Pole piece folding detection method and system

Abstract

The invention discloses a pole piece turnover detection method and system, and relates to the technical field of batteries. The method comprises the following steps: acquiring a lamination image obtained by performing image acquisition on a lamination by a camera arranged at a lamination position at a preset time, wherein the preset time at least comprises: after positive and negative pole pieces are laminated and before the positive pole pieces are clamped by a lamination feeding clamping jaw and after the positive pole pieces are clamped by the lamination feeding clamping jaw; detecting an included angle between a pole piece transverse edge and a pole piece vertical edge of the positive pole piece in the lamination image at each preset time; the transverse edge of the pole piece is an edge line of the pole piece in the horizontal direction, and the vertical edge of the pole piece is an edge line of the pole piece in the vertical direction; and according to the included angle between the pole piece transverse edge and the pole piece vertical edge of the positive pole piece in the lamination image at each preset time, determining the folding state of the positive pole piece at the preset time. A complete detection closed loop is formed, and the whole process of the folding defect generation of the positive plate is monitored, so that the pole plate with the folding defect can be conveniently intercepted in time, and the production quality of a battery cell is further improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of batteries, in particular to a tab folding detection method and system. BACKGROUND

[0002] A feeding and stacking integrated machine is an integrated machine with automatic feeding and stacking functions, and is mainly used in environments such as electronic component mounting or battery cell assembly, to stack tabs and other materials to form a complete structure. For example, a feeding and stacking integrated machine for lithium battery manufacturing combines positive tabs, negative tabs and separators to obtain a stacked tab, and the stacked tab is grabbed from a conveyor belt or a tab combining manipulator, and is stacked together with other stacked tabs according to a specific stacking rule in a stacking state to form a cell.

[0003] However, during the cell assembly process, the positive tab in the stacked tab is folded due to factors such as equipment movement and tab material, and the tab folding directly affects the quality, safety and performance of the cell. Therefore, there is an urgent need for a tab folding detection method to detect the folding state of the tab in a timely manner during the cell production process to ensure the quality of the cell. SUMMARY

[0004] In view of the above problems, the present application is proposed to provide a tab folding detection method and system to detect the folding state of the tab in a timely manner during the cell production process. The specific scheme is as follows:

[0005] The first aspect of the application provides a tab folding detection method, comprising:

[0006] obtaining a stacked tab image obtained by a camera arranged at a tab combining position for image acquisition of the stacked tab at a preset time, wherein the preset time at least includes: after the positive and negative tabs are combined and before the feeding and stacking gripper clamps the positive tab, and after the feeding and stacking gripper clamps the positive tab;

[0007] detecting the angle between the tab horizontal edge and the tab vertical edge of the positive tab in the stacked tab image of each preset time; the tab horizontal edge is the edge line of the tab in the horizontal direction, and the tab vertical edge is the edge line of the tab in the vertical direction;

[0008] determining the folding state of the positive tab at the preset time according to the angle between the tab horizontal edge and the tab vertical edge of the positive tab in the stacked tab image of each preset time.

[0009] The second aspect of the application provides a tab folding detection system, comprising: a camera and a visual detection module;

[0010] The camera is arranged at a laminating position of the laminating and feeding integrated machine, and is configured to collect laminated sheet images at preset time points, obtain laminated sheet images at different preset time points, and transmit the laminated sheet images at different preset time points to the visual detection module.

[0011] The visual detection module is configured to receive the laminated sheet images at different preset time points transmitted by the camera, and process the laminated sheet images at different preset time points according to the positive sheet folding detection method described in any one of the preceding first aspects of the present application, to obtain the folding state of the positive sheet at the preset time points.

[0012] By means of the above technical solution, the laminated sheet images of the laminated sheet are collected at two key time points before and after the clamping action of the feeding and laminating clamp at the laminating (laminated sheet) position, and the folding state of the positive sheet at the two key time points is determined by detecting the included angle between the horizontal and vertical edges of the positive sheet in the laminated sheet image. The folding state of the positive sheet before clamping is detected, which can identify the deformation of the positive sheet itself, such as folding or natural drooping. The folding state of the positive sheet after clamping is detected, which can confirm whether new folding or aggravation of the original deformation is caused during the operation of the feeding and laminating clamp. Based on this, a complete detection closed loop is formed, which realizes the whole process monitoring of the folding defect of the positive sheet, so as to facilitate the timely interception of the positive sheet with folding defect and further improve the production quality of the battery cell. BRIEF DESCRIPTION OF DRAWINGS

[0013] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of preferred embodiments, and are not meant to limit the present application. Moreover, the same reference numerals in the attached drawings indicate the same or similar components. In the drawings:

[0014] Figure 1 A system architecture schematic diagram for implementing the positive sheet folding detection method provided by the embodiments of the present application;

[0015] Figure 2 A flowchart schematic diagram of a positive sheet folding detection method provided by the embodiments of the present application;

[0016] Figure 3 An example diagram of laminated sheet images in different folding states provided by the embodiments of the present application;

[0017] Figure 4 An example diagram of angle detection provided by the embodiments of the present application;

[0018] Figure 5 An example diagram of positive sheet drooping provided by the embodiments of the present application;

[0019] Figure 6An example diagram of several models provided for an embodiment of the present application;

[0020] Figure 7 An example diagram of tail area detection provided for an embodiment of the present application;

[0021] Figure 8 A structural schematic diagram of an electronic device provided for an embodiment of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0023] In the diaphragm unwinding process of the feeding and stacking integrated machine, the vacuum chuck moves the positive electrode sheet and the negative electrode sheet to the two sides of the diaphragm, respectively, the mechanical clamping jaw (hereinafter also referred to as the feeding and stacking clamping jaw) clamps the positive and negative electrode sheets from the two sides, carries the diaphragm together to the lamination table (sheet combining table) to perform sheet lamination. However, in the process of clamping and feeding the mechanical clamping jaw, the positive electrode sheet may be sagging or folded, which directly affects the quality of the battery cell finally formed by the electrode sheet. In order to accurately detect the folding of the positive electrode sheet in the above process, the present application provides an electrode sheet folding detection method, system and related equipment.

[0024] The present application provides an electrode sheet folding detection method, which can be applied to an electrode sheet folding detection system deployed in a feeding and stacking integrated machine. Referring to Figure 1 , the present application provides a system architecture schematic diagram for implementing the electrode sheet folding detection method, which includes a camera and a visual detection module. The camera is arranged at the positive and negative sheet combining position of the feeding and stacking integrated machine, used to collect images of the laminated sheet at a preset time, obtain laminated sheet images at different preset times, and transmit the laminated sheet images at different preset times to the visual detection module. The visual detection module is used to receive the laminated sheet images at different preset times transmitted by the camera, and implement the electrode sheet folding detection method based on the laminated sheet images at different preset times.

[0025] Next, the product form of the camera and the visual detection module in the electrode sheet folding detection system will be described.

[0026] The camera can use the existing camera in the feeding and stacking integrated machine, such as the OH detection camera and the CCD camera arranged at the sheet combining position. Optionally, the image acquisition program suitable for the electrode sheet folding detection method is introduced to the existing camera. In this way, through the functional upgrade of the existing equipment (camera), the existing equipment assets of the feeding and stacking integrated machine are maximally utilized, and the electrode sheet folding detection and the production quality improvement of the battery cell are realized at a low cost.

[0027] In a possible implementation, the camera can be arranged vertically to the splicing position, for example, directly below the splicing position, to capture the stack image from directly below the stack.

[0028] The visual detection module can be implemented by software, hardware, or a combination thereof. Optionally, the visual detection module can be embedded in the core processor of the stack feeding and splicing integrated machine in hardware form or independent of the core processor, or stored in the memory of the stack feeding and splicing integrated machine in software form, so as to enable the core processor to call and execute the specific steps of the stack folding detection method.

[0029] The stack folding detection method is exemplarily described by taking the visual detection module arranged in the stack feeding and splicing integrated machine as an example. Figure 2 The stack folding detection method specifically includes the following steps.

[0030] In step S100, a stack image captured by a camera arranged at a splicing position at a preset time is obtained. The preset time includes at least: after the positive and negative electrode plates are spliced and before the stack feeding and splicing clamp holds the positive electrode plate, and after the stack feeding and splicing clamp holds the positive electrode plate.

[0031] In this step, the camera arranged at the splicing position captures the stack image at the preset time to capture the key state change of the stack.

[0032] Specifically, in the present application, the camera takes two photos from directly below the stack. The first photo is taken after the positive and negative electrode plates are spliced and before the stack feeding and splicing clamp holds the positive electrode plate. The second photo is taken after the stack feeding and splicing clamp holds the positive electrode plate. It can be understood that before the stack feeding and splicing clamp grabs the stack, the positive electrode plate will appear to be drooping on the splicing robot, which causes the stack feeding and splicing clamp to completely flip the positive electrode plate when it holds the positive electrode plate. Alternatively, after the stack feeding and splicing clamp grabs the stack and sends the stack to the stacking table, the positive electrode plate tail may be blown by the air flow generated by the movement of the stack feeding and splicing clamp, resulting in the folding of the electrode plate. Therefore, using only the stack image obtained by the first photo may not accurately detect the final folding state of the positive electrode plate. Therefore, the present application takes photos at two preset times to achieve full monitoring of the folding state changes such as folding and drooping of the stack, and to avoid missing the transient defects in the holding process.

[0033] The first photo is used to detect whether the initial stack state has a potential folding risk, for example, drooping. The second photo is used to verify the effect of the holding on the electrode plate and to detect whether the electrode plate is completely folded due to mechanical operation.

[0034] Optionally, the camera can be configured with a corresponding execution program to automatically trigger image acquisition at a preset timing. For example, the camera captures the laminates in the shooting range in real time, detects the device states such as the splicing machine, the laminate feeding clamp, and the like placed in the current picture, and automatically triggers the image acquisition function when each device state corresponds to the preset timing of each device state, thereby obtaining the laminate image in the current picture. For example, when it is detected that the laminate feeding clamp starts to move or the splicing machine laminate is sent to the close position of the laminate feeding clamp, i.e., after splicing and before clamping, the camera triggers shooting, thereby collecting the laminate image at the current time.

[0035] In another possible implementation, a control module in the laminate folding detection system can also be used to detect the clamping state of the laminate feeding clamp, determine whether the current timing is the preset timing based on the clamping state, and control the camera to collect the image of the laminate when the current timing is the preset timing. It can be understood that the control module can receive the real-time state of the splicing machine, the laminate feeding clamp, and the like in real time, determine whether the current timing is the preset shooting timing by detecting the real-time state fed back by the current laminate feeding clamp, and send a shooting instruction to the camera to collect the image of the laminate in the shooting range if the current timing is the preset shooting timing.

[0036] Step S110, detecting the included angle between the laminate horizontal edge and the laminate vertical edge of the positive electrode in the laminate image of each preset timing; the laminate horizontal edge is the edge line of the laminate in the horizontal direction, and the laminate vertical edge is the edge line of the laminate in the vertical direction.

[0037] Step S120, determining the folding state of the positive electrode at the preset timing according to the included angle between the laminate horizontal edge and the laminate vertical edge of the positive electrode in the laminate image of each preset timing.

[0038] Reference Figure 3 The present application provides an example of a laminate image with different folding states. As shown in the figure, when the positive electrode in the laminate does not fold or sag, the imaging of the positive electrode in the laminate image is shown as "normal electrode" on the left side of the figure, with straight edges, flat surfaces, and standard right angles at the corners, regular overall shape, and smooth surface. During the laminating process, such normal electrodes can be accurately stacked together to ensure the structural integrity and performance stability of the battery.

[0039] Figure 3 As shown in the middle, the positive electrode is in a completely folded state. One side edge of the positive electrode has been folded and deformed, forming an irregular bevel. This folding may be caused by mechanical stress, improper operation, or material properties. Folded electrodes may reduce the contact area with adjacent electrodes during laminating, which may cause uneven local current density and affect the performance and safety of the battery, and may even cause internal short circuit.

[0040] Figure 3 The positive electrode sheet shown on the right side is in a sagging state, which can be understood as an intermediate state between the fully folded state and the normal state (no folding state). Among them, one corner of the electrode sheet appears sagging deformation, and the approximate bending angle of the sagging part is marked with "R" in the figure. This sagging may be due to insufficient rigidity of the electrode sheet itself, gravity or external force during the stacking process. The sagging electrode sheet will cause uneven gaps between the electrode sheets during stacking, affecting the infiltration of electrolyte and ion transmission, and thus reducing the performance and cycle life of the battery.

[0041] Therefore, the stacking images of the positive electrode sheet in different folding states have certain differences in imaging, and the embodiments of the present application utilize these image differences to calculate the included angle R between the horizontal edge and the vertical edge of the positive electrode sheet in the stacking image corresponding to each preset time through an image processing algorithm, and determine the folding state of the positive electrode sheet according to the included angle value. It can be understood that the folding state in the present embodiment can include: no folding state, fully folded state and sagging state.

[0042] According to the detected positive electrode sheet folding state at different preset times, different operations are performed. It can be understood that in the case that the folding state of the positive electrode sheet at the preset time meets the preset alarm condition, a folding warning information is generated. In the preset time after the positive and negative electrode sheets are stacked and before the stacking and clamping of the stacking and clamping jaw, the folding or sagging of the positive electrode sheet can generate a warning information to prompt the stacking and clamping all-in-one machine or the operator to pay attention to the change of the folding state of the positive electrode sheet after the stacking and clamping of the stacking and clamping jaw. If it is still in a fully folded or sagging state, the stack will be discarded. Thus, according to the whole process monitoring result of the folding of the positive electrode sheet in the stack, the waste sheet is timely warned and intercepted, and the quality of the final assembled battery cell is improved.

[0043] In summary, the present application collects the stacking images of the stack at two key times before and after the clamping action of the stacking and clamping jaw in the stacking (stacking) station, and judges the folding state of the positive electrode sheet at the two key times by detecting the included angle between the horizontal and vertical edges of the positive electrode sheet in the stacking image. Among them, detecting the folding state of the positive electrode sheet before clamping can identify the deformation conditions such as folding or natural sagging existing in the positive electrode sheet itself, and detecting the folding state of the positive electrode sheet after clamping can confirm whether new folding or aggravation of the original deformation is caused in the operation process of the stacking and clamping jaw. Based on this, a complete detection closed loop is formed to realize the whole process monitoring of the folding defect of the positive electrode sheet, so as to facilitate the timely interception of the electrode sheet with folding defect and further improve the production quality of the battery cell.

[0044] Next, the other possible implementations of the electrode sheet folding detection method provided by the present application are described through the following embodiments.

[0045] In a possible implementation, the step S110 detecting the included angle between the positive plate horizontal edge and the positive plate vertical edge in the positive plate image of each preset occasion comprises: generating a horizontal edge detection frame and a vertical edge detection frame according to the positive plate image; determining key horizontal edge position information located in the horizontal edge detection frame and key vertical edge position information located in the vertical edge detection frame in the positive plate image; and calculating the included angle between the positive plate horizontal edge and the positive plate vertical edge based on the key horizontal edge position information and the key vertical edge position information.

[0046] Firstly, the horizontal edge and the vertical edge of the positive plate are located on the collected positive plate image by a specific algorithm or method, that is, a corresponding detection frame is generated. The detection frame can help to locate the positions of the horizontal edge and the vertical edge more accurately, and prepare for obtaining the key horizontal edge position information and the key vertical edge position information.

[0047] Optionally, a traditional image processing algorithm such as an edge detection algorithm can be used to first detect all edge information in the positive plate image, that is, to locate the specific position of the positive plate in the image. Then, the straight lines are detected by means such as Hough transformation according to the shape and position characteristics of the positive plate, and the key horizontal edge detection frame and the key vertical edge detection frame are determined according to the detected straight line information, in combination with the approximate position and direction of the positive plate. Alternatively, a deep learning target detection model can also be used for detection, such as a target detection model trained by using a large number of positive plate images labeled with the positions of the horizontal edge and the vertical edge of the positive plate. The collected positive plate image is input into the model, and the coordinate information of the horizontal edge detection frame and the vertical edge detection frame is directly output by the model, or the key horizontal edge position information and the key vertical edge position information are directly output.

[0048] In a possible implementation, the horizontal edge detection frame and the vertical edge detection frame are generated according to the positive plate image, which comprises: positioning and detecting the positive plate horizontal edge and the positive plate vertical edge in the positive plate image based on the pre-set initial horizontal edge detection frame and the initial vertical edge detection frame to obtain the position information of the positive plate horizontal edge and the positive plate vertical edge in the positive plate image respectively; and generating a horizontal edge detection frame for locating the key horizontal edge and a vertical edge detection frame for locating the key vertical edge of the positive plate vertical edge according to the position information of the positive plate horizontal edge and the positive plate vertical edge in the positive plate image respectively, wherein the distance between the edge close to the included angle in the horizontal edge detection frame and the positive plate vertical edge is a first preset distance, and the distance between the edge close to the included angle in the vertical edge detection frame and the positive plate horizontal edge is a second preset distance.

[0049] With reference to the following Figure 4 , the embodiment provided by the present application provides an example diagram of the included angle detection, and the embodiment is described in detail.

[0050] It can be understood that, with reference to the following Figure 3From the folded state of the pole piece and the sag state of the pole piece, it can be seen that even if the pole piece is folded or sagged, the offset angle of the horizontal edge and the vertical edge compared to the normal pole piece is very small. If the horizontal edge detection frame and the vertical edge detection frame are far away from the folding position or the sagging position, the final angle between the horizontal edge and the vertical edge is the same as the angle between the horizontal edge and the vertical edge of the normal pole piece, which leads to missed detection. Therefore, in the embodiment of the present application, the initial horizontal edge detection frame and the initial vertical edge detection frame are used to locate the position of the pole piece in the current detected pole piece image, and then the horizontal edge detection frame for detecting the key horizontal edge with a small tilt and the vertical edge detection frame for detecting the key vertical edge with a small tilt are generated at the key position close to the folding or sagging position (R angle) according to the located pole piece position. For example, Figure 4 The horizontal edge detection frame and the vertical edge detection frame can cover part of the two inclined edges forming the R angle (i.e. the key horizontal edge and the key vertical edge), so as to improve the detection accuracy of the angle R and improve the detection accuracy of the folded state of the pole piece.

[0051] The pre-set initial horizontal edge detection frame and the initial vertical edge detection frame are set based on the prior knowledge of the approximate position and shape of the positive pole piece. The initial horizontal edge detection frame and the initial vertical edge detection frame are used for positioning detection operation on the pole piece image, and the image processing algorithm (such as edge detection, template matching, etc.) is used to preliminarily locate the accurate position information of the pole piece horizontal edge and the pole piece vertical edge in the pole piece image.

[0052] For example, the scale-invariant feature transform (SIFT) algorithm can be used to extract the feature points of the initial horizontal edge detection frame and the pole piece edge, and then the feature matching is used to determine the accurate position of the pole piece horizontal edge in the image. Alternatively, a template image containing the initial horizontal edge detection frame of the located pole piece horizontal edge can also be used for sliding matching on the pole piece image, and the position with the highest matching degree is found as the position of the pole piece edge. The above method can also be used to detect the position information of the vertical edge in the initial vertical edge detection frame, which will not be described here.

[0053] After obtaining the position information of the pole piece horizontal edge and the pole piece vertical edge, new horizontal edge detection frames for detecting key horizontal edges and vertical edge detection frames for detecting key vertical edges are generated according to the pre-set rules. Specifically, the distance between the edge close to the angle in the horizontal edge detection frame and the pole piece vertical edge (i.e. the first pre-set distance) and the distance between the edge close to the angle in the vertical edge detection frame and the pole piece horizontal edge (i.e. the second pre-set distance) are specified to ensure that the horizontal edge detection frame and the vertical edge detection frame can accurately locate the folded area, while avoiding containing too much irrelevant information.

[0054] The first pre-set distance and the second pre-set distance can be determined according to the size of the R angle generated in the folded and sagged state. For example, referring to Figure 4In the process of tail drooping to folding, the maximum folding size of the horizontal edge and the vertical edge of the R angle is 4mm, and the corresponding first preset distance can be set to 5mm, the horizontal width of the horizontal edge detection frame can be set to 2-4mm, and the horizontal edge detection frame is designed to be as close to the R angle as possible without covering the R angle, and to cover part of the horizontal inclined edge (i.e. the key horizontal edge) constituting the R angle. Similarly, the vertical length of the vertical edge detection frame can be set to be greater than the horizontal width of the horizontal detection frame, and the second preset distance also needs to be set to meet the condition that the vertical edge detection frame is as close to the R angle as possible without covering the R angle, and to cover part of the vertical inclined edge (i.e. the key vertical edge) constituting the R angle.

[0055] After generating the horizontal edge detection frame and the vertical edge detection frame, the specific position information of the key horizontal edge and the key vertical edge needs to be extracted from the detection frame. These position information can accurately describe the position of the horizontal edge and the vertical edge in the image, which is the key data for subsequent calculation of the included angle.

[0056] Alternatively, the key horizontal edge position and the key vertical edge position can be determined by analyzing the pixel distribution and change in the horizontal edge detection frame and the vertical edge detection frame. Or in the horizontal edge detection frame and the vertical edge detection frame, a straight line fitting algorithm such as least squares method is used to fit the edge points in the detection frame to obtain the equation representing the key horizontal edge and the key vertical edge. The slope and intercept of the edge line can be extracted from the equation, which can accurately describe the position information of the key horizontal edge and the key vertical edge.

[0057] After obtaining the position information of the key horizontal edge and the key vertical edge, the R included angle between them is calculated by using mathematical calculation method. This included angle is an important basis for judging whether the pole piece is folded and the folding degree. Alternatively, according to the equation of the key horizontal edge and the key vertical edge obtained by fitting, the included angle formed by the key horizontal edge and the key vertical edge is calculated according to the slope of the two equations. Or, according to the position information of the key horizontal edge and the key vertical edge, the direction vectors of the two edges are determined, and the included angle between the two direction vectors is calculated by using the dot product of the vectors.

[0058] Further, according to the included angle between the pole piece horizontal edge and the pole piece vertical edge of the positive pole piece in the pole piece image of each preset time obtained by the above process, the folding state of the positive pole piece at the preset time is determined.

[0059] In one possible implementation, step S120, determining the folding state of the positive pole piece at the preset time, includes: calculating the difference between the included angle between the pole piece horizontal edge and the pole piece vertical edge in the pole piece image of each preset time and the preset angle to obtain an angle difference; determining the folding state corresponding to the angle difference according to the angle difference threshold range corresponding to different folding states, as the folding state of the pole piece at the preset time.

[0060] The preset angle represents a standard included angle between the horizontal edge and the vertical edge of the pole piece in a normal state (no folding state), such as 90°. By subtracting the actual detected included angle from the preset angle, the deviation of the actual included angle relative to the standard included angle can be quantified, and the deviation, that is, the angle difference, is an important basis for judging whether the pole piece is folded and the folding degree.

[0061] The process of setting the angle difference threshold range corresponding to different folding states includes: constructing a geometric model of the tail part of the positive pole piece in advance, the geometric model being a triangle constructed according to the projection result of the tail part of the positive pole piece, the bottom edge of the triangle being equivalent to the root of the tail part of the positive pole piece, and the hypotenuse being equivalent to the equivalent projection distance of the root of the tail part to the tip of the tail part on the horizontal plane; adjusting the length of the bottom edge and the hypotenuse of the triangle according to the drooping shape of the tail part of the pole piece under different folding states to obtain the geometric model under different folding states; calculating the top angle of the triangle in the geometric model under different folding states to determine the included angle range corresponding to different folding states; and determining the angle difference threshold range corresponding to different folding states based on the difference between the included angle range corresponding to different folding states and the preset angle.

[0062] Referring to Figure 5 , an example diagram of the drooping of the pole piece is provided to explain the construction principle of the above geometric model in detail.

[0063] The embodiment of the present application constructs a geometric model of the tail part of the positive pole piece in advance, and the model is a triangle constructed according to the projection result of the tail part of the positive pole piece. In this example, an isosceles triangle is taken as an example. Referring to Figure 5 , the part enclosed by the circular frame in the figure shows the shape of the tail part of the positive pole piece, which can be abstracted as an isosceles triangle from the projection result. The bottom edge of the isosceles triangle is equivalent to the root of the tail part of the positive pole piece, that is, the line segment L corresponding to the position where the tail part of the positive pole piece starts to droop; and the hypotenuse is equivalent to the equivalent projection distance of the root of the tail part to the tip of the tail part on the horizontal plane, that is, the horizontal projection length (d1, d2) from the starting point of the droop to the end of the droop (the length a of the drooping edge). It can be understood that, in actual camera imaging, the equivalent projection distance of the root of the tail part to the tip of the tail part on the horizontal plane and the equivalent focal length of the isosceles triangle are equivalent to the R angle of the isosceles triangle. Figure 1

[0064] Referring to the projection process in the figure, the tail part of the positive pole piece is drooped on the horizontal plane to form a dashed line projection, and the dashed line projection is the geometric model of the isosceles triangle constructed in this embodiment. The state of the tail part of the positive pole piece being drooped can be realized by adjusting the droop height h and the droop length a, and the projection edge d1 of the droop length a is calculated by using the geometric function of the triangle, as shown in the following formula (1).

[0065] ​ (1)

[0066] Based on this, the projection edges of the two sagging edges of the sagging part of the positive electrode tab tail are calculated, that is, d1 and d2. Based on d1, d2 and the length L of the sagging root, the projection result of the sagging part of the positive electrode tab tail is obtained, and further, the top angle θ of the sagging shape is determined by using the geometric function of the triangle, which is used as the angle value of the folding state corresponding to the geometric model.

[0067] Based on the above principle, according to the sagging shape of the tab tail under different folding states, the length of the base and the hypotenuse of the isosceles triangle are adjusted to obtain the geometric model under different folding states, so as to measure the top angle θ in the isosceles triangle, which is equivalent to the R angle in Figure 3 Figure 6 Referring to, the length of the sagging edge a of the sagging part of the positive electrode tab tail is unchanged, that is, the length of the sagging root L is unchanged, and as the sagging height increases, the projection edge d1 of the sagging edge a becomes smaller, the length of the isosceles triangle becomes shorter, and the top angle becomes larger until the length of the projection edge is insufficient to form a triangle. The sagging height represents the folding degree of the tab tail, so it can be seen that the greater the folding degree of the tail, the larger the top angle.

[0068] Based on this, the angle range corresponding to different folding states is determined. That is, the angle between the horizontal and vertical edges of the positive electrode tab tail is 90° when the positive electrode tab tail is not folded, the angle between the horizontal and vertical edges of the positive electrode tab tail is 180° when the positive electrode tab tail is completely folded, and the angle between the horizontal and vertical edges of the positive electrode tab tail is between 90° and 180° when the positive electrode tab tail is sagging.

[0069] In one possible implementation, according to the angle range corresponding to different folding states, the folding state corresponding to the angle between the tab horizontal edge and the tab vertical edge in the lamination image of each preset opportunity is determined as the folding state of the tab at the preset opportunity.

[0070] In another possible implementation, the angle range is subtracted from the preset angle to obtain the angle difference threshold range under different folding states, such as: the angle difference threshold range corresponding to the no-folding state is 0° when the positive electrode tab tail is not folded; the angle difference threshold range corresponding to the completely folded state is 90° when the positive electrode tab tail is completely folded; and the angle difference threshold range corresponding to the tab tail sagging state is greater than 0° and less than 90° when the positive electrode tab tail is sagging. Among them, sagging is a special case of folding, so the angle difference can not only represent the folding state, but also reflect the folding degree.

[0071] Based on this, the angle difference between the angle between the tab horizontal edge and the tab vertical edge in the lamination image of each preset opportunity and the preset angle is matched with the angle difference threshold range corresponding to different folding states to determine the folding state corresponding to the angle difference as the folding state of the tab at the preset opportunity.

[0072] In a possible implementation, the folding detection of the positive plate can also be implemented by using the gray scale information in the laminated plate image. Specifically, the detection process can include: performing gray scale detection on the tail region of the positive plate of the laminated plate image at each preset timing to obtain a gray scale statistical value of the tail region of the positive plate; and determining the folding state of the plate if the gray scale statistical value meets a preset gray scale condition corresponding to different folding states.

[0073] First, for the laminated plate images collected at preset different preset timings, the tail region of the positive plate is focused on, and a gray scale statistical value of the region is obtained by using a preset gray scale detection method. The gray scale value reflects the brightness of the pixels in the image. The surface light reflection of the tail region of the positive plate may change under different folding states, thereby causing differences in the gray scale value. Therefore, the gray scale statistical value can be used as a characteristic parameter for judging the folding state.

[0074] It can be understood that when the tail region of the positive plate is folded, the folded part includes the positive plate, the separator, and the negative plate. The colors of the separator and the negative plate are usually brighter than that of the positive plate. Therefore, due to the addition of the positive plate (the folded tail region of the positive plate) under the folding state, the gray scale value of the tail region of the positive plate deviates from the normal value. Based on this, if it is detected that the gray scale statistical value of the tail region of the plate deviates from a preset threshold value, the folding state of the current positive plate can be determined.

[0075] Specifically, the position of the tail region of the positive plate in the laminated plate image needs to be accurately determined. The tail region of the positive plate in the laminated plate image can be located by using an edge detection algorithm (such as a Canny operator), a target detection model (such as a YOLO series), or the like.

[0076] In a possible implementation, the gray scale detection is performed on the tail region of the positive plate of the laminated plate image at each preset timing to obtain a gray scale statistical value of the tail region of the positive plate, including: locating the tail region of the positive plate in the laminated plate image, and generating a tail detection frame in the tail region; and performing statistics on the gray scale values of the pixel points in the tail detection frame in the laminated plate image to obtain the gray scale statistical value of the tail region of the positive plate.

[0077] First, the tail position of the positive plate in the laminated plate image is accurately located. It can be understood that the positive plate has specific morphological and positional characteristics in the laminated plate image. By locating the tail region, the specific region for subsequent gray scale detection can be determined. Further, a tail detection frame is generated for the region to clearly define the range of the tail region, limit the pixel points to be analyzed within the frame, exclude the interference of other irrelevant regions, and improve the accuracy and efficiency of the gray scale detection.

[0078] Optionally, a large number of standard images of the tail part of the positive electrode sheet can be collected in advance as a template library. When processing a new stacking image, a feature matching algorithm (such as SIFT, SURF, etc.) is used to match the feature points in the stacking image with the feature points in the template library. The position of the tail part of the positive electrode sheet in the image is determined according to the matching result, and a tail detection frame is generated.

[0079] Optionally, referring to the following Figure 7 The tail part of the positive electrode sheet in the stacking image is located, and a tail detection frame is generated in the tail part region. The method comprises: detecting a region in the stacking image in a preset horizontal edge detection frame to determine the position information of the electrode sheet horizontal edge; and adjusting the position of a preset initial tail detection frame according to the position information of the electrode sheet horizontal edge to obtain a tail detection frame, the tail detection frame containing a region where the tail part of the positive electrode sheet in the stacking image is located.

[0080] The preset horizontal edge detection frame is a specific region preset in the stacking image, which focuses on the position range where the electrode sheet horizontal edge may exist. By detecting in this region, the specific position information of the electrode sheet horizontal edge can be accurately located, providing key coordinate basis for the generation of the subsequent tail detection frame.

[0081] The initial tail detection frame is a template frame preset based on the approximate tail part position of the electrode sheet. The horizontal edge position information obtained is used to adjust the coordinate parameters of the initial frame through geometric relationship (such as the relative position rule of the horizontal edge and the tail part), so that the initial frame accurately covers the actual region of the tail part of the positive electrode sheet in the stacking image.

[0082] For example, according to the horizontal edge position coordinates, the offset of the tail detection frame is calculated. If the horizontal edge is located at the image coordinates (x1, y1), and the vertical distance between the horizontal edge and the tail part is known as m, the y coordinate of the initial tail detection frame can be adjusted to y1+m, and the x coordinate can be translated left and right according to the tail part width to ensure that the detection frame contains the tail part region.

[0083] After the tail detection frame is determined, the gray value of all pixel points in the tail part region is extracted and statistically calculated. The obtained gray statistical value can be the gray average value or the gray variance, and optionally, the gray median value can also be selected.

[0084] The gray average value μ is the arithmetic average value of the gray values of all pixel points in the tail part region, which is calculated as shown in formula (2); and the gray variance σ 2 is a statistical quantity for measuring the dispersion degree of the pixel gray values in the tail part region, which is calculated as shown in formula (3).

[0085] (2)

[0086] (3)

[0087] wherein, N represents the total number of pixel points in the tail region, and Ii represents the gray value of the i-th pixel point in the tail region (usually 0-255).

[0088] Further, the gray scale statistical value is compared with preset gray scale conditions (such as gray scale statistical value interval, threshold reference value, etc.) corresponding to different folding states, to determine the folding state of the positive plate tail in the laminated image taken at the preset time.

[0089] In a possible implementation, the above-mentioned manner of using the angle R between the positive plate horizontal edge and the positive plate vertical edge in the laminated image and the manner of using the gray scale statistical value of the positive plate tail are parallel manners for implementing positive plate folding detection, and can be applied simultaneously. When the above-mentioned two manners are used simultaneously for folding detection of the positive plate, as long as any one of the two manners detects that the positive plate is folded, a prompt information that the positive plate is folded can be sent to the front end.

[0090] Similarly, in the case where the folding state of the positive plate at the preset time meets the preset discard condition, the detected laminated plate is discarded, and the preset discard condition can include that the angle between the positive plate horizontal edge and the positive plate vertical edge is within a preset angle range, or the gray scale statistical value of the positive plate tail exceeds a preset range value. For example, the gray scale average value μ of the positive plate is usually within the range of 30-50, and if the detected gray scale statistical value exceeds the range, it proves that the current positive plate is folded; the angle R between the positive plate horizontal edge and the positive plate vertical edge in the laminated image is within the interval (94°, 180°]; or the difference between the angle R and 90° is within the interval (4°, 90°).

[0091] It can be understood that the preset angle range and the preset range value in the above-mentioned preset discard condition can be set according to the actual detection accuracy requirement and the actual parameter state of the positive plate, and are not uniquely limited herein.

[0092] The electronic device provided in the embodiments of the present application can include but is not limited to fixed terminals such as mobile phones, tablet computers, teaching large screens, wearable devices, etc. Figure 8 As shown in FIG. 1, an electronic device is shown, which shows a structural schematic diagram suitable for implementing the electronic device in the embodiments of the present application. The electronic device in the embodiments of the present application can include but is not limited to fixed terminals such as mobile phones, tablet computers, teaching large screens, wearable devices, etc. Figure 8 The electronic device shown in FIG. 1 is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0093] As shown in FIG. 1, an electronic device is shown, which shows a structural schematic diagram suitable for implementing the electronic device in the embodiments of the present application. The electronic device in the embodiments of the present application can include but is not limited to fixed terminals such as mobile phones, tablet computers, teaching large screens, wearable devices, etc. Figure 8As shown, the electronic device can include a processing device (e.g., a central processor, a graphics processor, etc.) 1 that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 2 or programs loaded from a storage device 8 into a random access memory (RAM) 3 to implement the tab folding detection method of the foregoing embodiments of the present application. In a state where the electronic device is powered on, various programs and data required for operation of the electronic device are also stored in the RAM 3. The processing device 1, the ROM 2, and the RAM 3 are connected to each other through a bus 4. An input / output (I / O) interface 5 is also connected to the bus 4.

[0094] Generally, the following devices can be connected to the I / O interface 5: input devices 6 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 7 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 8 including, for example, a memory card, a hard disk, etc.; and communication devices 9. The communication devices 9 can allow the electronic device to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 8 An electronic device having various devices is shown, but it should be understood that it is not required to implement or have all of the illustrated devices. More or less devices can alternatively be implemented or included.

[0095] The embodiments of the present application also provide a computer program product including computer readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the tab folding detection methods provided by the embodiments of the present application.

[0096] The embodiments of the present application also provide a computer readable storage medium carrying one or more computer programs, which, when executed by an electronic device, can cause the electronic device to implement any of the tab folding detection methods provided by the embodiments of the present application.

[0097] It should be further noted that the device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., they can be located in one place, or distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments. In addition, the connection relationship between the modules in the device embodiments provided by the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines.

[0098] Those skilled in the art can clearly understand, through the description of the foregoing embodiments, that the present application can be implemented by means of software and the necessary universal hardware, and of course can also be implemented by means of special hardware including special integrated circuits, special CPUs, special memories, special components, etc. Generally, functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure for implementing the same function can be various, such as analog circuits, digital circuits, or special circuits. However, for the present application, software program implementation is a better embodiment. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer's floppy disk, U disk, mobile hard disk, ROM, RAM, magnetic disk or optical disk, etc., including a plurality of instructions for causing a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in various embodiments of the present application.

[0099] In the above embodiments, all or part can be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part can be implemented in the form of a computer program product.

[0100] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, training device or data center to another through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. integrated with one or more available media. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)), etc.

[0101] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Each embodiment can be combined as needed, and the same or similar parts refer to each other.

Claims

1. A method for detecting electrode flipping, characterized in that, include: The image of the stacked sheets is obtained by the camera set at the stacking position capturing images of the stacked sheets at a preset time. The preset time includes at least: after the positive and negative electrode sheets are stacked and before the stacking jaws hold the stacked sheets, and after the stacking jaws hold the stacked sheets. Detect the angle between the horizontal edge and the vertical edge of the positive electrode in the stacked image at each preset timing; the horizontal edge of the electrode is the edge line of the electrode in the horizontal direction, and the vertical edge of the electrode is the edge line of the electrode in the vertical direction; The folding state of the positive electrode at each preset timing is determined based on the angle between the horizontal and vertical edges of the positive electrode in the stacking image at each preset timing.

2. The electrode flipping detection method according to claim 1, characterized in that, The detection of the angle between the horizontal and vertical edges of the positive electrode in the stacked image at each preset timing includes: Generate horizontal and vertical detection boxes based on the stacked images; Determine the key horizontal edge position information located in the horizontal edge detection box and the key vertical edge position information located in the vertical edge detection box in the stacked image; Based on the key horizontal edge position information and the key vertical edge position information, the included angle between the horizontal edge and the vertical edge of the electrode is calculated.

3. The electrode flipping detection method according to claim 2, characterized in that, The step of generating horizontal and vertical detection boxes based on the stacked images includes: Based on the pre-set initial horizontal edge detection box and initial vertical edge detection box, the horizontal edge and vertical edge of the electrode in the stacked image are located and detected to obtain the position information of the horizontal edge and vertical edge of the electrode in the stacked image respectively; Based on the position information of the horizontal edge and the vertical edge of the electrode in the stacked image, a horizontal edge detection box for locating the key horizontal edge and a vertical edge detection box for locating the key vertical edge of the electrode are generated. The distance between the edge of the horizontal edge detection box near the included angle and the vertical edge of the electrode is a first preset distance, and the distance between the edge of the vertical edge detection box near the included angle and the horizontal edge of the electrode is a second preset distance.

4. The electrode flipping detection method according to claim 1, characterized in that, Determining the folding state of the positive electrode at each preset timing based on the angle between the horizontal and vertical edges of the positive electrode in the stacked image at each preset timing includes: The angle difference is obtained by subtracting the angle between the horizontal and vertical edges of the electrode in the stacked image of each preset timing from the preset angle. Based on the angle difference threshold range corresponding to different folding states, the folding state corresponding to the angle difference is determined as the folding state of the electrode at the preset time.

5. The electrode flipping detection method according to claim 4, characterized in that, The process of setting the angle difference threshold range corresponding to the different folding states includes: A geometric model of the drooping tail of the positive electrode is pre-constructed. The geometric model is a triangle constructed based on the projection result of the drooping tail of the positive electrode. The base of the triangle is equivalent to the drooping root of the drooping tail of the positive electrode, and the hypotenuse is equivalent to the equivalent projection distance from the drooping root to the drooping tip on the horizontal plane. Based on the drooping shape of the tail of the electrode in different folding states, the length of the base and the hypotenuse of the triangle are adjusted to obtain the geometric model in different folding states; Calculate the vertex angle of the triangle in the geometric model under different folding states, and determine the range of included angles corresponding to different folding states; Based on the difference between the included angle range corresponding to the different folding states and the preset angle, the angle difference threshold range corresponding to the different folding states is determined.

6. The electrode folding detection method according to claim 5, characterized in that, The folding states include at least: no folding state, fully folded state, and electrode tail drooping state; when the preset angle is 90°: The angle difference threshold range corresponding to the non-flipping state is 0°; The angle difference threshold range corresponding to the fully folded state is 90°; The angle difference threshold range corresponding to the drooping state of the electrode tail is greater than 0° and less than 90°.

7. The electrode flipping detection method according to claim 5, characterized in that, Determining the folding state of the positive electrode at each preset timing based on the angle between the horizontal and vertical edges of the positive electrode in the stacked image at each preset timing includes: Based on the included angle range corresponding to different folding states, the flipping state corresponding to the included angle between the horizontal edge and the vertical edge of the positive electrode in the stacked image at each preset timing is determined as the folding state of the positive electrode at the preset timing.

8. The electrode flipping detection method according to claim 1, characterized in that, Also includes: Gray-scale detection is performed on the tail region of the positive electrode in the stacked image at each preset timing to obtain the gray-scale statistical value of the tail region of the positive electrode. When the grayscale statistical value meets the preset grayscale conditions corresponding to different folding states, the positive electrode sheet is in the folding state at the preset time.

9. The electrode flipping detection method according to claim 8, characterized in that, The step of performing grayscale detection on the tail region of the positive electrode in the stacked image at each preset timing to obtain the grayscale statistical value of the tail region of the positive electrode includes: Locate the tail of the positive electrode in the stacked image, and generate a tail detection box in the tail region; The grayscale values ​​of the pixels located in the tail detection box in the stacked image are statistically analyzed to obtain the grayscale statistical value of the tail of the positive electrode sheet.

10. The electrode flipping detection method according to claim 9, characterized in that, The step of locating the tail of the positive electrode in the stacked image and generating a tail detection box in the tail region includes: The region within the preset horizontal edge detection box in the stacked image is detected to determine the position information of the horizontal edge of the electrode. Based on the position information of the horizontal edge of the electrode, the position of the preset initial tail detection box is adjusted to obtain the tail detection box, which includes the area where the positive electrode tail is located in the stacked image.

11. The electrode flipping detection method according to claim 8, characterized in that, The grayscale statistical value is at least the grayscale average or grayscale variance.

12. The electrode flipping detection method according to any one of claims 1-11, characterized in that, Also includes: When the folding state of the positive electrode at the preset time meets the preset alarm conditions, a folding warning message is generated.

13. The electrode flipping detection method according to any one of claims 8-11, characterized in that, Also includes: If the positive electrode sheet meets the preset discard conditions when it is folded at the preset time, the detected stack of sheets is discarded. The preset discard conditions include: in the current folded state, the angle between the horizontal edge and the vertical edge of the positive electrode sheet is within a preset angle range, or the gray value of the tail of the positive electrode sheet exceeds a preset range value.

14. A system for detecting electrode flipping, characterized in that, include: Camera and vision inspection module; The camera is set at the stacking position of the integrated stacking machine and is used to acquire images of the stacked sheets at preset times to obtain stacked sheet images at different preset times, and transmit the stacked sheet images at different preset times to the visual detection module. The visual detection module is used to receive stacked images at different preset times transmitted by the camera, and to process the stacked images at different preset times according to the electrode flipping detection method according to any one of claims 1-13, so as to obtain the flipping state of the positive electrode at the preset time.

15. The electrode flipping detection system according to claim 14, characterized in that, The camera is positioned vertically to the assembly position.

16. The electrode flipping detection system according to claim 14 or 15, characterized in that, The camera is a CCD camera.

17. The electrode flipping detection system according to claim 14, characterized in that, Also includes: The control module is used to detect the gripping state of the stacking gripper and determine whether the current moment is a preset timing based on the gripping state. When the current moment is the preset timing, the camera is controlled to acquire images of the stacked sheets. The preset timing includes: after the positive and negative electrode sheets are combined and before the stacking jaws hold the stacked sheets, and after the stacking jaws hold the stacked sheets.

Images