CN117218057A - New energy battery pole welding line defect detection method, equipment and storage medium - Google Patents
New energy battery pole welding line defect detection method, equipment and storage medium Download PDFInfo
- Publication number
- CN117218057A CN117218057A CN202310957857.1A CN202310957857A CN117218057A CN 117218057 A CN117218057 A CN 117218057A CN 202310957857 A CN202310957857 A CN 202310957857A CN 117218057 A CN117218057 A CN 117218057A
- Authority
- CN
- China
- Prior art keywords
- battery pole
- welding
- pixel
- point cloud
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003466 welding Methods 0.000 title claims abstract description 162
- 230000007547 defect Effects 0.000 title claims abstract description 54
- 238000001514 detection method Methods 0.000 title claims abstract description 52
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 30
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000004880 explosion Methods 0.000 claims description 24
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 11
- 238000013461 design Methods 0.000 claims description 3
- 238000005476 soldering Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 7
- 238000013135 deep learning Methods 0.000 description 5
- 238000006073 displacement reaction Methods 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 229910000679 solder Inorganic materials 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Abstract
The invention discloses a new energy battery pole welding seam defect detection method, equipment and a storage medium, wherein the method comprises the following steps: step 1, acquiring point cloud data of a measured surface of a battery module and converting the point cloud data into a binary image; step 2, positioning welding seam areas of each battery pole in the binary image; step 3, processing point cloud data of a welding line area of the battery pole by adopting a mask operation method, and detecting welding line defects by combining a set threshold value; the device comprises a processor and a memory storing program instructions, the program instructions being stored in a storage medium, the program instructions being read and executed to perform the steps of the detection method. The invention can carry out defect detection after identifying and positioning the battery pole column region, and obtain the type information, the position information and the geometric information of each weld defect, and has the advantage of strong usability.
Description
Technical Field
The invention relates to the field of battery welding seam detection methods, in particular to a new energy battery pole welding seam defect detection method, equipment and a storage medium.
Background
The battery is an important power component of the new energy vehicle, and the battery needs to detect the welding line of the electrode column after production. The prior art mainly comprises manual detection, acoustic flaw detection, 2D image detection and deep learning detection for detecting the welding seam of the electrode column on the battery. Wherein:
the reliability of the manual detection method is influenced by subjective factors of detection personnel, so that the detection result is full of uncertainty, for example, the detection personnel is tired or is not carefully observed, and the detection omission phenomenon is easy to occur for tiny defects; the personal proficiency influences the efficiency of detection, and the differentiation leads to realizing the pipelined production difficultly, and manual detection still can consume higher human cost in addition.
In the acoustic flaw detection, the eddy current and ultrasonic flaw detection equipment are adopted for detection, the detection precision is higher, but the complex detection mode and the high price can not classify the weld defects of the battery, and the information such as the width and the area of the defects can not be obtained, so that obvious limitations exist.
In the 2D image detection, a welding line area of a new energy battery pole is shot through a camera, the shot image is analyzed to extract the characteristics of specific defects, a special defect matching template is formulated based on the characteristics, and the defects caused in the welding line of the battery are detected through the formulated template. Because of the complexity and variety of defects generated during the welding process, external illumination has a high impact on the quality of the acquired images. Schemes that rely solely on 2D images for defect detection are not able to accomplish efficient detection.
The deep learning detection mainly adopts a deep learning method to detect, a large amount of training data is needed to carry out recognition training of weld defect characteristics on the recognition model, the problems of large data set acquisition difficulty, high development cost and long period exist, and the recognition model for the deep learning detection needs to be retrained after a scene is replaced, cannot be reused and has no universality.
The above-mentioned manual detection, acoustic flaw detection, 2D image detection, and deep learning detection methods have problems, respectively, which makes it difficult to consider efficiency, cost, and effectiveness in application. In order to solve this problem,
the prior art also adopts a 3D image detection method for detection. In the 3D image detection method, a linear array structure light sensor or an area array structure light sensor is used for scanning a welding line area of a new energy battery pole, point cloud data of the welding line area and the surrounding area of the welding line area are obtained, and the point cloud data are analyzed to obtain welding line defect information. For example, in the 'new energy cylindrical battery top sealing weld 3D defect detection method and system' of the application number 202210953067.1, the area array structure light sensor is used for acquiring the battery top point cloud data, and the weld defect is detected through an algorithm, so that whether the weld defect exists in the battery pole can be accurately judged. However, the type information, the position and the geometric information of the weld defects cannot be obtained by the method, so that the method still has great limitation in application.
Disclosure of Invention
The invention provides a new energy battery pole welding seam defect detection method, equipment and a storage medium, which are used for solving the problem that the welding seam defect type, the position and the geometric information cannot be obtained in the 3D image detection method in the prior art.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
the new energy battery pole welding line defect detection method comprises the following steps:
step 1, acquiring point cloud data of a measured surface of a battery module and converting the point cloud data into a binary image;
step 2, determining a pole pixel area screening threshold range, counting all connected areas of the binary image obtained in the step 1, reserving a plurality of connected areas with pixel areas in the pole pixel area screening threshold range, and performing external ellipse fitting on each connected area to obtain corresponding fitted ellipses;
calculating the elliptical confidence of each fitting ellipse, sorting the fitting ellipses from large to small according to the elliptical confidence, and reserving the first X fitting ellipses in the sorting, wherein X is the number of poles of a known battery module, each reserved fitting ellipse corresponds to each battery pole of a measured surface of the battery module, and the average semi-major axis and the average semi-minor axis of each reserved fitting ellipse are obtained;
determining the pixel width W and the pixel height H of each battery pole theoretical welding line area and the maximum welding offset pixel distance O of each battery pole welding line in the X direction 1 Maximum welding offset pixel distance O of welding seam in Y direction 2 The method comprises the steps of carrying out a first treatment on the surface of the Taking the center of each reserved fitting ellipse as the center, and taking the pixel width W+O of the corresponding battery post in the binary image 1 The pixel height is H+O 2 As the actual weld area of each battery post obtained by positioning in the binary image;
step 3, a mask operation method is adopted, a mask M is firstly constructed for the welding line area of the battery pole, which is determined in the step 2, the number of lines of the mask M is equal to the pixel height H of the theoretical welding line area of the battery pole in the binary image, the number of columns of the mask M is equal to the pixel width W of the theoretical welding line area of the battery pole in the binary image, and the value of each line of the mask M is 0;
drawing a concentric elliptical ring formed by two ellipses by taking the center of the mask M as the center of a circle, wherein the elliptical major axis of the inner ring is 2 times of the average semi-major axis, the elliptical minor axis of the inner ring is 2 times of the average semi-minor axis, the elliptical major axis of the outer ring is equal to the pixel width W of the theoretical welding line area of the corresponding battery pole, the elliptical minor axis of the outer ring is equal to the pixel height H of the theoretical welding line area of the corresponding battery pole, and the value in the concentric elliptical ring in the mask M is 1;
taking absolute values of depth information of each point cloud data of an actual welding line region of each battery pole in the binary image, and filling the absolute values to H+O row by row 2 Row, W+O 1 In the empty matrix of the columns, then carrying out convolution calculation with the corresponding mask to obtain a convolution result;
calculating Euclidean distance between the center coordinates of the reserved fitting ellipse corresponding to each battery pole and the row coordinates of the maximum value in the corresponding convolution result, comparing the calculated Euclidean distance with a set bias welding judgment threshold, and judging that bias welding exists in the welding seam of the corresponding battery pole if the calculated Euclidean distance is larger than the set bias welding judgment threshold, wherein the obtained Euclidean distance is the welding seam offset distance.
In the further step 1, firstly, screening is performed based on the depth of each point cloud data, invalid point cloud data in the point cloud data are removed, effective point cloud data are reserved, then trend removal processing is performed on the effective point cloud data to obtain trend-removed processed point cloud data, and finally the trend-removed processed point cloud data are converted into binary images.
In the further step 2, a pole pixel area screening threshold range is determined according to the design area of a single battery pole on the tested surface of the battery module.
In a further step 2, determining the pixel width W and the pixel height H of the theoretical weld joint area of each battery pole according to the average semi-major axis, the average semi-minor axis, the ideal weld joint width, the X-direction pixel scale and the Y-direction pixel scale; determining the maximum welding offset pixel distance O of each battery pole welding seam in the X direction according to the maximum deviation empirical values of the welding seam in the X direction and the Y direction, the X-direction pixel scale and the Y-direction pixel scale 1 Maximum welding offset pixel distance O of welding seam in Y direction 2 The pixel width of the corresponding battery pole in the binary image is W+O 1 The pixel height is H+O 2 As the actual weld area for each cell post.
Further, step 3 further includes: for each battery pole, if the calculated Euclidean distance is smaller than a set partial welding judgment threshold value, taking a row coordinate of the maximum value in a convolution result as a center, and taking a region with the width equal to the pixel width W of a theoretical welding line region of the corresponding battery pole and the height equal to the pixel height H of the theoretical welding line region of the corresponding battery pole as an accurate welding line region in the binary image;
filling depth information of each point cloud data of each battery pole accurate welding line area row by row to a matrix N of H rows and W columns, performing point multiplication calculation with a corresponding mask M, and combining a point multiplication calculation result with a set explosion point depth judgment threshold K 2 Comparing, and then storing the comparison result in a binary image I' of the H rows and the W columns;
traversing the point cloud data of the accurate welding seam area row by row and column by column, and traversing depth information dZ in each piece of point cloud data ij And the set threshold K for determining the depth of the explosion point 2 Comparing, if depth information dZ of the point cloud data obtained by traversing ij Greater than K 2 The gray value of the ith row and j column in the binary image I' is 0, and if the depth information dZ of the point cloud data obtained by traversing ij Less than or equal to K 2 The gray value of the ith row and j column in the binary image I' is 255;
counting all connected areas in the binary image I' and calculating the pixel area s of each connected area z Pixel area s of each connected region z And the set explosion point area judgment threshold K 3 Comparing, reserving s z ≥K 3 If s is z ≥K 3 If the number of the connected areas is greater than 0, judging that the explosion points exist in the accurate welding seam area, and calculating the minimum value z of depth information in the point cloud data of each reserved connected area min With the corresponding position coordinates (x ij ,y ij ) In which the position coordinates (x ij ,y ij ) Namely the coordinates of the explosion point corresponding to the welding line of the battery pole, and the depth of the explosion point corresponding to the welding line of the battery pole is-z min 。
Further, step 3 further includes: the statistical matrix N is larger than a set weld depth judgment threshold K 4 (K) 4 Element 0) number q 1 And less than-K 4 Number of elements q 2 Counting the number Q of elements with the median value of 1 in the mask M corresponding to each battery poleThe method comprises the steps of carrying out a first treatment on the surface of the Calculating the corresponding welding seam area proportion R of each battery pole 0 ＝(q 1 +q 2 ) Q, and comparing the calculated result of the welding seam area proportion with a set missing welding judgment threshold value K 5 Comparing if R 0 ≤K 5 Judging that the corresponding battery pole is in the missing welding state, and obtaining the welding line area proportion R of the battery pole in the missing welding state 0 。
Further, step 3 further includes: calculating the proportion of each battery pole above the reference planeAnd the calculated proportion higher than the reference plane is compared with a set false soldering judgment threshold K 6 Comparing if R 1 ≤K 6 Judging the corresponding battery post as a cold joint, and obtaining the ratio R of the center coordinate of the cold joint battery post to the center coordinate higher than the reference plane 1 The method comprises the steps of carrying out a first treatment on the surface of the If R is 1 ＞K 6 And judging that the corresponding battery post welding line does not detect the explosion point as a qualified welding line.
The electronic equipment comprises a processor and a memory, wherein program instructions which can be read and operated by the processor are stored in the memory, and when the program instructions are read and operated by the processor, the steps 1-3 of the new energy battery pole welding seam defect detection method are executed.
And the storage medium is stored with program instructions, and when the program instructions are read and run, the steps 1-3 of the new energy battery pole welding line defect detection method are executed.
Compared with the prior art, the invention has the beneficial effects that:
(1) The invention can automatically identify and position the battery pole post area in the new energy battery module, then detect the defects, and obtain the type information of each weld defect, namely judge whether the partial welding, the missing welding, the false welding and the explosion point exist.
(2) The invention can output the position information and the geometric information of each type of weld defects, and can reduce the misjudgment problem caused by subjectivity;
(3) When the scene change is measured, the battery modules with different sizes and types can be detected by only modifying the setting parameters; the user can modify the judging conditions of each defect detection according to the actual scene requirement, so the invention has the advantage of strong usability.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention.
Fig. 2 is a brightness map of a battery module according to an embodiment of the invention.
Fig. 3 is a point cloud depth map obtained according to point cloud data in a first embodiment of the present invention.
Fig. 4 is a binary image obtained in the first embodiment of the present invention.
Detailed Description
In order to make the technical solution of the present invention better understood by those skilled in the art, the following detailed description will be given with reference to the accompanying drawings and examples, by which the technical means are applied to solve the technical problem, and the implementation process for achieving the corresponding technical effects can be fully understood and implemented. The embodiment of the invention and the characteristics in the embodiment can be mutually combined on the premise of no conflict, and the formed technical scheme is within the protection scope of the invention.
It will be apparent that the described embodiments are merely some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It is noted that the terms "comprises" and "comprising," and any variations thereof, in the description and claims of the present invention and in the foregoing figures, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
Example 1
As shown in fig. 1, the embodiment discloses a new energy battery pole welding seam defect detection method, which comprises the following steps:
step 1, acquiring point cloud data of a measured surface of a battery module and converting the point cloud data into a binary image, wherein the process is as follows:
(1.1) acquiring data.
And placing the new energy battery module on a displacement platform, adopting a linear array structure light sensor, fixing the linear array structure light sensor right above the battery module, and keeping the laser surface of the linear array structure light sensor vertical to the measured surface of the battery module. The linear array structure optical sensor is enabled to work, the battery module is enabled to move on the displacement platform along a unidirectional uniform speed, and point cloud data P of a measured surface of the battery module is obtained through the linear array structure optical sensor scanning module 1 (x ij ,y ij ,z ij ) And obtaining a point cloud depth map according to the point cloud data, wherein the obtained point cloud depth map of the battery module is shown in fig. 3 according to the point cloud data, and the obtained point cloud brightness map of the battery module is shown in fig. 2.
In this embodiment, the obtained point cloud data P 1 (x ij ,y ij ,z ij ) Homogenizing data for m rows and n columns, where i and j represent the position index of the row and column where the point cloud is located, x ij X coordinate, y representing the j-th column point of the i-th row ij Representing the y coordinate, z of the j-th column point of the i-th row ij Depth information indicating the i-th row and j-th column point cloud, (i=1, m, j=1, …, n). The x coordinates of the points in the same column are equal, and the y coordinates of the points in the same row are equal. And the point cloud data is according to the depth information z ij Screening to remove depth information z ij Exceeding the depth measurement range or occluded region z ij As the data of the invalid value nan, the remaining point cloud data is reserved as valid point cloud data.
(1.2), trending treatment.
Firstly, fitting a space plane equation F of a measured surface of the battery module: is provided with effective point cloud data P 1 (x ij ,x ij ,z ij ) The number of the effective point cloud data is k, and the x coordinate, the y coordinate and the depth information z of the effective point cloud data are calculated ij Mean value of [ ((L))
X coordinate, y coordinate and depth information z of k effective point cloud coordinates ij Respectively corresponding to subtracting the mean value%) Then sequentially ordering according to rows and columns to obtain a matrix A of k rows and 3 columns k3 . For matrix A k3 Singular value decomposition is carried out to obtain an orthogonal matrix V with 3 rows and 3 columns 33 Singular value decomposition and orthometric matrix V 33 Can be expressed as:
wherein U is kk Is A k3 Is a left singular matrix, a k-row and k-column orthogonal matrix, and the columns are represented by A k3 A k3 T Is comprised of feature vectors of (a). S is S k3 Is A k3 Is a singular value matrix, a k row and 3 column diagonal matrix, and diagonal elements are A k3 A k3 T Is arranged in decreasing order. V (V) 33 Is A k3 Right singular matrix of (3) row 3 column orthogonal matrix, column is represented by A k3 T A k3 Is comprised of feature vectors of (a). r is (r) 11 、r 22 、r 33 V respectively 33 Is a diagonal coefficient of (c).
According to an orthogonal matrix V 33 4 coefficients of the spatial plane equation F can be obtained, and the spatial plane equation F can be expressed as:
F:ax+by+cz+d＝0，
wherein a, b, c, d are coefficients of the equation, and the orthogonal matrix V 33 The relationship of (2) can be expressed as:
effective point cloud data P 1 (x ij ,y ij ,z ij ) X coordinate x of midpoint ij And y coordinates y ij Coordinate brought into space plane equationF, obtaining the depth value fz of the fitting plane ij Effective point cloud data P 1 (x ij ,y ij ,z ij ) Depth information z of (2) ij Depth value fz to the fitting plane ij Subtracting the points from each other to obtain trending point cloud data P 2 (x ij ,y ij ,dz ij ) Wherein dZ ij ＝z ij -fz ij 。
(1.3) obtaining a binary image.
An 8-bit single-channel image I with m rows and n columns is constructed, and a binarization threshold T is set. Traversing valid point cloud data P row by row and column by column 2 (x ij ,y ij ,dz ij ) If dz ij Setting the gray value of the ith row and j column of the image I to be 0; if dz ij T or dz is less than or equal to ij As the invalid value nan, the gray value of the ith row and j column of the image I is set to 255, thereby converting the valid point cloud data into a binary image as shown in fig. 4.
Step 2, positioning welding seam areas of each battery pole in the binary image obtained in the step 1, wherein the process is as follows:
firstly, determining a pole pixel area screening threshold T based on a design area S of a battery pole on a tested surface of a battery module 1 And T is 2 . Specifically, after S is converted into a pixel area s=s/(dx×dy), the pixel area is selected as a threshold T 1 ＝s-Δ，T 2 =s+Δ. Wherein dx is the known X-direction pixel dimension, dy is the known Y-direction pixel dimension, and the unit is mm/pixel; delta is a pixel deviation value, determined from empirical values.
Counting all connected areas of the binary image obtained in the step 1, and reserving pixel areas s p A plurality of connected regions within the polar pixel area screening threshold range, i.e. the reserved pixel area accords with T 1 ≤s p ≤T 2 Is provided.
Carrying out external ellipse fitting on each communication region to obtain corresponding fitted ellipses, and setting the center coordinates C of each fitted ellipse p (x p ,y p ) Semi-long axis A p And a half minor axis B p . Calculating the ellipse confidence K of each fitting ellipse p ，K p The calculation formula of (2) is as follows:
obtaining the confidence coefficient K of each fitting ellipse p Then, each fitting ellipse is determined according to the ellipse confidence degree K p And (3) sorting from large to small, and reserving the first X fitting ellipses in the sorting, wherein X is the number of poles of the known battery module. Each reserved fitting ellipse corresponds to each battery post of the measured surface of the battery module respectively, and the circle center C of each reserved fitting ellipse is obtained in the binary image p Coordinates of (x) p ,y p ) Semi-long axis A p And a half minor axis B p Specific values of (3).
Based on each of the remaining semimajor axes A of the fitted ellipses p And a half minor axis B p Calculating the average semi-major axis of each retention fit ellipseAnd average semi-minor axis>Then according to said average semimajor axis +.>Average semi-minor axis->And determining the pixel width W and the pixel height H of a theoretical region of the welding seam of the corresponding battery pole by combining the actual width L of the welding seam of each battery pole on the detected surface of the battery module. The specific calculation formula is as follows: w=2 x (a+l/dx), h=2 x (b+l/dy), wherein a is the average semi-major axis +.>Is given in pixels; b is the average semi-minor axis->Is given in pixels; l is the known ideal weld width in mm; dx is the known X-direction pixel dimension and dy is the known Y-direction pixel dimension in mm/pixel.
Taking the center of each reserved fitting ellipse as the center to obtain the area of the pixel width W and the pixel height H of the corresponding battery pole, positioning in the binary image to obtain the welding line area of each battery pole, and setting the data of each point cloud of the welding line area of each battery pole as P p (，x ij Point y ij Cloud, d number z ij According) is P p (x) ij Sit y ij Label, dP is z pij ()x ij ,y ij ,dz ij )。
In this embodiment, considering that partial welding may occur, it is also necessary to further locate each battery post weld region in combination with the maximum welding partial distance O of the battery post weld. Specifically, according to the maximum welding offset distance O, calculating to obtain the X-direction maximum welding offset pixel distance O of each battery post welding seam 1 And Y-direction maximum welding offset pixel distance O 2 . The specific calculation formula is as follows: o1=2xo1/dx, o2=2xo2/dy, where O1 is the known maximum displacement empirical value of the weld in the X-direction and O2 is the known maximum displacement empirical value of the weld in the Y-direction, in mm.
Then the circle center C of the fitting ellipse is reserved in each binary image p (is x p In y p Heart), the corresponding battery post in the binary image takes a pixel width of w+o 1 The pixel height is H+O 2 As the actual weld area of each cell post located in the binary image.
And step 3, processing the point cloud data of the welding line area of the battery pole by adopting a mask operation method so as to detect the welding line defect and obtain the type, position and geometric information of the welding line defect.
In this embodiment, a bias welding determination threshold K is set 1 Threshold value K for determining depth of explosion point 2 (K 2 Less than 0), a burst area judgment threshold value K 3 Weld depth determination threshold K 4 (K 4 > 0) miss-welding judgment threshold K 5 Threshold value K for determining cold joint 6 Weld defect detection is performed based on these thresholds. The thresholds are all judging indexes given by a user for different weld defects, and the thresholds are obtained based on the indexes.
The weld defect detection process is as follows:
(3.1) partial welding detection.
Firstly, respectively constructing masks M for the welding line areas of the battery pole, wherein the row number of the masks M is equal to the pixel height H of the theoretical welding line areas of the battery pole in the binary image, the column number of the masks M is equal to the pixel width W of the theoretical welding line areas of the battery pole in the binary image, and the value of each row and each column in the masks M is 0.
Drawing a concentric elliptical ring formed by two ellipses by taking the center of the mask M as the center of a circle, wherein the average semi-major axis of the elliptical major axis of the inner ring is 2 timesThe average semi-minor axis of the inner ring having an elliptic minor axis of 2 times +.>The elliptical major axis of the outer ring is equal to the pixel width W of the theoretical weld zone of the corresponding battery post, the elliptical minor axis of the outer ring is equal to the pixel height H of the theoretical weld zone of the corresponding battery post, and the value in the concentric elliptical ring in the mask M is 1.
The point cloud data P of the actual welding seam area of each battery pole in the binary image are obtained p The absolute value of the (depth) information is taken and then filled into H+O row by row 2 Row, W+O 1 And in the empty matrix of the columns, performing convolution calculation with the corresponding mask to obtain a convolution result.
Calculating the center C of the reserved fitting ellipse corresponding to each battery pole p Coordinates (x) p ,y p ) And the row-column coordinates (x pmax ,y pmax ) The Euclidean distance D between the two welding points, and then the calculated Euclidean distance D and a set partial welding judgment threshold K 1 Comparing, if the calculated Euclidean distance D is larger than the set partial welding judgmentThreshold value K 1 Judging that the welding seam defect exists in the welding seam of the corresponding battery pole, wherein the type of the welding seam defect is offset welding, and the central coordinate of the battery pole corresponding to the offset welding is the circle center C of the corresponding reserved fitting ellipse p Coordinates (x) p ,y p ) The obtained Euclidean distance D is the offset distance of the welding seam.
(3.2), burst point detection.
For each battery pole, if the calculated Euclidean distance D is smaller than the set partial welding judgment threshold K 1 Then the row and column coordinates (x) pmax ,y pmax ) Taking a region with the width equal to the pixel width W of the theoretical welding seam region of the corresponding battery pole and the height equal to the pixel height H of the theoretical welding seam region of the corresponding battery pole as an accurate welding seam region in the binary image, and setting each point cloud data of the accurate welding seam region as P '' p The point cloud data is P' p Is (x) ij ,y ij ,dz ij )。
The respective point cloud data P 'of each battery pole accurate welding line area' p Depth information dz of (2) ij Filling the matrix N of H rows and W columns row by row, performing dot multiplication calculation with the corresponding mask M, and determining the dot multiplication calculation result and the set explosion dot depth judgment threshold K 2 The comparison is then performed, and the comparison result is stored in the binary image I' of the H rows and W columns.
Traversing the point cloud data of the accurate welding seam area row by row and column by column, and traversing depth information dz in each piece of point cloud data ij And the set threshold K for determining the depth of the explosion point 2 Comparing depth information dz of the point cloud data obtained by traversing ij Greater than K 2 Setting the gray value of the ith row and j column in the binary image I' to 0, and if the depth information dZ of the point cloud data obtained by traversing ij Less than or equal to K 2 The gray value of the ith row j column in the binary image I' is set to 255.
Counting all connected areas in the binary image I' and calculating the pixel area s of each connected area z Pixel area s of each connected region z And the set explosion point area judgment threshold K 3 Comparing and protectingLeave s z ≥K 3 If s is z ≥K 3 If the number of the connected areas is greater than 0, determining that a explosion point exists in the accurate welding line area, namely, the welding line defect type corresponding to the battery pole exists is the explosion point. Calculating minimum value z of depth information in point cloud data of each reserved connected area min With the corresponding position coordinates (x ij ,y ij ) In which the position coordinates (x ij ,y ij ) Namely the coordinates of the explosion point corresponding to the welding line of the battery pole, and the depth of the explosion point corresponding to the welding line of the battery pole is-z min 。
(3.3) detecting the missing welding.
The statistical matrix N is larger than a set weld depth judgment threshold K 4 Number of elements q 1 And less than-K 4 Number of elements q 2 Counting the number Q of elements with the median value of 1 in the mask M corresponding to each battery pole; calculating the corresponding welding seam area proportion R of each battery pole 0 ＝(q 1 +q 2 ) Q, and comparing the calculated result of the welding seam area proportion with a set missing welding judgment threshold value K 5 Comparing if R 0 ≤K 5 Judging that the corresponding battery pole has weld defects, wherein the type of the weld defects is missing welding, and the central coordinate of the battery pole corresponding to missing welding is the circle center C of the corresponding reserved fitting ellipse p Coordinates (x) p ,y p ) And obtain the welding seam area proportion R of the battery pole post of the missing welding 0 。
(3.4) detecting the cold joint.
Calculating the proportion of each battery pole above the reference planeAnd the calculated proportion R higher than the reference plane 1 And the set false solder judgment threshold K 6 Comparing if R 1 ≤K 6 Judging that the corresponding battery post has weld defects, wherein the type of the weld defects is virtual welding, and the center coordinate of the battery post corresponding to the virtual welding is the center C of the corresponding reserved fitting ellipse p Coordinates (x) p ,y p ) And obtain the dummy welded battery postRatio R of the heart coordinates to above the reference plane 1 The method comprises the steps of carrying out a first treatment on the surface of the If R is 1 ＞K 6 And judging that the corresponding battery post welding line does not detect the explosion point as a qualified welding line.
Example two
The embodiment discloses an electronic device for realizing the new energy battery pole welding seam defect detection method according to the first embodiment, wherein the electronic device comprises a processor and a memory. The storage medium of the memory has stored therein program instructions that can be read and executed by the processor or by an external device. And when the program instructions are read and run by the processor, executing the steps 1-3 of the new energy battery pole welding seam defect detection method in the embodiment.
The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, and the examples described herein are merely illustrative of the preferred embodiments of the present invention and are not intended to limit the spirit and scope of the present invention. The individual technical features described in the above-described embodiments may be combined in any suitable manner without contradiction, and such combination should also be regarded as the disclosure of the present disclosure as long as it does not deviate from the idea of the present invention. The various possible combinations of the invention are not described in detail in order to avoid unnecessary repetition.
The present invention is not limited to the specific details of the above embodiments, and various modifications and improvements made by those skilled in the art to the technical solution of the present invention should fall within the protection scope of the present invention without departing from the scope of the technical concept of the present invention, and the technical content of the present invention is fully described in the claims.
Claims (9)
1. The new energy battery pole welding line defect detection method is characterized by comprising the following steps of:
step 1, acquiring point cloud data of a measured surface of a battery module and converting the point cloud data into a binary image;
step 2, determining a pole pixel area screening threshold range, counting all connected areas of the binary image obtained in the step 1, reserving a plurality of connected areas with pixel areas in the pole pixel area screening threshold range, and performing external ellipse fitting on each connected area to obtain corresponding fitted ellipses;
calculating the elliptical confidence of each fitting ellipse, sorting the fitting ellipses from large to small according to the elliptical confidence, and reserving the first X fitting ellipses in the sorting, wherein X is the number of poles of a known battery module, each reserved fitting ellipse corresponds to each battery pole of a measured surface of the battery module, and the average semi-major axis and the average semi-minor axis of each reserved fitting ellipse are obtained;
determining the pixel width W and the pixel height H of each battery pole theoretical welding line area and the maximum welding offset pixel distance O of each battery pole welding line in the X direction 1 Maximum welding offset pixel distance O of welding seam in Y direction 2 The method comprises the steps of carrying out a first treatment on the surface of the Taking the center of each reserved fitting ellipse as the center, and taking the pixel width W+O of the corresponding battery post in the binary image 1 The pixel height is H+O 2 As the actual weld area of each battery post obtained by positioning in the binary image;
step 3, a mask operation method is adopted, a mask M is firstly constructed for the welding line area of the battery pole, which is determined in the step 2, the number of lines of the mask M is equal to the pixel height H of the theoretical welding line area of the battery pole in the binary image, the number of columns of the mask M is equal to the pixel width W of the theoretical welding line area of the battery pole in the binary image, and the value of each line of the mask M is 0;
drawing a concentric elliptical ring formed by two ellipses by taking the center of the mask M as the center of a circle, wherein the elliptical major axis of the inner ring is 2 times of the average semi-major axis, the elliptical minor axis of the inner ring is 2 times of the average semi-minor axis, the elliptical major axis of the outer ring is equal to the pixel width W of the theoretical welding line area of the corresponding battery pole, the elliptical minor axis of the outer ring is equal to the pixel height H of the theoretical welding line area of the corresponding battery pole, and the value in the concentric elliptical ring in the mask M is 1;
taking absolute values of depth information of each point cloud data of an actual welding line region of each battery pole in the binary image, and filling the absolute values to H+O row by row 2 Row, W+O 1 Empty of columnsIn the matrix, then carrying out convolution calculation with the corresponding mask to obtain a convolution result;
calculating Euclidean distance between the center coordinates of the reserved fitting ellipse corresponding to each battery pole and the row coordinates of the maximum value in the corresponding convolution result, comparing the calculated Euclidean distance with a set bias welding judgment threshold, and judging that bias welding exists in the welding seam of the corresponding battery pole if the calculated Euclidean distance is larger than the set bias welding judgment threshold, wherein the obtained Euclidean distance is the welding seam offset distance.
2. The method for detecting the welding seam defect of the new energy battery pole is characterized in that in the step 1, firstly, screening is conducted based on the depth of each point cloud data, invalid point cloud data in the point cloud data are removed, effective point cloud data are reserved, then trend removal processing is conducted on the effective point cloud data to obtain trend-removed point cloud data, and finally the trend-removed point cloud data are converted into binary images.
3. The method for detecting the welding seam defect of the electrode column of the new energy battery according to claim 1, wherein in the step 2, a pixel area screening threshold range of the electrode column is determined according to the design area of the single battery electrode column on the detected surface of the battery module.
4. The method for detecting the welding seam defect of the new energy battery pole according to claim 1, wherein in the step 2, the pixel width W and the pixel height H of the theoretical welding seam area of each battery pole are determined according to the average semi-major axis, the average semi-minor axis, the ideal welding seam width, the X-direction pixel scale and the Y-direction pixel scale; determining the maximum welding offset pixel distance O of each battery pole welding seam in the X direction according to the maximum deviation empirical values of the welding seam in the X direction and the Y direction, the X-direction pixel scale and the Y-direction pixel scale 1 Maximum welding offset pixel distance O of welding seam in Y direction 2 The pixel width of the corresponding battery pole in the binary image is W+O 1 The pixel height is H+O 2 As the actual weld area for each cell post.
5. The method for detecting a weld defect of a new energy battery post according to claim 1, wherein step 3 further comprises: for each battery pole, if the calculated Euclidean distance is smaller than a set partial welding judgment threshold value, taking a row coordinate of the maximum value in a convolution result as a center, and taking a region with the width equal to the pixel width W of a theoretical welding line region of the corresponding battery pole and the height equal to the pixel height H of the theoretical welding line region of the corresponding battery pole as an accurate welding line region in the binary image;
filling depth information of each point cloud data of each battery pole accurate welding line area row by row to a matrix N of H rows and W columns, performing point multiplication calculation with a corresponding mask M, and combining a point multiplication calculation result with a set explosion point depth judgment threshold K 2 Comparing, and then storing the comparison result in a binary image I' of the H rows and the W columns;
traversing the point cloud data of the accurate welding seam area row by row and column by column, and traversing depth information dZ in each piece of point cloud data ij And the set threshold K for determining the depth of the explosion point 2 Comparing, if depth information dZ of the point cloud data obtained by traversing ij Greater than K 2 The gray value of the ith row and j column in the binary image I' is 0, and if the depth information dZ of the point cloud data obtained by traversing ij Less than or equal to K 2 The gray value of the ith row and j column in the binary image I' is 255;
counting all connected areas in the binary image I' and calculating the pixel area s of each connected area z Pixel area s of each connected region z And the set explosion point area judgment threshold K 3 Comparing, reserving s z ≥K 3 If s is z ≥K 3 If the number of the connected areas is greater than 0, judging that the explosion points exist in the accurate welding seam area, and calculating the minimum value z of depth information in the point cloud data of each reserved connected area min With the corresponding position coordinates (x ij ,y ij ) In which the position coordinates (x ij ,y ij ) Namely the coordinates of the explosion point corresponding to the welding line of the battery pole, and the depth of the explosion point corresponding to the welding line of the battery pole is-z min 。
6. The method for detecting a weld defect of a new energy battery post according to claim 5, wherein step 3 further comprises: the statistical matrix N is larger than a set weld depth judgment threshold K 4 (K) 4 Element 0) number q 1 And less than-K 4 Number of elements q 2 Counting the number Q of elements with the median value of 1 in the mask M corresponding to each battery pole; calculating the corresponding welding seam area proportion R of each battery pole 0 ＝(q 1 +q 2 ) Q, and comparing the calculated result of the welding seam area proportion with a set missing welding judgment threshold value K 5 Comparing if R 0 ≤K 5 Judging that the corresponding battery pole is in the missing welding state, and obtaining the welding line area proportion R of the battery pole in the missing welding state 0 。
7. The method for detecting a weld defect of a new energy battery post according to claim 6, wherein step 3 further comprises: calculating the proportion of each battery pole above the reference planeAnd the calculated proportion higher than the reference plane is compared with a set false soldering judgment threshold K 6 Comparing if R 1 ≤K 6 Judging the corresponding battery post as a cold joint, and obtaining the ratio R of the center coordinate of the cold joint battery post to the center coordinate higher than the reference plane 1 The method comprises the steps of carrying out a first treatment on the surface of the If R is 1 ＞K 6 And judging that the corresponding battery post welding line does not detect the explosion point as a qualified welding line.
8. The electronic device comprises a processor and a memory, wherein the memory stores program instructions which can be read and operated by the processor, and the method is characterized in that when the program instructions are read and operated by the processor, the steps 1-3 of the new energy battery pole welding defect detection method are executed.
9. A storage medium storing program instructions, wherein the program instructions, when read and executed, perform steps 1-3 of the new energy battery post weld defect detection method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310957857.1A CN117218057A (en) | 2023-08-01 | 2023-08-01 | New energy battery pole welding line defect detection method, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310957857.1A CN117218057A (en) | 2023-08-01 | 2023-08-01 | New energy battery pole welding line defect detection method, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117218057A true CN117218057A (en) | 2023-12-12 |
Family
ID=89037746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310957857.1A Pending CN117218057A (en) | 2023-08-01 | 2023-08-01 | New energy battery pole welding line defect detection method, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117218057A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117783147A (en) * | 2024-02-27 | 2024-03-29 | 宁德时代新能源科技股份有限公司 | Welding detection method and system |
-
2023
- 2023-08-01 CN CN202310957857.1A patent/CN117218057A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117783147A (en) * | 2024-02-27 | 2024-03-29 | 宁德时代新能源科技股份有限公司 | Welding detection method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111307818B (en) | Visual online detection method for laser welding spot of lithium battery tab | |
| CN105678689B (en) | High-precision map data registration relation determining method and device | |
| US20200364849A1 (en) | Method and device for automatically drawing structural cracks and precisely measuring widths thereof | |
| CN111815572B (en) | Method for detecting welding quality of lithium battery based on convolutional neural network | |
| CN111507976B (en) | Defect detection method and system based on multi-angle imaging | |
| CN117218057A (en) | New energy battery pole welding line defect detection method, equipment and storage medium | |
| CN113920107A (en) | Insulator damage detection method based on improved yolov5 algorithm | |
| CN110490936A (en) | Scaling method, device, equipment and the readable storage medium storing program for executing of vehicle camera | |
| CN115156093A (en) | Battery shell defect detection method, system and device | |
| CN112700489B (en) | Ship-based video image sea ice thickness measurement method and system based on deep learning | |
| CN113936004B (en) | Lithium battery welding quality detection method, device and system | |
| CN112414623A (en) | Method and system for detecting part air tightness leakage defect based on artificial intelligence | |
| CN110390261A (en) | Object detection method, device, computer readable storage medium and electronic equipment | |
| CN115752969A (en) | Method, system and equipment for detecting sealing performance of aluminum foil seal | |
| CN112924037A (en) | Infrared body temperature detection system and detection method based on image registration | |
| CN115984360B (en) | Method and system for calculating length of dry beach based on image processing | |
| CN116385390A (en) | Method and device for detecting rubberizing quality, electronic equipment and storage medium | |
| CN116452519A (en) | Tyre bulge flaw detection method based on multi-mode information fusion | |
| CN112750113B (en) | Glass bottle defect detection method and device based on deep learning and linear detection | |
| CN115330705A (en) | Skin paint surface defect detection method based on adaptive weighting template NCC | |
| CN107123105A (en) | Images match defect inspection method based on FAST algorithms | |
| CN116579955B (en) | New energy battery cell weld reflection point denoising and point cloud complement method and system | |
| CN110910450A (en) | Method for carrying out 3D target detection based on mixed feature perception neural network | |
| CN117455917B (en) | Establishment of false alarm library of etched lead frame and false alarm on-line judging and screening method | |
| CN115060731B (en) | Method for detecting scratch and metal exposure of negative membrane particles by variance algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |