Vacuum laser welding online monitoring method

Abstract

The invention discloses an online monitoring method for vacuum laser welding, which is characterized in that secondary current signals of vacuum laser welding are collected, association rules of the secondary current signals, weld forming and penetration depth are analyzed, and then the weld appearance of vacuum laser welding is actively controlled based on the secondary current signals. And finally, online closed-loop monitoring of the vacuum laser welding process is effectively realized.

Description

Technical Field

[0001] The invention relates to an online monitoring method for vacuum laser welding. Background Art

[0002] The vacuum laser welding process is a welding technology that uses a laser beam to heat, melt and connect the workpiece with high energy density in a vacuum environment. Since it is difficult to intervene manually in the vacuum laser welding process, when process defects such as poor forming occur, it must be able to perform complete closed-loop automatic control. In this regard, domestic and foreign peers have carried out a lot of research work. Osaka University in Japan observed the dynamic behavior of the pinhole and molten pool and the defect formation process of high-power vacuum laser welding through in-situ X-ray high-speed imaging technology; at the same time, the flow behavior of the molten pool surface was observed by high-speed imaging. Harbin Institute of Technology and Shanghai Jiaotong University in China have also carried out experimental observation and research on the flow behavior of the molten pool through cooperation with Osaka University. However, through in-situ X-ray experiments or high-speed imaging observations, it is difficult to meet the needs of real-time adjustment of process parameters due to the slow speed of image signal processing, so it is difficult to achieve closed-loop monitoring of the vacuum laser welding process. Moreover, during vacuum laser welding, due to the presence of metal vapor in the vacuum chamber, the online monitoring of the welding process is achieved through high-speed photography and other methods. The lens is easily blocked and contaminated by vapor, making it very difficult to monitor the vacuum laser welding process online through visual means. Therefore, the existing vacuum laser welding monitoring methods are difficult to effectively realize online closed-loop monitoring of the vacuum laser welding process and are difficult to meet the needs of real-time adjustment of process parameters. Summary of the invention

[0003] To address the defects of the prior art, the present invention provides an online monitoring method for vacuum laser welding, comprising the following steps: 1) collecting secondary current signals of vacuum laser welding; 2) analyzing the correlation between the secondary current signals and weld formation and penetration; 3) actively controlling the weld morphology of vacuum laser welding based on the secondary current signals.

[0004] Preferably, the vacuum laser welding online monitoring method of the present invention comprises the following steps:

[0005] 1) Collect the secondary current signal of the vacuum laser welding process;

[0006] 2) Based on the collected secondary current signal, the orthogonal experimental design method was used to carry out multi-process vacuum laser welding experiments, analyze the correlation between process parameters and weld formation and penetration, and apply the Taguchi algorithm to establish the intrinsic relationship between process parameters and weld formation and penetration; based on the signal collection electrode, signal transmission and amplification system and computer online real-time analysis technology, analyze and establish the correlation between secondary current and process parameters; based on the above analysis results, obtain the correlation law between process parameters, secondary current and weld formation;

[0007] 3) Using support vector machine, linear and nonlinear regression relationships are set, and the secondary current characteristic values ​​and weld characteristic maps under specific process parameters are compared and analyzed to obtain the characteristics of weld morphology and penetration depth of the process parameters corresponding to the electrical signal, and put them into the supervision area for supervision and classification; based on the support vector machine, the original pattern space is mapped to a high-dimensional feature space to determine whether the weld formation and penetration depth corresponding to the secondary current signal are ideal. If not, the optimal classification hyperplane is found in the feature space, and the changes in the laser parameters are fed back to control the welding process again, thereby achieving active control of the weld morphology.

[0008] For more specific details of the vacuum laser welding online monitoring method of the present invention, please refer to the specific implementation method.

[0009] The advantages and beneficial effects of the present invention are: providing an online monitoring method for vacuum laser welding, which collects the secondary current signal of vacuum laser welding, and analyzes the correlation between the secondary current signal and weld formation and penetration depth, and then actively controls the weld morphology of vacuum laser welding based on the secondary current signal, and finally effectively realizes online closed-loop monitoring of the vacuum laser welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0011] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0012] The technical solutions specifically implemented in the present invention are as follows:

[0013] like Figure 1 As shown, the present invention provides a method for online monitoring of vacuum laser welding weld formation based on secondary current signals, comprising the following steps:

[0014] 1) Collecting secondary current signals during vacuum laser welding; During vacuum laser welding, in a vacuum chamber, a large number of free electrons are generated inside the metal vapor and plasma formed by the evaporation of the metal material under laser irradiation, and these free electrons are secondary electrons, and the current formed by the secondary electrons is the secondary current; a thin copper plate is set as a secondary current collector in the vacuum chamber of vacuum laser welding, and the thin copper plate is connected to the top of the welding plate, so that a potential difference is generated between the thin copper plate and the workpiece, and an external load resistor is connected to form a loop; the thin copper plate outputs the collected secondary current to the outside, and a voltage drop is generated on the load resistor, so as to realize real-time collection of the secondary current signal during the welding process;

[0015] 2) Based on the collected secondary current signal, the orthogonal experimental design method was used to carry out multi-process vacuum laser welding experiments, analyze the correlation between process parameters and weld formation and penetration, and apply the Taguchi algorithm to establish the intrinsic relationship between process parameters and weld formation and penetration; based on the signal collection electrode, signal transmission and amplification system and computer online real-time analysis technology, analyze and establish the correlation between secondary current and process parameters; based on the above analysis results, obtain the correlation law between process parameters, secondary current and weld formation;

[0016] 3) Using support vector machine, linear and nonlinear regression relationships are set, and the secondary current characteristic values ​​and weld characteristic maps under specific process parameters are compared and analyzed to obtain the characteristics of weld morphology and penetration depth of the process parameters corresponding to the electrical signal, and put them into the supervision area for supervision and classification; based on the support vector machine, the original pattern space is mapped to a high-dimensional feature space to determine whether the weld formation and penetration depth corresponding to the secondary current signal are ideal. If not, the optimal classification hyperplane is found in the feature space, and the changes in the laser parameters are fed back to control the welding process again, thereby achieving active control of the weld morphology.

[0017] The vacuum laser welding online monitoring method of the present invention specifically comprises the following steps:

[0018] 1) Collect secondary current signals of vacuum laser welding

[0019] During vacuum laser welding, in a vacuum chamber, a large number of free electrons are generated inside the metal vapor and plasma formed by the evaporation of the metal material under laser irradiation. These free electrons are secondary electrons, and the current formed by the secondary electrons is the secondary current. A thin copper plate is set as a secondary current collector in the vacuum chamber of vacuum laser welding, and the thin copper plate is connected to the top of the welding plate. A potential difference is generated between the thin copper plate and the workpiece, and an external load resistor is connected to form a loop. The thin copper plate outputs the collected secondary current to the outside, generating a voltage drop on the load resistor, thereby realizing real-time collection of the secondary current signal during the welding process. The collected secondary current signal is transmitted and amplified. The amplified secondary current signal is subjected to digital processing such as Fourier transform. The digitally processed secondary current signal is displayed online, and the signal spectrum and characteristic values ​​are analyzed in real time, thereby finally completing the detection of the secondary current signal during the vacuum laser welding process.

[0020] Based on the high-power laser welding platform, a detection device with high light transmittance, high temperature resistance, high voltage resistance and secondary current signal collection is used; by using high-transmittance quartz glass, the penetration of the device to the laser is improved; a 15mm 304 stainless steel thick-walled cavity is used to form the overall structure of the detection device, which improves the device's high-pressure and high-temperature resistance, while also improving the device's conductivity; a flexible hinge constraint design is used to achieve miniaturization and lightweight of the laser welding device; the airtightness requirements of the device are met by using a high-temperature resistant annular sealing strip; a detection device pressure regulation system that integrates a vacuum pump, a pressure reducing valve, a stop valve, a vacuum valve, an exhaust valve and a vacuum gauge is used to achieve regulation of the internal pressure of the closed cavity during welding;

[0021] 2) Analyze the correlation between secondary current signal and weld formation and penetration

[0022] Based on the detected secondary current signal of vacuum laser welding, the orthogonal experimental design method is used to carry out vacuum laser welding experiments under multiple process conditions (including environmental pressure, laser power, welding speed, defocusing amount, collector distance). The ultra-depth three-dimensional microscope, SEM, and metallographic corrosion technology are used to obtain the weld surface morphology, weld cross-sectional morphology and deep melting map under different processes, analyze the correlation between process parameters and weld formation and melting depth, establish the intrinsic connection between process parameters and weld formation and melting depth, and establish a welding process control method; based on the signal collector, signal transmission and amplification system, and computer online real-time analysis technology, the original input electrical signal is decomposed into the eigenmode function through empirical mode decomposition and time domain analysis is performed to obtain the current size. The formula is as follows:

[0023]

[0024] Analyze the correlation between the secondary current and the process parameters to obtain a stable process control method for the current signal generated during vacuum laser welding; Based on the current signal control method and the correlation between weld morphology, penetration and process parameters, analyze the change of current signal under each process parameter to obtain the correlation between current signal and weld formation and penetration;

[0025] 3) Active control of weld morphology of vacuum laser welding based on secondary current signal

[0026] A vacuum laser welding experiment was carried out. Combined with semiconductor laser backlight source, narrowband filtering and image processing technology, the collected electrical signals were filtered and amplified by an op amp, compared with cluttered signals by a comparator, and shaped by a shaper. The main frequency of the signal and the energy ratio of the main frequency area were selected as feature parameters to characterize the signal, remove welding interference signals, and obtain specific current signal output values ​​based on feature parameter extraction. A support vector machine was used to set linear and nonlinear regression relationships, and the secondary current signal feature values ​​and weld feature maps under specific process parameters were compared and analyzed to obtain the weld morphology and penetration characteristics of the process parameters corresponding to the electrical signal, and the features were placed in the supervision area for supervision and classification. Based on the support vector machine, the original pattern space was mapped to a high-dimensional feature space to determine whether the weld formation and penetration corresponding to the current signal were ideal. If not, the optimal classification hyperplane was found in the feature space, and the changes in the laser parameters were feedback-controlled to stabilize the welding process again, thereby achieving active control of the weld morphology.

[0027] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A vacuum laser welding online monitoring method, characterized in that: The following steps are involved: 1) Collect the secondary current signal of vacuum laser welding; 2) Analyze the correlation between the secondary current signal and weld formation and penetration; 3) Actively control the weld morphology of vacuum laser welding based on the secondary current signal.

2. The vacuum laser welding online monitoring method according to claim 1, characterized in that: The following steps are involved: 1) Collect the secondary current signal of vacuum laser welding process; 2) Based on the collected secondary current signals, the orthogonal experimental design method was used to carry out multi-process vacuum laser welding experiments, analyze the correlation between process parameters and weld formation and penetration, and apply the Taguchi algorithm to establish the intrinsic relationship between process parameters and weld formation and penetration; Based on the signal collecting electrode, signal transmission and amplification system and computer online real-time analysis technology, the correlation between the secondary current and the process parameters is analyzed and established; based on the above analysis results, the correlation law between the process parameters, the secondary current and the weld formation is obtained; 3) Using support vector machine, linear and nonlinear regression relationships are set, and the secondary current characteristic values ​​and weld characteristic maps under specific process parameters are compared and analyzed to obtain the characteristics of weld morphology and penetration depth of the process parameters corresponding to the electrical signal, and put them into the supervision area for supervision and classification; based on the support vector machine, the original pattern space is mapped to a high-dimensional feature space to determine whether the weld formation and penetration depth corresponding to the secondary current signal are ideal. If not, the optimal classification hyperplane is found in the feature space, and the changes in the laser parameters are fed back to control the welding process again, thereby achieving active control of the weld morphology.

3. The vacuum laser welding online monitoring method according to claim 1, characterized in that: The specific steps include: 1) Collect secondary current signals of vacuum laser welding A thin copper plate as a secondary current collector is set in the vacuum chamber of vacuum laser welding, and the thin copper plate is connected to the top of the welding plate, so that a potential difference is generated between the thin copper plate and the workpiece, and an external load resistor is connected to form a loop; the thin copper plate outputs the collected secondary current to the outside, generating a voltage drop on the load resistor, so as to realize real-time collection of the secondary current signal during the welding process; the collected secondary current signal is transmitted and amplified; the amplified secondary current signal is subjected to digital processing such as Fourier transform; The digitally processed secondary current signal is displayed online and the signal spectrum and characteristic value are analyzed in real time, and finally the secondary current signal of the vacuum laser welding process is detected; 2) Analyze the correlation between secondary current signal and weld formation and penetration Based on the detected secondary current signal of vacuum laser welding, the orthogonal experimental design method was used to carry out vacuum laser welding experiments under multiple process conditions. The ultra-depth three-dimensional microscope, SEM, and metallographic corrosion technology were used to obtain the weld surface morphology, weld cross-sectional morphology, and deep melting map under different processes. The correlation between process parameters and weld formation and melting depth was analyzed, and the intrinsic connection between process parameters and weld formation and melting depth was established, and a welding process control method was established. Based on the signal collecting electrode, signal transmission and amplification system and computer online real-time analysis technology, the original input electrical signal is decomposed into the eigenmode function through empirical mode decomposition and time domain analysis is performed to obtain the current size; Analyze the correlation between secondary current and process parameters to obtain a stable process control method for the current signal generated during vacuum laser welding; Based on the current signal control method and the correlation between weld morphology, penetration and process parameters, the current signal changes under each process parameter are analyzed to obtain the correlation between the current signal and weld formation and penetration; 3) Active control of weld morphology of vacuum laser welding based on secondary current signal Carry out vacuum laser welding experiments, combine semiconductor laser backlight source, narrowband filtering, and image processing technology, filter and amplify the collected electrical signals through the op amp, compare the messy signals through the comparator, and perform signal shaping through the shaper. Select the signal's main frequency and main frequency area energy ratio as characteristic parameters to characterize the signal, remove welding interference signals, and obtain specific current signal output values ​​based on characteristic parameter extraction; A support vector machine is used to set up linear and nonlinear regression relationships, and the characteristic values ​​of the secondary current signal and the weld characteristic map under specific process parameters are compared and analyzed to obtain the characteristics of the weld morphology and penetration depth of the process parameters corresponding to the electrical signal, and put them into the supervision area for supervision and classification; based on the support vector machine, the original pattern space is mapped to a high-dimensional feature space to determine whether the weld formation and penetration depth corresponding to the current signal are ideal. If not, the optimal classification hyperplane is found in the feature space, and the changes in the laser parameters are fed back to control the welding process to stabilize the welding process again, thereby achieving active control of the weld morphology.

4. The vacuum laser welding online monitoring method according to claim 3 is characterized in that: In step 1), based on a high-power laser welding platform, a detection device with high light transmittance, high temperature resistance, high voltage resistance, and secondary current signal collection is used.

5. The vacuum laser welding online monitoring method according to claim 4, characterized in that: In step 1), high-transmittance quartz glass is used to improve the penetration of the device to laser light.

6. The vacuum laser welding online monitoring method according to claim 5, characterized in that: In step 1), a 15 mm thick-walled 304 stainless steel cavity is used to form the overall structure of the detection device, thereby improving the high pressure and high temperature resistance of the device and improving the conductivity of the device.

7. The vacuum laser welding online monitoring method according to claim 6, characterized in that: In step 1), a flexible hinge constraint design is adopted to achieve miniaturization and lightweight of the laser welding device.

8. The vacuum laser welding online monitoring method according to claim 7, characterized in that: In step 1), the airtightness requirement of the device is achieved by using a high temperature resistant annular sealing strip.

9. The vacuum laser welding online monitoring method according to claim 8, characterized in that: In step 1), a detection device pressure regulating system integrating a vacuum pump, a pressure reducing valve, a stop valve, a vacuum valve, an exhaust valve, and a vacuum gauge is used to adjust the internal pressure of the closed cavity during the welding process.

10. The vacuum laser welding online monitoring method according to claim 3, characterized in that: In step 2), the multiple process conditions include ambient pressure, laser power, welding speed, defocusing amount, and collector distance.

Images