding slag to be identified in the two-dimensional image with the first preset welding slag size to obtain a two-dimensional detection result, and compares the second size of the welding slag to be identified in the three-dimensional image with the second preset welding slag size to obtain a three-dimensional detection result.

[0254] In an embodiment of the present application, the host computer can compare the first size of the welding slag to be identified in the two-dimensional image or the area outside the two-dimensional weld bead with the first preset welding slag size to obtain a two-dimensional detection result; at the same time, the host computer can compare the second size of the welding slag to be identified in the three-dimensional image or the area outside the three-dimensional weld bead with the second preset welding slag size to obtain a three-dimensional detection result.

[0255] In an embodiment of the present application, the two-dimensional detection result can be described using a first size greater than, equal to, or less than a first preset slag size; correspondingly, the three-dimensional detection result can be described using a second size greater than, equal to, or less than a second preset slag size.

[0256] In an embodiment of the present application, the first size may include: the first length of the welding slag to be identified, the first width of the welding slag to be identified; correspondingly, the first preset welding slag size may include: the preset welding slag length, the preset welding slag width; at the same time, the second size may include: the second length of the welding slag to be identified, the second width of the welding slag to be identified and the first height of the welding slag to be identified; correspondingly, the second preset welding slag size includes: the preset welding slag length, the preset welding slag width and the preset welding slag height.

[0257] Here, the units corresponding to the first size and the second size can be expressed in millimeters.

[0258] Continuing from the above description, after performing abnormality detection on the two-dimensional weld bead outer area and the three-dimensional weld bead outer area simultaneously, such as whether welding slag is present, and determining the two-dimensional detection results and the three-dimensional detection results, the upper computer in S103 obtains the welding detection results of the battery cell based on the two-dimensional detection results and the three-dimensional detection results, which may include the following C2:

[0259] C2. When the two-dimensional detection result indicates that the first size is larger than the first preset slag size, and / or the three-dimensional detection result indicates that the second size is larger than the second preset slag size, the host computer determines that the welding detection result of the battery cell is that slag exists in the welding area of the battery cell.

[0260] In the embodiment of the present application, the host computer determines that the welding detection result of the battery cell is abnormal, that is, there is welding slag in the welding area of the battery cell (the area outside the weld bead inside it) that affects the welding quality.

[0261] In the embodiment of the present application, the host computer determines that the welding detection result of the battery cell is that there is welding slag in the welding area of the battery cell, which may include the following three situations:

[0262] Case (1): the two-dimensional detection result indicates that the first size is larger than the first preset welding slag size, and the three-dimensional detection result indicates that the second size is larger than the second preset welding slag size.

[0263] Case (2): The two-dimensional detection result indicates that the first size is larger than the first preset welding slag size.

[0264] Case (3): The three-dimensional detection result indicates that the second size is larger than the second preset welding slag size.

[0265] It is understood that through the above steps, the host computer determines that welding slag that affects welding quality is present in the welding area of the battery cell when the first size of the welding slag to be identified in the two-dimensional image is larger than the first preset welding slag size, and / or the second size of the welding slag to be identified in the three-dimensional image is larger than the second preset welding slag size. In this way, based on the results of welding slag identification in the two-dimensional image and the three-dimensional image of different dimensions, further judgment of the welding slag to be identified based on the preset welding slag size can improve the accuracy of welding slag identification in the welding area of the battery cell.

[0266] Continuing with the above description, the dimensions of the welding slag to be identified in the length and width directions are extracted from the first dimension, and the dimension of the welding slag to be identified in the height direction is extracted from the second dimension, and the subsequent dimension judgment of the welding slag to be identified is performed accordingly; that is, when the first dimension includes: a first length and a first width, the first preset welding slag dimension includes: a preset welding slag length and a preset welding slag width, and correspondingly, the second dimension includes: a first height, and the second preset welding slag dimension includes: a preset welding slag height, the implementation of determining in C2 that the first dimension is greater than the first preset welding slag dimension and the second dimension is greater than the second preset welding slag dimension may include C21 and C22:

[0267] C21. When the first length is greater than the preset slag length and / or the first width is greater than the preset slag width, the host computer determines that the first size is greater than the first preset slag size.

[0268] In the embodiment of the present application, the host computer determines that the first size is greater than the first preset slag size in the following three situations:

[0269] Case (1): the first length is greater than the preset slag length, and the first width is greater than the preset slag width.

[0270] Case (2): The first length is greater than the preset welding slag length.

[0271] Case (3): The first width is greater than the preset slag width.

[0272] C22. When the first height is greater than the preset slag height, the host computer determines that the second size is greater than the second preset slag size.

[0273] In the embodiment of the present application, the preset welding slag length, the preset welding slag width and the preset welding slag height can be determined according to the actual needs of the battery cell in the welding result detection link, and the present application does not impose any restrictions on this.

[0274] It should be noted that if the second size includes: the second length of the welding slag to be identified, the second width of the welding slag to be identified and the first height of the welding slag to be identified; correspondingly, the second preset welding slag size includes: the preset welding slag height, the preset welding slag width and the preset welding slag length.

[0275] It is understood that through the above steps, by performing specific size comparisons, a comparison result of a first size of the weld slag to be identified in the two-dimensional image and a first preset weld slag size is further determined, and a comparison result of a second size of the weld slag to be identified in the three-dimensional image and a second preset weld slag size is determined. This can make the comparison result between the relevant sizes (first size, second size) and the preset weld slag sizes (first preset weld slag size, second preset weld slag size) more accurate.

[0276] In some embodiments of the present application, the two-dimensional welding area includes a two-dimensional weld bead area, and the three-dimensional welding area includes a three-dimensional weld bead area; the host computer in S104 performs abnormality detection on the two-dimensional welding area and the three-dimensional welding area respectively to obtain the two-dimensional detection result and the three-dimensional detection result, which may include S403 and S404:

[0277] S403 , the host computer extracts weld bead contours from the two-dimensional weld bead area and the three-dimensional weld bead area respectively to obtain a two-dimensional arc contour and a three-dimensional arc contour.

[0278] In an embodiment of the present application, the host computer can use semantic segmentation to extract weld bead contours for the two-dimensional weld bead area and the three-dimensional weld bead area respectively to obtain a two-dimensional arc contour within the two-dimensional weld bead area and a three-dimensional arc contour within the three-dimensional weld bead area.

[0279] It should be noted that the two-dimensional arc profile is the figure constituting the weld bead within the two-dimensional weld bead area, or the line of the outer edge of the weld bead; correspondingly, the three-dimensional arc profile is the figure constituting the weld bead within the three-dimensional weld bead area, or the line of the outer edge of the weld bead.

[0280] In the embodiment of the present application, the relevant attribute information of the two-dimensional arc profile and the relevant attribute information of the three-dimensional arc profile can be the same or different, such as: the length corresponding to the weld bead, the width corresponding to the weld bead, the specific position of the weld bead, etc.

[0281] S404: The host computer performs abnormality detection on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a two-dimensional detection result and a three-dimensional detection result.

[0282] In the embodiments of the present application, the host computer can perform anomaly detection on the two-dimensional arc profile and the three-dimensional arc profile, such as whether the weld bead profile is offset, whether the weld bead size is abnormal (the weld bead is too long or too short, the weld bead is too wide or not wide enough, the weld bead is too high or not high enough), and thus obtain corresponding two-dimensional detection results and three-dimensional detection results. Here, the two-dimensional detection results and the three-dimensional detection results can be represented by: whether the weld bead is offset or not offset; or whether the weld bead size is abnormal or normal.

[0283] It should be noted that if the weld bead is offset or has abnormal weld bead dimensions, it will also have a significant impact on welding quality, such as welding deformation, welding material waste, or poor welding stability.

[0284] It can be understood that through the above steps, the host computer performs anomaly detection on the two-dimensional arc profile corresponding to the two-dimensional weld bead area and the three-dimensional arc profile corresponding to the three-dimensional weld bead area, respectively, to obtain two-dimensional detection results and three-dimensional detection results. This not only provides a parameter basis for subsequent inspection of weld offset and size, but also improves the accuracy of determining anomaly detection of the relevant weld bead by taking the two detection results (two-dimensional detection result and three-dimensional detection result) into account.

[0285] In the embodiment of the present application, since the positive electrode of the battery cell itself has a liquid injection hole, which is located at the center of the positive electrode top cover of the battery cell, when the welding area of the battery cell is the positive electrode welding area of the battery cell, the content presented in the two-dimensional image and the three-dimensional image acquired by the host computer includes the liquid injection hole of the battery cell; thus, the two-dimensional detection result and the three-dimensional detection result can be determined by determining the first distance between the two-dimensional arc contour and the liquid injection hole, the second distance between the three-dimensional arc contour and the liquid injection hole, and the preset offset distance; that is, the host computer in S404 performs abnormality detection on the two-dimensional arc contour and the three-dimensional arc contour respectively, and obtains the implementation of the two-dimensional detection result and the three-dimensional detection result, including the following D1 and D2:

[0286] D1. The host computer determines a first distance between the center of the two-dimensional arc contour and the center of the liquid injection hole of the cylindrical battery cell, and determines a second distance between the three-dimensional arc contour and the center of the liquid injection hole.

[0287] In the embodiment of the present application, the units corresponding to the first distance and the second distance can be expressed in mm or cm; and the specific numerical values corresponding to the first distance and the second distance can be the same or different.

[0288] Here, the liquid injection hole of the battery cell is usually circular in practical applications.

[0289] In an embodiment of the present application, the first reference point can be the first center coordinate of the two-dimensional arc profile in the two-dimensional image (two-dimensional weld bead area), and the second center coordinate of the three-dimensional arc profile in the three-dimensional image (three-dimensional weld bead area), so that the distance between the first center coordinate and the first comparison coordinate of the injection hole center in the two-dimensional image is determined as the first distance; and the distance between the second center coordinate and the second comparison coordinate of the injection hole center in the three-dimensional image is determined as the second distance. Here, the first distance can include distance values in two dimensions, and the second distance can include distance values in three dimensions.

[0290] In the embodiment of the present application, the implementation of the host computer in D1 determining the first distance and the second distance may include the following D11 and D12:

[0291] D11. The host computer determines the first center coordinates corresponding to the two-dimensional arc contour in the two-dimensional image and the second center coordinates corresponding to the three-dimensional arc contour in the three-dimensional image.

[0292] In the embodiment of the present application, the first center coordinates can be represented by (x1, y1); correspondingly, the second center coordinates can be represented by (x2, y2, z2).

[0293] In an embodiment of the present application, the upper computer can first fit the center line of the two-dimensional arc contour to obtain the two-dimensional center line, and then identify the center coordinates corresponding to the two-dimensional center line in the two-dimensional image as the first circle center coordinates; correspondingly, the upper computer also fits the center line of the three-dimensional arc contour to obtain the three-dimensional center line, and then identifies the center coordinates corresponding to the three-dimensional center line in the three-dimensional image as the second circle center coordinates.

[0294] D12. The host computer determines the distance between the first circle center coordinate and the first comparison coordinate as the first distance, and determines the distance between the second circle center coordinate and the second comparison coordinate as the second distance.

[0295] The first comparison coordinates are the coordinates of the center of the injection hole in the two-dimensional image, and the second comparison coordinates are the coordinates of the center of the injection hole in the three-dimensional image.

[0296] In the embodiment of the present application, the first distance may include: distance difference 1 in the X-axis direction and distance difference 2 in the Y-axis direction. Correspondingly, the second distance may include: distance difference 3 in the X-axis direction, distance difference 4 in the Y-axis direction, and distance difference 5 in the Z-axis direction.

[0297] In an embodiment of the present application, the host computer can directly use traditional algorithms, such as threshold segmentation, to extract the center of the injection hole in the two-dimensional image and the three-dimensional image respectively, and then use the operator of the center of the circle to obtain the center of the circle in the machine vision algorithm package (such as HALCON), for example, the area center (area_center), to calculate the first comparison coordinates (X1, Y1) of the center of the injection hole in the two-dimensional image and the second comparison coordinates (X2, Y2, Z2) in the three-dimensional image.

[0298] It can be understood that through the above steps, the host computer determines the distance between the first center coordinate of the two-dimensional arc contour in the two-dimensional image and the coordinate corresponding to the center of the injection hole; and the distance between the second center coordinate of the three-dimensional arc contour in the three-dimensional image and the coordinate corresponding to the center of the injection hole as the first distance and the second distance, respectively. This allows for more convenient determination of distance values for subsequent weld bead offset identification, thereby improving the accuracy and efficiency of subsequent weld bead offset detection.

[0299] D2. The host computer compares the first distance with the preset offset distance to obtain a two-dimensional detection result, and compares the second distance with the preset offset distance to obtain a three-dimensional detection result.

[0300] In an embodiment of the present application, the preset offset distance may include: a first offset distance corresponding to the X-axis direction, a second offset distance corresponding to the Y-axis direction, and a third offset distance corresponding to the Z-axis direction; exemplarily, [lx, ly, lz].

[0301] Continuing from the above description, the first distance is compared with the preset offset distance, which can be: distance difference 1 is compared with lx, and distance difference 2 is compared with ly, to obtain a two-dimensional detection result; similarly, the second distance is compared with the preset offset distance, which can be: distance difference 3 is compared with lx, distance difference 4 is compared with ly, and distance difference 5 is compared with lz, to obtain a three-dimensional detection result.

[0302] Continuing from the above description, when abnormality detection is performed on both the two-dimensional weld bead area and the three-dimensional weld bead area, such as whether the weld bead is offset, after determining the two-dimensional detection results and the three-dimensional detection results, the upper computer in S104 obtains the welding detection results of the battery cell based on the two-dimensional detection results and the three-dimensional detection results, which may include the following D3:

[0303] D3. When the two-dimensional detection result indicates that the first distance is greater than the preset offset distance, and the three-dimensional detection result indicates that the second distance is greater than the preset offset distance, the host computer determines that the welding detection result of the battery cell is that the weld bead of the battery cell is offset.

[0304] In the embodiment of the present application, the host computer determines that the welding detection result of the battery cell is abnormal, that is, there is a welding offset in the weld of the welding area (the weld area inside it) of the battery cell.

[0305] Continuing with the above description, a 2D detection result indicating that the first distance is greater than the preset offset distance can refer to the following: distance difference 1 is greater than lx; and / or distance difference 2 is greater than ly. Similarly, a 3D detection result indicating that the second distance is greater than the preset offset distance can refer to the following: distance difference 3 is greater than lx; and / or distance difference 4 is greater than ly; and / or distance difference 5 is greater than lz.

[0306] In the embodiments of the present application, the battery cell weld bead is offset, which indicates that the weld bead within the positive electrode welding area of the battery cell is offset from a preset position. Alternatively, the offset means that the actual weld bead of the battery cell exceeds the distance corresponding to the preset weld bead position by more than the preset offset distance.

[0307] Here, see Figure 8 , wherein 801 is a two-dimensional weld bead area identified from a two-dimensional image, or a three-dimensional weld bead area identified from a three-dimensional image (a two-dimensional projection of a three-dimensional weld bead area); 802 is a two-dimensional arc profile extracted from the two-dimensional weld bead area, or a three-dimensional arc profile extracted from the three-dimensional weld bead area; 803 is the first center coordinate corresponding to the two-dimensional arc profile in the two-dimensional image (two-dimensional weld bead area), or the second center coordinate corresponding to the three-dimensional arc profile in the three-dimensional image (three-dimensional weld bead area); 804 is the first comparison coordinate of the liquid injection hole of the battery cell in the two-dimensional image (two-dimensional weld bead area), or the second comparison coordinate of the three-dimensional arc profile in the three-dimensional image (three-dimensional weld bead area); 805 is a schematic diagram of positionally comparing the first center coordinate with the first comparison coordinate, or a schematic diagram of positionally comparing the second center coordinate with the second comparison coordinate.

[0308] It can be understood that through the above steps, the host computer determines that the weld bead of the battery cell is offset if the distance between the two-dimensional arc contour and the battery cell's injection hole in the two-dimensional image is greater than the preset offset distance, and the distance between the three-dimensional arc contour and the battery cell's injection hole in the three-dimensional image is also greater than the preset offset distance. In this way, by taking the sum of these two detection results, the accuracy of the weld bead offset detection can be improved.

[0309] In the embodiment of the present application, based on the comparison between the two-dimensional size of the two-dimensional arc profile and the first preset weld bead size range, and the comparison between the three-dimensional size of the three-dimensional arc profile and the second preset weld bead size range, a two-dimensional detection result and a three-dimensional detection result are obtained; that is, the host computer in S404 performs abnormality detection on the two-dimensional arc profile and the three-dimensional arc profile respectively, and the implementation of obtaining the two-dimensional detection result and the three-dimensional detection result includes the following E1 and E2:

[0310] E1. The host computer determines the two-dimensional size of the two-dimensional arc profile and the three-dimensional size of the three-dimensional arc profile.

[0311] In an embodiment of the present application, the two-dimensional size of a two-dimensional arc profile may refer to the size corresponding to the two-dimensional arc profile in the length direction, and / or the size corresponding to the two-dimensional arc profile in the width direction; correspondingly, the three-dimensional size of a three-dimensional arc profile may refer to the size corresponding to the three-dimensional arc profile in the length direction, and / or the size corresponding to the three-dimensional arc profile in the width direction, and / or the size corresponding to the three-dimensional arc profile in the height (depth) direction.

[0312] In an embodiment of the present application, the host computer also uses semantic segmentation to extract the weld bead contour (two-dimensional arc contour and three-dimensional arc contour) to fit the center line of the contour and perform geometric transformation to obtain a straight weld bead, so that the length of the weld bead can be calculated. At the same time, the straight weld bead is equally segmented into equal pixels to calculate the width set of the weld bead.

[0313] In the embodiment of the present application, the upper computer in E1 determines the two-dimensional size of the two-dimensional arc profile and the three-dimensional size of the three-dimensional arc profile, which may include the following E11 and E12:

[0314] E11. The host computer performs geometric transformation on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a first contour to be measured and a second contour to be measured that form a straight line.

[0315] In an embodiment of the present application, the host computer performs geometric transformation on the two-dimensional arc contour and the three-dimensional arc contour respectively, that is, semantic segmentation can be used to extract the weld contour, fit the center line of the contour area, and perform geometric transformation to obtain a straight weld (the first contour to be measured and the second contour to be measured).

[0316] E12. The host computer determines the two-dimensional size corresponding to the first contour to be measured and the three-dimensional size corresponding to the second contour to be measured.

[0317] In an embodiment of the present application, the host computer directly measures the length and width of the first contour to be measured that forms a straight line, and simultaneously measures the length and width of the second contour to be measured that forms a straight line to obtain two-dimensional and three-dimensional dimensions.

[0318] It should be noted that the two-dimensional dimensions corresponding to the first contour to be measured include: the length of the first contour to be measured, the width of the first contour to be measured in each segment (here, it can refer to multiple sub-segments obtained by dividing the first contour to be measured); correspondingly, the three-dimensional dimensions corresponding to the second contour to be measured include: the length of the second contour to be measured, the width of the second contour to be measured in each segment (here, it can refer to multiple sub-segments obtained by dividing the second contour to be measured) and the height.

[0319] It can be understood that through the above steps, the host computer measures and determines the two-dimensional dimensions corresponding to the first straight line contour to be measured, and the three-dimensional dimensions corresponding to the second straight line contour to be measured. This allows for relatively accurate determination of the dimensions corresponding to the two-dimensional and three-dimensional arc contours, providing a more accurate parameter reference for subsequent determination of whether the weld bead dimensions of the battery cell are abnormal.

[0320] Continuing with the above description, based on the two-dimensional image being a two-dimensional image and the three-dimensional image being a three-dimensional image, the second length corresponding to the first contour to be measured, and the maximum width and minimum width corresponding to the first contour to be measured can be extracted in the two-dimensional dimension; and the maximum height and minimum height corresponding to the second contour to be measured can be extracted in the three-dimensional dimension, and the subsequent determination of the dimensions of the contours to be measured (the first contour to be measured and the second contour to be measured) can be performed accordingly; that is, the two-dimensional dimension includes: the second length, the maximum width, and the minimum width, and the three-dimensional dimension includes: the maximum height and the minimum height. The determination of the two-dimensional dimension and the three-dimensional dimension in E12 can include the following C1 to C4:

[0321] E121. The host computer determines a second length corresponding to the first contour to be measured.

[0322] E122. The host computer performs equal pixel segmentation on the first contour to be measured, and determines the width of each pixel obtained by the segmentation to obtain a width set.

[0323] E123. The host computer performs equal pixel segmentation on the second contour to be measured, and determines the height of each pixel obtained by the segmentation to obtain a thickness set.

[0324] E124. The host computer determines the maximum width and minimum width of the width set, and the maximum height and minimum height of the thickness set.

[0325] In an embodiment of the present application, the host computer can use a related measurement algorithm to measure the second length corresponding to the first contour to be measured, and at the same time perform equal pixel segmentation and width measurement on the first contour to be measured, so as to obtain the maximum width and minimum width corresponding to the first contour to be measured; correspondingly, the host computer can also perform equal pixel segmentation and height measurement on the second contour to be measured, so as to obtain the maximum height and minimum height corresponding to the second contour to be measured.

[0326] Here, see Figure 9 , 901 can be used to characterize a two-dimensional arc profile or a three-dimensional arc profile (a two-dimensional projection of a three-dimensional arc profile); 902 is the first contour to be measured or the second contour to be measured that forms a straight line; here, the two-dimensional arc profile or the three-dimensional arc profile is unfolded to obtain the corresponding contour skeleton, which can be used to calculate or measure the length of the contour. At the same time, 903 is a schematic diagram of performing equal pixel segmentation on the first contour to be measured or the second contour to be measured; here, the width value or height value corresponding to each pixel after the equal pixel segmentation can be calculated and measured; the height value is for the second contour to be measured.

[0327] It should be noted that the host computer can also perform statistical analysis on the widths in the width set corresponding to the first contour to be measured to determine whether the width of the first contour to be measured is uniform. Similarly, the host computer can also perform statistical analysis on the heights in the height set corresponding to the second contour to be measured to determine whether the height of the second contour to be measured is uniform.

[0328] It can be understood that, through the above steps, by providing the specific dimensional information included in the two-dimensional and three-dimensional dimensions, a relatively accurate parameter basis can be provided for subsequent size-based identification of whether the first and second contours to be measured have dimensional anomalies, thereby improving the accuracy of subsequent identification of whether the weld bead dimensions of the battery cell have abnormalities.

[0329] E2. The host computer compares the two-dimensional size with the first preset weld bead size range to obtain a two-dimensional detection result, and compares the three-dimensional size with the second preset weld bead size range to obtain a three-dimensional detection result.

[0330] In the embodiment of the present application, the first preset weld bead size range includes but is not limited to: a first preset weld bead length range, a first preset weld bead width range; correspondingly, the second preset weld bead size range includes: a second preset weld bead height range, etc.

[0331] Continuing with the above description, after simultaneously performing abnormal detection on the weld bead dimensions within the two-dimensional weld bead area and the weld bead dimensions within the three-dimensional weld bead area and determining the two-dimensional detection results and the three-dimensional detection results, the host computer in S103 obtains the welding detection results of the battery cell based on the two-dimensional detection results and the three-dimensional detection results, which can include E3:

[0332] E3. When the two-dimensional detection result indicates that the two-dimensional dimension is outside the first preset weld bead size range, and / or the three-dimensional dimension result indicates that the three-dimensional dimension is outside the second preset weld bead size range, the host computer determines that the welding detection result of the battery cell is that the weld bead size of the battery cell is abnormal.

[0333] In the embodiment of the present application, the host computer determines that the welding detection result of the battery cell is abnormal, that is, there is an abnormality in the weld size within the welding area of the battery cell (the area outside the weld inside it).

[0334] In the embodiment of the present application, the host computer determines that the welding detection result of the battery cell is that the weld bead size of the battery cell is abnormal, which may include the following three situations:

[0335] Case (1): the two-dimensional detection result indicates that the two-dimensional size is outside the first preset weld bead size range, and the three-dimensional detection result indicates that the three-dimensional size is outside the second preset weld bead size range.

[0336] Case (2): The two-dimensional detection result indicates that the two-dimensional size is outside the first preset weld size range.

[0337] Case (3): The three-dimensional detection result indicates that the three-dimensional size is outside the second preset weld size range.

[0338] Here, on the basis that the two-dimensional dimension includes: the second length, the maximum width, and the minimum width, and the three-dimensional dimension includes: the maximum height and the minimum height, if the first preset weld bead size range includes: the weld bead length range and the weld bead width range, and the second preset weld bead size includes: the weld bead height range; then the implementation of determining in E3 that the two-dimensional dimension is outside the first preset weld bead size range and the three-dimensional dimension is outside the second preset weld bead size range may include the following E31 and E32:

[0339] E31. When the second length is outside the weld length range, and / or the maximum width is outside the weld width range, and / or the minimum width is outside the weld width range, the host computer determines that the two-dimensional dimension is outside the first preset weld size range.

[0340] E32. When the maximum height is outside the weld bead height range, and / or the minimum height is outside the weld bead height range, the host computer determines that the three-dimensional dimension is outside the second preset weld bead dimension range.

[0341] In an embodiment of the present application, the host computer can identify whether the two-dimensional dimension is outside the first preset weld bead size range through the following three situations, namely: the second length is outside the weld bead length range, the maximum width is outside the weld bead width range, and the minimum width is outside the weld bead width range; correspondingly, the host computer can identify whether the three-dimensional dimension is outside the second preset weld bead size range through the following two situations, namely: the maximum height is outside the weld bead height range, and / or the minimum height is outside the weld bead height range.

[0342] It can be understood that the above steps provide comparison results between different parameters to determine whether the two-dimensional dimension is outside the first preset weld bead dimension range and whether the three-dimensional dimension is outside the second preset weld bead dimension range. This can make the dimension comparison logic in the weld bead dimension anomaly determination process clearer and more accurate.

[0343] It should be noted that abnormal weld size of battery cells also has an important impact on the welding quality of battery cells, such as poor welding stability, poor welding surface presentation, and reduced mechanical properties of the weld.

[0344] It can be understood that through the above steps, the host computer determines whether the battery cell's weld bead size is abnormal based on the comparison results between the two-dimensional dimensions of the two-dimensional arc profile and the first preset weld bead size range, and the comparison results between the three-dimensional dimensions of the three-dimensional arc profile and the second preset weld bead size range. In this way, by simultaneously identifying weld bead size anomalies for both the two-dimensional and three-dimensional arc profiles, the accuracy of identifying weld bead size anomalies can be improved.

[0345] In some embodiments of the present application, the two-dimensional welding area includes a two-dimensional weld bead area, and the three-dimensional welding area includes a three-dimensional weld bead area; the host computer in S104 performs abnormality detection on the two-dimensional welding area and the three-dimensional welding area respectively to obtain the two-dimensional detection result and the three-dimensional detection result, and may further include S405 and S406:

[0346] S405 , the host computer performs pixel value recognition on the pixel points in the two-dimensional weld bead area and the three-dimensional weld bead area respectively, and obtains a first recognition result and a second recognition result.

[0347] In an embodiment of the present application, the host computer performs pixel value recognition on all pixel points in the two-dimensional weld bead area to obtain a first pixel value set corresponding to multiple pixel values, that is, a first recognition result; correspondingly, the host computer performs pixel value recognition on all pixel points in the three-dimensional weld bead area to obtain a second pixel value set corresponding to multiple pixel values, that is, a second recognition result.

[0348] It should be noted that the values of the pixels corresponding to the weld bead are usually different from the values of the pixels corresponding to the non-weld bead.

[0349] S406. When the first recognition result indicates that the third area in the two-dimensional weld area has a preset brightness area, and the second recognition result indicates that the fourth area has a preset brightness area, the host computer determines the two-dimensional detection result and the three-dimensional detection result based on the preset brightness area, the two-dimensional image, and the three-dimensional image.

[0350] Among them, the pixel value of the pixel point located in the preset brightness area is less than the preset pixel value; the fourth area is the area obtained by mapping the third area to the three-dimensional image based on the mapping relationship; the mapping relationship is the corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for alignment.

[0351] In the embodiment of the present application, when the host computer detects that the weld bead area (two-dimensional weld bead area and three-dimensional weld bead area) has a preset brightness area, it can further identify the preset brightness area to determine the corresponding two-dimensional detection results and three-dimensional detection results. Here, by further identifying the preset brightness area, that is, when the preset brightness area meets certain conditions, it can be determined that the preset brightness area is an abnormal area within the weld bead area.

[0352] In the embodiments of the present application, the mapping relationship between the two-dimensional image and the three-dimensional image is to import the two images into the same coordinate system for registration to obtain a corresponding mapping relationship. The registration of the two-dimensional image and the three-dimensional image into the same coordinate system can be based on a reference object, such as a certain outline or position of a battery cell that exists in both the two-dimensional image and the three-dimensional image. Here, the same coordinate system can be a two-dimensional plane coordinate system or a three-dimensional coordinate system.

[0353] It should be noted that the mapping relationship between the two-dimensional image and the three-dimensional image is the mapping relationship between the two-dimensional weld bead area and the three-dimensional weld bead area.

[0354] In the embodiment of the present application, there is an abnormal preset brightness area in the weld area, and the preset brightness area may be a pinhole, a burst point, etc. that may exist in the weld.

[0355] In the embodiment of the present application, the pixel value of each pixel within the preset brightness area may be less than the preset pixel value, or the average pixel value within the preset brightness area may be less than the preset pixel value; wherein the average pixel value is the average value obtained by summing the pixel values of all pixels within the preset brightness area. Here, the preset pixel value can be determined according to the actual production requirements of the battery cell.

[0356] It can be understood that through the above steps, the host computer first identifies the pixel values of the two-dimensional weld bead area and the three-dimensional weld bead area to obtain a first recognition result and a second recognition result. Then, if the two-dimensional weld bead area and the three-dimensional weld bead area both have preset brightness areas, the two-dimensional detection result and the three-dimensional detection result can be further determined based on the preset brightness areas, the two-dimensional image, and the three-dimensional image. In this way, by simultaneously identifying anomalies and determining the results of the two-dimensional weld bead area and the three-dimensional weld bead area, the accuracy of the two-dimensional and three-dimensional detection results for welding anomaly detection can be improved.

[0357] In the embodiment of the present application, the host computer in S406 determines the corresponding two-dimensional detection result and three-dimensional detection result based on the preset brightness area, the two-dimensional image, and the three-dimensional image, and may include the following F1:

[0358] F1. The host computer compares the size of the preset brightness area in the two-dimensional image with the first preset interval size to obtain a two-dimensional detection result, and compares the size of the preset brightness area in the three-dimensional image with the second preset interval size to obtain a three-dimensional detection result.

[0359] In an embodiment of the present application, the host computer compares the size of the preset brightness area in the two-dimensional image or the two-dimensional weld area with the first preset interval size to obtain a two-dimensional detection result; at the same time, the host computer compares the size of the preset brightness area in the three-dimensional image or the three-dimensional weld area with the second preset interval size to obtain a three-dimensional detection result.

[0360] Here, the two-dimensional detection result can be described by using the size of the preset brightness area in the two-dimensional image to be greater than, equal to, or less than the first preset interval size. Correspondingly, the three-dimensional detection result can be described by using the size of the preset brightness area in the three-dimensional image to be greater than, equal to, or less than the second preset interval size.

[0361] In the embodiment of the present application, since the dimensions of the two-dimensional image and the three-dimensional image are different, the dimensions corresponding to the sizes included in the first preset interval size and the second preset interval size are different.

[0362] Continuing from the above description, pixel value recognition is performed on both the two-dimensional weld bead area and the three-dimensional weld bead area. For example, when there are cracks or pinholes in the weld, after determining the two-dimensional detection results and the three-dimensional detection results, the upper computer in S104 obtains the welding detection results of the battery cell based on the two-dimensional detection results and the three-dimensional detection results. This may include the following F2:

[0363] F2. When the two-dimensional detection result indicates that the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size, and / or the three-dimensional detection result indicates that the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size, the host computer determines that the welding detection result of the battery cell is that there is a welding abnormality in the weld of the battery cell.

[0364] In an embodiment of the present application, the host computer determines that the welding detection result of the battery cell is a detection abnormality, that is, there is a welding abnormality in the weld of the welding area of the battery cell (the area outside the weld inside it), such as: there are explosion spots, pinholes, etc. in the welding.

[0365] In the embodiment of the present application, the host computer determines that the welding detection result of the battery cell is that there is a welding abnormality in the weld of the battery cell, which may include the following three situations:

[0366] Case (1): the two-dimensional detection result indicates that the size of the preset brightness area in the two-dimensional image is larger than the first preset interval size, and the three-dimensional detection result indicates that the size of the preset brightness area in the three-dimensional image is larger than the second preset interval size.

[0367] Case (2): The two-dimensional detection result indicates that the size of the preset brightness area in the two-dimensional image is larger than the first preset interval size.

[0368] Case (3): The three-dimensional detection result indicates that the size of the preset brightness area in the three-dimensional image is larger than the second preset interval size.

[0369] It is understood that through the above steps, the host computer determines that the welding detection result of the battery cell is a welding abnormality when the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size, and / or the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size. In this way, based on the combination of the welding results identified in the two-dimensional image and the three-dimensional image of different dimensions, the weld bead of the battery cell is further judged based on the preset interval size, which can improve the accuracy of the welding result identification of the battery cell.

[0370] Continuing with the above description, on the basis that the two-dimensional image is a two-dimensional image and the three-dimensional image is a three-dimensional image, the sizes of the preset brightness area in the length and width directions can be extracted from the size of the preset brightness area in the two-dimensional image, and the size of the preset brightness area in the height direction can be extracted from the size of the preset brightness area in the three-dimensional image, and subsequent size determination of the preset brightness area is performed accordingly; that is, when the size of the preset brightness area in the two-dimensional image includes: a third length and a third width, the size of the preset brightness area in the three-dimensional image includes: a third height, the first preset interval size includes: a preset brightness area length and a preset brightness area width, and the second preset interval size includes: a preset brightness area height, the implementation of determining that the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size and the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size in the above F2 may include F21 and F22:

[0371] F21. When the third length is greater than the length of the preset brightness area and / or the third width is greater than the width of the preset brightness area, the host computer determines that the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size.

[0372] F22: When the third height is greater than the height of the preset brightness area, the host computer determines that the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size.

[0373] In the embodiment of the present application, the implementation logic here can refer to the above description of the upper computer determining that the first size is larger than the first preset welding slag size, and the second size is larger than the second preset welding slag size, etc., and will not be repeated here.

[0374] It can be understood that through the above steps, through specific size comparison, the comparison results between the relevant sizes (the size of the preset brightness area in the two-dimensional image, the size of the preset brightness area in the three-dimensional image) and the preset interval sizes (the first preset interval size, the second preset interval size) can be made more accurate.

[0375] In the embodiments of the present application, the above-mentioned detection of whether the weld bead size in the battery core weld bead area is abnormal, whether the weld bead is offset, whether the weld bead is abnormal (such as the presence of explosion points, pinholes, protrusions, etc.), and the detection of welding slag in the area outside the battery core weld bead all involve two-dimensional images and three-dimensional images of different dimensions. Figure 10 , wherein the steps corresponding to the detection method include the following:

[0376] S1001, turntable loading, that is, the battery cell arrives at the loading station, first inspection station and second inspection station corresponding to the weld inspection process. Here, the battery cell may arrive at the 2D inspection station and the 3D inspection station.

[0377] S1002, signal triggering, that is, after detecting that the battery cell arrives at the first inspection station and / or the second inspection station, the controller sends a corresponding acquisition signal to the host computer, so that the host computer controls the relevant camera structure to acquire an image of the battery cell.

[0378] S1003, Image Capture, involves the host computer capturing images of the welding area of the battery cell in response to acquisition signals sent by the controller. This includes using a 3D camera structure to capture images of the battery cell in the 3D inspection station, and using a 2D camera structure to capture images of the battery cell in the 2D inspection station. The order of the 2D and 3D inspection stations can be determined based on actual production requirements and is not limited here.

[0379] S1004, detection results, that is, the host computer detects and analyzes the collected images (two-dimensional images and three-dimensional images) to obtain corresponding detection results, such as: detecting that the weld of the battery cell is offset, the weld size of the battery cell is abnormal, there is welding slag in the area outside the weld of the battery cell, and there are explosion points, protrusions or pinholes in the weld of the battery cell.

[0380] S1005: Result feedback, that is, the host computer feeds back the determined detection result to the controller, or the industrial computer where the controller is located.

[0381] S1006, turntable unloading, that is, after receiving the image acquisition end signal sent by the host computer, the controller continues to control the operation of the carrying battery cell until it reaches the unloading station to realize the unloading of the battery cell.

[0382] In the embodiment of the present application, the host computer feeds back relevant feedback of the welding detection result as abnormal detection to the controller, so that the controller performs corresponding operations, that is, the weld detection method provided in the embodiment of the present application can also be as follows G1 and G2:

[0383] G1. When the welding detection result is characterized as abnormal detection, the host computer sends a marking signal to the controller.

[0384] The abnormality detected includes at least one of the following: welding slag exists in the welding area of the battery cell, the weld bead of the battery cell is offset, the weld bead size of the battery cell is abnormal, and the weld bead of the battery cell has a welding abnormality.

[0385] In the embodiment of the present application, the host computer can send a marking signal carrying abnormality information to the controller when the welding detection result is abnormal. Here, the host computer can send the marking signal to the controller via the industrial computer where it is located.

[0386] In the embodiments of the present application, the detection abnormalities include but are not limited to the multiple detection abnormalities mentioned above, such as: the presence of welding slag in the welding area of the battery cell, the offset of the weld bead of the battery cell, the abnormal size of the weld bead of the battery cell, and the welding abnormality of the weld bead of the battery cell.

[0387] G2. The controller controls the marking mechanism to mark the battery cell based on the marking signal.

[0388] In an embodiment of the present application, the controller can control a marking mechanism based on the received marking signal, such as a marking structure carrying a marking pen, to mark the battery cell as abnormal, or control a marking mechanism to attach an abnormal label to the battery cell.

[0389] It should be noted that the marking mechanism may be a marking mechanism deployed on an industrial computer, wherein the marking signal is used to mark the battery cell.

[0390] It is understood that through the above steps, if the welding test result is abnormal, the host computer can send a marking signal to the controller, so that the controller can promptly mark the battery cell. In this way, the relevant marking operation can be executed in a timely manner when an abnormality is detected, so as to identify defective products and issue an alarm, thereby quickly preventing the outflow of defective products and the resulting large-scale scrapping.

[0391] Continuing from the above description, the host computer provided in the embodiment of the present application can also execute the following H1 and H2:

[0392] H1. When the number of abnormal detections received reaches the preset number, the host computer sends a shutdown signal to the controller.

[0393] In the embodiment of the present application, the preset number of times may be 10 times, or 5 times. The specific value corresponding to the preset number of times is determined by the production process of the battery cell in actual application.

[0394] In an embodiment of the present application, the upper computer detects that the welding detection result of the battery cell is a detection abnormality. If the next battery cell arrives at the detection station (the first detection station and the second detection station) subsequently, the upper computer will also capture an image of the next battery cell under the action of the relevant acquisition signal sent by the controller, and perform abnormal detection of welding information on the captured image to obtain the welding detection result of the next battery cell as a detection abnormality. The number of detection abnormalities can be added by 1, and so on, to count the number of detection abnormalities.

[0395] It should be noted that, here, the host computer may send a shutdown signal to the controller through the industrial computer where it is located only when the number of abnormal detections received by the host computer reaches a preset number within a preset time period.

[0396] H2. The controller stops working and / or detects the cause of the abnormality based on the shutdown signal.

[0397] In an embodiment of the present application, the controller may stop the operation of the welding detection system in which it is located, or stop the operation of the host computer based on the shutdown signal; and / or detect and analyze the cause of the abnormality.

[0398] It is understood that through the above steps, when the host computer receives a preset number of abnormal detections, it sends a shutdown signal to the controller, causing the controller to stop operating and / or detect the cause of the abnormality. This allows the weld detection system to automatically perform logical protection if the relevant abnormal detection occurs multiple times.

[0399] To implement the weld detection method of the embodiment of the present application, see Figure 11 The embodiment of the present application further provides a weld detection system. The weld detection system 1100 includes: a controller 1101, a host computer 1102, and a camera 1103, wherein:

[0400] When the cylindrical battery cell arrives at the first inspection station, the camera 1103 is configured to capture an image of the welding area of the cylindrical battery cell based on the first acquisition signal received from the controller 1101 to obtain a two-dimensional image;

[0401] When the cylindrical battery cell reaches the second inspection station, the camera 1103 is further configured to capture an image of the welding area based on the second acquisition signal sent by the controller 1101 to obtain a three-dimensional image;

[0402] The host computer 1102 is further configured to determine, based on the two-dimensional image and the three-dimensional image, a circular arc-shaped two-dimensional welding area in the two-dimensional image and a circular arc-shaped three-dimensional welding area in the three-dimensional image;

[0403] The host computer 1102 is further configured to perform abnormality detection on the two-dimensional welding area and the three-dimensional welding area respectively to obtain two-dimensional detection results and three-dimensional detection results; and obtain the welding detection results of the battery cell based on the two-dimensional detection results and the three-dimensional detection results.

[0404] It should be noted that the camera 1103 includes a two-dimensional camera and a three-dimensional camera; the two-dimensional camera is installed at the first inspection station and is configured to rotate through a turret to collect a two-dimensional image of the welding area of the battery cell when the battery cell reaches the first inspection station; the three-dimensional camera is installed at the second inspection station, and the second inspection station is provided with a rotating motor;

[0405] The upper computer 1102 is configured to, after the two-dimensional camera has completed the acquisition, rotate the turret, and when the battery cell reaches the second inspection station, control the rotation of the rotating motor to enable the three-dimensional camera to acquire a three-dimensional image of the welding area of the battery cell; and at the same time as arriving at the second inspection station, determine the two-dimensional positive electrode welding area in the two-dimensional image, perform abnormality detection on the weld in the two-dimensional positive electrode welding area, and obtain a two-dimensional detection result; and determine the three-dimensional positive electrode welding area in the three-dimensional image; perform abnormality detection on the weld in the three-dimensional positive electrode welding area to obtain a three-dimensional detection result; and obtain the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result.

[0406] In an embodiment of the present application, the weld detection system 1100 also includes: a turntable 1103, on which a plurality of cups for holding batteries are carried; the camera includes a two-dimensional camera and a three-dimensional camera; when the cylindrical battery cell is placed on the first cup at the loading station, the controller 1101 is also configured to control the rotation of the turntable 1103 to drive the first cup to operate; when the first cup operates to the first detection station, a first acquisition signal is sent to the two-dimensional camera set at the first detection station.

[0407] In an embodiment of the present application, a two-dimensional camera is provided at the first inspection station; when the first support cup is transferred to the first inspection station, the controller 1101 is also configured to send a first acquisition signal to the two-dimensional camera; the two-dimensional camera is also configured to perform image acquisition on the welding area based on the first acquisition signal to obtain a two-dimensional image.

[0408] In an embodiment of the present application, a three-dimensional camera, a rotating motor and an encoder are provided on the second inspection station; when the cylindrical battery cell arrives at the second inspection station, the controller 1101 is also configured to send a second acquisition signal to the rotating motor; the rotating motor is configured to drive the first support cup, the cylindrical battery cell in the first support cup and the encoder to rotate in response to the second acquisition signal; the encoder is configured to generate and send a pulse signal to the three-dimensional camera during the rotation process; the three-dimensional camera is configured to capture an image of the welding area based on the pulse signal to obtain a three-dimensional image.

[0409] In an embodiment of the present application, the host computer 1102 is further configured to send an end signal to the controller 1101 after the 3D camera completes image acquisition of the welding area; the controller 1101 is further configured to control the rotation of the turntable 1103 based on the end signal to transfer the first support cup to the unloading station to achieve battery cell unloading. The battery cell enters the loading station through the rotation of the turret, moves through different stations, and is unloaded from the unloading station; and at least a first inspection station and a second inspection station are included between the loading station and the unloading station.

[0410] It should be noted that the turntable can be understood as a turret for transferring battery cells, and the battery cells can be placed through a supporting cup, etc.

[0411] Wherein, when the battery cell is at the loading station, the controller 1101 is configured to send a first acquisition signal to the host computer when the battery cell arrives at the first detection station;

[0412] The host computer 1102 is further configured to control the two-dimensional camera to capture an image of the positive electrode welding area of the battery cell based on the first acquisition signal to obtain a two-dimensional image;

[0413] The controller 1101 is further configured to send a second acquisition signal to the host computer when the battery cell arrives at the second detection station;

[0414] The host computer 1102 is further configured to acquire an image of the positive electrode welding area based on the second acquisition signal to obtain a three-dimensional image.

[0415] In addition, the host computer 1102 is also configured to control the two-dimensional camera to capture images of the positive electrode welding area of the battery cell based on the first acquisition signal to obtain a two-dimensional image; and, based on the second acquisition signal, to control the rotation of the rotating motor so that the three-dimensional camera set on the second detection station captures a three-dimensional image; after the three-dimensional camera finishes capturing the three-dimensional image, the host computer sends an end signal to the controller; the controller 1101 is also configured to control the rotation of the battery cell based on the end signal to move the battery cell to the unloading station to realize battery cell unloading.

[0416] In an embodiment of the present application, the host computer 1102 is further configured to control the rotating motor to drive the battery cell and the encoder to rotate based on the second acquisition signal; the encoder is configured to generate and send a pulse signal to the three-dimensional camera during the rotation process; the three-dimensional camera is configured to capture an image of the positive electrode welding area based on the pulse signal to obtain a three-dimensional image.

[0417] In an embodiment of the present application, the host computer 1102 is further configured to determine a two-dimensional welding area in a two-dimensional image and a three-dimensional welding area in a three-dimensional image; perform abnormality detection on the two-dimensional welding area and the three-dimensional welding area respectively to obtain a two-dimensional detection result and a three-dimensional detection result.

[0418] In an embodiment of the present application, the host computer 1102 is further configured to perform semantic segmentation on the two-dimensional image and the three-dimensional image respectively to obtain a two-dimensional welding area and a three-dimensional welding area; or, the host computer 1102 is further configured to perform semantic segmentation on the two-dimensional image to obtain a two-dimensional welding area, and then map the two-dimensional welding area to the three-dimensional image to obtain a three-dimensional welding area; or, the host computer 1102 is further configured to perform semantic segmentation on the three-dimensional image to obtain a three-dimensional welding area, and then map the three-dimensional welding area to the two-dimensional image to obtain a two-dimensional welding area.

[0419] In an embodiment of the present application, the two-dimensional welding area includes a two-dimensional area outside the weld bead, and the three-dimensional welding area includes a three-dimensional area outside the weld bead; the upper computer 1102 is also configured to use a trained model to perform welding slag detection on the two-dimensional area outside the weld bead and the three-dimensional area outside the weld bead, respectively, to obtain a two-dimensional welding slag result and a three-dimensional welding slag result; wherein the trained model is obtained by training based on a two-dimensional sample image and a three-dimensional sample image with labeled welding slag information; the dimensions of the two-dimensional sample image and the three-dimensional sample image are different; when the two-dimensional welding slag result indicates that there is welding slag to be identified in the first area of the two-dimensional weld bead area, and the three-dimensional welding slag result indicates that there is welding slag to be identified in the second area, the two-dimensional detection result and the three-dimensional detection result are determined based on the welding slag to be identified, the two-dimensional image and the three-dimensional image; wherein the second area is the area obtained by mapping the first area to the three-dimensional image based on a mapping relationship; the mapping relationship is the corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for registration.

[0420] In an embodiment of the present application, the host computer 1102 is further configured to compare the first size of the welding slag to be identified in the two-dimensional image with the first preset welding slag size to obtain a two-dimensional detection result, and to compare the second size of the welding slag to be identified in the three-dimensional image with the second preset welding slag size to obtain a three-dimensional detection result; when the two-dimensional detection result indicates that the first size is larger than the first preset welding slag size, and / or the three-dimensional detection result indicates that the second size is larger than the second preset welding slag size, it is determined that the welding detection result of the battery cell is that there is welding slag in the welding area of the battery cell.

[0421] In an embodiment of the present application, the first size includes: a first length and a first width, the first preset slag size includes: a preset slag length and a preset slag width, the second size includes: a first height, and the second preset slag size includes: a preset slag height; when the first length is greater than the preset slag length, and / or the first width is greater than the preset slag width, the upper computer 1102 is also configured to determine that the first size is greater than the first preset slag size; when the first height is greater than the preset slag height, the upper computer 1102 is also configured to determine that the second size is greater than the second preset slag size.

[0422] In an embodiment of the present application, the two-dimensional welding area includes a two-dimensional weld bead area, and the three-dimensional welding area includes a three-dimensional weld bead area; the host computer 1102 is also configured to extract the weld bead contours of the two-dimensional weld bead area and the three-dimensional weld bead area respectively to obtain a two-dimensional arc contour and a three-dimensional arc contour; perform abnormality detection on the two-dimensional arc contour and the three-dimensional arc contour respectively to obtain a two-dimensional detection result and a three-dimensional detection result.

[0423] The welding area is the positive electrode welding area of the battery cell; the upper computer 1102 is also configured to determine a first distance between the center of the two-dimensional arc contour and the center of the liquid injection hole of the battery cell, and determine a second distance between the center of the three-dimensional arc contour and the center of the liquid injection hole; compare the first distance with the preset offset distance to obtain a two-dimensional detection result, and compare the second distance with the preset offset distance to obtain a three-dimensional detection result; when the two-dimensional detection result indicates that the first distance is greater than the preset offset distance, and the three-dimensional detection result indicates that the second distance is greater than the preset offset distance, it is determined that the welding detection result of the battery cell is that there is an offset in the weld of the battery cell.

[0424] In an embodiment of the present application, the host computer 1102 is further configured to perform centerline fitting processing on the two-dimensional arc contour and the three-dimensional arc contour, respectively, to obtain the two-dimensional center line of the two-dimensional arc contour and the three-dimensional center line of the three-dimensional arc contour; the host computer 1102 is further configured to determine the first center coordinates corresponding to the two-dimensional center line in the two-dimensional image and the second center coordinates corresponding to the three-dimensional center line in the three-dimensional image; determine the distance between the first center coordinate and the first comparison coordinate as the first distance, and determine the distance between the second center coordinate and the second comparison coordinate as the second distance; wherein the first comparison coordinate is the coordinate of the center of the injection hole in the two-dimensional image, and the second comparison coordinate is the coordinate of the center of the injection hole in the three-dimensional image.

[0425] In an embodiment of the present application, the host computer 1102 is further configured to determine the two-dimensional size of the two-dimensional arc profile and the three-dimensional size of the three-dimensional arc profile; compare the two-dimensional size with the first preset weld size range to obtain a two-dimensional detection result, and compare the three-dimensional size with the second preset weld size range to obtain a three-dimensional detection result; when the two-dimensional detection result indicates that the two-dimensional size is outside the first preset weld size range, and / or the three-dimensional detection result indicates that the three-dimensional size is outside the second preset weld size range, it is determined that the welding detection result of the battery cell is that the weld size of the battery cell is abnormal.

[0426] In an embodiment of the present application, the host computer 1102 is also configured to perform geometric transformations on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a first contour to be measured and a second contour to be measured that form a straight line; determine the two-dimensional size corresponding to the first contour to be measured and the three-dimensional size corresponding to the second contour to be measured.

[0427] In an embodiment of the present application, the two-dimensional dimensions include: a second length, a maximum width, and a minimum width, and the three-dimensional dimensions include: a maximum height and a minimum height; the host computer 1102 is further configured to determine the second length corresponding to the first contour to be measured; perform equal pixel segmentation on the first contour to be measured, and determine the width of each pixel obtained by segmentation to obtain a width set; perform equal pixel segmentation on the second contour to be measured, and determine the height of each pixel obtained by segmentation to obtain a thickness set; determine the maximum width and minimum width in the width set, respectively, and determine the maximum height and minimum height in the thickness set.

[0428] In an embodiment of the present application, the first preset weld bead size range includes: a weld bead length range and a weld bead width range, and the second preset weld bead size includes: a weld bead height range; when the second length is outside the weld bead length range, and / or the maximum width is outside the weld bead width range, and / or the minimum width is outside the weld bead width range, the host computer 1102 is also configured to determine that the two-dimensional dimension is outside the first preset weld bead size range; when the maximum height is outside the weld bead height range, and / or the minimum height is outside the weld bead height range, the host computer 1102 is also configured to determine that the three-dimensional dimension is outside the second preset weld bead size range.

[0429] In an embodiment of the present application, the two-dimensional welding area includes a two-dimensional weld area, and the three-dimensional welding area includes a three-dimensional weld area; the host computer 1102 is further configured to perform pixel value recognition on pixel points in the two-dimensional weld area and the three-dimensional weld area, respectively, to obtain a first recognition result and a second recognition result; when the first recognition result indicates that there is a preset brightness area in the third area of the two-dimensional weld area, and the second recognition result indicates that there is a preset brightness area in the fourth area, the two-dimensional detection result and the three-dimensional detection result are determined based on the preset brightness area, the two-dimensional image and the three-dimensional image; wherein the pixel value of the pixel point located in the preset brightness area is less than the preset pixel value; the fourth area is the area obtained by mapping the third area to the three-dimensional image based on the mapping relationship; the mapping relationship is the corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for registration.

[0430] In an embodiment of the present application, the host computer 1102 is further configured to compare the size of the preset brightness area in the two-dimensional image with the first preset interval size to obtain a two-dimensional detection result, and to compare the size of the preset brightness area in the three-dimensional image with the second preset interval size to obtain a three-dimensional detection result; when the two-dimensional detection result indicates that the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size, and / or the three-dimensional detection result indicates that the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size, it is determined that the welding detection result of the battery cell is that there is a welding abnormality in the weld of the battery cell.

[0431] In an embodiment of the present application, the size of the preset brightness area in the two-dimensional image includes: a third length and a third width, the size of the preset brightness area in the three-dimensional image includes: a third height, the first preset interval size includes: a preset brightness area length and a preset brightness area width, and the second preset interval size includes: a preset brightness area height; when the third length is greater than the preset brightness area length, and / or the third width is greater than the preset brightness area width, the host computer 1102 is further configured to determine that the size of the preset brightness area in the two-dimensional image is greater than the first preset interval size; when the third height is greater than the preset brightness area height, the host computer 1102 is further configured to determine that the size of the preset brightness area in the three-dimensional image is greater than the second preset interval size.

[0432] In an embodiment of the present application, when the welding detection result is characterized as a detection abnormality, the host computer 1102 is further configured to send a marking signal to the controller 1101; wherein the detection abnormality includes at least one of the following: the presence of welding slag in the welding area of the battery cell, the presence of an offset in the weld bead of the battery cell, the abnormal size of the weld bead of the battery cell, and the presence of a welding abnormality in the weld bead of the battery cell; the controller 1101 is further configured to control the marking mechanism to mark the battery cell based on the marking signal.

[0433] In an embodiment of the present application, when the number of received abnormality detections reaches a preset number, the host computer 1102 is also configured to send a shutdown signal to the controller 1101; the controller 1101 is also configured to stop working and / or detect the cause of the abnormality based on the shutdown signal.

[0434] It should be noted that the description of the corresponding embodiment of the weld detection system is similar to the description of the above-mentioned method embodiment, and has similar beneficial effects as the method embodiment. For any technical details not disclosed in the system embodiment of this application, please refer to the description of the method embodiment of this application for understanding.

[0435] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned steps / processes does not mean the order of execution, and the execution order of each step / process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned serial numbers of the embodiments of the present application are for description only and do not represent the advantages and disadvantages of the embodiments.

[0436] This application uses descriptions of directions or positional relationships such as "up", "down", "top", "bottom", "front", "back", "inside" and "outside" to facilitate the description of this application, and does not indicate or imply that the device referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, it should not be understood as limiting the scope of protection of this application.

[0437] It should also be noted that, in the description of this application, unless otherwise specified or limited, the terms "installed," "connected," and "connected" should be understood broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to direct connections or indirect connections through an intermediary. Those skilled in the art will understand the specific meanings of these terms in this application based on the specific circumstances.

[0438] It should be noted that, in this application, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0439] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0440] The units described above as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, the functional units in the various embodiments of the present application may all be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0441] The above are only implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any changes or replacements that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed in this application should be covered by the protection scope of the present application.

Claims

1. A weld detection system, characterized in that: The weld detection system includes: a host computer, a turret, the turret transmits multiple battery cells, a two-dimensional camera, and a three-dimensional camera; wherein: The two-dimensional camera is installed at the first inspection station and is configured to rotate through the turret to collect a two-dimensional image of the welding area of the battery cell when the battery cell reaches the first inspection station; The three-dimensional camera is installed at a second inspection station, and a rotating motor is provided on the second inspection station; The host computer is configured to, after the two-dimensional camera completes the acquisition, rotate the turret, and when the battery cell reaches the second inspection station, control the rotation of the rotating motor to enable the three-dimensional camera to acquire a three-dimensional image of the welding area of the battery cell; and at the same time as arriving at the second inspection station, determine the two-dimensional positive electrode welding area in the two-dimensional image, perform abnormality detection on the weld of the two-dimensional positive electrode welding area, and obtain a two-dimensional detection result; and determine the three-dimensional positive electrode welding area in the three-dimensional image; perform abnormality detection on the weld of the three-dimensional positive electrode welding area, and obtain a three-dimensional detection result; and obtain the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result.

2. The detection system according to claim 1, characterized in that The weld detection system further includes: a controller; When the battery cell is at the loading station, the controller is configured to send a first acquisition signal to the host computer when the battery cell reaches the first detection station; The host computer is further configured to control the two-dimensional camera to capture an image of the positive electrode welding area of the battery cell based on the first acquisition signal to obtain the two-dimensional image; The controller is further configured to send a second acquisition signal to the host computer when the battery cell arrives at the second detection station; The host computer is further configured to acquire an image of the positive electrode welding area based on the second acquisition signal to obtain the three-dimensional image.

3. The detection system according to claim 2, characterized in that The host computer is further configured to control a two-dimensional camera to acquire an image of the positive electrode welding area of the battery cell based on the first acquisition signal to obtain the two-dimensional image; and Based on the second acquisition signal, the three-dimensional camera provided on the second inspection station acquires the three-dimensional image by controlling the rotation of the rotary motor; After the three-dimensional camera finishes acquiring the three-dimensional image, the host computer sends an end signal to the controller; The controller is further configured to control the rotation of the battery cell based on the end signal, so as to move the battery cell to a unloading station to realize unloading of the battery cell.

4. The detection system according to claim 2 or 3, characterized in that: The second detection station is provided with an encoder; The host computer is further configured to control the rotary motor to drive the battery cell and the encoder to rotate based on the second collected signal; The encoder is configured to generate and send a pulse signal to the three-dimensional camera during the rotation process; The three-dimensional camera is configured to capture an image of the positive electrode welding area based on the pulse signal to obtain the three-dimensional image.

5. The detection system according to any one of claims 1 to 4, characterized in that: Through the rotation of the turret, the battery cell enters from the loading station, moves on different stations, and is unloaded from the unloading station; wherein, at least a first detection station and a second detection station are included between the loading station and the unloading station.

6. The detection system according to any one of claims 1 to 5, characterized in that: The two-dimensional positive electrode welding region includes a two-dimensional weld bead region, and the three-dimensional positive electrode welding region includes a three-dimensional weld bead region; The host computer is further configured to extract weld bead contours from the two-dimensional weld bead area and the three-dimensional weld bead area respectively to obtain a two-dimensional arc contour and a three-dimensional arc contour; The host computer is further configured to determine a first distance between a center of the two-dimensional arc-shaped contour and a center of the liquid injection hole of the battery cell, and to determine a second distance between the three-dimensional arc-shaped contour and the center of the liquid injection hole; The host computer is further configured to compare the first distance with a preset offset distance to obtain the two-dimensional detection result, and to compare the second distance with the preset offset distance to obtain the three-dimensional detection result.

7. The detection method according to claim 6, characterized in that The host computer is further configured to determine that the welding detection result of the battery cell is that there is an offset in the weld of the battery cell when the two-dimensional detection result indicates that the first distance is greater than the preset offset distance, and the three-dimensional detection result indicates that the second distance is greater than the preset offset distance.

8. The detection method according to claim 6, characterized in that The host computer is further configured to perform centerline fitting processing on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a two-dimensional center line of the two-dimensional arc profile and a three-dimensional center line of the three-dimensional arc profile; respectively determining a first center coordinate of the two-dimensional center line in the two-dimensional image and a second center coordinate of the three-dimensional center line in the three-dimensional image; Determine the distance between the first circle center coordinate and the first comparison coordinate as the first distance, and determine the distance between the second circle center coordinate and the second comparison coordinate as the second distance; The first comparison coordinates are coordinates of the center of the injection hole in the two-dimensional image, and the second comparison coordinates are coordinates of the center of the injection hole in the three-dimensional image.

9. The detection method according to claim 6, characterized in that The host computer is further configured to determine the two-dimensional size of the two-dimensional arc profile and the three-dimensional size of the three-dimensional arc profile; Comparing the two-dimensional size with a first preset weld bead size range to obtain the two-dimensional detection result, and comparing the three-dimensional size with a second preset weld bead size range to obtain the three-dimensional detection result; The host computer is further configured to determine that the welding detection result of the battery cell is that the weld size of the battery cell is abnormal when the two-dimensional detection result indicates that the two-dimensional dimension is outside the first preset weld size range, and / or when the three-dimensional detection result indicates that the three-dimensional dimension is outside the second preset weld size range.

10. The detection method according to any one of claims 1 to 9, characterized in that: The host computer is further configured to send a marking signal to the controller when the welding detection result is characterized as a detection abnormality; The abnormality detected includes at least one of the following: welding slag exists in the welding area of the battery cell, the weld bead of the battery cell is offset, the weld bead size of the battery cell is abnormal, and the weld bead of the battery cell has a welding abnormality; The controller is further configured to control a marking mechanism to mark the battery cell based on the marking signal.

11. A method for detecting a weld, characterized in that: The detection method comprises: When the battery cell arrives at the first inspection station, collecting a two-dimensional image of the welding area of the battery cell; When the battery cell arrives at the second inspection station, a three-dimensional image of the welding area of the battery cell is collected, and a two-dimensional positive electrode welding area in the two-dimensional image is determined, and an abnormality detection is performed on the weld bead of the two-dimensional positive electrode welding area to obtain a two-dimensional detection result; Determining a three-dimensional positive electrode welding area in the three-dimensional image; performing abnormality detection on the weld bead of the three-dimensional positive electrode welding area to obtain a three-dimensional detection result; A welding detection result of the battery cell is obtained based on the two-dimensional detection result and the three-dimensional detection result.

12. The detection method according to claim 11, characterized in that The collecting of the two-dimensional image of the welding area of the battery cell includes: When the battery cell arrives at the first inspection station, an image of the welding area of the battery cell is captured based on the received first acquisition signal to obtain the two-dimensional image.

13. The detection method according to claim 11 or 12, characterized in that: The collecting of the three-dimensional image of the welding area of the battery cell includes: When the battery cell reaches the second inspection station, image acquisition is performed on the welding area based on the received second acquisition signal to obtain the three-dimensional image.

14. The detection method according to claim 13, characterized in that The detection method further comprises: After the three-dimensional image acquisition is completed, an end signal is sent, and the end signal is used to control the transportation of the battery cell to the unloading station to realize the unloading of the battery cell.

15. The detection method according to claim 13 or 14, characterized in that: The step of performing image acquisition on the welding area based on the received second acquisition signal to obtain the three-dimensional image includes: In response to the second acquisition signal, controlling the battery cell to rotate and generate a pulse signal; Based on the pulse signal, an image of the welding area is collected to obtain the three-dimensional image.

16. The detection method according to any one of claims 11 to 15, characterized in that: The two-dimensional positive electrode welding area includes a two-dimensional weld bead outer area, and the three-dimensional positive electrode welding area includes a three-dimensional weld bead outer area; The performing abnormality detection on the weld bead of the two-dimensional positive electrode welding area to obtain a two-dimensional detection result, and the performing abnormality detection on the weld bead of the three-dimensional positive electrode welding area to obtain a three-dimensional detection result, include: Using a trained model, performing welding slag detection on the two-dimensional weld bead outer region and the three-dimensional weld bead outer region, respectively, to obtain the two-dimensional welding slag result and the three-dimensional welding slag result; wherein the trained model is trained based on the two-dimensional sample images and the three-dimensional sample images labeled with welding slag information; When the two-dimensional welding slag result indicates that the welding slag to be identified exists in a first area outside the two-dimensional weld bead area, and the three-dimensional welding slag result indicates that the welding slag to be identified exists in a second area outside the three-dimensional weld bead area, the two-dimensional detection result is obtained based on a comparison between a first size of the welding slag to be identified in the two-dimensional image and a first preset welding slag size, and the three-dimensional detection result is obtained based on a comparison between a second size of the welding slag to be identified in the three-dimensional image and a second preset welding slag size; The obtaining of the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result includes: When the two-dimensional detection result indicates that the first size is larger than the first preset slag size, and / or the three-dimensional detection result indicates that the second size is larger than the second preset slag size, determining that the welding detection result of the battery cell is that slag is present in the welding area of the battery cell; The second area is an area obtained by mapping the first area to the three-dimensional image based on a mapping relationship; the mapping relationship is a corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for registration.

17. The detection method according to any one of claims 11 to 15, characterized in that: The two-dimensional positive electrode welding area includes a two-dimensional weld bead area, and the three-dimensional positive electrode welding area includes a three-dimensional weld bead area; performing abnormality detection on the weld bead of the two-dimensional positive electrode welding area to obtain a two-dimensional detection result, and performing abnormality detection on the weld bead of the three-dimensional positive electrode welding area to obtain a three-dimensional detection result, include: The weld bead contours are extracted from the two-dimensional weld bead area and the three-dimensional weld bead area respectively to obtain a two-dimensional arc contour and a three-dimensional arc contour; Determining a first distance between a center of the two-dimensional arc-shaped contour and a center of a liquid injection hole of the battery cell, and determining a second distance between the three-dimensional arc-shaped contour and the center of the liquid injection hole; The first distance is compared with a preset offset distance to obtain the two-dimensional detection result, and the second distance is compared with the preset offset distance to obtain the three-dimensional detection result.

18. The detection method according to claim 17, characterized in that: The obtaining of the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result includes: When the two-dimensional detection result indicates that the first distance is greater than the preset offset distance, and the three-dimensional detection result indicates that the second distance is greater than the preset offset distance, it is determined that the welding detection result of the battery cell is that the weld of the battery cell is offset.

19. The detection method according to claim 17, characterized in that: Determining a first distance between the center of the two-dimensional arc-shaped contour and the center of the liquid injection hole of the battery cell, and determining a second distance between the three-dimensional arc-shaped contour and the center of the liquid injection hole, includes: performing centerline fitting processing on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a two-dimensional centerline of the two-dimensional arc profile and a three-dimensional centerline of the three-dimensional arc profile; respectively determining a first center coordinate of the two-dimensional center line in the two-dimensional image and a second center coordinate of the three-dimensional center line in the three-dimensional image; Determine the distance between the first circle center coordinate and the first comparison coordinate as the first distance, and determine the distance between the second circle center coordinate and the second comparison coordinate as the second distance; The first comparison coordinates are coordinates of the center of the injection hole in the two-dimensional image, and the second comparison coordinates are coordinates of the center of the injection hole in the three-dimensional image.

20. The detection method according to claim 17, characterized in that The detection method further comprises: determining a two-dimensional size of the two-dimensional arc profile and a three-dimensional size of the three-dimensional arc profile; Comparing the two-dimensional size with a first preset weld bead size range to obtain the two-dimensional detection result, and comparing the three-dimensional size with a second preset weld bead size range to obtain the three-dimensional detection result; The obtaining of the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result includes: When the two-dimensional detection result indicates that the two-dimensional dimension is outside the first preset weld bead size range, and / or the three-dimensional detection result indicates that the three-dimensional dimension is outside the second preset weld bead size range, it is determined that the welding detection result of the battery cell is that the weld bead size of the battery cell is abnormal.

21. The detection method according to claim 20, characterized in that Determining the two-dimensional size of the two-dimensional arc profile and the three-dimensional size of the three-dimensional arc profile includes: Performing geometric transformation on the two-dimensional arc profile and the three-dimensional arc profile respectively to obtain a first contour to be measured and a second contour to be measured that form a straight line; The two-dimensional size corresponding to the first contour to be measured and the three-dimensional size corresponding to the second contour to be measured are determined.

22. The detection method according to claim 21, characterized in that The two-dimensional dimensions include: a second length, a maximum width, and a minimum width; the three-dimensional dimensions include: a maximum height and a minimum height; The detection method further comprises: determining a second length corresponding to the first contour to be measured; Performing equal pixel segmentation on the first contour to be measured, and determining the width of each pixel obtained by the segmentation to obtain a width set; performing equal pixel segmentation on the second contour to be measured, and determining the height of each pixel obtained by the segmentation to obtain a thickness set; The maximum width and the minimum width in the width set are determined respectively, and the maximum height and the minimum height in the thickness set are determined.

23. The detection method according to any one of claims 11 to 15, characterized in that The two-dimensional positive electrode welding region includes a two-dimensional weld bead region, and the three-dimensional positive electrode welding region includes a three-dimensional weld bead region; The performing abnormality detection on the weld bead of the two-dimensional positive electrode welding area to obtain a two-dimensional detection result, and the performing abnormality detection on the weld bead of the three-dimensional positive electrode welding area to obtain a three-dimensional detection result, include: Respectively performing pixel value recognition on pixel points in the two-dimensional positive electrode weld bead region and the three-dimensional positive electrode weld bead region to obtain a first recognition result and a second recognition result; When the first recognition result indicates that a preset brightness area exists in the third area of the two-dimensional positive weld bead area, and the second recognition result indicates that the preset brightness area exists in the fourth area of the three-dimensional positive weld bead area, the two-dimensional detection result is obtained based on a comparison between a size of the preset brightness area in the two-dimensional image and a first preset interval size; and the three-dimensional detection result is obtained based on a comparison between a size of the preset brightness area in the three-dimensional image and a second preset interval size. The pixel values of the pixels within the preset brightness area are less than the preset pixel value; the fourth area is an area obtained by mapping the third area to the three-dimensional image based on a mapping relationship; and the mapping relationship is a corresponding relationship obtained by importing the two-dimensional image and the three-dimensional image into the same coordinate system for registration; The obtaining of the welding detection result of the battery cell based on the two-dimensional detection result and the three-dimensional detection result includes: When the two-dimensional detection result indicates that the size of the preset brightness area in the two-dimensional image is larger than the first preset interval size, and / or the three-dimensional detection result indicates that the size of the preset brightness area in the three-dimensional image is larger than the second preset interval size, it is determined that the welding detection result of the battery cell is that there is a welding abnormality in the weld of the battery cell.

24. The detection method according to any one of claims 11 to 23, characterized in that: The detection method further comprises: When the welding detection result is characterized as a detection abnormality, generating a marking signal; the marking signal is used to mark the battery cell; The detected abnormality includes at least one of the following: welding slag exists in the welding area of the battery cell, the weld bead of the battery cell is offset, the weld bead size of the battery cell is abnormal, and the weld bead of the battery cell has a welding abnormality.

25. The detection method according to claim 24, characterized in that The detection method further comprises: When the number of times the abnormality is detected reaches a preset number, a shutdown signal is generated; the shutdown signal is used to indicate stopping work and / or detecting the cause of the abnormality.

Images