Coaxial weld seam tracking system and method for oscillating laser welding based on frequency domain OCT detection

Abstract

The present invention proposes a coaxial weld tracking method for oscillating laser welding based on frequency domain OCT detection, comprising the following steps: obtaining the height value in the normal direction of the weld; performing point cloud reconstruction on the weld splicing morphology using the obtained surface height value H1 of the weld; extracting weld information through the point cloud reconstruction atlas; and providing a matching coaxial weld tracking system for oscillating laser welding based on frequency domain OCT detection, comprising a robot, an oscillating welding head, a turntable, a control cabinet, a monitoring system host, a sample arm and a reference arm, which compares the extracted weld information with a preset weld standard value, corrects the motion trajectory of the welding head in real time, and realizes the weld tracking function.

Description

Technical Field

[0001] The present invention relates to the technical field of automated laser welding, and in particular to a system and method for tracking a coaxial weld seam of oscillating laser welding based on frequency domain OCT detection. Background Art

[0002] In recent years, with the continuous development of laser welding technology, especially swing laser welding process technology, the high requirements of traditional laser welding process for weld splicing have been overcome, making laser welding process technology increasingly widely used in various industries such as aerospace, rail transportation, construction, and medical treatment. However, at present, the laser welding process mainly relies on manual assembly and teaching methods on welding manufacturing workpieces with complex surfaces, many welds, and high quality requirements. It is difficult to ensure the accessibility of the process, the accuracy of the welding position, and the reliability of the product welding quality. The current weld tracking generally uses visual, current, electromagnetic, and mechanical sensors to provide weld structure information, and through data processing, the position of the laser welding head can be adjusted in real time.

[0003] At present, traditional weld visual tracking methods are usually based on morphological image processing. Such methods are cumbersome, have high algorithm complexity, and have poor adaptability to environmental changes and noise interference. With the rapid development of deep learning and target tracking technology, they have also been applied to the field of weld visual tracking, but most of these methods are difficult to balance accuracy and speed. On the other hand, most of the current robot path planning research focuses on point-to-point obstacle avoidance path planning, and lacks research on path planning with directional constraints between multiple points involved in the path planning problem of weld tracking of complex components. In addition, the key point of welding automation and intelligence lies in the real-time acquisition and intelligent control of the dynamic characteristic information of the molten pool during the welding process, which is also the difficulty and hot spot of welding research. Most of the welding equipment or teaching and reproducing robots currently used, in the actual welding process, due to the constant parameter configuration and fixed robot motion trajectory, cannot make corrections to these changes and interferences in real time when encountering interference from factors such as misalignment, gap changes, abnormal heat dissipation conditions, and thermal deformation of the workpiece, resulting in a decrease in the quality of weld formation and cannot meet high standards of welding requirements. Summary of the invention

[0004] The present invention proposes a coaxial weld tracking system and method for oscillating laser welding based on frequency domain OCT detection, which solves the problem in the prior art that there is a lack of a method and device that can obtain dynamic characteristic information of the molten pool in real time and intelligently control it during the welding process. The technical solution of the present invention is implemented as follows:

[0005] The coaxial weld tracking method of oscillating laser welding based on frequency domain OCT detection includes the following steps: obtaining the height value of the weld in the normal direction; reconstructing the weld joint morphology by point cloud using the surface height value H1 of the weld obtained; extracting weld information through the point cloud reconstruction atlas; comparing the extracted weld information with the preset weld standard value, correcting the motion trajectory of the welding head in real time, and realizing the weld tracking function.

[0006] As a preferred technical solution, the weld information includes the left and right edge positions C1 of the weld, the center position C2 of the weld, the joint gap width D, and the weld point position height H2.

[0007] As a preferred technical solution, the gap width D of the joint is calculated through the height difference information reflected by the two-dimensional morphology of the weld.

[0008] As a preferred technical solution, the surface height of the workpiece before welding is obtained by pre-measurement, and then the height H2 of the position of the welding point is calculated by trigonometric function.

[0009] As a preferred technical solution, the weld tracking function is realized by outputting the deviation in the form of analog quantity to the swing welding head and controlling the center offset of the swing welding to correct the movement trajectory of the welding head in real time.

[0010] As a preferred technical solution, the deviation value of the weld trajectory is calculated by comparing the weld center position obtained by detection with the current weld center position and real-time tracking and calibration is carried out, which specifically includes the following steps:

[0011] Step S1: The weld tracking system of the invention converts the deviation value of the weld center position into an analog signal;

[0012] Step S2: The weld tracking system transmits the analog signal to the analog port of the swing welding head in the laser welding platform at a specific time according to the welding speed, the detection lead distance and the system response speed;

[0013] Step S3: The swing welding head adjusts the position of the welding laser center according to the received analog value

[0014] Step S4: The welding center position is offset to the weld center position to realize welding trajectory deviation correction and weld tracking functions.

[0015] As a preferred technical solution, the height value H1 in the normal direction of the weld is obtained by the following steps: acquiring a surface detection signal via a spectrometer, extracting a weld normal height signal, and acquiring a position signal via the weld normal height signal.

[0016] As a preferred technical solution, extracting the normal height signal of the weld specifically includes: removing DC terms, wavelength calibration, CCD compensation, resampling, inverse Fourier transform and coordinate calibration; obtaining the position signal specifically includes the following steps: lateral scanning data extraction, calibration position corresponding time, noise filtering and position signal processing steps.

[0017] The oscillating laser welding coaxial weld tracking system based on frequency domain OCT detection is used to execute the above-mentioned oscillating laser welding coaxial weld tracking method based on frequency domain OCT detection, including a robot, an oscillating welding head, a turntable, a control cabinet, a monitoring system host, a sample arm and a reference arm. The sample arm is provided with a two-dimensional galvanometer, through which the normal direction of the workpiece weld on the turntable is scanned. The monitoring system host outputs the deviation amount in the form of an analog amount to the oscillating welding head, controls the center offset of the oscillating welding to correct the motion trajectory of the welding head in real time, and realizes the weld tracking function. The control process is executed by the robot.

[0018] As a preferred technical solution, the standard aperture of the oscillating welding head is greater than 40 mm, so the prerequisite scanning distance and scanning width of the weld are greater than 15 mm.

[0019] Compared with the existing technology, this solution has the following beneficial effects: it provides a coaxial weld tracking system and intelligent algorithm suitable for swing laser welding, and develops a high-performance coaxial weld tracking technology and system based on the optical low-coherence principle technology (OCT), and combines the weld feature signal extraction technology, weld splicing extraction technology, and robot control technology to achieve intelligent laser welding of complex structural parts. It solves the problem in the existing technology that there is a lack of a method and equipment that can obtain and intelligently control the dynamic feature information of the molten pool in real time during the welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 It is a structural schematic diagram of the coaxial weld seam tracking system of oscillating laser welding based on frequency domain OCT detection of the present invention;

[0022] Figure 2 This is a schematic diagram of the spectral domain OCT principle of the present invention;

[0023] Figure 3 It is a flow chart of processing the height value of the normal direction of the weld seam of a part of the present invention;

[0024] Figure 4 It is a test interface for the height value of the weld seam of the component in the normal direction of the present invention;

[0025] Figure 5 It is the characteristic point cloud data of the weld interface of the present invention;

[0026] Figure 6 Point cloud data diagram of the groove detection of the present invention.

[0027] Description of reference numerals:

[0028] 1. Turntable; 2. Reference arm; 3. Sample arm; 4. Swing welding head; 5. Robot; 6. Control cabinet; 7. System monitoring host. DETAILED DESCRIPTION

[0029] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Reference Figure 1 , a oscillating laser welding coaxial weld tracking system based on frequency domain OCT detection, which is used to execute the above-mentioned oscillating laser welding coaxial weld tracking method based on frequency domain OCT detection, including a robot, an oscillating welding head, a turntable, a control cabinet, a monitoring system host, a sample arm and a reference arm, the sample arm is provided with a two-dimensional galvanometer, through which the normal direction of the workpiece weld on the turntable is scanned, the monitoring system host outputs the deviation amount in the form of an analog amount to the oscillating welding head, controls the center offset of the oscillating welding to correct the motion trajectory of the welding head in real time, and realizes the weld tracking function, and the control process is performed by the robot. And the standard aperture of the oscillating welding head used in this scheme is greater than 40mm, so the prerequisite scanning distance and scanning width of the weld are greater than 15mm, the lateral resolution of the system depends on the size of the laser spot detected by the system, which is less than 0.05mm, the weld recognition accuracy of the system can reach ±0.05mm, and can identify butt welds with a gap ≥0.1mm, and realize the weld tracking function of a weld gap of 0.1mm.

[0031] This system uses spectral domain optical low coherence technology. Its technical principle Figure 2As shown. This technology uses the principle of spectrum interference technology. Its core components are broadband light source, Michelson interferometer and spectrometer. The light emitted from the broadband light source reaches the sample arm and the detection arm after passing through the dichroic mirror, and is reflected at the reference mirror and the sample surface respectively, and returns along the original light path. An interference light signal carrying the sample surface morphology information is generated at the dichroic mirror and output to the spectrometer (interference signal spectrum acquisition unit). The sample backscattered light returned from the sample arm can be regarded as the superposition of multiple monochromatic light waves. After interfering with the reference light of similar composition, the interference light spectrum output by the interferometer is received by the spectrometer, and then expanded by the grating of the spectrometer according to the wavelength, and then sampled by the linear CCD of the spectrometer to obtain a one-dimensional interference spectrum, which is finally stored in the computer in the form of an interference spectrum. After the light returned from the reference arm and the sample arm interferes, it is divided into interference spectrum signals of different wavelengths in the fast spectrometer, which are photographed and received by the linear array CCD, and finally the height information of the sample surface can be obtained after fast Fourier transform.

[0032] The specific detection working principle is as follows: the two-dimensional galvanometer installed on the sample arm in the invention patent controls the detection laser to scan in the normal direction of the workpiece weld to obtain the height information in the normal direction of the weld. The surface height value of the weld can be used to reconstruct the point cloud of the weld splicing morphology, thereby obtaining the left and right edge positions of the weld, the center position of the weld, the width of the splicing gap and other weld information.

[0033] The system in the invention patent knows the surface height of the workpiece before welding, and can use trigonometric functions to determine the height of the welding point. The height difference information reflected by the two-dimensional morphology of the weld can be used to calculate the gap width of the joint. For workpieces greater than the threshold range, the system issues an alarm message. The gap position of the joint is generally the center position of the weld. By comparing the center position with the preset weld trajectory, the system can output the deviation in the form of an analog quantity to the swing welding head, control the center offset of the swing welding, and correct the movement trajectory of the welding head in real time to realize the weld tracking function.

[0034] During the welding process, the system controls the detection light to scan the front of the welding pool, and the scanning direction is perpendicular to the direction of the weld. The collected two-dimensional point cloud is shown in the figure, and the center position of the weld can be identified. The detection system compares the center position of the weld obtained by detection with the current welding center position and calculates the deviation value of the weld trajectory. The specific process is:

[0035] (1) The weld tracking system of the invention converts the deviation value of the weld center position into an analog signal;

[0036] (2) The weld tracking system transmits the analog signal to the analog port of the swing welding head in the laser welding platform at a specific time according to the welding speed, the detection lead distance and the system response speed;

[0037] (3) The oscillating welding head adjusts the position of the welding laser center according to the analog value received

[0038] (4) The welding center position is offset to the weld center position to realize welding trajectory deviation correction and weld tracking functions.

[0039] Among them, Figure 3 As shown, extracting the normal height signal of the weld specifically includes: removing the DC term, wavelength calibration, CCD compensation, resampling, inverse Fourier transform and coordinate calibration; obtaining the position signal specifically includes the following steps: lateral scanning data extraction, calibration position corresponding time, noise filtering and position signal processing steps.

[0040] Figures 4 to 6 To monitor the system host part test results and system interface, among which, Figure 4 It is a test interface for the height value of the weld seam of the component in the normal direction of the present invention; Figure 5 It is the characteristic point cloud data of the weld interface of the present invention; Figure 6 Point cloud data diagram of the groove detection of the present invention.

[0041] This technical solution patent aims to develop a coaxial weld tracking system and intelligent algorithm suitable for swing laser welding. Based on the optical low-coherence principle technology (OCT), a high-performance coaxial weld tracking technology and system is invented. Combined with the weld characteristic signal extraction technology, weld splicing extraction technology, and robot control technology, intelligent laser welding of complex structural parts is realized.

[0042] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A coaxial weld seam tracking method for oscillating laser welding based on frequency domain OCT detection, characterized in that: The following steps are involved: Obtain the height value H1 in the normal direction of the weld; use the obtained surface height value H1 of the weld to reconstruct the weld joint morphology through point cloud; extract the weld information through the point cloud reconstruction atlas; compare the extracted weld information with the preset weld standard value, and then correct the movement trajectory of the welding head in real time to realize the weld tracking function.

2. The method for coaxial weld tracking of oscillating laser welding based on frequency domain OCT detection according to claim 1, characterized in that: The weld information includes the left and right edge positions C1 of the weld, the center position C2 of the weld, the gap width D of the weld, and the height H2 of the weld point.

3. The method for coaxial weld tracking of oscillating laser welding based on frequency domain OCT detection as claimed in claim 2, characterized in that: The gap width D of the joint is calculated by the height difference information reflected by the two-dimensional morphology of the weld.

4. The method for coaxial weld tracking of oscillating laser welding based on frequency domain OCT detection as claimed in claim 2, characterized in that: The workpiece surface height before welding is obtained by pre-measurement, and then the height H2 of the position of the welding point is calculated by trigonometric function.

5. The method for coaxial weld tracking of oscillating laser welding based on frequency domain OCT detection according to claim 1, characterized in that: By outputting the deviation in the form of analog quantity to the swing welding head, the center offset of the swing welding is controlled to correct the movement trajectory of the welding head in real time, thereby realizing the weld tracking function.

6. The method for coaxial weld tracking of oscillating laser welding based on frequency domain OCT detection as claimed in claim 5, characterized in that: The deviation value of the weld trajectory is calculated by comparing the weld center position obtained by detection with the current weld center position and real-time tracking calibration is carried out, which specifically includes the following steps: Step S1: The weld tracking system of the invention converts the deviation value of the weld center position into an analog signal; Step S2: The weld tracking system transmits the analog signal to the analog port of the swing welding head in the laser welding platform at a specific time according to the welding speed, the detection lead distance and the system response speed; Step S3: The swing welding head adjusts the position of the welding laser center according to the received analog value Step S4: The welding center position is offset to the weld center position to realize welding trajectory deviation correction and weld tracking functions.

7. The method for coaxial weld tracking of oscillating laser welding based on frequency domain OCT detection according to claim 1, characterized in that: The height value H1 in the normal direction of the weld is obtained by the following steps: acquiring a surface detection signal via a spectrometer, extracting a weld normal height signal, and acquiring a position signal through the weld normal height signal.

8. The method for coaxial weld tracking of oscillating laser welding based on frequency domain OCT detection according to claim 7, characterized in that: Extracting the normal height signal of the weld specifically includes: removing the DC term, wavelength calibration, CCD compensation, resampling, inverse Fourier transform and coordinate calibration; obtaining the position signal specifically includes the following steps: lateral scanning data extraction, calibration position corresponding time, noise filtering and position signal processing steps.

9. A oscillating laser welding coaxial weld seam tracking system based on frequency domain OCT detection, which is used to execute the oscillating laser welding coaxial weld seam tracking method based on frequency domain OCT detection as described in any one of claims 1 to 8 above, characterized in that: The invention comprises a robot, an oscillating welding head, a turntable, a control cabinet, a monitoring system host, a sample arm and a reference arm. The sample arm is provided with a two-dimensional galvanometer, through which the normal direction of the weld of the workpiece on the turntable is scanned. The monitoring system host outputs the deviation in the form of an analog quantity to the oscillating welding head, controls the center offset of the oscillating welding, and corrects the motion trajectory of the welding head in real time to realize the weld tracking function. The control process is executed by the robot.

10. The oscillating laser welding coaxial weld seam tracking system based on frequency domain OCT detection according to claim 9, characterized in that: The standard aperture of the oscillating welding head is greater than 40 mm, and the prerequisite scanning distance and scanning width of the weld are greater than 15 mm.

Images