A method, device and system for detecting cold welds of welds

Abstract

The present invention relates to a method for detecting a cold weld of a weld, a detection device and a detection system thereof, wherein the detection method comprises the following steps: S1: intercepting m edge contours at a certain distance along the extension direction of the weld from a weld depth map, wherein m≥3, the edge contours are composed of a top cover contour, a shell side wall contour and a weld contour, and the plane where the edge contours are located is perpendicular to the extension direction of the weld; S2: performing feature extraction processing on each edge contour obtained to obtain the point distance value, the first angle value, the weld fullness value and the missing area value of each edge contour, and obtain the reference point distance value, the minimum point distance value and the reference missing area value; S3: judging whether the weld has a cold weld according to the obtained reference point distance value, the minimum point distance value, the reference missing area value, the first angle value and the weld fullness value. The present invention can accurately detect whether the weld between the battery top cover and the shell side wall has a cold weld, and the detection accuracy is high.

Description

Technical Field

[0001] The present invention relates to the field of visual algorithm detection, and in particular to a method for detecting a weld joint defect, a device for detecting a weld joint defect for implementing the method, and a detection system having the device for detecting a weld joint defect. Background Art

[0002] With the rapid growth of the new energy vehicle market and the continuous expansion of energy storage demand, the application prospects of power batteries are broad. Figure 1 As shown, the power battery generally includes a shell 2 and a top cover 1 covering the shell 2. During the assembly process of the power battery, the top cover 1 and the shell side wall 2 need to be welded to form a weld 3 at the junction of the two. During the welding process, due to improper operation, the weld 3 may have a cold weld. The generation of cold welds will bring great safety hazards to the battery. In order to eliminate the safety hazards caused by cold welds, it is necessary to inspect the assembled power battery. At present, the common inspection methods are manual visual inspection or machine vision inspection. However, manual visual inspection is inefficient, and the weld size is small and cannot be clearly seen, and small defects are difficult to detect in time; compared with manual visual inspection, the existing machine vision inspection has a higher inspection efficiency, but the existing machine vision inspection mostly obtains the cross-section of the weld at one or several positions and extracts the weld width and weld height, and then compares and analyzes the extracted weld width and weld height with the preset value to determine whether the weld has a cold weld. However, this method is often prone to misjudgment and has poor detection accuracy. Summary of the invention

[0003] Based on this, the purpose of the present invention is, on the one hand, to provide a method for detecting a cold weld in a weld, which can detect whether there is a cold weld in the weld between the top cover of a power battery and the side wall of a shell, and is not prone to misjudgment and has high detection accuracy.

[0004] A method for detecting a cold weld of a weld comprises the following steps:

[0005] S1: intercept m edge contours at a certain distance from the weld depth map along the extension direction of the weld, where m≥3, and the edge contours are composed of the top cover contour, the shell side wall contour and the weld contour, and the plane where the edge contours are located is perpendicular to the extension direction of the weld;

[0006] S2: Perform feature extraction processing on each edge contour obtained to obtain the point distance value, the first angle value, the weld fullness value and the missing area value of each edge contour, and obtain the reference point distance value, the minimum point distance value and the reference missing area value;

[0007] S3: Determine whether there is a cold weld in the weld according to the obtained reference point distance value, the minimum point distance value, the reference missing area value, the first angle value and the weld fullness value.

[0008] Further, in step S2, the reference point distance value, the minimum point distance value and the reference missing area value are obtained in the following manner: one of the edge contours is taken as the reference edge contour, the point distance value of the reference edge contour is the reference point distance value, and its missing area value is the reference missing area value; the minimum value of the point distance values ​​of the n edge contours before and after the reference edge contour along the extension direction of the weld is obtained as the minimum point distance value, wherein n≥1.

[0009] Further, step S3 includes: performing at least one of the following comparative analyses:

[0010] a) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a first preset value, and whether the reference missing area value is greater than a second preset value, and if so, the weld is a wave-type cold weld;

[0011] b) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a third preset value, and if so, the weld is a groove type cold weld;

[0012] c) Counting the number of edge contours among the m edge contours whose first angle value is greater than a fourth preset value and whose weld fullness value is less than a fifth preset value, and determining whether the number is greater than or equal to a sixth preset value; if so, the weld is a flat cold weld.

[0013] Furthermore, the first preset value is 0.15-0.2 mm, and the second preset value is 0.01-0.02 mm. 2 The third preset value is 0.28-0.33 mm, the fourth preset value is 60-85°, the fifth preset value ranges from 0.1-0.15, and the sixth preset value is 3-8.

[0014] In another aspect, the present invention provides a device for detecting a weld defect, comprising:

[0015] A contour extraction unit, used to intercept m edge contours spaced at a certain distance from the weld depth map along the extension direction of the weld, wherein m≥3, the edge contours are composed of a top cover contour, a shell side wall contour and a weld contour, and the plane where the edge contours are located is perpendicular to the extension direction of the weld;

[0016] A feature extraction unit is used to perform feature extraction processing on each edge contour to obtain a point distance value, a first angle value, a weld fullness value and a missing area value of each edge contour, and obtain a reference point distance value, a minimum point distance value and a reference missing area value;

[0017] The feature analysis unit is used to determine whether there is a cold weld in the weld according to the obtained reference point distance value, the minimum point distance value, the reference missing area value, the first angle value and the weld fullness value.

[0018] Furthermore, the reference point distance value, the minimum point distance value and the reference missing area value are obtained in the following manner:

[0019] Take one of the edge contours as the reference edge contour, the point distance value of the reference edge contour is the reference point distance value, and its missing area value is the reference missing area value; obtain the minimum value of the point distance values ​​of the n edge contours before and after the reference edge contour along the extension direction of the weld, which is the minimum point distance value, where n≥1.

[0020] Furthermore, the feature analysis unit is used to perform at least one of the following comparative analyses:

[0021] a) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a first preset value, and whether the reference missing area value is greater than a second preset value, and if so, the weld is a wave-type cold weld;

[0022] b) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a third preset value, and if so, the weld is a groove type cold weld;

[0023] c) Counting the number of edge contours among the m edge contours whose first angle value is greater than a fourth preset value and whose weld fullness value is less than a fifth preset value, and determining whether the number is greater than or equal to a sixth preset value; if so, the weld is a flat cold weld.

[0024] Furthermore, the first preset value is 0.15-0.2 mm, and the second preset value is 0.01-0.02 mm. 2 The third preset value is 0.28-0.33 mm, the fourth preset value is 60-85°, the fifth preset value ranges from 0.1-0.15, and the sixth preset value is 3-8.

[0025] On the other hand, the present invention provides a detection system for weld defects, comprising an image acquisition device and any one of the above-mentioned detection devices; the image acquisition device forms a weld depth map and transmits it to the detection device, wherein the weld depth map shows welds on a top cover, a shell side wall, and a junction between the top cover and the shell side wall; the detection device captures an edge contour from the weld depth map.

[0026] Furthermore, the weld depth map is formed in the following manner: the image acquisition device scans the weld at the junction of the top cover and the shell side wall at a certain angle to obtain a weld depth map including the top cover, the shell side wall and the weld.

[0027] The method, device and system for detecting cold welds of welds of the present invention extract the features that distinguish the morphology of normal welds from the morphology of various types of defective welds: point spacing, first angle, weld fullness and missing area, and then analyze whether there is a cold weld in the weld according to the values ​​of these features. It can detect whether there is a cold weld in the weld between the battery top cover and the side wall of the shell, and it is not easy to make misjudgment and the detection accuracy is high.

[0028] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a structural schematic diagram of an existing power battery;

[0030] Figure 2 The weld depth diagram of different types of virtual welds at the junction of the power battery top cover and the shell side wall: Figure 2 (a) is a wave-type cold solder joint; Figure 2 (b) is a groove type cold solder joint; Figure 2 (c) Flat solder joint;

[0031] Figure 3 A schematic diagram of the structure of an exemplary detection system of the present invention;

[0032] Figure 4 is an exemplary weld depth map;

[0033] Figure 5 for Figure 4 Edge profile of section K (a);

[0034] Figure 6 for Figure 4 Edge profile of section K (ii);

[0035] Figure 7 An exemplary edge contour diagram of a cold solder joint;

[0036] Figure 8 is a detection flow chart of an exemplary detection device of the present invention;

[0037] Fig. 9 This is a schematic diagram of the first position and the second position of the weld depth map and the corresponding edge profile diagram of Example 1 of the present invention: Fig. 9 (a) Schematic diagram of the first position of the weld depth map; Fig. 9(b) a schematic diagram of the second position of the weld depth map; Fig. 9 (c) is the edge profile of the first position of the weld; Fig. 9 (d) is the edge profile of the second position of the weld;

[0038] Fig.10 This is a schematic diagram of the first position and the second position of the weld depth map and the corresponding edge contour map of Example 2 of the present invention: Fig.10 (a) Schematic diagram of the first position of the weld depth map; Fig.10 (b) a schematic diagram of the second position of the weld depth map; Fig.10 (c) is the edge profile of the first position of the weld; Fig.10 (d) is the edge profile of the second position of the weld;

[0039] Fig.11 Schematic diagram of the first position, the second position and the third position of the weld depth map and the corresponding edge contour diagram of Example 3 of the present invention: Fig.11 (a) Schematic diagram of the first position of the weld depth map; Fig.11 (b) a schematic diagram of the second position of the weld depth map; Fig.11 (c) A schematic diagram of the third position of the weld depth map; Fig.11 (d) is the edge profile of the first position of the weld; Fig.11 (e) is the edge profile of the second position of the weld; Fig.11 (f) is the edge profile of the weld at the third position;

[0040] Reference numerals:

[0041] 1. Top cover; 1a. Top cover profile; 2. Shell side wall; 2a. Shell side wall profile; 3. Weld; 3a. Weld profile; K. Section;

[0042] L1, first fitting line; L2, second fitting line; L3, connecting line; O, intersection point; Q1, first inflection point; Q2, second inflection point; A1, first area; A2, second area; E, extreme point;

[0043] Dist, dot distance; A lost , missing area: α1, first angle; α2, second angle;

[0044] K 11 , the cross section of the first position of the weld in Example 1; K 12 , the cross section of the second position of the weld in Example 1; K 22 , the cross section of the first position of the weld in Example 2; K 22 , the cross section of the second position of the weld in Example 2; K 31 , the cross section of the first position of the weld in Example 3; K32 , the cross section of the second position of the weld in Example 3; K 33 , the cross section of the weld at the third position in Example 3. DETAILED DESCRIPTION

[0045] It should be clear that the described embodiments are only some embodiments of the embodiments of the present application, rather than all embodiments. Based on the embodiments in the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.

[0046] The present invention has found that the existing analysis method of extracting weld width and weld height is prone to misjudgment. The reason is that there are many existing virtual weld forms, including Figure 2 (a) The wavy solder joint shown in Figure 2 (b) The groove type cold solder joint and Figure 2 (c) As shown in the figure, the existing method of extracting the width and height of a certain position of the weld cannot detect the above-mentioned types of welds, so it is easy to make misjudgments and the detection accuracy is poor.

[0047] In order to solve the above problems, the present invention compares and analyzes the morphology of normal welds with the morphology of wavy cold welds, groove cold welds and flat cold welds, extracts the features that are different in morphology between normal welds and wavy cold welds, groove cold welds and flat cold welds, namely, point spacing, missing area, first angle and weld fullness, and then judges whether there is a cold weld in the weld according to these features. Based on this, the present invention first intercepts multiple cross-sectional views at a certain distance from the weld depth map, and then extracts the above features (namely point spacing, missing area, first angle and weld fullness) from each cross-sectional view, obtains the point spacing value, missing area value, first angle value and weld fullness value in each cross-sectional view, and further obtains the reference point spacing value, the minimum point spacing value and the reference missing area value based on this, and then compares these feature values ​​with the preset values ​​to determine whether there is a cold weld in the weld. If there is a cold weld in the weld, the type of the cold weld is further determined according to the comparison results of the values ​​of these features with the preset values.

[0048] Based on the above improvement ideas, the solution of the present invention is described in detail below in conjunction with the accompanying drawings.

[0049] Figure 3 The schematic structure of the detection system of the present invention is shown. Figure 3 As shown, the detection system includes an image acquisition device and a detection device.

[0050] The image acquisition device is specifically a 3D depth camera, which is arranged near the detection station. When the image acquisition device scans the weld 3 at the junction of the top cover 1 and the shell side wall 2 at a certain angle, the following can be obtained: Figure 4 The weld depth map shown shows the top cover 1, the shell side wall 2 and the weld 3 at the junction of the top cover 1 and the shell side wall 2. The formed weld depth map is then transmitted to the detection device for further analysis.

[0051] The detection device is used to analyze whether there is a cold weld in the weld 3 by analyzing the weld depth map obtained by the image acquisition device, specifically comprising:

[0052] A contour extraction unit is used to extract m edge contours spaced at a certain distance h from the weld depth map along the extension direction of the weld 3, wherein m ≥ 3, the edge contour is composed of the top cover contour 1a, the shell side wall contour 2a and the weld contour 3a, and the plane where the edge contour is located is perpendicular K to the extension direction of the weld 3;

[0053] Here, the interval distance h=1mm. If the interval distance is small, more samples are taken and the detection time is longer; if the interval distance is large, it is easy to skip the defect area, resulting in incorrect detection results.

[0054] The feature extraction unit is used to perform feature extraction processing on each edge contour to obtain the point distance Dist, the first angle α1, the weld fullness p and the missing area A of each edge contour. lost , calculate the point distance value, first angle value, weld fullness value and missing area value of each edge contour;

[0055] Specifically, Figure 5-7 As shown, the point distance value is extracted by fitting the top cover profile 1a and the shell side wall profile 2a respectively, forming a first fitting line L1 and a second fitting line L2, and making the first fitting line L1 and the second fitting line L2 intersect at an intersection point O, and the shortest distance between the intersection point O and the weld profile 3a is the point distance Dist;

[0056] Here, the fitting method of the first fitting line L1 and the second fitting line L2 is specifically as follows: the first fitting line L1 is formed by selecting points with an angle range of 30-60° based on the top cover profile 1a, and the second fitting line L2 is formed by selecting points with an angle range of 30-60° based on the shell side wall profile 2a. After the first fitting line L1 is formed, if the distance between less than 30% of the fitting points and the first fitting line L1 is less than 0.02mm, it is considered that the fitting is abnormal, and the 10 points before and after the fitting points need to be removed to re-fit the first fitting line L1; similarly, after the second fitting line L2 is formed, if the distance between less than 30% of the fitting points and the second fitting line L2 is less than 0.02mm, it is considered that the fitting is abnormal, and the 10 points before and after the fitting points need to be removed to re-fit the second fitting line L2.

[0057] The first angle value is obtained in the following manner: the boundary point between the top cover profile 1a and the weld profile 3a is taken as the first inflection point Q1, the boundary point between the shell side wall profile 2a and the weld profile 3a is taken as the second inflection point Q2, the line between the first inflection point Q1 and the second inflection point Q2 is taken as the connecting line L3, the first angle α1 is formed between the first fitting line L1 and the connecting line L3, and the second angle α2 is formed between the second fitting line L2 and the connecting line L3;

[0058] Here, the first inflection point Q1 and the second inflection point Q2 are determined as follows: taking the point of the weld contour 3a directly below the intersection O as the starting point, if the distance between 5 consecutive points and the top cover contour 1a is less than 0.012mm, it is considered to be the first inflection point Q1; similarly, if the distance between 5 consecutive points and the shell side wall contour 2a is less than 0.012mm, it is considered to be the second inflection point Q2.

[0059] The method for extracting the weld fullness value is as follows: the area enclosed by the connecting line L3 and the weld contour 3a is taken as the first area A1, and the ratio between the first area A1 and the connecting line L3 is the weld fullness value;

[0060] The missing area value is obtained by taking the point on the weld contour 3a that is farthest from the connecting line L3 as the extreme point E, the triangular area surrounded by the extreme point E, the first inflection point Q1 and the second inflection point Q2 as the second area A2, and the area difference set between the first area A1 and the second area A2 as the missing area A lost (The second area A2 of the normal weld 3 is completely within the first area A1, so the missing area A of the normal weld 3 is lost is 0).

[0061] Take one of the edge contours as the reference edge contour, the point distance of the reference edge contour as the reference point distance, and the missing area as the reference missing area value, and calculate the reference point distance value and the reference missing area value;

[0062] The point distance values ​​of n edge contours before and after the reference edge contour along the extension direction of the weld 3 are calculated, and the minimum value among them is taken as the minimum point distance value, wherein n≥1.

[0063] A feature analysis unit is used to determine whether the weld 3 has a cold weld according to the obtained reference point distance value, minimum point distance value, reference missing area value, first angle value and weld fullness value;

[0064] Specifically, the feature analysis unit is used to perform at least one of the following comparative analyses to determine whether the weld 3 has a cold weld:

[0065] a) Determine whether the difference between the reference point distance value and the minimum point distance value is greater than the first preset value a, and determine whether the reference missing area value is greater than the second preset value b, wherein the first preset value a is 0.15-0.2mm and the second preset value b is 0.01-0.02mm 2 ;

[0066] Specifically, if the difference between the reference point distance value and the minimum point distance value is greater than the first preset value a, and the reference missing area value is greater than the second preset value b, it is determined that there is a cold weld in the weld 3, and the cold weld is a wave-type cold weld, and the detection ends;

[0067] If the difference between the reference point distance value and the minimum point distance value is less than or equal to the first preset value a, and / or the reference missing area value is less than or equal to the second preset value b, then comparative analysis b) is performed.

[0068] b) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a third preset value c, wherein the third preset value c is 0.28-0.33 mm;

[0069] Specifically, if the difference between the reference point distance value and the minimum point distance value is greater than the third preset value c, it is determined that there is a cold weld in the weld 3, and the cold weld is a groove type cold weld, and the detection ends;

[0070] If the difference between the reference point distance value and the minimum point distance value is less than or equal to the third preset value c, then comparative analysis c) is performed.

[0071] c) counting the number of edge contours whose first angle value is greater than a fourth preset value d and whose weld fullness value is less than a fifth preset value e among the m edge contours, and determining whether the number is greater than or equal to a sixth preset value f, wherein the fourth preset value d is 60-85°, the fifth preset value e is 0.1-0.15, and the sixth preset value f is 3-8;

[0072] Specifically, if the number of edge contours whose first angle value is greater than the fourth preset value d and whose weld fullness value is less than the fifth preset value e among the m edge contours is greater than or equal to the sixth preset value f, it is determined that there is a cold weld in the weld 3, and the cold weld is a flat cold weld, and the detection ends;

[0073] If the number of edge contours whose first angle value is greater than the fourth preset value d and whose weld fullness value is less than the fifth preset value e among the m edge contours is less than the sixth preset value f, it is determined that there is no cold weld in the weld 3 and the detection ends.

[0074] That is to say, if any of the above comparative analysis a), b) and c) is met, it can be determined that weld 3 has a cold weld, and further:

[0075] If it meets the comparison analysis a), the type of cold solder joint is determined to be a wave-type cold solder joint;

[0076] If it meets the comparison analysis b), the type of cold solder joint is determined to be a groove cold solder joint;

[0077] If it meets the comparison analysis c), the type of cold solder joint is determined to be a flat cold solder joint.

[0078] When the above comparative analysis a), b) and c) do not meet the requirements, it can be determined that there is no cold weld in the weld 3.

[0079] The workflow of the above detection system is as follows Figure 8 shown.

[0080] Here, the execution order of comparison analysis a), b) and c) is not fixed:

[0081] Comparative analysis a) can be performed first. If it does not meet the requirements of comparative analysis a), it means that weld 3 does not have a wavy cold weld, and then comparative analysis b) can be performed; if it does not meet the requirements of comparative analysis b), it means that weld 3 does not have a grooved weld, and then comparative analysis c) can be performed. If it does not meet the requirements of comparative analysis c), it means that weld 3 does not have a flat cold weld, and then it can be determined that weld 3 does not have a cold weld.

[0082] Alternatively, comparative analysis b) may be performed first, and if comparative analysis b) does not meet the requirements, then proceed to c); if comparative analysis c) does not meet the requirements, then proceed to comparative analysis a); or comparative analysis c) may be performed first, and if comparative analysis c) does not meet the requirements, then proceed to comparative analysis a); if comparative analysis a) does not meet the requirements, then proceed to comparative analysis b).

[0083] In any of the above execution sequences, as long as any one of the comparison analyses a), b) and c) is met, it can be determined that the weld 3 has a cold weld, and the detection ends.

[0084] It should be noted that the same position of the same weld 3 may meet the conditions of different types of cold welds at the same time. For example, if there is a wavy cold weld at a certain position of the weld 3, and if the wavy cold weld is seriously concave, it will also meet the conditions of a groove cold weld. That is, the structural characteristics of the same position of the same weld 3 may meet both comparative analysis a) and comparative analysis b). Therefore, in practical applications, after the first comparative analysis confirms that there is a cold weld in the weld 3, the detection may not be terminated immediately, but the remaining comparative analysis may be performed to further confirm the types of all cold welds in the same weld. Figure 8 Taking the inspection process shown as an example, when the comparison analysis a) is met, it can be determined that there is a wavy cold weld in the weld 3; at this time, the comparison analysis b) can be continued. If it is met, it can be determined that there is a groove-type cold weld in the weld 3; at this time, the comparison analysis c) can be continued. If it also meets the comparison analysis b), it can be determined that there is a flat cold weld in the weld 3.

[0085] Since the weld 3 is long, different types of cold welds may exist at different positions of the same weld 3. Therefore, in practical applications, after confirming that a certain section of the weld 3 has a cold weld through the above process, the above detection process can be repeated to detect other positions of the weld 3 as needed. Since the detection process has been described in detail above, it will not be repeated here.

[0086] It should also be noted that, since power batteries have different specifications, and the sizes of normal welds of power batteries of different specifications are also different, the numerical ranges of the first preset value a, the second preset value b, the third preset value c, the fourth preset value d, the fifth preset value e, and the sixth preset value f need to be adjusted according to the specifications of the power battery and actual needs in actual application. The numerical ranges of the first preset value a, the second preset value b, the third preset value c, the fourth preset value d, the fifth preset value e, and the sixth preset value f are empirical values ​​obtained by the present invention through research and long-term practice. If the set threshold range is exceeded, the weld defects will be more obvious, and thus can be used to distinguish defects of different levels.

[0087] The following further specifically describes the detection method implemented by the above detection system through three implementation regulations for detecting wavy cold solder joints, groove cold solder joints and flat cold solder joints respectively.

[0088] Example 1

[0089] As Fig. 9 Taking the wave-type cold solder joint shown in (a)-9(d) as an example, the first preset value a is set to 0.18 mm and the second preset value b is set to 0.015 mm 2 ;

[0090] The above detection device obtains Fig. 9 (a) The first section K 11 The missing area A of the edge contour lost The value is 0.024mm 2 , the point distance value is 0.456mm; Fig. 9 (b) The second section K 12 The first section K 11 The minimum point distance value among the 6 sections before and after is 0.230mm;

[0091] First section K 11 Missing area A lost The value is 0.024mm 2 , which is greater than the second preset value b, and the first profile K 11 With the first section K 12 The difference in the point distance Dist is 0.226 mm, which is greater than the first preset value a, and it is determined that there is a cold weld in the weld 3, and the cold weld is specifically a wavy cold weld.

[0092] Example 2

[0093] As Fig.10 Taking the groove type cold solder joint shown in (a)-10(d) as an example, the third preset value c is set to 0.3 mm;

[0094] The above detection device obtains Fig.10 (a) The first section K 21 The point distance value of the edge contour is 0.740mm; Fig.10 (b) The second section K 22 The first section K 21 The minimum point distance value of the six sections before and after is 0.386mm. 21 With the second section K 22 The point distance difference is 0.354 mm, which is greater than the third preset value c, then it is determined that there is a cold weld in the weld 3, and the cold weld is specifically a groove type cold weld.

[0095] Example 3

[0096] As Fig.11 Taking the groove type cold solder joint shown in (a)-11(f) as an example, the fourth preset value d is set to 60°, the fifth preset value e is set to 0.1, and the sixth preset value f is set to 3;

[0097] The above detection device obtains Fig.11 (a) The first section K 31 The first bending angle α1 of the edge profile is 65.924°, and the weld fullness value is 0.020; Fig.11 (b) The second section K 32 The first bending angle α1 is 65.525°, and the weld fullness value is 0.028; Fig.11 (c) The third section K 33 The first bending angle α1 is 63.541°, and its weld fullness value is 0.029, both of which satisfy that the first bending angle α1 is greater than the fourth preset value d and the fullness value is less than the fifth preset value e. The cumulative number is 3, reaching the sixth preset value f, then it is determined that there is a cold weld in the weld 3, and the cold weld is specifically a flat cold weld.

[0098] Based on the same inventive concept, the present application also provides an electronic device, which may be a terminal device such as a server, a desktop computing device or a mobile computing device (for example, a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). The device includes one or more processors and a memory, wherein the processor is used to execute a program to implement a weld-based cold weld detection method; and the memory is used to store a computer program executable by the processor.

[0099] Based on the same inventive concept, the present application also provides a computer-readable storage medium, corresponding to the aforementioned embodiment of the weld-based cold weld detection method, and the computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the entity relationship extraction method recorded in any of the above embodiments.

[0100] The present application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0101] Compared with the prior art, the weld cold weld detection method, detection device and detection system of the present invention extract the features that distinguish the morphology of normal welds from the morphology of various types of defective welds: point spacing, first angle, weld fullness and missing area, calculate the values ​​of these features, and then compare the values ​​of these features with preset values ​​to determine whether there is a cold weld in the weld. It can accurately detect whether there is a cold weld in the weld between the battery top cover and the side wall of the shell; if there is a cold weld in the weld, the type of the cold weld is further determined based on the comparison results of the values ​​of these features with the preset values.

[0102] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, and the present invention is also intended to include these modifications and modifications.

Claims

1. A method for detecting a weld joint, used to detect the weld joint between a power battery top cover and a shell side wall, characterized in that: The following steps are involved: S1: intercept m edge contours at a certain distance from the weld depth map along the extension direction of the weld, where m≥3, and the edge contours are composed of the top cover contour, the shell side wall contour and the weld contour, and the plane where the edge contours are located is perpendicular to the extension direction of the weld; S2: Perform feature extraction processing on each edge contour obtained to obtain the point distance value, the first angle value, the weld fullness value and the missing area value of each edge contour, and obtain the reference point distance value, the minimum point distance value and the reference missing area value; The point distance value is extracted by fitting the top cover profile and the shell side wall profile respectively to form a first fitting line and a second fitting line, and the first fitting line and the second fitting line intersect at an intersection point, and the shortest distance between the intersection point and the weld profile is the point distance; The first angle value is obtained by taking the dividing point between the top cover contour and the weld contour as the first inflection point, the dividing point between the shell side wall contour and the weld contour as the second inflection point, the line between the first inflection point and the second inflection point as the connecting line, and the first fitting line and the connecting line form a first angle; The method for extracting the weld fullness value is: taking the area enclosed by the connecting line and the weld contour as the first area, and the ratio between the first area and the connecting line as the weld fullness value; The missing area value is obtained by taking the point on the weld contour that is farthest from the connecting line as the extreme point, the area surrounded by the extreme point, the first inflection point and the second inflection point as the second area, and the area difference set between the first area and the second area as the missing area; The method for obtaining the reference point distance value, the minimum point distance value and the reference missing area value is as follows: Take one of the edge contours as the reference edge contour, the point distance value of the reference edge contour is the reference point distance value, and the missing area value thereof is the reference missing area value; Obtaining the minimum value of the midpoint distances of n edge contours before and after the reference edge contour along the extension direction of the weld as the minimum point distance value, wherein n≥1; S3: Determine whether there is a cold weld in the weld according to the obtained reference point distance value, the minimum point distance value, the reference missing area value, the first angle value and the weld fullness value.

2. The method for detecting a cold weld of a weld according to claim 1, characterized in that: Step S3 includes: performing at least one of the following comparative analyses: a) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a first preset value, and whether the reference missing area value is greater than a second preset value, and if so, the weld is a wave-type cold weld; b) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a third preset value, and if so, the weld is a groove type cold weld; c) Counting the number of edge contours among the m edge contours whose first angle value is greater than a fourth preset value and whose weld fullness value is less than a fifth preset value, and determining whether the number is greater than or equal to a sixth preset value; if so, the weld is a flat cold weld.

3. The method for detecting a cold weld of a weld according to claim 2, characterized in that: The first preset value is 0.15-0.2 mm, and the second preset value is 0.01-0.02 mm 2 The third preset value is 0.28-0.33 mm, the fourth preset value is 60-85°, the fifth preset value ranges from 0.1-0.15, and the sixth preset value is 3-8.

4. A device for detecting a weld defect, characterized in that: include: A contour extraction unit, used to intercept m edge contours spaced at a certain distance from the weld depth map along the extension direction of the weld, wherein m≥3, the edge contours are composed of a top cover contour, a shell side wall contour and a weld contour, and the plane where the edge contours are located is perpendicular to the extension direction of the weld; A feature extraction unit is used to perform feature extraction processing on each edge contour to obtain a point distance value, a first angle value, a weld fullness value and a missing area value of each edge contour, and obtain a reference point distance value, a minimum point distance value and a reference missing area value; The point distance value is extracted by fitting the top cover profile and the shell side wall profile respectively to form a first fitting line and a second fitting line, and the first fitting line and the second fitting line intersect at an intersection point, and the shortest distance between the intersection point and the weld profile is the point distance; The first angle value is obtained by taking the dividing point between the top cover contour and the weld contour as the first inflection point, the dividing point between the shell side wall contour and the weld contour as the second inflection point, the line between the first inflection point and the second inflection point as the connecting line, and the first fitting line and the connecting line form a first angle; The method for extracting the weld fullness value is: taking the area enclosed by the connecting line and the weld contour as the first area, and the ratio between the first area and the connecting line as the weld fullness value; The missing area value is obtained by taking the point on the weld contour that is farthest from the connecting line as the extreme point, the area surrounded by the extreme point, the first inflection point and the second inflection point as the second area, and the area difference set between the first area and the second area as the missing area; The reference point distance value, the minimum point distance value and the reference missing area value are obtained in the following manner: Take one of the edge contours as a reference edge contour, the point distance value of the reference edge contour is the reference point distance value, and the missing area value thereof is the reference missing area value; Obtaining the minimum value of the midpoint distances of n edge contours before and after the reference edge contour along the extension direction of the weld as the minimum point distance value, wherein n≥1; The feature analysis unit is used to determine whether there is a cold weld in the weld according to the obtained reference point distance value, the minimum point distance value, the reference missing area value, the first angle value and the weld fullness value.

5. The device for detecting a poor weld according to claim 4, characterized in that: The feature analysis unit is used to perform at least one of the following comparative analyses: a) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a first preset value, and whether the reference missing area value is greater than a second preset value, and if so, the weld is a wave-type cold weld; b) determining whether the difference between the reference point distance value and the minimum point distance value is greater than a third preset value, and if so, the weld is a groove type cold weld; c) Counting the number of edge contours among the m edge contours whose first angle value is greater than a fourth preset value and whose weld fullness value is less than a fifth preset value, and determining whether the number is greater than or equal to a sixth preset value; if so, the weld is a flat cold weld.

6. The device for detecting a poor weld according to claim 5, characterized in that: The first preset value is 0.15-0.2 mm, and the second preset value is 0.01-0.02 mm 2 The third preset value is 0.28-0.33 mm, the fourth preset value is 60-85°, the fifth preset value ranges from 0.1-0.15, and the sixth preset value is 3-8.

7. A detection system for weld defects, characterized in that: It comprises an image acquisition device and a detection device as claimed in any one of claims 4 to 6; The image acquisition device forms a weld depth map and transmits it to the detection device, wherein the weld depth map shows welds on the top cover, the side wall of the shell, and the junction between the top cover and the side wall of the shell; The detection device captures an edge profile from the weld depth map.

8. The detection system for weld defects according to claim 7, characterized in that: The weld depth map is formed in the following manner: the image acquisition device scans the weld at the junction of the top cover and the shell side wall at a certain angle to obtain a weld depth map including the top cover, the shell side wall and the weld.

Images