Spot welding method and system for indicative fuse production

Abstract

The invention relates to the technical field of resistance welding, and provides a spot welding method and system for indicative fuse production, and the method comprises the steps: carrying out the test welding of a fuse, and obtaining a welding spot image; carrying out edge detection on the welding spot image, carrying out welding spot outer contour roundness and smoothness analysis and welding spot internal gradient distribution analysis, and determining the welding spot outer contour quality and the welding spot surface defect degree of test welding; the welding spot outer contour quality and the welding spot surface defect degree of test welding are judged, so that electric welding parameters are adjusted; constructing a welding quality regression model through multiple times of test welding and adjustment of electric welding parameters; and for test welding, electric welding parameters are continuously adjusted through the welding quality regression model, and spot welding of follow-up fuse production is carried out. The invention aims to optimize the process of improving the spot welding effect by adjusting the parameter control of the electric welding machine through multiple times of test welding.

Description

Technical Field

[0001] The invention relates to the technical field of resistance welding, and in particular to a spot welding method and system for producing indicating fuses. Background Art

[0002] As modern electrical equipment becomes increasingly complex and sophisticated, the importance of fuses as circuit protection components becomes increasingly prominent; and indicating fuses, because they can intuitively display the melting state, have been widely used and received attention. However, in the production process of indicating fuses, the quality of the spot welding process directly affects the stability and reliability of their performance; when welding indicating fuses, traditional spot welding methods often face problems such as inaccurate parameter control and poor welding consistency; improper parameter settings, such as too high or too low current, time or pressure, will lead to defects such as insufficient welding strength, uneven welding, welding spatter and welding cracks, which will in turn affect the electrical and mechanical properties of the fuse; in addition, problems such as overheating of materials and too fast cooling during welding will also lead to a decline in fuse performance, and even serious problems such as welding burn-through and excessive welding residues.

[0003] Before formal production, it is necessary to select multiple fuse materials as test welding samples, and judge the spot welding effect and quality by observing the appearance of the solder joints, so as to fine-tune the currently set welding machine parameters; however, the welding machine parameters are adjusted by conventional manual observation of the solder joint appearance, and this process is repeated until the expected effect is achieved. The adjustment process is too subjective, and the change in the appearance of the solder joint after each adjustment cannot be measured by a fixed standard; it is necessary to establish a relevant model for the solder joint appearance and perform fitting regression analysis until the regression converges as a judgment of the expected effect or the best effect. Summary of the invention

[0004] The present invention provides a spot welding method and system for the production of indicating fuses to solve the problem that the spot welding improvement effect is inaccurate due to manual adjustment of the electric welding machine during multiple test welding. The technical solution adopted is as follows: The present invention proposes a spot welding method for producing an indicating fuse, the method comprising the following steps: Test welding of fuse to obtain welding spot image; Through edge detection of the solder joint image, the roundness and smoothness of the solder joint outer contour and the gradient distribution analysis inside the solder joint are analyzed to determine the outer contour quality of the solder joint and the surface defect degree of the solder joint; Determine the outer contour quality and surface defect degree of the weld spot during the test welding, and adjust the welding parameters accordingly; build a welding quality regression model through multiple test welding and adjustment of welding parameters; The welding parameters are continuously adjusted for the trial welding through the welding quality regression model.

[0005] Optionally, the welding spot image is subjected to edge detection to perform roundness and smoothness analysis of the outer contour of the welding spot, as well as gradient distribution analysis inside the welding spot, to determine the outer contour quality of the welding spot of the test welding and the surface defect degree of the welding spot, including the specific method of: The outer contour edge and several internal edge pixel points are obtained by edge detection of the welding spot image; Analyze the roundness and corner point distribution of the outer contour edge to determine the quality of the outer contour of the weld; The gradient comparison and edge length analysis on different edges of the internal edge pixels are performed to determine the surface defect degree of the solder joint.

[0006] Optionally, the outer contour edge and several inner edge pixel points are obtained by: Obtain several edge pixel points from the welding point image by edge detection; Connect adjacent edge pixels to obtain several edges; The length of each edge is obtained, and the edge with the longest length among all closed edges is used as the outer contour edge of the solder joint image; and several edge pixel points inside the outer contour edge are used as the internal edge pixel points of the solder joint image.

[0007] Optionally, the analysis of the roundness and corner point distribution of the outer contour edge to determine the outer contour quality of the solder joint includes the following specific methods: Get the roundness of the outer contour edge; Obtain several corner points of the outer contour edge by corner point detection; obtain the ratio of the number of corner points to the number of pixel points on the outer contour edge, and subtract the difference obtained by subtracting the ratio from 1 as the contour smoothness of the outer contour edge; The average of the roundness of the outer contour edge and the contour smoothness is taken as the outer contour quality of the weld spot of the test welding.

[0008] Optionally, the obtaining of the roundness of the outer contour edge includes the following specific methods: The area of ​​the enclosed area of ​​the outer contour edge is obtained, and the roundness of the outer contour edge is calculated as follows: in, Indicates the roundness of the outer contour edge. Indicates the length of the outer contour edge, Indicates the area of ​​the region enclosed by the outer contour edge.

[0009] Optionally, the gradient comparison and edge length analysis on different edges of the internal edge pixels to determine the surface defect of the solder joint includes the following specific methods: Obtain a number of edges formed by internal edge pixels, which are recorded as internal edges of the welding point; Obtain the length of the inner edge of each solder joint and the gradient mean of all inner edge pixel points on any inner edge of the solder joint as the overall gradient of the inner edge of the solder joint; The surface defect degree of the solder joint is determined by the overall gradient and length of the internal edge of the solder joint.

[0010] Optionally, the method of determining the surface defect degree of the solder joint by the overall gradient and length of the internal edge of the solder joint includes the following specific methods: in, Indicates the surface defect of the solder joint. Indicates the number of internal edges of the solder joint, Indicates The length of the inner edge of the strip weld, Indicates The overall gradient of the inner edge of the strip weld, represents the overall gradient mean of the internal edges of all solder joints, represents the weight normalization function.

[0011] Optionally, the welding quality regression model is constructed by multiple welding trials and adjusting welding parameters, and the specific method includes: After the adjusted welding parameters are obtained, the fuse is re-tested, and a welding spot image of the new test welding is obtained, the outer contour quality of the welding spot and the surface defect degree of the welding spot of the test welding are obtained, and the welding parameters are adjusted to obtain the second adjusted welding parameters; Similarly, several test welds are conducted to obtain the weld point outer contour quality and weld point surface defect degree of the test welds, and various welding parameters of each test weld are recorded, including welding current, welding time, electrode pressure, preheating temperature and cooling speed; Obtaining the standard deviation of the welding parameters corresponding to any welding parameter under multiple welding trials, weight normalizing the standard deviations of all welding parameters, and using the obtained results as the regression weights of each welding parameter; The multiple welding parameters of any welding trial are input into the model. The preset weight of each welding parameter in the model is its corresponding regression weight. At the same time, the weld outer contour quality and weld surface defect degree of the welding trial are input into the model. The output feature is the adjustment action of each welding parameter, including the parameter value and adjustment direction of each adjustment. The model is trained to build a welding quality regression model.

[0012] Optionally, the welding test parameters are continuously adjusted by a welding quality regression model, including the specific method of: After the welding quality regression model outputs the adjustment action, a trial welding is performed according to the adjusted electric welding parameters, and the outer contour quality of the weld point and the surface defect degree of the weld point of the current trial welding are obtained; Based on the welding quality regression model, combined with the welding parameters of the current trial welding, the outer contour quality of the weld spot and the surface defect degree of the weld spot, the adjustment action continues to be output, and the welding parameters are adjusted continuously by analogy until the outer contour quality and the surface defect degree of the weld spot of a trial welding meet the standards. The corresponding welding parameters are used as the welding parameters for spot welding in the fuse production process.

[0013] The present invention also proposes a spot welding system for producing indicative fuses, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the above method when executing the computer program.

[0014] The beneficial effects of the present invention are as follows: the present invention quantifies the welding quality of the test welding, and constructs a welding quality regression model in combination with the electric welding parameters, thereby outputting the adjustment action of the electric welding parameters; wherein the edge detection is performed on the weld spot image of the test welding, and the appearance quality of the weld spot is analyzed by using a machine vision method, and the outer contour quality of the weld spot is determined by analyzing whether the outer contour of the weld spot fits the circle and is smooth, and the gradient difference and distribution inside the weld spot are analyzed to quantify whether there are defects such as cracks, thereby reflecting the surface quality of the weld spot; by judging the outer contour quality of the weld spot and the surface defect degree of the weld spot under multiple test weldings, adjusting the electric welding parameters and constructing a welding quality regression model, the welding quality regression model can output the adjusted electric welding parameters, and through continuous test welding, provide a basis for subsequent spot welding quality to meet the standards; thereby the spot welding quality is rapidly improved while continuously meeting the standards, and finally adjusted to suitable electric welding parameters to ensure the spot welding effect of indicative fuse production. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 A schematic flow chart of a spot welding method for producing an indicative fuse provided by an embodiment of the present invention; Figure 2 is the solder joint image; Figure 3 is the outer contour edge image; Figure 4 is the inner contour edge image. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 , which shows a flow chart of a spot welding method for producing an indicative fuse provided by an embodiment of the present invention, the method comprising the following steps: Step S001, test welding a fuse to obtain a welding point image.

[0019] The purpose of this embodiment is to quantify the welding quality of the weld spot during the trial welding process through machine vision, and to build a model to quantify the adjustment action of the weld parameters by adjusting the weld parameters corresponding to the trial welding, so as to improve the accuracy of the spot welding quality during the trial welding. First, it is necessary to prepare the welder and perform welding, and collect weld spot images at the same time.

[0020] Specifically, the material and fusing capacity of the indicative fuse to be welded are determined. In this embodiment, 300A current and 2.5 to 3 seconds of fusing time are expected specifications; initial welding parameters are set based on the expected specifications of the indicative fuse, including welding current, welding time, electrode pressure, preheating temperature and post-heating time; the various welding parameters are not specifically limited in this embodiment, and are set based on experience during the initial trial welding; it should be noted that since the subsequent process is to adjust the welding parameters to improve the spot welding quality, the welding parameters can be set based on experience during the initial trial welding.

[0021] Furthermore, the welding position of the fuse and the terminal is determined, and at the same time, the welding point is ensured to be within the specified tolerance range. The welding position is directly facing the industrial camera, and a small industrial camera is fixedly installed on the electric welder to collect images of the welding point; Figure 2 As shown, after the industrial camera is installed, it is necessary to ensure that the soldering point is located in the center of the lens. Figure 2 The central circle portion is assumed to be the soldering point area. While ensuring the integrity of the soldering point area, the zoom factor is used to keep a certain distance from the edge. In this embodiment, the distance is set to 10% of the short side length of the image.

[0022] Furthermore, one end of the fuse is aligned with the terminal and placed between the welding heads of the electric welder; the electric welder is started, and when the welding head reaches the welding temperature, pressure is applied to make the fuse and the terminal contact until the welding is completed; the set welding time and electrode pressure are maintained to ensure a firm connection between the fuse and the terminal; after the welding is completed, an image of the solder joint is collected by an industrial camera, and an image of the solder joint of the test welding is obtained by grayscale processing.

[0023] Step S002: Perform edge detection on the solder joint image to analyze the roundness and smoothness of the solder joint outer contour and the gradient distribution inside the solder joint to determine the quality of the solder joint outer contour and the surface defect degree of the solder joint.

[0024] It should be noted that an ideal solder joint should be round or oval with smooth edges, without obvious sharp corners or notches, and the solder joint surface should be smooth and the color should be evenly distributed. On the contrary, if the solder joint surface is rough and cracks or irregular shapes appear, it is usually caused by unstable parameters during the spot welding process and excessive temperature. In this case, it is necessary to perform an outer contour quality analysis on the roundness and smoothness of the outer contour of the solder joint in the solder joint image, and at the same time, analyze the gradient changes and distribution inside the solder joint in the solder joint image, so as to quantify whether the solder joint surface is smooth by smoothing the internal gradient.

[0025] Preferably, in one embodiment of the present invention, the step includes: The outer contour edge and several internal edge pixel points are obtained by edge detection of the welding spot image; Analyze the roundness and corner point distribution of the outer contour edge to determine the quality of the outer contour of the weld; The gradient comparison and edge length analysis on different edges of the internal edge pixels are performed to determine the surface defect degree of the solder joint.

[0026] Preferably, in one embodiment of the present invention, the outer contour edge and a plurality of internal edge pixel points are obtained by edge detection of the welding spot image, and the specific method includes: The welding point image is detected and a number of edge pixel points are obtained by using the Canny edge detection algorithm; wherein the Canny edge detection is a well-known technology and will not be described in detail in this embodiment; adjacent edge pixel points are connected (eight-neighborhood adjacent), that is, the connected edge pixel points are connected, and a number of curves are obtained by connection, each curve is an edge, and a number of edges are obtained.

[0027] Furthermore, the length of each edge is obtained, and the edge with the longest length among all closed edges is used as the outer contour edge of the solder joint image; and several edge pixel points inside the outer contour edge are used as internal edge pixel points of the solder joint image.

[0028] Preferably, in one embodiment of the present invention, the roundness and corner point distribution of the outer contour edge are analyzed to determine the outer contour quality of the solder joint, including the specific method of: The Shoelace formula (Gaussian area formula) is used to obtain the area of ​​the enclosed area of ​​the outer contour edge. The roundness of the outer contour edge is calculated as follows: in, Indicates the roundness of the outer contour edge. Indicates the length of the outer contour edge, Indicates the area of ​​the region enclosed by the outer contour edge.

[0029] It should be noted that the closer the roundness is to 1, the closer the outer contour edge is to a circle, and the more standard the overall shape of the outer contour of the solder joint is, without considering whether the outer contour of the solder joint is smooth.

[0030] Furthermore, a corner point detection algorithm is used to obtain several corner points of the outer contour edge. The corner point detection algorithm is a well-known technology and will not be described in detail in this embodiment. The ratio of the number of corner points to the number of pixel points on the outer contour edge is obtained, and the difference obtained by subtracting the ratio from 1 is used as the contour smoothness of the outer contour edge.

[0031] It should be noted that the fewer the corner points on the outer contour edge, the smoother the outer contour edge is, the outer contour of the weld point has fewer sharp corners and notches, is less rough, and the quality of the corresponding outer contour of the weld point is higher.

[0032] Furthermore, the average of the roundness of the outer contour edge and the contour smoothness is taken as the outer contour quality of the weld spot of the test welding.

[0033] Preferably, in one embodiment of the present invention, gradient comparison and edge length analysis are performed on different edges of internal edge pixels to determine the surface defect of the solder joint, including the following specific methods: Obtain several edges formed by internal edge pixels and record them as internal edges of solder joints; obtain the length of each internal edge of solder joints and the gradient mean of all internal edge pixels on any internal edge of solder joints as the overall gradient of the internal edge of the solder joints. The calculation method of solder joint surface defect degree is: in, Indicates the surface defect of the solder joint. Indicates the number of internal edges of the solder joint, Indicates The length of the inner edge of the strip weld, Indicates The overall gradient of the inner edge of the strip weld, represents the overall gradient mean of the internal edges of all solder joints, Represents the weight normalization function, and the normalized object is the length of the internal edges of all welds.

[0034] It should be noted that, usually, during the spot welding process, heat is transferred through the welding area, resulting in a temperature difference between the welding spot and the surrounding base material, and this temperature difference causes the generation of thermal stress. The uneven distribution of thermal stress will cause local stress concentration, which may exceed the yield limit of the material and lead to the generation of cracks. The cracks appear in the image in a tortuous and winding form, and show grayscale variation characteristics in the image. In this case, the edge detection algorithm can be used to extract the gradient of the corresponding edge in the image, and the deviation of the edge pixel on the edge and the overall gradient of the edge can be further calculated by the gradient to reflect the edge. The degree of tortuosity of the corresponding edge, and the crack usually has a certain length, and the longer the length, the more severe the crack. Therefore, in the present application, when the overall gradient of the internal edge of the weld in the image deviates, and the larger the deviation, the more it conforms to the morphological characteristics of the crack, and when the length of the crack is longer, the worse the surface quality of the weld, that is, the greater the defect, that is, for the internal edge of the weld, the greater the deviation of the overall gradient, and the longer the length, the more likely it is an obvious crack or defect, then there are more obvious cracks and other defects inside the weld, the rougher the weld surface, the greater the surface defect of the weld, and the worse the surface quality of the weld.

[0035] At this point, the edge detection of the solder joint image of the test welding is carried out, and the appearance quality of the solder joint is analyzed by using machine vision methods. The quality of the solder joint outer contour is determined by analyzing whether the solder joint outer contour fits the circle and is smooth. The gradient difference and distribution inside the solder joint are analyzed to quantify whether there are defects such as cracks, thereby reflecting the surface quality of the solder joint.

[0036] Step S003, judging the outer contour quality of the weld spot and the surface defect degree of the weld spot in the test welding, and adjusting the welding parameters accordingly; constructing a welding quality regression model through multiple test welding and adjusting the welding parameters.

[0037] It should be noted that if the shape of the weld is irregular, that is, it does not fit the circle or ellipse, or there are sharp corners, spatter, etc. on the edge contour of the weld, it is usually caused by excessive welding current causing the weld to overheat, or excessive electrode pressure causing poor contact of the weld. In this case, it is necessary to reduce the welding current and increase the electrode pressure. When the surface defects of the weld are large, reflecting the poor surface quality of the weld, it is usually caused by excessive welding current or long welding time, which leads to overheating of the weld and crack defects or rough surface.

[0038] Preferably, in one embodiment of the present invention, the outer contour quality of the weld spot and the surface defect degree of the weld spot of the test welding are judged to adjust the welding parameters, and the specific method includes: The contour threshold is preset. In this embodiment, the contour threshold is described as 0.8. If the outer contour quality of the weld point of the test welding is less than the contour threshold, the welding parameters need to be adjusted, as follows: 1. Welding current: 1) Adjustment direction: reduce; 2) Adjustment range: reduce the current value by 5-10% and observe the change in solder joint quality; in this embodiment, the adjustment range is 5% of the current value.

[0039] 2. Welding time: 1) Adjustment direction: shorten; 2) Adjustment range: The adjustment is reduced by 0.1-0.2 seconds, depending on the actual welding time; the adjustment range in this embodiment is 0.1 seconds.

[0040] 3. Electrode pressure: 1) Adjustment direction: increase; 2) Adjustment range: adjust and increase the pressure value by 5-10% to ensure good contact between the electrode and the welding point; in this embodiment, the adjustment range is 5% of the pressure value.

[0041] After adjusting the welding parameters based on the quality of the outer contour of the weld, a defect threshold is preset. In this embodiment, the defect threshold is described as 0.2. If the surface defect of the weld of the test weld is greater than the defect threshold, it is necessary to continue to adjust the welding parameters based on the already adjusted welding parameters, as follows: 1. Welding current: 1) Adjustment direction: reduce; 2) Adjustment range: reduce the current value by 5-10% to avoid overheating of the solder joint; in this embodiment, the adjustment range is 5% of the current value.

[0042] 2. Welding time: 1) Adjustment direction: shorten; 2) Adjustment range: Adjust to reduce 0.1-0.2 seconds to reduce heat input; the adjustment range in this embodiment is 0.1 seconds.

[0043] 3. Preheating temperature and post-heating time: 1) Adjustment direction: increase preheating temperature and postheating time.

[0044] 2) Adjustment range: adjust and increase the preheating temperature by 10-20°C and the post-heating time by 0.5-1 minute; the adjustment range in this embodiment is to increase the preheating temperature by 10°C and the post-heating time by 1 minute.

[0045] Then, based on the initial welding trial, the adjusted welding parameters are obtained as the first adjusted welding parameters.

[0046] Preferably, in one embodiment of the present invention, a welding quality regression model is constructed by multiple welding trials and adjusting welding parameters, and the specific method includes: After obtaining the welding parameters adjusted for the first time, the fuse is re-tested and the weld spot image of the new test welding is obtained. The weld spot outer contour quality and weld spot surface defect degree of the test welding are obtained according to the above method, and the welding parameters are adjusted to obtain the welding parameters adjusted for the second time. Similarly, several test weldings are performed and the weld spot outer contour quality and weld spot surface defect degree of the test weldings are obtained. At the same time, various welding parameters of each test welding are recorded, including welding current, welding time, electrode pressure, preheating temperature and cooling rate (obtained through post-heating time). In this embodiment, a total of five test weldings are performed to construct a welding quality regression model.

[0047] Furthermore, the standard deviation of the welding parameters corresponding to any welding parameter in multiple welding trials is obtained, the standard deviation of all welding parameters is weighted normalized, and the result is used as the regression weight of each welding parameter; the present embodiment adopts the Lasoo regression model, and multiple welding parameters of any welding trial are input into the model, and the preset weight of each welding parameter in the model is its corresponding regression weight, and the weld outer contour quality and weld surface defect degree of the welding trial are input into the model at the same time; the output feature is the adjustment action of each welding parameter, including the parameter value and adjustment direction (for example, increase or decrease) of each adjustment; the model is used for training, and the welding parameters of each welding trial and the corresponding weld outer contour quality and weld surface defect degree are input into the regression model, and the adjustment action of the welding parameters for the next welding trial is output, and finally the construction of the welding quality regression model is completed.

[0048] At this point, the quality of the weld contour and the surface defects of the weld are judged through multiple trial welding, the welding parameters are adjusted and a welding quality regression model is constructed based on this. The welding quality regression model can output the adjusted welding parameters and, through continuous trial welding, provide a basis for the subsequent spot welding quality to meet the standards.

[0049] Step S004: continuously adjusting the welding parameters for the test welding through the welding quality regression model, and performing spot welding for subsequent fuse production.

[0050] After the welding quality regression model outputs the adjustment action, a trial welding is performed according to the adjusted welding parameters, and the weld outer contour quality and weld surface defect degree of the current trial welding are obtained; based on the welding quality regression model, combined with the welding parameters of the current trial welding and the weld outer contour quality and weld surface defect degree, the adjustment action continues to be output, and the welding parameters are adjusted continuously by analogy until the weld outer contour quality and weld surface defect degree of a trial welding meet the standards, that is, the weld outer contour quality is greater than or equal to the contour threshold and the weld surface defect degree is less than or equal to the defect threshold, then the corresponding welding parameters are used as the welding parameters for spot welding in the fuse production process, and spot welding is performed for subsequent fuse production.

[0051] At this point, the welding quality of the test welding is quantified, and a welding quality regression model is constructed in combination with the welding parameters to output the adjustment actions of the welding parameters, so that the spot welding quality can be rapidly improved while continuously meeting the standards, and finally the appropriate welding parameters are adjusted to ensure the spot welding effect of the indicative fuse production.

[0052] In a specific embodiment of the present invention, a spot welding method for indicative fuse production is provided, in which edge detection is performed on a welding spot image by the method in step S002, and an outer contour edge and an inner contour edge in the welding spot image are obtained respectively, such as Figure 2 Shown is an image of a solder joint. Figure 3 is the image corresponding to the outer contour edge, Figure 4 The inner contour edge image is obtained by using the step S001 to step S004 in the spot welding method for producing indicative fuses, and the data calculation results corresponding to the gradient value, the inner edge length, the roundness, the smoothness and the defect degree in the welding spot image are shown in Table 1: Table 1: Calculation results of the data corresponding to the gradient value, internal edge length, roundness, smoothness and defectivity in the solder joint image.

[0053] Another embodiment of the present invention provides a spot welding system for producing indicative fuses, the system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the above method steps S001 to S004 are implemented.

[0054] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A spot welding method for the production of indicative fuses, characterized in that: The method comprises the following steps: Test welding of fuse to obtain welding spot image; Through edge detection of the solder joint image, the roundness and smoothness of the solder joint outer contour and the gradient distribution analysis inside the solder joint are analyzed to determine the outer contour quality of the solder joint and the surface defect degree of the solder joint; Determine the outer contour quality and surface defect degree of the weld spot during the test welding, and adjust the welding parameters accordingly; build a welding quality regression model through multiple test welding and adjustment of welding parameters; The welding parameters are continuously adjusted for the trial welding through the welding quality regression model.

2. The spot welding method for producing an indicating fuse according to claim 1, characterized in that: The welding spot image is subjected to edge detection, and the roundness and smoothness analysis of the outer contour of the welding spot and the gradient distribution analysis inside the welding spot are performed to determine the outer contour quality of the welding spot and the surface defect degree of the welding spot of the test welding, including the specific method as follows: The outer contour edge and several internal edge pixel points are obtained by edge detection of the welding spot image; Analyze the roundness and corner point distribution of the outer contour edge to determine the quality of the outer contour of the weld; The gradient comparison and edge length analysis on different edges of the internal edge pixels are performed to determine the surface defect degree of the solder joint.

3. The spot welding method for producing an indicating fuse according to claim 2, characterized in that: The specific method for obtaining the outer contour edge and several internal edge pixel points is as follows: Obtain several edge pixel points from the welding point image by edge detection; Connect adjacent edge pixels to obtain several edges; The length of each edge is obtained, and the edge with the largest length among all closed edges is used as the outer contour edge of the solder joint image; and several edge pixel points inside the outer contour edge are used as the inner edge pixel points of the solder joint image.

4. The spot welding method for producing an indicating fuse according to claim 2, characterized in that: The analysis of the roundness and corner point distribution of the outer contour edge to determine the outer contour quality of the solder joint includes the following specific methods: Get the roundness of the outer contour edge; Obtain several corner points of the outer contour edge by corner point detection; obtain the ratio of the number of corner points to the number of pixel points on the outer contour edge, and subtract the difference obtained by subtracting the ratio from 1 as the contour smoothness of the outer contour edge; The average of the roundness of the outer contour edge and the contour smoothness is taken as the outer contour quality of the weld spot of the test welding.

5. The spot welding method for producing an indicating fuse according to claim 4, characterized in that: The specific method for obtaining the roundness of the outer contour edge includes: The area of ​​the enclosed area of ​​the outer contour edge is obtained, and the roundness of the outer contour edge is calculated as follows: in, Indicates the roundness of the outer contour edge. Indicates the length of the outer contour edge, Indicates the area of ​​the region enclosed by the outer contour edge.

6. The spot welding method for producing an indicating fuse according to claim 2, characterized in that: The specific method of comparing the gradients on different edges and analyzing the edge lengths of the internal edge pixels to determine the surface defect of the solder joint includes: Obtain a number of edges formed by internal edge pixels, which are recorded as internal edges of the welding point; Obtain the length of the inner edge of each solder joint and the gradient mean of all inner edge pixel points on any inner edge of the solder joint as the overall gradient of the inner edge of the solder joint; The surface defect degree of the solder joint is determined by the overall gradient and length of the internal edge of the solder joint.

7. The spot welding method for producing an indicating fuse according to claim 6, characterized in that: The specific method of determining the surface defect of the solder joint by the overall gradient and length of the internal edge of the solder joint is as follows: in, Indicates the surface defect of the solder joint. Indicates the number of internal edges of the solder joint, Indicates The length of the inner edge of the strip weld, Indicates The overall gradient of the inner edge of the strip weld, represents the overall gradient mean of the internal edges of all solder joints, represents the weight normalization function.

8. The spot welding method for producing an indicating fuse according to claim 2, characterized in that: The welding quality regression model is constructed by multiple welding trials and adjusting welding parameters, and the specific method includes: After the adjusted welding parameters are obtained, the fuse is re-tested, and a welding spot image of the new test welding is obtained, the outer contour quality of the welding spot and the surface defect degree of the welding spot of the test welding are obtained, and the welding parameters are adjusted to obtain the second adjusted welding parameters; Similarly, several test welds are conducted to obtain the weld point outer contour quality and weld point surface defect degree of the test welds, and various welding parameters of each test weld are recorded, including welding current, welding time, electrode pressure, preheating temperature and cooling speed; Obtaining the standard deviation of the welding parameters corresponding to any welding parameter under multiple welding trials, weight normalizing the standard deviations of all welding parameters, and using the obtained results as the regression weights of each welding parameter; Input multiple welding parameters of any welding test into the model, and the preset weight of each welding parameter in the model is its corresponding regression weight. At the same time, input the weld point outer contour quality and weld point surface defect degree of the welding test into the model; The output features are the adjustment actions of each welding parameter, including the parameter value and adjustment direction of each adjustment. The model is trained to build a welding quality regression model.

9. The spot welding method for producing an indicating fuse according to claim 2, characterized in that: The specific method of continuously adjusting the welding parameters of the test welding by the welding quality regression model is as follows: After the welding quality regression model outputs the adjustment action, a trial welding is performed according to the adjusted electric welding parameters, and the outer contour quality of the weld point and the surface defect degree of the weld point of the current trial welding are obtained; Based on the welding quality regression model, combined with the welding parameters of the current trial welding, the outer contour quality of the weld spot and the surface defect degree of the weld spot, the adjustment action continues to be output, and the welding parameters are adjusted continuously by analogy until the outer contour quality and the surface defect degree of the weld spot of a trial welding meet the standards. The corresponding welding parameters are used as the welding parameters for spot welding in the fuse production process.

10. A spot welding system for the production of indicative fuses, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the spot welding method for producing an indicating fuse according to any one of claims 1 to 9 are implemented.

Images