CN116930194B

CN116930194B - Defect detection system and method for friction stir welding, electronic equipment and medium

Info

Publication number: CN116930194B
Application number: CN202311181285.9A
Authority: CN
Inventors: 王银柏; 徐武松; 朱国强; 钱春峰
Original assignee: Zhangjiagang Sifu Security Technology Co ltd
Current assignee: Zhangjiagang Sifu Security Technology Co ltd
Priority date: 2023-09-14
Filing date: 2023-09-14
Publication date: 2023-12-08
Anticipated expiration: 2043-09-14
Also published as: CN116930194A

Abstract

The application discloses a friction stir welding defect detection system, a friction stir welding defect detection method, electronic equipment and a medium, which are used for improving the defect detection efficiency of friction stir welding. The defect detection method comprises the following steps: placing a workpiece to be welded on a bearing platform, and setting initial positions of a friction stir welding module, a light-emitting element and an acquisition camera; starting and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along a welding path; controlling the light-emitting element to emit illumination light, and controlling the acquisition camera to acquire a real-time detection image; generating a weld edge fitting line on the real-time detection image; generating a search frame by taking a weld seam edge fitting line as a common edge; dividing the first and second defect search boxes to generate square boxes; gray level analysis is carried out on the square boxes, and gray level analysis results are generated; determining the center point of the gradual change box, and analyzing; and determining that the burr defects exist when the irregular linear central lattice exists in the first defect search frame and the second defect search frame.

Description

Defect detection system and method for friction stir welding, electronic equipment and medium

Technical Field

The embodiment of the application relates to the field of friction stir welding, in particular to a defect detection system, a defect detection method, electronic equipment and a defect detection medium for friction stir welding.

Background

With the continuous development of modern industry, the reliable connection between different materials or different materials is inevitably not separated in the fields of ship manufacturing, transportation, aerospace, ocean engineering, petrochemical industry and the like. Welding is an important connection mode, and the forming quality of the welding seam determines the safety and reliability of the member in service.

Friction stir welding (Friction stir welding, FSW for short) is a solid phase connection technology invented by British welding research in 1991, has the advantages of low welding heat input, no pollution in the welding process, excellent joint quality, simple operation and the like, and is widely applied to connection of materials such as aluminum alloy, magnesium alloy, copper alloy and the like. Compared with the traditional fusion welding, the friction stir welding temperature is lower than the solidus line, the materials are not melted, and joint defects (such as air holes, hot cracks, slag inclusion and the like) caused by melting the materials can be effectively avoided. In addition, the method has the advantages of low joint residual stress, small deformation of the welded workpiece, fine joint microstructure, no alloy element burning loss, high joint strength and the like. Called the "most revolutionary connection technology in the 21 st century".

And secondly, compared with the original welding process, the technology has the advantages of simple principle, less control parameters, reduced artificial factor influence in the welding process, improved safety conditions and welding quality, wide application in the fields of aerospace manufacturing industry, ship manufacturing industry, nuclear energy industry and the like, and very important research on friction stir welding technology at home and abroad, wherein the research is mainly focused on friction heat generating models of stirring heads and workpieces, simulation and prediction of the friction stir welding process, migration of weld joint formation and thermoplastic metal, influence of friction stir welding acting force, mechanism of defect formation in the weld joint, residual stress and deformation caused by friction stir welding and other directions.

In the friction stir welding process, the temperature of the stirring pin, the rotating speed of the stirring pin, the downward pressing degree of the stirring pin, the rotating speed of the shaft shoulder, the downward pressing degree of the shaft shoulder and the temperature of the shaft shoulder are required to be strictly set according to the material of the workpiece, so that great test is brought to detection and control. If these parameters cannot be accurately controlled during operation, numerous defects such as burrs, key holes, surface depressions, burrs, furrows and the like are caused on the workpieces to be welded, and other defects are mainly caused by mismatching of the parameters during welding except that the key holes are non-parametric defects. In the prior art, the welding parameters of the workpiece to be welded are unsuitable, which can lead to rough lines on the surface of the welding line (the lines are normally uniform scale lines), and the workpiece to be welded has a burr sense, which is called as a burr defect.

The reasons for forming burrs are various, namely the viscosity of the material of the workpiece to be welded is too high, the material is adhered to the shaft shoulder during welding, when the stirring pin and the shaft shoulder rotate in the melted workpiece, the material is splashed, and burrs are formed due to adhesion; secondly, the surface of the workpiece to be welded is not treated cleanly before welding, so that the workpiece to be welded is adhered to the impurity and the shaft shoulder in the welding process, and burrs are formed due to adhesion in the material splashing process; thirdly, as the temperature of the stirring pin and the shaft shoulder is too high, the heat input of the workpiece to be welded is too high in the welding process, and burrs are formed due to too high melting degree.

Among the above-mentioned three kinds of burr defect's formation reasons, the material viscosity of waiting to weld the work piece is too big, and we can select stirring needle and the shaft shoulder of specific material, reduce waiting to weld the adhesion of work piece to clear up waiting to weld the surface of work piece before the welding, reduce the incorporation of impurity. However, if the temperature of the stirring pin and the shaft shoulder is too high, strict detection control technology is required. At present, the defect detection method of burrs mainly monitors the state of a welding line in real time by manpower, in the welding process, when welding scraps are timely treated, whether the defects of the burrs are generated or not is judged manually, and the temperature of a stirring pin and a shaft shoulder is adjusted manually, so that the defect detection mode greatly reduces the defect detection efficiency of friction stir welding.

Disclosure of Invention

The application discloses a friction stir welding defect detection system, a friction stir welding defect detection method, electronic equipment and a medium, which are used for improving the defect detection efficiency of friction stir welding.

The first aspect of the present application provides a method for detecting a defect in friction stir welding, comprising:

placing a workpiece to be welded on a bearing platform, setting initial positions of a friction stir welding module, a light-emitting element and an acquisition camera according to a preset welding path, wherein the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module;

starting a friction stir welding module, and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along a welding path;

when the acquisition camera moves to a preset point, controlling the light-emitting element to emit illumination light perpendicular to a workpiece to be welded, and controlling the acquisition camera to shoot a welded section of area to acquire a real-time detection image;

generating a weld joint edge fitting line on a real-time detection image according to the diameter of a shaft shoulder and a welding path of the friction stir welding module;

generating search frames at equal distances on two sides of a weld edge fit line by taking the weld edge fit line as a common edge, wherein the weld edge fit line is used for dividing a workpiece area to be welded on a real-time detection image into a weld area and a non-weld area, the search frames on the weld area side are first defect search frames, and the search frames on the non-weld area side are second defect search frames;

Dividing the first defect search frame and the second defect search frame to generate square boxes of M pixels, wherein M is an integer greater than 1;

carrying out gray level analysis on each square box to generate a gray level analysis result;

determining the center points of the gradual change boxes in the first defect search box and the second defect search box according to the gray level analysis result, and analyzing the coordinates of the center points, wherein the gradual change boxes are square boxes with pixel gray level gradual change conditions;

when the first defect search frame and the second defect search frame both have irregular linear central lattices, determining that burrs appear on the workpieces to be welded.

Optionally, the welding path is a straight line;

generating a weld edge fitting line on the real-time detection image according to the shoulder diameter and the welding path of the friction stir welding module, comprising:

generating a linear welding seam edge on a real-time detection image according to the diameter of a shaft shoulder of the friction stir welding module and a welding path;

equidistant rectangular frames are generated on the edges of the linear welding seams, the rectangular frames are cut, square frames of N x N pixel points are generated, the rectangular frames comprise welding seam areas and non-welding seam areas, and N is an odd number larger than 1;

Gray level analysis is carried out on each square frame, the center point of the gradual change frame is determined, and defect-free rectangular frames with the dispersion degree lower than a preset value are selected from all rectangular frames according to the coordinates of the center point;

a weld edge fit line is generated on the real-time inspection image from the center point of each defect-free rectangular frame.

Optionally, the welding path is an arc;

generating an arc welding seam edge on a real-time detection image according to the diameter of a shaft shoulder of the friction stir welding module and a welding path;

generating equidistant rectangular frames on the edges of arc welding seams, cutting the rectangular frames to generate square frames of J pixels, wherein the rectangular frames comprise welding seam areas and non-welding seam areas, and J is an odd number larger than 1;

carrying out gray average value calculation on each square box in the rectangular box, determining square boxes with gray gradient changes, and carrying out curve fitting according to the center point coordinates of the square boxes with gray gradient changes to generate an initial arc line;

and screening out the outlier points according to the initial arc line, and performing curve fitting through the residual coordinates of the central points to generate a weld edge fitting line.

Optionally, two sides of the light-emitting element are respectively provided with an irradiation plate, and an included angle exists between the irradiation plates and the light-emitting element;

after generating the weld bead edge fitting line on the real-time detection image according to the shoulder diameter and the welding path of the friction stir welding module, the defect detection method further includes:

the irradiation plate is lightened, the acquisition camera is controlled to shoot a welded section of area, and a depth detection image is obtained;

marking a weld edge fitting line on the depth detection image;

generating a depth search area on the depth detection image according to a weld edge fitting line, wherein the weld edge fitting line is the central line of the depth search area, and divides the depth search area into an inner weld area and an outer weld area;

dividing the depth search area to generate square boxes of L pixels, carrying out gray analysis on each square box, wherein L is an integer greater than 1;

determining a first distance from a gradient box with the largest weld edge fitting line in the weld inner area and a second distance from the gradient box with the largest weld edge fitting line in the weld outer area;

and when the total distance of the first distance and the second distance is larger than the preset distance, determining that the workpiece to be welded has the defect of concave surface.

Optionally, after controlling the light emitting element to emit the irradiation light perpendicular to the workpiece to be welded and controlling the acquisition camera to shoot a welded section of the area, acquiring the real-time detection image, the defect detection method further comprises:

generating two stirring pin fitting lines on the real-time detection image according to the diameter of the stirring pin and the welding path of the stirring friction welding module, wherein the area between the two stirring pin fitting lines is a stirring pin area;

generating a furrow search frame in a stirring needle area by taking a stirring needle fitting line as an edge, and cutting the furrow search frame to generate square boxes of K pixels, wherein K is an odd number larger than 1;

carrying out gray average value calculation on each square box, and determining a gray change box, wherein the difference between the gray average value of the gray change box and the gray average value of more than one adjacent square boxes is larger than a preset gray difference threshold;

and generating a furrow area according to the central point coordinates of the gray scale change square frame, and determining that the workpiece to be welded has a furrow defect.

Optionally, after determining the center point of the gradient frame in the first defect search frame and the second defect search frame according to the gray level analysis result and performing the analysis of the coordinates of the center point, the defect detection method further includes:

When the first defect search frame only has an arc-shaped central lattice and the second defect search frame has a block-shaped central lattice, determining that the workpiece to be welded has the defect of flash.

In a second aspect, the present application provides a friction stir welding defect detection system, comprising:

the placing unit is used for placing the workpiece to be welded on the bearing platform, and setting initial positions of the friction stir welding module, the light-emitting element and the acquisition camera according to a preset welding path, wherein the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module;

the starting unit is used for starting the friction stir welding module and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along a welding path;

the first acquisition unit is used for controlling the light-emitting element to emit illumination light perpendicular to a workpiece to be welded when the acquisition camera moves to a preset point position, and controlling the acquisition camera to shoot a welded section of area to acquire a real-time detection image;

the first generation unit is used for generating a welding seam edge fitting line on the real-time detection image according to the diameter of the shaft shoulder and the welding path of the friction stir welding module;

the second generation unit is used for generating search frames at equal distance on two sides of the weld seam edge fitting line by taking the weld seam edge fitting line as a common edge, wherein the weld seam edge fitting line is used for dividing a workpiece area to be welded on a real-time detection image into a weld joint area and a non-weld joint area, the search frames on the side of the weld joint area are first defect search frames, and the search frames on the side of the non-weld joint area are second defect search frames;

The third generating unit is used for dividing the first defect search frame and the second defect search frame to generate square boxes of M x M pixel points, wherein M is an integer greater than 1;

the fourth generation unit is used for carrying out gray level analysis on each square box and generating a gray level analysis result;

the first determining unit is used for determining the center point of the gradual change box in the first defect searching box and the second defect searching box according to the gray level analysis result, analyzing the coordinates of the center point, wherein the gradual change box is a square box with the gradual change condition of the pixel gray level;

and the second determining unit is used for determining that the workpiece to be welded has burrs when the irregular linear central lattice exists in the first defect searching frame and the second defect searching frame.

Optionally, the welding path is a straight line;

a first generation unit including:

Optionally, the welding path is an arc;

a first generation unit including:

Optionally, the two sides of the light-emitting element are respectively provided with an irradiation plate, and an included angle exists between the irradiation plate and the light-emitting element;

After the first generating unit, the defect detecting apparatus further includes:

the second acquisition unit is used for lighting the irradiation plate, controlling the acquisition camera to shoot a welded section of area, and acquiring a depth detection image;

the marking unit is used for marking a weld edge fitting line on the depth detection image;

a fifth generation unit for generating a depth search area on the depth detection image according to a weld edge fitting line, the weld edge fitting line being a center line of the depth search area, the weld edge fitting line dividing the depth search area into an in-weld area and an out-weld area;

a sixth generating unit, configured to segment the depth search area, generate square boxes of l×l pixel points, perform gray analysis on each square box, and L is an integer greater than 1;

a third determining unit, configured to determine a first distance from a gradient box with a maximum weld edge fitting line in an area within the weld and a second distance from the gradient box with a maximum weld edge fitting line in an area outside the weld;

and the fourth determining unit is used for determining that the workpiece to be welded has the defect of concave surface when the total distance of the first distance and the second distance is larger than the preset distance.

Optionally, after the acquisition unit, the defect detection device further comprises:

the seventh generation unit is used for generating two stirring pin fitting lines on the real-time detection image according to the diameter and the welding path of the stirring pin of the stirring friction welding module, and the area between the two stirring pin fitting lines is a stirring pin area;

an eighth generating unit, configured to generate a furrow search frame in a stirring needle area with a stirring needle fitting line as an edge, and segment the furrow search frame to generate square boxes with k×k pixels, where K is an odd number greater than 1;

a fifth determining unit, configured to perform gray average value calculation on each square box, and determine a gray change box, where a difference between a gray average value of the gray change box and a gray average value of more than one neighboring square box is greater than a preset gray difference threshold;

and the sixth determining unit is used for generating a furrow area according to the central point coordinates of the gray level change box and determining that the workpiece to be welded has a furrow defect.

Optionally, after the first determining unit, the defect detecting apparatus further includes:

and a seventh determining unit, configured to determine that the workpiece to be welded has a flash defect when the first defect search frame has only the arc-shaped central lattice and the second defect search frame has the block-shaped central lattice.

A fourth aspect of the present application provides an electronic device comprising:

a processor, a memory, an input-output unit, and a bus;

the processor is connected with the memory, the input/output unit and the bus;

the memory holds a program that the processor invokes to perform the method of defect detection as in the first aspect and any of the optional friction stir welding methods of the first aspect.

A fifth aspect of the present application provides a computer readable storage medium having a program stored thereon, which when executed on a computer performs the method of defect detection for friction stir welding as in the first aspect and any of the optional methods of friction stir welding of the first aspect.

From the above technical solutions, the embodiment of the present application has the following advantages:

in the application, firstly, a workpiece to be welded is placed on a bearing platform, and the initial positions of a friction stir welding module, a light-emitting element and a collecting camera are set according to a preset welding path, wherein the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module. The friction stir welding module is placed first, then the luminous element and the acquisition camera, wherein the whole length of the luminous element is larger than the diameter of the shaft shoulder of the friction stir welding module so as to enable the luminous element to cover the whole welding line.

Next, the friction stir welding module is started, and the friction stir welding module, the light emitting element, and the acquisition camera are controlled to move along the welding path in sequence. When the acquisition camera moves to a preset point, the light-emitting element is controlled to emit illumination light perpendicular to the workpiece to be welded, and the acquisition camera is controlled to shoot a welded section of area, so that a real-time detection image is obtained. Each point location is collected at least once, an image with the cleanest surface is selected as a real-time detection image, wherein the irradiation light of the light-emitting element is determined according to the color of the workpiece to be welded, and the pixel points which can enable the non-welded area of the workpiece to be welded to generate larger contrast with the welding line and enable the two to be fed back clearly in the real-time detection image are mainly achieved. And generating weld joint edge fit lines on the real-time detection image according to the shaft shoulder diameter and the welding path of the friction stir welding module, wherein as the size of the weld joint is similar to the shaft shoulder diameter, two weld joint edge fit lines can be generated in advance according to the welding path (central line) and the shaft shoulder diameter, and the part between the two weld joint edge fit lines is the weld joint. And generating search frames at equal distances on two sides of the weld seam edge fitting line by taking the weld seam edge fitting line as a common edge, wherein the weld seam edge fitting line is used for dividing a workpiece area to be welded on the real-time detection image into a weld joint area and a non-weld joint area, the search frames on the side of the weld joint area are first defect search frames, and the search frames on the side of the non-weld joint area are second defect search frames. That is, if a burr defect occurs, a splash-state solidification waste is mainly generated in the non-weld region, and the splash-state solidification waste is in a long and thin strip shape, and the splash-state solidification waste is irradiated to the welding region and generates a strip-shaped shadow in the welding region and also generates a strip-shaped shadow in the non-welding region because the light emitting device is longer than the welding seam. And then the first defect search frame and the second defect search frame are segmented to generate square boxes of M pixels, wherein M is an integer greater than 1. And carrying out gray level analysis on each square box to generate a gray level analysis result. And determining the central points of the gradual change boxes in the first defect search box and the second defect search box according to the gray level analysis result, analyzing coordinates of the central points, wherein the gradual change boxes are square boxes with pixel gray level gradual change conditions. When linear central lattices exist in the first defect search frame and the second defect search frame, determining that burrs appear on the workpieces to be welded. Because the workpiece to be welded in the conventional state forms arc-shaped grains in the welding area, and the non-welding area is mainly cut grains or is free of grains (straight lines or smooth and free of lattices), after gray analysis is carried out, the central lattices generated in the welding area can show regular arc-shaped central lattices, and the non-welding area is regular straight central lattices or almost free of lattices. Therefore, the first defect search frame and the second defect search frame are segmented to generate square boxes of M pixels, gray algorithm analysis is carried out on each square box to generate gray analysis results, gradient boxes and gradient box center point coordinates in the first defect search frame and the second defect search frame are determined according to the gray analysis results, center point coordinates are analyzed, algorithm fitting or marking can be carried out on the coordinates of the center points, and by comparing preset normal lattices, what center lattice appears can be determined. If a linear center lattice exists in both the first defect search box and the second defect search box, it is determined that the workpiece to be welded has a burr defect in the welding area. According to the scheme, the gray level algorithm and the dot matrix comparison algorithm are utilized, manual detection is not needed, detection is carried out in a pure automatic mode, and the defect detection efficiency of friction stir welding is improved to a great extent.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic view showing a first embodiment of a method for detecting a defect in friction stir welding according to the present application;

FIG. 2 is a schematic diagram showing a first stage of a second embodiment of a method for detecting defects in friction stir welding according to the present application;

FIG. 3 is a schematic diagram showing a second stage of a second embodiment of a method for detecting defects in friction stir welding according to the present application;

FIG. 4 is a schematic diagram showing a third stage of a second embodiment of a method for detecting defects in friction stir welding according to the present application;

FIG. 5 is a schematic diagram showing a third exemplary first stage of a method for detecting a defect in friction stir welding according to the present application;

FIG. 6 is a schematic diagram showing a second stage of a third embodiment of a method for detecting defects in friction stir welding according to the present application;

FIG. 7 is a schematic diagram showing a third stage of a third embodiment of a method for detecting defects in friction stir welding according to the present application;

FIG. 8 is a schematic diagram of one embodiment of a defect detection system for friction stir welding of the present application;

FIG. 9 is a schematic diagram of another embodiment of a friction stir welding defect detection system according to the present application;

FIG. 10 is a schematic diagram of an embodiment of an electronic device of the present application.

Detailed Description

In the following description, for purposes of explanation and not limitation, specific details are set forth such as the particular system architecture, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

It should also be understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.

As used in the present description and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".

Furthermore, the terms "first," "second," "third," and the like in the description of the present specification and in the appended claims, are used for distinguishing between descriptions and not necessarily for indicating or implying a relative importance.

Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.

In the prior art, friction stir welding (Friction stir welding is called FSW for short) is a solid phase connection technology invented by UK welding research in 1991, has the advantages of low welding heat input, no pollution in the welding process, excellent joint quality, simple operation and the like, and is widely applied to connection of materials such as aluminum alloy, magnesium alloy, copper alloy and the like. Compared with the traditional fusion welding, the friction stir welding temperature is lower than the solidus line, the materials are not melted, and joint defects (such as air holes, hot cracks, slag inclusion and the like) caused by melting the materials can be effectively avoided. In addition, the method has the advantages of low joint residual stress, small deformation of the welded workpiece, fine joint microstructure, no alloy element burning loss, high joint strength and the like. Called the "most revolutionary connection technology in the 21 st century".

Based on the above, the application discloses a friction stir welding defect detection system, a friction stir welding defect detection method, electronic equipment and a medium, which are used for improving the defect detection efficiency of friction stir welding.

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

The method of the present application may be applied to a server, a device, a terminal, or other devices having logic processing capabilities, and the present application is not limited thereto. For convenience of description, the following description will take an execution body as an example of a terminal.

Referring to fig. 1, the present application provides a first embodiment of a method for detecting defects in friction stir welding, which includes:

101. placing a workpiece to be welded on a bearing platform, setting initial positions of a friction stir welding module, a light-emitting element and an acquisition camera according to a preset welding path, wherein the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module;

The terminal places the work piece of waiting to weld on accepting the platform for the gap aligns the preset route, sets up friction stir welding module, light emitting component and gathers the initial position of camera according to the welding route that presets, and light emitting component's length is greater than friction stir welding module's shaft shoulder diameter. The friction stir welding module is placed first, then the luminous element and the acquisition camera, wherein the whole length of the luminous element is larger than the diameter of the shaft shoulder of the friction stir welding module so as to enable the luminous element to cover the whole welding line.

102. Starting a friction stir welding module, and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along a welding path;

the friction stir welding module, the light emitting element and the acquisition camera are displaced according to a preset path, so that after the friction stir welding module finishes welding, the acquisition camera and the light emitting element can reach the welding line area.

103. When the acquisition camera moves to a preset point, controlling the light-emitting element to emit illumination light perpendicular to a workpiece to be welded, and controlling the acquisition camera to shoot a welded section of area to acquire a real-time detection image;

when the acquisition camera moves to a preset point position, the terminal controls the light-emitting element to emit illumination light perpendicular to the workpiece to be welded, and controls the acquisition camera to shoot a welded section of area, so that a real-time detection image is obtained. Each point location is collected at least once, an image with the cleanest surface is selected as a real-time detection image, wherein the irradiation light of the light-emitting element is determined according to the color of the workpiece to be welded, and the pixel points which can enable the non-welded area of the workpiece to be welded to generate larger contrast with the welding line and enable the two to be fed back clearly in the real-time detection image are mainly achieved.

104. Generating a weld joint edge fitting line on a real-time detection image according to the diameter of a shaft shoulder and a welding path of the friction stir welding module;

and generating weld joint edge fit lines on the real-time detection image by the terminal according to the shaft shoulder diameter and the welding path of the friction stir welding module, wherein as the size of the weld joint is similar to the shaft shoulder diameter, two weld joint edge fit lines can be generated in advance according to the welding path (central line) and the shaft shoulder diameter, and the part between the two weld joint edge fit lines is the weld joint.

105. Generating search frames at equal distances on two sides of a weld edge fit line by taking the weld edge fit line as a common edge, wherein the weld edge fit line is used for dividing a workpiece area to be welded on a real-time detection image into a weld area and a non-weld area, the search frames on the weld area side are first defect search frames, and the search frames on the non-weld area side are second defect search frames;

the terminal uses a weld seam edge fitting line as a common edge, search frames are generated at equal distance on two sides of the weld seam edge fitting line, the weld seam edge fitting line is used for dividing a workpiece area to be welded on a real-time detection image into a weld seam area and a non-weld seam area, the search frames on the side of the weld seam area are first defect search frames, and the search frames on the side of the non-weld seam area are second defect search frames. That is, if a burr defect occurs, a splash-state solidification waste is mainly generated in the non-weld region, and the splash-state solidification waste is in a long and thin strip shape, and the splash-state solidification waste is irradiated to the welding region and generates a strip-shaped shadow in the welding region and also generates a strip-shaped shadow in the non-welding region because the light emitting device is longer than the welding seam.

In this embodiment, firstly, selecting multiple sections of the weld seam edge fitting line to have the same edge length, and generating a search box from the edge length to the inside of the weld seam and the outside of the weld seam.

106. Dividing the first defect search frame and the second defect search frame to generate square boxes of M pixels, wherein M is an integer greater than 1;

and the terminal then cuts the first defect search frame and the second defect search frame to generate square boxes of M pixels, wherein M is an integer greater than 1. And gray level analysis is carried out on each square box to generate a gray level analysis result, the square boxes are composed of pixel points, and all the pixel points in the same square box are compared by comparing the differences of the pixel points, so that the gray level analysis is not required to be carried out integrally, the workload is reduced, and the defect detection rate is increased.

107. Carrying out gray level analysis on each square box to generate a gray level analysis result;

108. determining the center points of the gradual change boxes in the first defect search box and the second defect search box according to the gray level analysis result, and analyzing the coordinates of the center points, wherein the gradual change boxes are square boxes with pixel gray level gradual change conditions;

And the terminal carries out gray level analysis on each square frame to generate a gray level analysis result, then the terminal determines a gradual change frame in the first defect search frame and the second defect search frame according to the gray level analysis result, determines the center point coordinate of the gradual change frame, carries out analysis on the center point coordinate, and the gradual change frame is a square frame with the gradual change condition of the pixel gray level.

Because the workpiece to be welded in the conventional state forms arc-shaped grains in the welding area, and the non-welding area is mainly cut grains or is free of grains (straight lines or smooth and free of lattices), after gray analysis is carried out, the central lattices generated in the welding area can show regular arc-shaped central lattices, and the non-welding area is regular straight central lattices or almost free of lattices.

109. When the first defect search frame and the second defect search frame both have irregular linear central lattices, determining that burrs appear on the workpieces to be welded.

The terminal tool analyzes the result, compares the difference between the lattice in the real-time detection image and the normal lattice, in particular to obtain a defect-free lattice, wherein the lattice in the non-welding area is a regular straight line central lattice, or a smooth surface lattice, or a pattern lattice with regular patterns, and can analyze which type the lattice belongs to by comparing the lattice rules on the lattice in the real-time detection image, the state of the surface to be welded can be input into a computer in advance to match the corresponding lattice, the workpiece to be welded belongs to the drying surface, the corresponding non-lattice is obtained, the straight line lattice is matched if the workpiece is cut, and the like. Therefore, the first defect search frame and the second defect search frame are segmented to generate square boxes of M pixels, gray algorithm analysis is carried out on each square box to generate gray analysis results, gradient boxes and gradient box center point coordinates in the first defect search frame and the second defect search frame are determined according to the gray analysis results, center point coordinates are analyzed, algorithm fitting or marking can be carried out on the coordinates of the center points, and by comparing preset normal lattices, what center lattice appears can be determined. If a linear center lattice exists in both the first defect search box and the second defect search box, it is determined that the workpiece to be welded has a burr defect in the welding area. It should be noted that, the second defect search box is not long and does not cover the movement area of the stirring pin, because the burr defect is entirely in the non-welding area, and the shadow of the burr defect is formed in the welding area, and these shadows appear because the light emitting element is longer than the welding seam, so that a part of the longer shadow of the burr irradiates the welding area, and the shadow of the burr does not grow into the movement area of the stirring pin.

In this embodiment, first, a workpiece to be welded is placed on a receiving platform, and an initial position of a friction stir welding module, a light emitting element and an acquisition camera is set according to a preset welding path, wherein the length of the light emitting element is greater than the diameter of a shaft shoulder of the friction stir welding module. The friction stir welding module is placed first, then the luminous element and the acquisition camera, wherein the whole length of the luminous element is larger than the diameter of the shaft shoulder of the friction stir welding module so as to enable the luminous element to cover the whole welding line.

Referring to fig. 2, 3 and 4, a second embodiment of a method for detecting defects in friction stir welding is provided, which includes:

201. placing a workpiece to be welded on a bearing platform, setting initial positions of a friction stir welding module, a light-emitting element and an acquisition camera according to a preset welding path, wherein the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module;

202. starting a friction stir welding module, and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along a welding path;

203. when the acquisition camera moves to a preset point, controlling the light-emitting element to emit illumination light perpendicular to a workpiece to be welded, and controlling the acquisition camera to shoot a welded section of area to acquire a real-time detection image;

in this embodiment, steps 201 to 203 are similar to steps 101 to 103, and are not described here.

204. Generating a linear welding seam edge on a real-time detection image according to the diameter of a shaft shoulder of the friction stir welding module and a welding path;

205. equidistant rectangular frames are generated on the edges of the linear welding seams, the rectangular frames are cut, square frames of N x N pixel points are generated, the rectangular frames comprise welding seam areas and non-welding seam areas, and N is an odd number larger than 1;

206. Gray level analysis is carried out on each square frame, the center point of the gradual change frame is determined, and defect-free rectangular frames with the dispersion degree lower than a preset value are selected from all rectangular frames according to the coordinates of the center point;

207. generating a weld edge fitting line on the real-time detection image according to the central point of each defect-free rectangular frame;

when the welding path is straight, the terminal generates a straight welding seam edge on the real-time detection image according to the welding path as a central line and the vertical central line as a length by taking the diameter of the shaft shoulder of the friction stir welding module. The straight weld edge is now the only initial edge line, which is still a distance, albeit close to the weld.

And then, the terminal generates equidistant rectangular frames on the edges of the linear welding lines, the rectangular frames are drawn by taking points on the edges of the linear welding lines as positioning points, the lengths of the rectangular frames can be manually selected, and only the real edge parts of the frame columns are needed, so that the detection calculation amount is reduced, and the calculation precision and efficiency are improved. At this time, the rectangular frame is further required to be segmented to generate square frames of N×N pixel points, the rectangular frame comprises a welding seam area and a non-welding seam area, and N is an odd number greater than 1. And carrying out gray level analysis on each square frame from top to bottom or from bottom to top, determining the coordinates of the center points of the gradual change frames and the gradual change frames, and screening defect-free rectangular frames with the dispersion degree lower than a preset value from all the rectangular frames according to the coordinates of the center points. And finally, the terminal generates a weld edge fit line on the real-time detection image according to the center point of each non-defective rectangular frame, wherein the weld edge fit line is relatively accurate.

208. Generating two stirring pin fitting lines on the real-time detection image according to the diameter of the stirring pin and the welding path of the stirring friction welding module, wherein the area between the two stirring pin fitting lines is a stirring pin area;

209. generating a furrow search frame in a stirring needle area by taking a stirring needle fitting line as an edge, and cutting the furrow search frame to generate square boxes of K pixels, wherein K is an odd number larger than 1;

210. carrying out gray average value calculation on each square box, and determining a gray change box, wherein the difference between the gray average value of the gray change box and the gray average value of more than one adjacent square boxes is larger than a preset gray difference threshold;

211. generating a furrow area according to the central point coordinates of the gray level change square frame, and determining that the workpiece to be welded has a furrow defect;

besides burr defects, furrow defects are very important defects in the friction stir welding process, and need to be detected in time.

The plow groove defect is a groove-shaped defect formed on the surface of a welding line on a workpiece to be welded, and becomes a plow groove, and the plow groove is positioned in the travelling area of the stirring pin.

The reason for forming furrows is as follows: in the friction stir welding process, the stirring pin is insufficient in heat input, so that the heat of the material is not reached, the cutting amount is insufficient in melting degree, and the flow is insufficient, so that the cutting amounts at two sides of the welding line are not fused, and a slit is opened to form a furrow.

Prevention measures are as follows: firstly, a proper stirring head shape is selected to ensure the full stirring of the welded material in the welding process. And secondly, proper technological parameters are selected to ensure sufficient welding heat input.

The first preventive measure is to manually configure the corresponding stirring pin in advance, but to select proper technological parameters, so that the sufficient welding heat input is ensured, and the rotation speed and temperature conduction of the stirring pin and the shaft shoulder are required to be detected and controlled in real time.

The terminal firstly generates two stirring pin fitting lines on a real-time detection image according to the diameter of a stirring pin and a welding path of the stirring friction welding module, and a region between the two stirring pin fitting lines is a stirring pin region. The principle of the generation is the same as that of step 204, and will not be described here again.

Then, the terminal takes the fitting line of the stirring needle as an edge, a furrow search frame is generated in the stirring needle area, the furrow search frame is segmented, square boxes of K pixels are generated, and K is an odd number larger than 1;

at this time, gray analysis is not needed according to the gradual change boxes, because accurate positioning effect is not needed at this time, the terminal calculates the gray average value of each square box and determines the gray change box, and the difference between the gray average value of the gray change box and the gray average value of more than one adjacent square box is larger than a preset gray difference threshold. Only the gray average value of the box is compared, the box with gray change is mainly determined, and the existence of the furrow defect is basically determined, because the detection area is the stirring pin defect, and the defect exists in the area and only has the furrow defect.

At the moment, the terminal needs to generate a furrow area according to the central point coordinates of the gray level change square frame, and the defect that the workpiece to be welded has furrows can be determined. And reporting through the system.

212. Generating search frames at equal distances on two sides of a weld edge fit line by taking the weld edge fit line as a common edge, wherein the weld edge fit line is used for dividing a workpiece area to be welded on a real-time detection image into a weld area and a non-weld area, the search frames on the weld area side are first defect search frames, and the search frames on the non-weld area side are second defect search frames;

213. dividing the first defect search frame and the second defect search frame to generate square boxes of M pixels, wherein M is an integer greater than 1;

214. carrying out gray level analysis on each square box to generate a gray level analysis result;

215. determining the center points of the gradual change boxes in the first defect search box and the second defect search box according to the gray level analysis result, and analyzing the coordinates of the center points, wherein the gradual change boxes are square boxes with pixel gray level gradual change conditions;

216. when irregular linear central lattices exist in the first defect search frame and the second defect search frame, determining that burrs appear on the workpiece to be welded;

In this embodiment, steps 212 to 216 are similar to steps 105 to 109 described above, and will not be described here.

217. When the first defect search frame only has an arc-shaped central lattice and the second defect search frame has a block-shaped central lattice, determining that the workpiece to be welded has the defect of flash;

in addition to burr rescue and furrow defect, flash defect is likely to occur in friction stir welding.

The flash defect is represented by metal rolled on both sides of the welded seam after welding, and is called flash.

The flash is formed because during friction stir welding, metal is extruded around the shoulder of the stirring tool to form the flash. Different from burrs, the workpiece to be welded in the deburring process is better in melting condition, and the main reason is that the shaft shoulder is excessively pressed down in the operation process, so that the melted part is extruded to form a block to cover a non-welding area, and the burrs are used for throwing out the part which is excessively melted to form a strip shape.

The prevention measures are that the pressing amount of the stirring tool is reasonably controlled in the welding process, the pressing amount is not excessively large, the thickness of the welded material is basically consistent, and the malpractice in the assembly process is avoided.

When the first defect search box only has the arc-shaped central lattice, it can be determined that no other lattice exists outside the welding area (except the stirring pin area) at the moment, and the second defect search box has the block-shaped central lattice, then it can be determined to belong to the flash defect.

218. The irradiation plate is lightened, the acquisition camera is controlled to shoot a welded section of area, and a depth detection image is obtained;

219. marking a weld edge fitting line on the depth detection image;

220. generating a depth search area on the depth detection image according to a weld edge fitting line, wherein the weld edge fitting line is the central line of the depth search area, and divides the depth search area into an inner weld area and an outer weld area;

221. dividing the depth search area to generate square boxes of L pixels, carrying out gray analysis on each square box, wherein L is an integer greater than 1;

222. determining a first distance from a gradient box with the largest weld edge fitting line in the weld inner area and a second distance from the gradient box with the largest weld edge fitting line in the weld outer area;

223. and when the total distance of the first distance and the second distance is larger than the preset distance, determining that the workpiece to be welded has the defect of concave surface.

The surface dishing defect is a phenomenon that the front surface of the weld joint is lower than the surface of the base material after the end of the stirring welding, and is called surface dishing.

The formation cause of the surface dishing defect: in the stirring welding process, the surface concave phenomenon is generated on the welding line due to the fact that a stirring tool is pressed down. This phenomenon is an inherent feature of friction stir welding and does not affect the performance of the welded joint when the amount of sag is within a reasonable range.

The prevention measure is to control the pressing amount of the stirring tool, and the pressing amount is contraindicated to be too large.

In this embodiment, the terminal performs auxiliary detection through the irradiation plates on two sides of the light emitting element, and an included angle exists between the irradiation plates and the light emitting element. At this time, the irradiation plate is required to be lightened, and the acquisition camera is controlled to shoot a welded section of area, so that a depth detection image is obtained. The purpose of illuminating the illuminated panel is to increase the shadow area if there is a surface sag where there are corners between the weld and the unwelded area where the light is not illuminated. The capturing camera still captures the shadows along the welding path from top to bottom.

The terminal marks a weld edge fitting line on the depth detection image. And generating a depth search area on the depth detection image according to the weld edge fitting line, wherein the weld edge fitting line is the central line of the depth search area, and divides the depth search area into an inner weld area and an outer weld area. Namely, a symmetrical rectangle is drawn by taking the weld edge fitting line as a central line. If surface dishing occurs, the rectangle (depth search area) encompasses the entire shadow portion.

The terminal segments the depth search area to generate square boxes of L pixels, gray analysis is carried out on each square box, and L is an integer greater than 1. And determining a first distance from the gradient box with the largest weld edge fitting line in the inner region of the weld and a second distance from the gradient box with the largest weld edge fitting line in the outer region of the weld. The purpose is to find the width of the shadow area, which is very small if no surface dip is present, and which is larger than a preset value if present. When the total distance of the first distance and the second distance is larger than the preset distance, the terminal can determine that the surface of the workpiece to be welded is concave.

Next, the friction stir welding module is started, and the friction stir welding module, the light emitting element, and the acquisition camera are controlled to move along the welding path in sequence. When the acquisition camera moves to a preset point, the light-emitting element is controlled to emit illumination light perpendicular to the workpiece to be welded, and the acquisition camera is controlled to shoot a welded section of area, so that a real-time detection image is obtained. Each point location is collected at least once, an image with the cleanest surface is selected as a real-time detection image, wherein the irradiation light of the light-emitting element is determined according to the color of the workpiece to be welded, and the pixel points which can enable the non-welded area of the workpiece to be welded to generate larger contrast with the welding line and enable the two to be fed back clearly in the real-time detection image are mainly achieved.

And generating a linear welding seam edge on the real-time detection image according to the diameter of the shaft shoulder and the welding path of the friction stir welding module. Equidistant rectangular frames are generated on the edges of the straight welding lines, the rectangular frames are cut, square frames of N x N pixel points are generated, the rectangular frames comprise welding line areas and non-welding line areas, and N is an odd number larger than 1. And carrying out gray level analysis on each square frame, determining the center point of the gradual change frame, and screening non-defective rectangular frames with the dispersion degree lower than a preset value from all the rectangular frames according to the coordinates of the center point. A weld edge fit line is generated on the real-time inspection image from the center point of each defect-free rectangular frame.

And generating two stirring pin fitting lines on the real-time detection image according to the diameter of the stirring pin and the welding path of the stirring friction welding module, wherein the area between the two stirring pin fitting lines is a stirring pin area. And generating a furrow search frame in the stirring needle area by taking the stirring needle fitting line as an edge, and cutting the furrow search frame to generate square boxes of K pixels, wherein K is an odd number larger than 1. And carrying out gray average value calculation on each square box, and determining a gray change box, wherein the difference between the gray average value of the gray change box and the gray average value of more than one adjacent square boxes is larger than a preset gray difference threshold. And generating a furrow area according to the central point coordinates of the gray scale change square frame, and determining that the workpiece to be welded has a furrow defect.

And generating search frames at equal distances on two sides of the weld seam edge fitting line by taking the weld seam edge fitting line as a common edge, wherein the weld seam edge fitting line is used for dividing a workpiece area to be welded on the real-time detection image into a weld joint area and a non-weld joint area, the search frames on the side of the weld joint area are first defect search frames, and the search frames on the side of the non-weld joint area are second defect search frames. That is, if a burr defect occurs, a splash-state solidification waste is mainly generated in the non-weld region, and the splash-state solidification waste is in a long and thin strip shape, and the splash-state solidification waste is irradiated to the welding region and generates a strip-shaped shadow in the welding region and also generates a strip-shaped shadow in the non-welding region because the light emitting device is longer than the welding seam. And then the first defect search frame and the second defect search frame are segmented to generate square boxes of M pixels, wherein M is an integer greater than 1. And carrying out gray level analysis on each square box to generate a gray level analysis result. And determining the central points of the gradual change boxes in the first defect search box and the second defect search box according to the gray level analysis result, analyzing coordinates of the central points, wherein the gradual change boxes are square boxes with pixel gray level gradual change conditions. When linear central lattices exist in the first defect search frame and the second defect search frame, determining that burrs appear on the workpieces to be welded. Because the workpiece to be welded in the conventional state forms arc-shaped grains in the welding area, and the non-welding area is mainly cut grains or is free of grains (straight lines or smooth and free of lattices), after gray analysis is carried out, the central lattices generated in the welding area can show regular arc-shaped central lattices, and the non-welding area is regular straight central lattices or almost free of lattices. Therefore, the first defect search frame and the second defect search frame are segmented to generate square boxes of M pixels, gray algorithm analysis is carried out on each square box to generate gray analysis results, gradient boxes and gradient box center point coordinates in the first defect search frame and the second defect search frame are determined according to the gray analysis results, center point coordinates are analyzed, algorithm fitting or marking can be carried out on the coordinates of the center points, and by comparing preset normal lattices, what center lattice appears can be determined. If a linear center lattice exists in both the first defect search box and the second defect search box, it is determined that the workpiece to be welded has a burr defect in the welding area.

And (5) lighting the irradiation plate, and controlling the acquisition camera to shoot a welded section of area to acquire a depth detection image. And marking a weld edge fitting line on the depth detection image. And generating a depth search area on the depth detection image according to the weld edge fitting line, wherein the weld edge fitting line is the central line of the depth search area, and divides the depth search area into an inner weld area and an outer weld area. And cutting the depth search area to generate square boxes of L pixels, carrying out gray analysis on each square box, wherein L is an integer greater than 1. And determining a first distance from the gradient box with the largest weld edge fitting line in the inner region of the weld and a second distance from the gradient box with the largest weld edge fitting line in the outer region of the weld. When the total distance of the first distance and the second distance is larger than the preset distance, determining that the workpiece to be welded has the defect of concave surface

According to the scheme, the gray level algorithm and the dot matrix comparison algorithm are utilized, manual detection is not needed, detection is carried out in a pure automatic mode, and the defect detection efficiency of friction stir welding is improved to a great extent.

Secondly, the embodiment also provides a method for generating the weld seam edge fitting line, which mainly aims at the region with the straight welding path, and an accurate critical point is found by detecting the gradual change condition of the square box, so that the more accurate weld seam edge fitting line is generated. The more accurate weld seam edge fitting line can accurately divide the area, and the defect detection precision is improved.

Secondly, in the embodiment, an effective and high-speed detection method is provided for the furrow defect, and the furrow defect area is obtained by selecting the stirring pin area and carrying out gray level detection and analysis range, so that the defect detection efficiency is improved.

Secondly, in the embodiment, a method for detecting the flash defect is further provided, whether the flash defect occurs can be determined by comparing the central dot matrix, and the defect detection efficiency is improved.

In the embodiment, a method for detecting the concave defect of the surface is further provided, and the width of the shadow is analyzed by irradiating the plate, so that whether the concave defect is generated or not is determined, and the defect detection efficiency is improved.

Referring to fig. 5, 6 and 7, a third embodiment of a method for detecting a defect in friction stir welding is provided, which includes:

301. Placing a workpiece to be welded on a bearing platform, setting initial positions of a friction stir welding module, a light-emitting element and an acquisition camera according to a preset welding path, wherein the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module;

302. starting a friction stir welding module, and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along a welding path;

303. when the acquisition camera moves to a preset point, controlling the light-emitting element to emit illumination light perpendicular to a workpiece to be welded, and controlling the acquisition camera to shoot a welded section of area to acquire a real-time detection image;

304. Generating an arc welding seam edge on a real-time detection image according to the diameter of a shaft shoulder of the friction stir welding module and a welding path;

305. generating equidistant rectangular frames on the edges of arc welding seams, cutting the rectangular frames to generate square frames of J pixels, wherein the rectangular frames comprise welding seam areas and non-welding seam areas, and J is an odd number larger than 1;

306. carrying out gray average value calculation on each square box in the rectangular box, determining square boxes with gray gradient changes, and carrying out curve fitting according to the center point coordinates of the square boxes with gray gradient changes to generate an initial arc line;

307. Screening out extreme points of outliers according to the initial arc line, and performing curve fitting through the residual coordinates of the central points to generate a weld edge fitting line;

in the embodiment, when the link path is an arc, the terminal generates an arc welding seam edge on the real-time detection image according to the shaft shoulder diameter of the friction stir welding module and the welding path. And generating equidistant rectangular frames on the edges of arc welding seams, cutting the rectangular frames to generate square frames of J-J pixel points, wherein the rectangular frames comprise welding seam areas and non-welding seam areas, and J is an odd number larger than 1. And then carrying out gray average value calculation on each square box in the rectangular box, determining the square box with gray gradient change, and carrying out curve fitting according to the center point coordinates of the square box with gray gradient change to generate an initial arc line. For use in friction stir welding, if the welding path is curved, the likelihood of edge chipping is increased, which is most likely to be detected, so the terminal screens off outlier points according to the initial arc and curve fits through the remaining center point coordinates to generate a weld edge fit line. Thereby removing these debris and increasing the accuracy of the weld edge fit line.

308. Generating two stirring pin fitting lines on the real-time detection image according to the diameter of the stirring pin and the welding path of the stirring friction welding module, wherein the area between the two stirring pin fitting lines is a stirring pin area;

309. generating a furrow search frame in a stirring needle area by taking a stirring needle fitting line as an edge, and cutting the furrow search frame to generate square boxes of K pixels, wherein K is an odd number larger than 1;

310. carrying out gray average value calculation on each square box, and determining a gray change box, wherein the difference between the gray average value of the gray change box and the gray average value of more than one adjacent square boxes is larger than a preset gray difference threshold;

311. generating a furrow area according to the central point coordinates of the gray level change square frame, and determining that the workpiece to be welded has a furrow defect;

312. generating search frames at equal distances on two sides of a weld edge fit line by taking the weld edge fit line as a common edge, wherein the weld edge fit line is used for dividing a workpiece area to be welded on a real-time detection image into a weld area and a non-weld area, the search frames on the weld area side are first defect search frames, and the search frames on the non-weld area side are second defect search frames;

313. Dividing the first defect search frame and the second defect search frame to generate square boxes of M pixels, wherein M is an integer greater than 1;

314. carrying out gray level analysis on each square box to generate a gray level analysis result;

315. determining the center points of the gradual change boxes in the first defect search box and the second defect search box according to the gray level analysis result, and analyzing the coordinates of the center points, wherein the gradual change boxes are square boxes with pixel gray level gradual change conditions;

316. when linear central lattices exist in the first defect search frame and the second defect search frame, determining that burrs appear on the workpiece to be welded;

317. when the first defect search frame only has an arc-shaped central lattice and the second defect search frame has a block-shaped central lattice, determining that the workpiece to be welded has the defect of flash;

318. the irradiation plate is lightened, the acquisition camera is controlled to shoot a welded section of area, and a depth detection image is obtained;

319. marking a weld edge fitting line on the depth detection image;

320. generating a depth search area on the depth detection image according to a weld edge fitting line, wherein the weld edge fitting line is the central line of the depth search area, and divides the depth search area into an inner weld area and an outer weld area;

321. Dividing the depth search area to generate square boxes of L pixels, carrying out gray analysis on each square box, wherein L is an integer greater than 1;

322. determining a first distance from a gradient box with the largest weld edge fitting line in the weld inner area and a second distance from the gradient box with the largest weld edge fitting line in the weld outer area;

323. and when the total distance of the first distance and the second distance is larger than the preset distance, determining that the workpiece to be welded has the defect of concave surface.

In this embodiment, steps 308 to 323 are similar to steps 208 to 223 described above, and will not be described here.

And generating an arc welding seam edge on the real-time detection image according to the shaft shoulder diameter and the welding path of the friction stir welding module. Equidistant rectangular frames are generated on the edges of arc welding seams, the rectangular frames are cut, square frames of J pixels are generated, welding seam areas and non-welding seam areas are included in the rectangular frames, and J is an odd number larger than 1. And carrying out gray average value calculation on each square box in the rectangular box, determining the square box with gray gradient change, and carrying out curve fitting according to the center point coordinates of the square box with gray gradient change to generate an initial arc line. And screening out the outlier points according to the initial arc line, and performing curve fitting through the residual coordinates of the central points to generate a weld edge fitting line.

Secondly, the embodiment also provides a method for generating the weld seam edge fitting line, which mainly aims at the area with the arc welding path, finds an accurate critical point by detecting the gradual change condition of the square box, and removes unreasonable point positions by screening out extreme points of outliers according to the initial arc, so as to generate the more accurate weld seam edge fitting line. The more accurate weld seam edge fitting line can accurately divide the area, and the defect detection precision is improved.

Referring to fig. 8, an embodiment of a friction stir welding defect detection system is provided, comprising:

the placing unit 401 is configured to place a workpiece to be welded on the receiving platform, and set an initial position of the friction stir welding module, the light emitting element and the acquisition camera according to a preset welding path, where the length of the light emitting element is greater than the diameter of a shaft shoulder of the friction stir welding module;

a starting unit 402, configured to start the friction stir welding module and control the friction stir welding module, the light emitting element, and the acquisition camera to sequentially move along a welding path;

the first acquisition unit 403 is configured to control the light emitting element to emit illumination light perpendicular to the workpiece to be welded when the acquisition camera moves to a preset point, and control the acquisition camera to shoot a welded section of area, so as to obtain a real-time detection image;

a first generating unit 404, configured to generate a weld edge fitting line on the real-time detection image according to the shoulder diameter and the welding path of the friction stir welding module;

a second generating unit 405, configured to generate search boxes at equal distances on both sides of a weld edge fitting line with the weld edge fitting line as a common edge, where the weld edge fitting line is configured to divide a workpiece area to be welded on a real-time detection image into a weld area and a non-weld area, the search boxes on the weld area side are first defect search boxes, and the search boxes on the non-weld area side are second defect search boxes;

A third generating unit 406, configured to segment the first defect search box and the second defect search box, generate square boxes of m×m pixel points, where M is an integer greater than 1;

a fourth generating unit 407, configured to perform gray level analysis on each square box, and generate a gray level analysis result;

a first determining unit 408, configured to determine a center point of a gradient box in the first defect search box and the second defect search box according to a gray level analysis result, and perform analysis of coordinates of the center point, where the gradient box is a square box with a pixel gray level gradient condition;

and a second determining unit 409, configured to determine that the workpiece to be welded has a burr defect when the linear center lattice exists in both the first defect search box and the second defect search box.

Referring to FIG. 9, another embodiment of a friction stir welding defect detection system is provided, comprising:

the placing unit 501 is configured to place a workpiece to be welded on the receiving platform, and set an initial position of the friction stir welding module, the light emitting element and the acquisition camera according to a preset welding path, where the length of the light emitting element is greater than the diameter of a shaft shoulder of the friction stir welding module;

The starting unit 502 is used for starting the friction stir welding module and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along a welding path;

the first acquisition unit 503 is configured to control the light emitting element to emit illumination light perpendicular to the workpiece to be welded when the acquisition camera moves to a preset point, and control the acquisition camera to shoot a welded section of area, so as to obtain a real-time detection image;

a first generating unit 504, configured to generate a weld edge fitting line on the real-time detection image according to the shoulder diameter and the welding path of the friction stir welding module;

optionally, the welding path is a straight line;

the first generation unit 504 includes:

Optionally, the welding path is an arc;

the first generation unit 504 includes:

A seventh generating unit 505, configured to generate two pin fitting lines on the real-time detection image according to the pin diameter and the welding path of the friction stir welding module, where a region between the two pin fitting lines is a pin region;

An eighth generating unit 506, configured to generate a furrow search frame in the stirring pin area with the stirring pin fitting line as a side, and segment the furrow search frame to generate square boxes with k×k pixels, where K is an odd number greater than 1;

a fifth determining unit 507, configured to perform a gray average value calculation on each square box, and determine a gray change box, where a difference between a gray average value of the gray change box and a gray average value of more than one neighboring square boxes is greater than a preset gray difference threshold;

a sixth determining unit 508, configured to generate a furrow area according to the coordinates of the center point of the gray scale change box, and determine that a defect of a furrow appears in the workpiece to be welded;

a second generating unit 509, configured to generate search frames equidistantly on two sides of a weld edge fitting line with the weld edge fitting line as a common edge, where the weld edge fitting line is used to divide a workpiece area to be welded on a real-time detection image into a weld area and a non-weld area, the search frame on the weld area side is a first defect search frame, and the search frame on the non-weld area side is a second defect search frame;

a third generating unit 510, configured to segment the first defect search box and the second defect search box, generate square boxes of m×m pixel points, where M is an integer greater than 1;

A fourth generation unit 511 for performing gray level analysis on each square block to generate a gray level analysis result;

a first determining unit 512, configured to determine a center point of a gradient frame in the first defect search frame and the second defect search frame according to a gray level analysis result, and perform analysis of coordinates of the center point, where the gradient frame is a square frame with a pixel gray level gradient condition;

a second determining unit 513, configured to determine that the workpiece to be welded has a burr defect when the linear center dot matrix exists in both the first defect search frame and the second defect search frame;

a seventh determining unit 514, configured to determine that the workpiece to be welded has a flash defect when the first defect search box has only the arc-shaped central lattice and the second defect search box has the block-shaped central lattice;

the second acquisition unit 515 is used for lighting the irradiation plate, controlling the acquisition camera to shoot a welded section of area, and acquiring a depth detection image;

a marking unit 516 for marking a weld edge fit line on the depth detection image;

a fifth generating unit 517, configured to generate a depth search area on the depth detection image according to a weld edge fitting line, where the weld edge fitting line is a center line of the depth search area, and the weld edge fitting line divides the depth search area into an intra-weld area and an outer-weld area;

A sixth generating unit 518, configured to segment the depth search area, generate square boxes of l×l pixels, perform gray analysis on each square box, and L is an integer greater than 1;

a third determining unit 519 for determining a first distance from a gradient box with the largest weld edge fit line in the region inside the weld and a second distance from the gradient box with the largest weld edge fit line in the region outside the weld;

and a fourth determining unit 520, configured to determine that the workpiece to be welded has a surface concave defect when the total distance of the first distance plus the second distance is greater than the preset distance.

Referring to fig. 10, the present application provides an electronic device, including:

a processor 601, a memory 603, an input-output unit 602, and a bus 604.

The processor 601 is connected to a memory 603, an input-output unit 602, and a bus 604.

The memory 603 stores a program that the processor 601 invokes to perform the method of detecting a defect in friction stir welding as in fig. 1, 2, 3, 4, 5, 6, and 7.

The present application provides a computer-readable storage medium having a program stored thereon, which when executed on a computer performs a method of detecting a defect in friction stir welding as in fig. 1, 2, 3, 4, 5, 6, and 7.

It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.

In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (RAM, random access memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Claims

1. A method of detecting a defect in friction stir welding, comprising:

placing a workpiece to be welded on a bearing platform, and setting initial positions of a friction stir welding module, a light-emitting element and an acquisition camera according to a preset welding path, wherein the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module;

starting the friction stir welding module and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along the welding path;

when the acquisition camera moves to a preset point, controlling the light-emitting element to emit illumination light perpendicular to the workpiece to be welded, and controlling the acquisition camera to shoot a welded section of area to acquire a real-time detection image;

generating a weld joint edge fitting line on the real-time detection image according to the shaft shoulder diameter of the friction stir welding module and the welding path;

generating search frames at equal distances on two sides of the weld seam edge fitting line by taking the weld seam edge fitting line as a common edge, wherein the weld seam edge fitting line is used for dividing a workpiece area to be welded on the real-time detection image into a weld seam area and a non-weld seam area, the search frames on the side of the weld seam area are first defect search frames, and the search frames on the side of the non-weld seam area are second defect search frames;

determining the center points of the gradual change boxes in the first defect search box and the second defect search box according to the gray level analysis result, and carrying out coordinate analysis on the center points, wherein the gradual change boxes are square boxes with pixel gray level gradual change conditions;

when the first defect search frame and the second defect search frame both have irregular linear central lattices, determining that burrs appear on the workpiece to be welded.

2. The defect detection method of claim 1, wherein the welding path is a straight line;

the generating a weld edge fitting line on the real-time detection image according to the shoulder diameter of the friction stir welding module and the welding path comprises the following steps:

generating a linear welding seam edge on the real-time detection image according to the shaft shoulder diameter of the friction stir welding module and the welding path;

generating equidistant rectangular frames on the edges of the linear welding seams, cutting the rectangular frames to generate square frames of N x N pixel points, wherein the rectangular frames comprise welding seam areas and non-welding seam areas, and N is an odd number larger than 1;

a weld edge fit line is generated on the real-time detection image based on the center point of each defect-free rectangular frame.

3. The defect detection method of claim 1, wherein the welding path is an arc;

generating arc welding seam edges on the real-time detection image according to the shaft shoulder diameter of the friction stir welding module and the welding path;

generating equidistant rectangular frames on the arc welding seam edges, and cutting the rectangular frames to generate square frames of J-J pixel points, wherein the rectangular frames comprise welding seam areas and non-welding seam areas, and J is an odd number larger than 1;

And screening out extreme points of the outliers according to the initial arc line, and performing curve fitting through the residual coordinates of the central points to generate a weld joint edge fitting line.

4. The defect detection method according to claim 1, wherein the light emitting element is provided with irradiation plates on both sides thereof, respectively, and an included angle exists between the irradiation plates and the light emitting element;

after generating a weld edge fit line on the real-time detection image according to the shoulder diameter of the friction stir welding module and the welding path, the defect detection method further includes:

marking the weld edge fitting line on the depth detection image;

generating a depth search area on the depth detection image according to the weld edge fitting line, wherein the weld edge fitting line is a central line of the depth search area and divides the depth search area into an inner weld area and an outer weld area;

Determining a first distance from a gradient box with the largest weld edge fit line in the weld joint inner area and a second distance from the gradient box with the largest weld edge fit line in the weld joint outer area;

and when the total distance of the first distance and the second distance is larger than a preset distance, determining that the workpiece to be welded has the defect of concave surface.

5. The defect detection method according to any one of claims 1 to 4, characterized in that after controlling the light emitting element to emit irradiation light perpendicular to the workpiece to be welded and controlling the acquisition camera to take a photograph of a welded segment of the region, the defect detection method further comprises:

generating two stirring pin fitting lines on the real-time detection image according to the diameter of the stirring pin of the stirring friction welding module and the welding path, wherein the area between the two stirring pin fitting lines is a stirring pin area;

generating a furrow search frame in the stirring needle area by taking the stirring needle fitting line as an edge, and cutting the furrow search frame to generate square boxes of K pixels, wherein K is an odd number larger than 1;

and generating a furrow area according to the central point coordinates of the gray level change square frame, and determining that the workpiece to be welded has a furrow defect.

6. The defect detection method according to any one of claims 1 to 4, characterized in that after determining the center points of gradation blocks in the first defect search box and the second defect search box from the gradation analysis result and performing analysis of center point coordinates, the defect detection method further comprises:

7. A friction stir welding defect detection system comprising:

the device comprises a placing unit, a friction stir welding module, a light-emitting element and an acquisition camera, wherein the placing unit is used for placing a workpiece to be welded on a receiving platform, and setting initial positions of the friction stir welding module, the light-emitting element and the acquisition camera according to a preset welding path, and the length of the light-emitting element is larger than the diameter of a shaft shoulder of the friction stir welding module;

The starting unit is used for starting the friction stir welding module and controlling the friction stir welding module, the light-emitting element and the acquisition camera to sequentially move along the welding path;

the first acquisition unit is used for controlling the light-emitting element to emit illumination light perpendicular to the workpiece to be welded when the acquisition camera moves to a preset point position, and controlling the acquisition camera to shoot a welded section of area to acquire a real-time detection image;

the first generation unit is used for generating a welding seam edge fitting line on the real-time detection image according to the shaft shoulder diameter of the friction stir welding module and the welding path;

the second generation unit is used for generating search frames at equal distance on two sides of the welding seam edge fitting line by taking the welding seam edge fitting line as a common edge, wherein the welding seam edge fitting line is used for dividing a workpiece area to be welded on the real-time detection image into a welding seam area and a non-welding seam area, the search frames on the side of the welding seam area are first defect search frames, and the search frames on the side of the non-welding seam area are second defect search frames;

a third generating unit, configured to segment the first defect search box and the second defect search box, and generate square boxes of m×m pixels, where M is an integer greater than 1;

the first determining unit is used for determining the center points of the gradual change boxes in the first defect searching box and the second defect searching box according to the gray level analysis result, and carrying out coordinate analysis on the center points, wherein the gradual change boxes are square boxes with the gradual change condition of pixel gray levels;

and the second determining unit is used for determining that the workpiece to be welded has the defect of burrs when the first defect searching frame and the second defect searching frame have irregular linear central lattices.

8. The defect detection system of claim 7, wherein the welding path is a straight line;

the first generation unit includes:

9. An electronic device, comprising a processor, a memory, an input/output unit and a bus, wherein the processor is connected with the memory, the input/output unit and the bus;

the memory holds a program, and the processor calls the program to execute the friction stir welding defect detecting method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a program which, when executed on a computer, performs the friction stir welding defect detection method according to any one of claims 1 to 6.