KR102680136B1 - Manufacturing method for a cowl cross - Google Patents

Abstract

The present invention relates to a method for inspecting the welding quality of cowl cross parts. More specifically, the present invention relates to a method of inspecting the welding quality of cowl cross parts without cutting the weld part, deviating from the classic method of inspecting the weld quality of the cut surface after cutting the weld part. Welding quality can be inspected, and as a result, the manpower and time required for inspection can be significantly reduced compared to the past, so it is related to a cowl cross parts welding quality inspection method that can easily and conveniently perform complete inspection of products as needed. .

Description

Cowl cross parts welding quality inspection method {Manufacturing method for a cowl cross}

The present invention relates to a method for inspecting the welding quality of cowl cross parts. More specifically, the present invention relates to a method of inspecting the welding quality of cowl cross parts without cutting the weld part, deviating from the classic method of inspecting the weld quality of the cut surface after cutting the weld part. Welding quality can be inspected, and as a result, the manpower and time required for inspection can be significantly reduced compared to the past, so it is related to a cowl cross parts welding quality inspection method that can easily and conveniently perform complete inspection of products as needed. .

A cowl cross (or cowl cross part) is a body made by assembling and manufacturing various parts (brackets, bolts, etc.) made by pressing a certain steel pipe by welding. It refers to a device that has been achieved.

The welding at this time is mainly arc welding such as CO2 and CMT, and it is known that about 20 to 30 small parts are assembled at each location.

Meanwhile, the quality of melt welding is determined by checking not only the apparent bead shape but also the shape of the weld cut cross section.

At this time, the existing cowl cross parts welding quality inspection method collects and cuts some samples from the cowl cross parts to check the quality of the welding, but the quality of unconfirmed parts is assessed using the sample test results, which not only reduces the reliability of the results but also makes them difficult to pass on. There is a problem that inspection is impossible and requires a lot of manpower and time for inspection.

Korea Intellectual Property Office Application No. 10-2015-0181567

The purpose of the present invention is to deviate from the classic method of inspecting the welding quality of the cut surface after cutting the welded portion, and to inspect the welding quality of the cowl cross part without cutting the welded portion, thereby reducing the manpower required for inspection. The aim is to provide a cowl cross parts welding quality inspection method that can significantly reduce the time compared to the conventional method and allow for easy and simple inspection of all products as needed.

The above purpose is a cowl cross parts welding device unit that welds cowl cross parts through a predetermined welding process; and a cowl cross finished parts welding quality inspection unit that inspects the welding quality of the cowl cross finished parts, which are welded assembled products, after going through the welding process of the cowl cross parts welding device unit, wherein the cowl cross parts welding device unit includes a plurality of parts. Dog moving devices; first and second movable loading devices of a dual structure, which are each movably mounted on the moving devices and are loaded with cowl cross parts seated on each device; The first and second cowls are provided in common between the first and second mobile loading devices, so that welding lines on the first and second cowl cross parts that are respectively seated on the first and second mobile loading devices can be confirmed. A common sensor robot equipped with a common laser vision sensor that scans cross parts; It is provided in common between the first and second mobile loading devices adjacent to the common sensor robot, and includes a common robot welder that performs compensation welding on the first and second cowl cross parts, and completes the cowl cross. The component welding quality inspection unit is provided with a base plate and a plurality of holding units provided at each location on the base plate to fix the assembled product for inspection of the assembled product after welding is completed. The final part where the assembled product is seated is provided. inspection jig device; an inspection laser vision sensor that scans the assembled product placed on the final inspection jig device in three dimensions using a common laser vision sensor; and a system controller that controls to inspect the welding quality of the cowl cross finished part based on the sensing value of the inspection laser vision sensor.

The system controller extracts the profile of the weld bead of the weld part of the cowl cross finished part detected by the inspection laser vision sensor, then classifies and preprocesses the data stored in the database based on the profile of the extracted weld bead, Using an artificial neural network, the tensile strength of the corresponding weld can be controlled to predict it.

The system controller controls the operation of the public robot welder and provides a plurality of welding data including welding torch angle, welding speed, welding current, and welding voltage during compensation welding of the cowl cross part by the public robot welder. It is possible to control the use of optimized welding correction condition data created based on this.

Meanwhile, the above object is a cowl cross parts welding step of welding cowl cross parts through a predetermined welding process by the cowl cross parts welding device unit; And a cowl cross finished parts welding quality inspection step of inspecting the welding quality of the cowl cross finished parts, which are welded assembled products, through the cowl cross finished parts welding quality inspection unit. The cowl cross parts welding step includes a plurality of moving ( A moving device installation step of installing a moving device; A mobile loading device mounting step of mounting a mobile loading device on each of the moving devices to form a dual structure; A first and second cowl cross parts loading step of loading cowl cross parts by seating them on dual first and second movable loading devices that are mounted on the moving device and capable of moving in position. ; Scanning the first cowl cross part loaded on the first mobile loading device using a common laser vision sensor mounted on a common sensor robot that commonly operates between the first and second mobile loading devices. First cowl cross component scanning step using a common laser vision sensor; A first cowl cross part weld line confirmation step of checking the weld line on the first cowl cross part loaded on the first mobile loading device based on scan information; A welding condition changing step of a common robot welder that changes the welding conditions of a common robot welder disposed adjacent to the common laser vision sensor based on welding line information of the identified first cowl cross part; And a first cowl cross part welding process step of performing compensation welding on the first cowl cross part using the common robot welder after the welding conditions are changed. A cowl cross part welding quality inspection method, characterized in that it includes It is also achieved by

The cowl cross finished part welding quality inspection step includes: an assembled product seating step of seating the cowl cross finished part on a predetermined final inspection jig device; A weld bead detection step of detecting a weld bead for a welded portion of the cowl cross finished part using a laser vision sensor; Weld bead profile extraction step of extracting the profile of the weld bead of the detected weld area; A data classification and preprocessing step of classifying and preprocessing data stored in the database based on the profile of the extracted weld bead; and a tensile strength prediction step of predicting the tensile strength of the corresponding welded portion using an artificial neural network, wherein when the common robot welder performs compensation welding on the first cowl cross part, it is simultaneously mounted on the common sensor robot. After the common laser vision sensor scans the second cowl cross part loaded on the second mobile loading device, the weld line on the second cowl cross part loaded on the second mobile loading device can be confirmed based on the scan information. there is.

According to the present invention, it is possible to inspect the welding quality of cowl cross parts without cutting the welded part, breaking away from the classical method of inspecting the welding quality of the cut surface after cutting the welded part, which reduces the manpower and manpower required for inspection. The time can be significantly reduced compared to before, which has the effect of enabling easy and simple inspection of all products as needed.

1 is a schematic layout of a cowl cross component welding quality inspection system according to an embodiment of the present invention.
Figures 2 and 3 are operational structural diagrams of the mobile loading device area in the cowl cross component welding quality inspection system of Figure 1.
Figure 4 is a perspective view of the first and second cowl cross parts loaded into the first and second mobile loading devices.
Figure 5 is a structural diagram of the final inspection jig device area in the cowl cross component welding quality inspection system of Figure 1.
FIG. 6 is a control block diagram of the cowl cross component welding quality inspection system of FIG. 1.
Figure 7 is a flowchart of a cowl cross component welding quality inspection method according to an embodiment of the present invention.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings.

However, the present invention is not limited to the embodiments disclosed below and will be implemented in various different forms.

In this specification, the present embodiments are provided to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention. And the present invention is only defined by the scope of the claims.

Accordingly, in some embodiments, well-known components, well-known operations and well-known techniques are not specifically described in order to avoid ambiguous interpretation of the present invention.

Like reference numerals refer to like elements throughout the specification. And the terms used (mentioned) in this specification are for describing embodiments and are not intended to limit the present invention.

In this specification, singular forms also include plural forms unless specifically stated in the phrase. Additionally, components and operations (operations) referred to as 'include (or, provided with)' do not exclude the presence or addition of one or more other components and operations.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings that can be commonly understood by those skilled in the art to which the present invention pertains.

Additionally, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless they are defined.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

Figure 1 is a schematic layout of a cowl cross parts welding quality inspection system according to an embodiment of the present invention, Figures 2 and 3 are operation structure diagrams of the mobile loading device area in the cowl cross parts welding quality inspection system of Figure 1; 4 is a perspective view of the first and second cowl cross parts loaded into the first and second mobile loading devices, and Figure 5 is a structural diagram of the final inspection jig device area in the cowl cross parts welding quality inspection system of Figure 1. 6 is a control block diagram of the cowl cross parts welding quality inspection system of FIG. 1, and FIG. 7 is a flowchart of the cowl cross parts welding quality inspection method according to an embodiment of the present invention.

Referring to these drawings, the present invention deviates from the classical method of inspecting the welding quality of the cut surface after cutting the welded portion, and instead inspects the welding quality of the cowl cross (10a, 10b, 10') parts without cutting the welded portion. It can be inspected, and as a result, the manpower and time required for inspection can be significantly reduced compared to before, making it possible to easily and conveniently perform a full inspection of products as needed. Hereinafter, the parts before welding are called cowl cross (10a, 10b) parts, and the parts after welding are called cowl cross finished parts (10').

The scope of the present invention, which can provide these effects, can be applied to the cowl cross parts welding quality inspection system as shown in FIGS. 1 to 6 and the cowl cross parts welding quality inspection method as shown in FIG. 7.

First, the wool cross parts welding quality inspection system according to this embodiment includes a cowl cross parts welding device unit 101 that welds cowl cross (10a, 10b, cowl cross) parts through a predetermined welding process, and a cowl cross parts welding device. After going through the welding process of the device unit 101, it includes a cowl cross finished part welding quality inspection unit 102 that inspects the welding quality of the cowl cross finished part 10', which is a welded assembled product.

First, the cowl cross parts welding device unit 101 is a series of equipment for welding cowl cross (10a, 10b, cowl cross) parts through a predetermined welding process as described above, and includes the first and second moving devices. (180a, 180b), first and second mobile loading devices (110a, 110b, loading device), common sensor robot 120 having a common laser vision sensor (123), common robot welder (130) may include.

The first and second moving devices (180a, 180b) are devices on which the first and second mobile loading devices (110a, 110b) are mounted, respectively, and the first and second moving devices (110a, 110b) are mounted on the first and second moving devices (110a, 110b). It serves to move in the direction of the arrow in FIGS. 1 and 2.

The first and second moving devices 180a and 180b may be used as linear motors, etc. The first and second moving devices 180a and 180b have the same structure.

The first and second mobile loading devices (110a, 110b) are movably mounted on the first and second moving devices (180a, 180b), respectively, and are mounted on the first and second cowl crosses (10a, 10b, Cowl Cross), respectively. ) It forms a place where parts are seated and loaded. Since the dual structure of the first and second mobile loading devices 110a and 110b is applied to this system, work efficiency and productivity can be increased.

As described above, the first and second cowl crosses 10a and 10b are made of steel and are assembly units in which dozens of parts are assembled by welding. That is, the first and second cowl crosses 10a and 10b are composed of a plurality of parts manufactured by a press process on a plurality of first and second pipes 11 and 12, as schematically shown in FIG. 3. It is a composite product assembled by welding.

To manufacture these first and second cowl crosses (10a, 10b), the first and second movable loading devices (110a, 110b), that is, the parts forming the first and second cowl crosses (10a, 10b) are loaded. First and second mobile loading devices 110a and 110b are provided to form a place for assembly by welding. The structures of the first and second mobile loading devices 110a and 110b are the same.

The first and second mobile loading devices (110a, 110b) form first and second cowl crosses (10a, 10b) including first and second pipes (11, 12) connected at different heights. It forms a place where the first and second cowl cross (10a, 10b) parts are loaded to weld the parts.

The first and second pipes (11, 12) that are initially loaded into the first and second mobile loading devices (110a, 110b) must be firmly positioned so as not to shake, and only then can the first and second pipes (11, 12) Numerous parts can be accurately welded by location.

To this end, the first and second mobile loading devices 110a and 110b may include a loading base 111 and a plurality of clamping units 113 provided at each location on the loading base 111.

The loading base 111 includes a plurality of clamping units 113 and supports the first and second cowl cross parts 10a and 10b.

And, the clamping units 113 clamp the first and second pipes 11 and 12 forming the first and second cowl crosses 10a and 10b or parts welded thereto at the corresponding positions. Accordingly, multiple clamping units 113 are provided and mounted on the loading base 111 with different sizes and directions.

The common sensor robot 120 is a robot provided in common between the first and second mobile loading devices 110a and 110b of a dual structure.

This common sensor robot 120 includes a common laser vision sensor (123, A laser vision sensor is installed.

The common laser vision sensor 123 is a non-contact sensor that searches for welding lines on a member or structure with a laser and automatically tracks the movement to accurately determine the position of the common robot welder 130 in real time. It is better than other sensors. The sensing value is accurate.

The common robot welder 130 is provided in common between the first and second mobile loading devices (110a, 110b) adjacent to the common sensor robot 120, and is installed on the first and second cowl cross (10a, 10b) parts. It is a device that performs compensation welding. Here, compensation welding includes welding the originally determined target position and welding a new welding position that has been re-tracked and corrected through the common laser vision sensor 123.

Next, the cowl cross finished parts welding quality inspection unit 102 inspects the welding quality of the cowl cross finished parts 10', which are welded assembled products, after going through the welding process of the cowl cross parts welding device unit 101. As a series of equipment, it may include a final inspection jig device 140, an inspection laser vision sensor 150, and a system controller 170.

As shown in FIG. 5, the final inspection jig device 140 is a device different from the first and second mobile loading devices 110a and 110b, and the cowl cross finished part 10', which is the final assembled product after welding is completed, is seated. It creates a place where

The final inspection jig device 140 may include a base plate 141 and a plurality of holding units 143 provided at each location on the base plate 141. The final assembled product for which compensation welding has been completed can be seated on the base plate 141 through the plurality of holding units 143. At this time, the holding units 143 hold the assembled product at the corresponding position. Accordingly, several holding units 143 are provided and mounted on the base plate 141 with different sizes and directions.

The inspection laser vision sensor 150 serves to scan the assembled product mounted on the final inspection jig device 140 in three dimensions with a laser. This inspection laser vision sensor 150 may include a scanner support 151 and a scanning module 152 that is provided at an end of the scanner support 151 and actually performs a scanning operation.

Considering that the length of the assembled product is long, the inspection laser vision sensor 150 in this embodiment needs to perform a scanning operation on the long assembled product mounted on the final inspection jig device 140. It is implemented as a mobile type, in other words, an electric type, rather than a fixed position type. For this purpose, the following configurations are added.

The inspection laser vision sensor 150 is connected to a second moving module 155 that supports the inspection laser vision sensor 150 so as to be movable.

The second moving module 155 can move along the X-axis and Y-axis, which are coordinates in FIG. 5. For this purpose, a second X-axis gantry 156 and a second Y-axis gantry 157 are provided.

The second X-axis gantry 156 is connected to the second moving module 155 and serves to move the second moving module 155 to a predetermined It is connected to the X-axis gantry 156 and serves to move the second X-axis gantry 156 along the Y-axis crossing the X-axis.

And a second up/down driver 158 is connected between the second moving module 155 and the vision device 150. The second up/down driving unit 158, which can be applied as a cylinder, serves to drive the vision device 150 up/down in the Z-axis, which is the vertical direction.

In addition to the second up/down driving unit 158, the second X-axis gantry 156 and the second Y-axis gantry 157 are applied, and as the second moving module 155 operates, the inspection laser vision sensor 150 )'s scanning module 152 can approach the assembled product and scan it three-dimensionally with a laser.

The system controller 170 controls the inspection of welding quality of the cowl cross finished part 10' based on the sensing value of the inspection laser vision sensor 150.

In other words, in this embodiment, the system controller 170 extracts the profile of the weld bead of the weld part of the cowl cross finished part 10' detected by the inspection laser vision sensor 150, and then extracts the profile of the extracted weld bead. Based on this, the data stored in the database is classified and pre-processed, and then an artificial neural network is used to predict and control the tensile strength of the corresponding weld zone.

Of course, the system controller 170 applied to this embodiment also controls the operation of the common robot welder 130. That is, optimized welding compensation created based on a plurality of welding data including welding torch angle, welding speed, welding current, and welding voltage during compensation welding of the cowl cross parts 10a and 10b by the common robot welder 130. Controls the use of conditional data.

The system controller 170 that performs this role may include a central processing unit (171, CPU), memory (172, MEMORY), and a support circuit (173, SUPPORT CIRCUIT).

In this embodiment, the central processing unit 171 is a variety of computer processors that can be applied industrially to inspect and control the welding quality of the cowl cross finished part 10' based on the sensing value of the inspection laser vision sensor 150. It could be one of:

Memory 172 (MEMORY) is connected to the central processing unit 171. Memory 172 is a computer-readable recording medium that can be installed locally or remotely, for example, in a readily available storage medium such as random access memory (RAM), ROM, floppy disk, hard disk, or any other digital storage form. It may be at least one memory.

The support circuit 173 (SUPPORT CIRCUIT) is combined with the central processing unit 171 to support typical operations of the processor. This support circuit 173 may include a cache, power supply, clock circuit, input/output circuit, subsystem, etc.

In this embodiment, the system controller 170 extracts the profile of the weld bead of the weld part of the cowl cross finished part 10' detected by the inspection laser vision sensor 150, and then extracts the profile of the weld bead based on the profile of the extracted weld bead. After classifying and preprocessing the data stored in the database, an artificial neural network is used to predict and control the tensile strength of the corresponding weld zone. This series of control processes can be stored in the memory 172. Typically, software routines may be stored in memory 172. Software routines may also be stored or executed by other central processing units (not shown).

Although the process according to the present invention has been described as being executed by software routines, it is also possible that at least some of the processes according to the present invention are performed by hardware. As such, the processes of the present invention may be implemented as software running on a computer system, as hardware such as an integrated circuit, or as a combination of software and hardware.

Meanwhile, the cowl cross parts welding quality inspection method according to this embodiment that can be performed by such a system involves welding the cowl cross parts (10a, 10b) through a predetermined welding process by the cowl cross parts welding device unit 101. A cowl cross part welding step (S100a) and a cowl cross finished part welding quality inspection step that inspects the welding quality of the cowl cross finished part (10'), which is an assembled product that has been welded through the cowl cross finished part welding quality inspection unit (101). (S100b) may be included.

First, the cowl cross parts welding step (S100a) is a process of welding the cowl cross (10a, 10b) parts through a predetermined welding process by the cowl cross parts welding device unit 101, and includes a moving device installation step (S111), Mobile loading device mounting step (S112), first and second cowl cross component loading step (S113), first cowl cross component scanning step using a common laser vision sensor (S114), first cowl cross component weld line confirmation step (S115) ), a welding condition change step of the common robot welder (S116), and a first cowl cross component compensation welding progress step (S117).

The moving device installation step (S111) is a process of installing a plurality of moving devices (180a, 180b). The first and second moving devices 180a and 180b may be installed so that the common sensor robot 120 and the common robot welder 130 are disposed between them.

The mobile loading device mounting step (S112) is a process of forming a dual structure by mounting the first and second mobile loading devices (110a, 110b) on the first and second moving devices (180a, 180b), respectively. am.

The first and second cowl cross parts loading step (S113) is a dual first and second movable loading device ( This is a process of seating and loading the first and second cowl cross (10a, 10b, Cowl Cross) parts on (110a, 110b) respectively.

The first cowl cross parts scanning step (S114) using a common laser vision sensor is a common laser vision sensor ( This is a process of scanning the first cowl cross (10a) component loaded on the first mobile loading device (110a) using 123, laser vision sensor.

The first cowl cross component weld line confirmation step (S115) is a process of confirming the weld line on the first cowl cross component (10a) loaded on the first mobile loading device (110a) based on scan information.

The welding condition change step (S116) of the common robot welder changes the welding conditions of the common robot welder 130 located adjacent to the common laser vision sensor 123 based on the welding line information of the confirmed first cowl cross (10a) part. It is a process of change.

The first cowl cross component compensation welding progress step (S117) is a process of performing compensation welding on the first cowl cross component (10a) using the common robot welder 130 after the welding conditions are changed.

Meanwhile, when the common robot welder 130 performs compensation welding on the first cowl cross (10a) part, at the same time, the common laser vision sensor 123 mounted on the common sensor robot 120 is connected to the second mobile as shown in FIG. 2. After scanning the second cowl cross part (10b) loaded on the loading device (110b), the weld line on the second cowl cross (10b) part loaded on the second mobile loading device (110b) is drawn based on the scan information. Confirm. Once confirmation is complete, based on this information, the common robot welder 130 performs compensation welding on the second cowl cross (10b) part as shown in Figure 1, and the common laser vision sensor 123 performs work on the opposite side. In this embodiment, workability can be improved because the common laser vision sensor 123 and the common robot welder 130 alternately work in opposite directions, thereby contributing to improved productivity. For reference, in the flow chart of FIG. 6, the sequence for the second mobile loading device 110b is omitted.

Next, the cowl cross finished part welding quality inspection step (S100b) is a process of inspecting the welding quality of the cowl cross finished part (10'), which is an assembled product that has been welded through the cowl cross finished part welding quality inspection unit 101, It may include an assembled product seating step (S121), a weld bead detection step (S122), a weld bead profile extraction step (S123), a data classification and preprocessing step (S124), and a tensile strength prediction step (S125).

The assembled product seating step (S121) is a process of seating the cowl cross finished part (10') on a predetermined final inspection jig device (140) as shown in FIG. 5.

The weld bead detection step (S122) is a process of detecting weld beads for the weld portion of the cowl cross finished part 10' using an inspection laser vision sensor (150).

The weld bead profile extraction step (S123) is a process of extracting the profile of the weld bead of the detected weld zone.

The data classification and preprocessing step (S124) is a process of classifying and preprocessing data stored in the database based on the profile of the extracted weld bead.

The tensile strength prediction step (S125) is a process that uses an artificial neural network to predict the tensile strength of the corresponding weld.

Welding of cowl cross finished parts including the assembled product seating step (S121), weld bead detection step (S122), weld bead profile extraction step (S123), data classification and preprocessing step (S124), and tensile strength prediction step (S125). In this embodiment, the quality inspection step (S120) is performed as is without cutting the cowl cross finished part (10'), unlike the prior art. Therefore, a complete inspection is possible as needed, and the manpower and time required for inspection can be significantly reduced compared to before.

According to this embodiment, which operates based on the structure described above, the cowl cross (10a, 10b, 10 ') The welding quality of parts can be inspected, and as a result, the manpower and time required for inspection can be significantly reduced compared to the past, making it possible to easily and conveniently perform a full inspection of products as needed.

As such, the present invention is not limited to the described embodiments, and it is obvious to those skilled in the art that various modifications and changes can be made without departing from the spirit and scope of the present invention. Therefore, it should be said that such modifications or variations fall within the scope of the claims of the present invention.

10a, 10b: 1st and 2nd cowl cross
101: Cowl cross parts welding device part
102: Cowl cross finished parts welding quality inspection department
110a, 110b: first and second mobile loading devices
111: loading base 113: clamping unit
120: Common sensor robot 123: Common laser vision sensor
130: Common robot welding 140: Final inspection jig device
150: Inspection laser vision sensor 170: System controller
180a, 180b: moving device

Claims (5)

delete delete delete A cowl cross parts welding step of welding cowl cross parts through a predetermined welding process by the cowl cross parts welding device unit; and
It includes a cowl cross finished parts welding quality inspection step in which the welding quality of cowl cross finished parts, which are assembled products with completed welding, is inspected through the cowl cross finished parts welding quality inspection department.
The cowl cross part welding step is,
A moving device installation step of installing a plurality of moving devices;
A mobile loading device mounting step of mounting a mobile loading device on each of the moving devices to form a dual structure;
A first and second cowl cross parts loading step of loading cowl cross parts by seating them on dual first and second movable loading devices that are mounted on the moving device and capable of moving in position. ;
Scanning the first cowl cross part loaded on the first mobile loading device using a common laser vision sensor mounted on a common sensor robot that commonly operates between the first and second mobile loading devices. First cowl cross part scanning step using a common laser vision sensor;
A first cowl cross part weld line confirmation step of checking the weld line on the first cowl cross part loaded on the first mobile loading device based on scan information;
A welding condition changing step of a common robot welder that changes the welding conditions of a common robot welder disposed adjacent to the common laser vision sensor based on welding line information of the identified first cowl cross part; and
Comprising a first cowl cross part compensation welding step of performing compensation welding on the first cowl cross part using the common robot welder after the welding conditions are changed,
The cowl cross finished parts welding quality inspection step is,
An assembled product seating step of seating the cowl cross completed part on a predetermined final inspection jig device;
A weld bead detection step of detecting a weld bead for a weld portion of the cowl cross finished part using a laser vision sensor;
Weld bead profile extraction step of extracting the profile of the weld bead of the detected weld area;
A data classification and preprocessing step of classifying and preprocessing data stored in the database based on the profile of the extracted weld bead; and
It includes a tensile strength prediction step that uses an artificial neural network to predict the tensile strength of the corresponding weld zone,
When the common robot welder performs compensation welding on the first cowl cross part, at the same time, the common laser vision sensor mounted on the common sensor robot scans the second cowl cross part loaded on the second mobile loading device. Afterwards, a cowl cross part welding quality inspection method characterized in that the weld line on the second cowl cross part loaded on the second mobile loading device is checked based on the scan information. delete

Images