A laser welding real-time monitoring method and system

Abstract

The application provides a laser welding real-time monitoring method and system, wherein the method comprises the following steps: collecting and analyzing optical radiation generated in a welding process to obtain first analysis data; analyzing reflected light of sample light coaxially coupled with a laser welding light path to obtain second analysis data; and comprehensively analyzing the first analysis data and the second analysis data to obtain monitoring data. The laser welding real-time monitoring method and system can perform welding quality inspection during the welding process, and facilitate correction of the welding action to ensure the accuracy and effectiveness of subsequent welding.

Description

Technical Field

[0001] This invention relates to the field of real-time monitoring technology, and in particular to a method and system for real-time monitoring of laser welding. Background Technology

[0002] Currently, the main problems with laser welding include welding defects such as spatter, porosity, cracks, incomplete welds, burn-through, and undercut. These defects lead to reduced workpiece strength, decreased sealing and conductivity. Traditional inspection methods, primarily post-weld inspections (e.g., airtightness testing, conductivity testing, and machine vision inspection), are inherently lagging. Furthermore, measurements of some core indicators, such as pull-out force and penetration depth, are destructive and cannot achieve 100% inspection. Therefore, there is an urgent need for a real-time monitoring method and system for laser welding that can perform inspections during the welding process. Summary of the Invention

[0003] One of the objectives of this invention is to provide a real-time monitoring method and system for laser welding, which performs welding quality inspection during the welding process, facilitating the correction of welding actions to ensure the accuracy and effectiveness of subsequent welding.

[0004] This invention provides a real-time monitoring method for laser welding, comprising:

[0005] The light radiation generated during the welding process was collected and analyzed to obtain the first analytical data;

[0006] The reflected light of the sample light coaxially coupled with the laser welding optical path is analyzed to obtain the second analytical data;

[0007] By combining the data from the first and second analyses, monitoring data is obtained.

[0008] Preferably, the light radiation is collected using a photoelectric detection sensor.

[0009] Preferably, the analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

[0010] Preferably, the sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

[0011] Preferably, the analysis process of the second analysis data includes: Fourier transform, algorithm processing, and image construction.

[0012] The present invention also provides a real-time monitoring system for laser welding, comprising: a first analysis module, a second analysis module, and a comprehensive analysis module; wherein, the first analysis module collects and analyzes the light radiation generated during the welding process to obtain first analysis data; the second analysis module analyzes the reflected light of the sample light coaxially coupled with the laser welding optical path to obtain second analysis data; and the comprehensive analysis module combines the first analysis data and the second analysis data to obtain monitoring data.

[0013] Preferably, the light radiation is collected using a photoelectric detection sensor.

[0014] Preferably, the analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

[0015] Preferably, the sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

[0016] Preferably, the analysis process of the second analysis data includes: Fourier transform, algorithm processing, and image construction.

[0017] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0020] Figure 1 This is a schematic diagram of a real-time monitoring method for laser welding according to an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of a monitoring device according to an embodiment of the present invention;

[0022] Figure 3 This is a data processing flowchart for real-time monitoring of laser welding in an embodiment of the present invention;

[0023] Figure 4 This is a data flow diagram of real-time monitoring of laser welding in an embodiment of the present invention;

[0024] Figure 5 This is a monitoring result output diagram in a specific application example of the present invention;

[0025] Figure 6 This is a schematic diagram of a real-time monitoring system for laser welding according to an embodiment of the present invention. Detailed Implementation

[0026] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0027] Example 1:

[0028] This invention provides a method for real-time monitoring of laser welding, such as... Figure 1 As shown, it includes:

[0029] Step 1: Collect and analyze the light radiation generated during the welding process to obtain the first analytical data;

[0030] Step 2: Analyze the reflected light of the sample light coaxially coupled with the laser welding optical path to obtain the second set of analytical data;

[0031] Step 3: Combine the data from the first and second analyses to obtain the monitoring data.

[0032] The collection of light radiation is achieved through photoelectric detection sensors.

[0033] The analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

[0034] The sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

[0035] The second analysis process of the data includes: Fourier transform, algorithm processing, and image construction.

[0036] This embodiment of a laser welding real-time monitoring method combines two measurement methods for accurate real-time monitoring. One method's principle is as follows: Figure 2As shown, the OCT measurement light sequentially passes through collimation module 2, scanning module 3, and dichroic mirror 7, acting on the surface of workpiece 8 and returning along the original path. It then interferes with the external reference light, and the light is collected by a spectrometer. After Fourier transform, algorithm processing, and image construction, real-time welding measurement is completed; this is the process of obtaining the second analysis data. That is, OCT measurement: The OCT (optical coherence tomography) weld depth monitoring system obtains keyhole depth information by coaxially coupling the sample light and the laser welding light path, based on the coherence of the broadband light source. It mainly includes a superluminescent diode (SLD), fiber coupler, reference arm, sample arm, and spectrometer. The sample arm is coaxially coupled to the laser welding light path through a fiber collimator, ensuring that the sample detection light is coaxially aligned with the center of the processing laser, allowing it to enter the bottom of the keyhole during welding to measure the depth. The second method works as follows: The visible light, reflected light, and infrared thermal radiation generated by the laser acting on the surface of the workpiece 8 through the collimation module 4, reflector 6, and dichroic mirror 7 are amplified, filtered, demodulated, and converted to A / D values ​​by the photoelectric detection module 5 after passing through the dichroic mirror 7 and reflector 6. Data processing and image construction are then performed to achieve real-time welding detection; this is the process of obtaining the first set of analytical data. Laser welding involves light radiation phenomena, including visible light (metal vapor), laser reflection (laser), and infrared light (molten pool thermal radiation). These radiated light signals reflect the welding status and whether defects are generated during the process. The photoelectric sensor converts the light radiation generated during welding into electrical signals. Analysis of these signals by the detection system allows for the determination of the workpiece welding quality, thus achieving defect detection and quality monitoring. Furthermore, the figure also shows the focusing module 1, which, in a specific application, such as... Figure 3 , 4 Figures 5 and 6 show the data processing flowchart, data flow diagram, and monitoring result output diagram, respectively. For monitoring, each detection channel will display three curves, representing: the upper package route, the average value, and the lower package route, respectively.

[0037] Example 2:

[0038] This invention provides a method for real-time monitoring of laser welding, comprising:

[0039] The light radiation generated during the welding process was collected and analyzed to obtain the first analytical data;

[0040] The reflected light of the sample light coaxially coupled with the laser welding optical path is analyzed to obtain the second analytical data;

[0041] By combining the data from the first and second analyses, monitoring data is obtained.

[0042] The collection of light radiation is achieved through photoelectric detection sensors.

[0043] The analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

[0044] The sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

[0045] The second analysis process of the data includes: Fourier transform, algorithm processing, and image construction.

[0046] To avoid interference between the first and second analysis data, the analysis timing must be allocated reasonably. Therefore, the real-time monitoring method for laser welding also includes:

[0047] Analyze the current welding conditions, and based on the analysis results, consult the pre-configured control table to determine the control parameters;

[0048] Based on the determined control parameters, the acquisition and analysis of optical radiation and the analysis of reflected light are controlled.

[0049] The control parameters include the time interval between two adjacent data acquisitions and the time span of a single data acquisition and analysis.

[0050] This embodiment proposes to control the acquisition and analysis of optical radiation and reflected light based on the analysis of welding conditions. Specifically, sample light is not emitted during the acquisition and analysis of optical radiation to avoid interference with the analysis of the first analysis data. Similarly, optical radiation is not acquired and analyzed during the acquisition and analysis of reflected light to avoid interference with the analysis of the second analysis data. The control parameters obtained through the control table analysis ensure that the acquisition and analysis of the two are carried out alternately.

[0051] The analysis of the current welding conditions includes: the power of the welding laser, the type of the object being welded, the size of the welding area, the depth of the welding area, and whether the current welding point is the starting or ending point.

[0052] In this embodiment, the control table enables the alternating, non-interfering operation of two detection methods at the start of laser welding. However, as welding progresses, various factors influence the process, necessitating adjustments based on different situations to ensure better detection. Specifically, adjustments can be made based on both the evaluation of detection effectiveness and the specific welding conditions. The adjustment steps are as follows: One alternating analysis is considered a round; feature extraction is performed on the data collected during the analysis of the two analysis methods in each round to obtain multiple round description parameters corresponding to the two analysis methods; using the current welding object and welding trajectory, from the weld... Retrieve the corresponding welding record from the historical record library; determine the quality data of the current welding point and a preset number of surrounding welding points from the welding record, quantify the quality data, and obtain multiple quality representation parameters; integrate the wheel description parameters and quality representation parameters of a preset number of rounds (any of 2 to 20) to obtain an analysis set; using the analysis set as a benchmark, retrieve the adjustment set from a pre-configured adjustment library; adjust the control parameters obtained from the control table analysis using the various control parameters in the adjustment set; wherein, the wheel description parameters include: parameters representing the deviation between the detection analysis value of each detection channel and the comprehensive analysis value, and parameters representing the comprehensive analysis value corresponding to a preset range. The parameters include position; the two analysis methods employ overlapping sampling multi-channel detection, which involves overlapping sampling of a data segment by sliding a sampling frame across the data, sampling multiple data points and inputting them into a detection channel for detection; then, the average of the detection analysis values ​​from each detection channel is taken to obtain the comprehensive analysis value; the interval range can be pre-configured according to the current welding conditions; the data integration rules in the analysis set include: the first and second rows of data are the wheel description parameters of the earliest round of the two analysis methods; the following rows are the wheel description parameters of other rounds sorted by time; after arranging the wheel description parameters, the last row is the quality representation parameter; in addition, quality... The analysis process for the representation parameters is as follows: Quality data of the same type at the same welding point are arranged in chronological order to form a data string. Feature extraction is performed on the data string. Based on the extracted data features, quality representation parameters are determined from a pre-configured quantification representation parameter library. The data features include: feature parameters representing the average value of the data, feature parameters representing situations (number of times) exceeding a pre-configured threshold range, and feature parameters representing the trend of change between data. The quality representation parameters can indirectly indicate the probability of welding anomalies at the current welding point. Accordingly, the control parameters for the detection time and the number of detection channels can be adjusted to ensure timely detection of anomalies.

[0053] In addition, active triggering adjustment can be performed to determine the optimal state of alternating between the two analysis methods. The control of active triggering adjustment requires the configuration of a trigger parameter. First, the trigger parameter is initialized (set to zero). Then, based on the pre-configured evaluation library, the position of the welding point in the pre-configured threshold space is analyzed and evaluated to obtain an evaluation value. The evaluation value is accumulated into the trigger parameter. When the trigger parameter is greater than the preset threshold, the active trigger parameter set is retrieved from the pre-configured active trigger library based on the data string of the evaluation value. Then, the trigger parameter is reset. In addition, the trigger parameter is reset when the control parameter has been adjusted.

[0054] Example 3:

[0055] This invention also provides a real-time monitoring system for laser welding, such as... Figure 6 As shown, it includes: a first analysis module 11, a second analysis module 12, and a comprehensive analysis module 13; wherein, the first analysis module 11 collects and analyzes the light radiation generated during the welding process to obtain first analysis data; the second analysis module 12 analyzes the reflected light of the sample light coaxially coupled with the laser welding optical path to obtain second analysis data; the comprehensive analysis module 13 combines the first analysis data and the second analysis data to obtain monitoring data.

[0056] The collection of light radiation is achieved through photoelectric detection sensors.

[0057] The analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

[0058] The sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

[0059] The second analysis process of the data includes: Fourier transform, algorithm processing, and image construction.

[0060] Example 4:

[0061] The present invention also provides a real-time monitoring system for laser welding, comprising: a first analysis module, a second analysis module, and a comprehensive analysis module; wherein, the first analysis module collects and analyzes the light radiation generated during the welding process to obtain first analysis data; the second analysis module analyzes the reflected light of the sample light coaxially coupled with the laser welding optical path to obtain second analysis data; and the comprehensive analysis module combines the first analysis data and the second analysis data to obtain monitoring data.

[0062] The collection of light radiation is achieved through photoelectric detection sensors.

[0063] The analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

[0064] The sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

[0065] The second analysis process of the data includes: Fourier transform, algorithm processing, and image construction.

[0066] To avoid mutual interference between the first and second analysis data, the analysis timing should be allocated reasonably. Therefore, the laser welding real-time monitoring system also includes a control module.

[0067] The control module performs the following operations:

[0068] Analyze the current welding conditions, and based on the analysis results, consult the pre-configured control table to determine the control parameters;

[0069] Based on the determined control parameters, the acquisition and analysis of optical radiation and the analysis of reflected light are controlled.

[0070] The control parameters include the time interval between two adjacent data acquisitions and the time span of a single data acquisition and analysis.

[0071] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for real-time monitoring of laser welding, characterized in that, include: The light radiation generated during the welding process was collected and analyzed to obtain the first analytical data; The reflected light of the sample light coaxially coupled with the laser welding optical path is analyzed to obtain the second analytical data; By combining the data from the first and second analyses, monitoring data is obtained. Real-time monitoring methods for laser welding also include: Analyze the current welding conditions, and based on the analysis results, consult the pre-configured control table to determine the control parameters; Based on the determined control parameters, the acquisition and analysis of optical radiation and the analysis of reflected light are controlled. The control parameters include: the time interval between two adjacent data acquisitions and analyses, and the time span of a single data acquisition and analysis. Adjustments to the control parameters include: One alternating analysis is considered as one round. Features are extracted from the data collected during the two analysis methods within a round to obtain multiple round description parameters corresponding to the two analysis methods. Based on the current welding object and welding trajectory, corresponding welding records are retrieved from the welding history database. Quality data corresponding to the current welding point and a preset number of surrounding welding points are determined from the welding records. This quality data is quantified to obtain multiple quality representation parameters. The round description parameters and quality representation parameters from the previous preset number of rounds are integrated to obtain an analysis set. Using the analysis set as a benchmark, an adjustment set is retrieved from a pre-configured adjustment database. The control parameters obtained from the control table analysis are adjusted using the various control parameters in the adjustment set. The round description parameters include: parameters representing the deviation between the detection analysis values ​​of each detection channel and the comprehensive analysis value; and parameters representing the position of the comprehensive analysis value corresponding to a preset interval range. The two analysis methods employ an overlapping sampling multi-channel detection method, i.e., sampling a segment of data... The template slides across the data for overlapping sampling, and multiple sampled data are input into a detection channel for detection. Then, the detection analysis values ​​of each detection channel are averaged to obtain a comprehensive analysis value. The interval range is pre-configured according to the current welding conditions. The data integration rules in the analysis set include: the first and second rows of data are the wheel description parameters of the earliest round of two analysis methods; the following rows are the wheel description parameters of other rounds sorted by time; after arranging the wheel description parameters, the last row is the quality representation parameter. The analysis process of the quality representation parameter is as follows: the same type of quality data at the same welding point is arranged in time order to form a data string, the data string is feature extracted, and based on the extracted data features, the quality representation parameter is determined from the pre-configured quantitative representation parameter library. The data features include: feature parameters representing the average value of the data, feature parameters representing the situation exceeding the pre-configured threshold range, and feature parameters representing the trend of change between data.

2. The real-time monitoring method for laser welding as described in claim 1, characterized in that, The collection of light radiation is achieved through photoelectric detection sensors.

3. The real-time monitoring method for laser welding as described in claim 2, characterized in that, The analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

4. The real-time monitoring method for laser welding as described in claim 1, characterized in that, The sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

5. The real-time monitoring method for laser welding as described in claim 1, characterized in that, The second analysis process of the data includes: Fourier transform, algorithm processing, and image construction.

6. A real-time monitoring system for laser welding, employing the method as described in any one of claims 1 to 5, characterized in that, include: The system comprises a first analysis module, a second analysis module, and a comprehensive analysis module; wherein, the first analysis module collects and analyzes the light radiation generated during the welding process to obtain the first analysis data; The second analysis module analyzes the reflected light of the sample light coaxially coupled with the laser welding optical path to obtain second analysis data; the comprehensive analysis module combines the first and second analysis data to obtain monitoring data.

7. The laser welding real-time monitoring system as described in claim 6, characterized in that, The collection of light radiation is achieved through photoelectric detection sensors.

8. The laser welding real-time monitoring system as described in claim 7, characterized in that, The analysis process of the first analysis data includes: amplification, filtering, demodulation, A / D conversion, followed by data processing and image construction.

9. The laser welding real-time monitoring system as described in claim 6, characterized in that, The sample light is coaxially coupled to the laser welding optical path through an optical fiber collimator; the reflected light is collected by a spectrometer.

10. The laser welding real-time monitoring system as described in claim 6, characterized in that, The second analysis process of the data includes: Fourier transform, algorithm processing, and image construction.

Images