Laser welding monitoring device

Abstract

The laser welding monitoring device is provided with: an acquisition interface for acquiring a signal corresponding to measurement light measured by irradiating a workpiece with laser light; an arithmetic circuit that determines a machining state on the basis of the acquired signal; and a storage device that stores, in advance, variation tendency information corresponding to a tendency in which the intensity of the measurement light varies in accordance with a position of a processing point in a region in which the laser light can be scanned by an optical system including the galvanometer mirror and the scanning lens. When an intensity signal indicating the intensity of measurement light received by scanning laser light by an optical system in a given welding mode is acquired through an acquisition interface, the arithmetic circuit corrects the acquired intensity signal based on variation tendency information for each intensity signal so as to suppress variation in intensity caused by the position of a machining point, and the welding precision is improved. The processing state is determined on the basis of the corrected intensity signal.

Description

Technical Field

[0001] The present disclosure relates to a laser welding monitoring device for monitoring a processing state based on laser welding. Background Art

[0002] As a conventional laser welding monitoring device, for example, Patent Document 1 discloses a laser welding monitoring device that can accurately detect the timing at which a laser welding machine starts welding a material to be welded and start monitoring of laser welding.

[0003] In the laser welding monitoring device of Patent Document 1, when the laser beam radiated by the processing head contacts the material to be welded (the butt joint surface of the metal plate), the light receiving unit receives the radiated light including the reflected light of the laser beam and the monitoring light based on thermal radiation which is light with a wavelength different from that of the reflected light. The spectroscopic unit spectroscopically separates the reflected light and the monitoring light contained in the radiated light, and converts the monitoring light into a first electrical signal. The trigger unit converts the reflected light into a second electrical signal, and outputs a trigger signal when the level of the second electrical signal is above a given threshold. When the trigger signal is input, the laser welding monitor starts to determine whether the laser welding of the material to be welded is proceeding normally based on the first electrical signal.

[0004] Prior Art Literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2021-137867

[0007] Patent Document 2: Japanese Patent No. 5671873 Summary of the invention

[0008] In order to move the laser at high speed and two-dimensionally during laser welding, an optical system composed of a galvanometer scanner (also called a galvanometer scanner) including a galvanometer mirror and a scanning lens (also called an fθ lens) is generally used. As a monitoring method for the processing point of welding by irradiating the workpiece with laser light by such a laser welding machine, there is a technology that receives light generated at the processing point to determine the processing status, such as Patent Document 1.

[0009] However, in the conventional structure described in Patent Document 1, a light receiving unit is provided outside the optical system for processing such as a galvanometer scanner and an fθ lens for laser irradiation. In such a structure, light is received from a wider area than the processing point, and it is difficult to accurately reflect the state of the processing point in the received light, and there is a problem that the accuracy of determining the processing state using the signal of the received light may be reduced.

[0010] On the other hand, in the conventional laser welding monitoring device described in Patent Document 2, the sensor part such as the photodiode is designed to receive light that has passed through the optical system for processing. In such a structure, for example, when an optical system including a scanning lens and a galvanometer mirror for scanning laser light is used, the angle and position of the laser light incident on the scanning lens change due to the scanning of the galvanometer mirror. Therefore, the light from the processing point generated by the irradiation of the laser light is dispersed in the scanning lens, and there is a problem that the signal intensity of the light that can be received changes depending on the position of the processing point.

[0011] The present invention aims to solve the above-mentioned conventional problems and has an object to provide a laser welding monitoring device capable of monitoring the processing state by laser welding with high accuracy even when the position of the processing point is changed by an optical system including a scanning lens and a galvanometer mirror.

[0012] A laser welding monitoring device according to one embodiment of the present disclosure is a laser welding monitoring device for monitoring a processing state based on laser welding. The laser welding monitoring device includes an acquisition interface, an operation circuit, and a storage device. The acquisition interface acquires a signal corresponding to a measurement light measured by irradiating a workpiece with a laser. The operation circuit performs operation processing to determine the processing state based on the signal acquired by the acquisition interface. The storage device pre-stores information on a change tendency corresponding to a tendency that the intensity of the measurement light changes according to the position of a processing point, which is a position of a processing point in an area where the laser can be scanned on the workpiece by an optical system including a galvanometer mirror and a scanning lens. The measurement light includes plasma light, heat radiation light, and reflected light of the laser generated at the processing point by irradiation of the laser. The measurement light is received by the optical system by scanning the laser through the optical system in a given welding mode and is split into plasma light, heat radiation light, and reflected light. After the measurement light is split, the acquisition interface acquires intensity signals indicating the respective intensities of the plasma light, the heat radiation light, and the reflected light. The calculation circuit corrects the intensity signal of each of the plasma light, thermal radiation light and reflected light based on the variation tendency information so as to suppress the variation of the intensity caused by the position of the processing point, and determines the processing state based on the corrected intensity signal.

[0013] According to the laser welding monitoring device disclosed in the present invention, the signal corresponding to the intensity of the measuring light is corrected so as to suppress the variation of the intensity of the measuring light caused by the position of the processing point. Thus, the influence on the intensity of the measuring light based on the position of the processing point can be suppressed, and for example, the processing state such as the welding state of the processing point can be monitored with high accuracy based on the signal of the corrected measuring light. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1It is a schematic diagram of the processing system in Embodiment 1 of the present disclosure.

[0015] Figure 2 This is a diagram illustrating the configuration of a monitoring device in the processing system according to the first embodiment.

[0016] Figure 3 This is a schematic diagram showing the trajectory of welding in the first embodiment as viewed from the upper surface of the workpiece.

[0017] Figure 4 This is a flowchart illustrating the operation of the monitoring device in the first embodiment.

[0018] Figure 5 This is a diagram for explaining waveform data of a signal before correction in the monitoring device according to the first embodiment.

[0019] Figure 6 This is a flowchart illustrating the pre-processing of the signal in the monitoring device according to the first embodiment.

[0020] Figure 7 This is a flowchart illustrating the correction processing for each processing position in the monitoring device according to the first embodiment.

[0021] Figure 8 This is a diagram for explaining changes in waveform data before and after correction in the first embodiment.

[0022] Fig. 9 This is a diagram for explaining waveform data after correction in the monitoring device according to the first embodiment.

[0023] Fig.10 This is a flowchart illustrating the process of creating a correction table in the monitoring device according to the first embodiment.

[0024] Fig.11 This is a diagram illustrating correction coefficients of waveform data in the first embodiment. DETAILED DESCRIPTION

[0025] (Method of the present disclosure)

[0026] The laser welding monitoring device according to the first aspect of the present disclosure monitors a processing state based on laser welding, and includes: an acquisition interface for acquiring a signal corresponding to a measurement light measured by irradiating a workpiece with laser light; an operation circuit for performing operation processing for determining the processing state based on the signal acquired by the acquisition interface; and a storage device for pre-storing change tendency information corresponding to a tendency of the intensity of the measurement light to change according to the position of a processing point, the position of the processing point being the position of the processing point in an area on the workpiece where the laser light can be scanned by an optical system including a galvanometer mirror and a scanning lens. The measurement light includes plasma light, thermal radiation light, and reflected light generated at the processing point by irradiation with the laser light. In a given welding mode, the laser is scanned by the optical system, and the measuring light is received via the optical system. When the measuring light is split into the plasma light, the thermal radiation light, and the reflected light, the acquisition interface acquires intensity signals representing the respective intensities of the plasma light, the thermal radiation light, and the reflected light. The operation circuit corrects the intensity signal for each of the plasma light, the thermal radiation light, and the reflected light based on the variation tendency information so as to suppress the variation in intensity caused by the position of the processing point, and determines the processing state based on the corrected intensity signal.

[0027] According to the second embodiment of the present disclosure, in the laser welding monitoring device described in the first embodiment, the given welding pattern includes multiple welds that make the processing conditions consistent except for the position of the processing point, and the change tendency information is generated based on the multiple intensity signals for each of the plasma light, the thermal radiation light, and the reflected light according to the measurement light from the trajectories of the multiple welds formed by the multiple welds in the scannable area.

[0028] According to the third embodiment of the present disclosure, in the laser welding monitoring device described in the second embodiment, the intensity signal that has undergone the correction has an intra-track average that is consistent with the inter-track average, and the inter-track average is an inter-track average based on the average of the intra-track averages calculated based on the intensities of each intensity signal within a given period in the multiple intensity signals corresponding to the multiple welding trajectories.

[0029] According to the 4th aspect of the present disclosure, in the laser welding monitoring device described in the 2nd aspect or the 3rd aspect, the approximate line that approximates the time variation of the intensity signal that has undergone the correction has the same inclination as the average value obtained by averaging the inclinations of the approximate lines of each of the multiple intensity signals.

[0030] According to the 5th aspect of the present disclosure, in the laser welding monitoring device described in the 2nd aspect or the 3rd aspect, the approximate line that approximates the time variation of the intensity signal that has undergone the correction has the same inclination as the inclination of the approximate line that approximates the time variation of the intensity of the laser output of the oscillating laser.

[0031] According to the sixth aspect of the present disclosure, in the laser welding monitoring device described in any one of the first to third aspects, when a parameter including the energy density (fluence) of the laser changes during welding, an approximate line that approximates the time variation of the intensity signal that has undergone the correction has the same inclination as the inclination determined according to the variation in the energy density.

[0032] According to a seventh aspect of the present disclosure, in the laser welding monitoring device described in any one of the second to sixth aspects, the variation tendency information includes a coefficient for correcting the intensity signal corresponding to each welding trajectory for each of the plurality of welding trajectories.

[0033] According to the 8th aspect of the present disclosure, in the laser welding monitoring device described in any one of the 2nd to 7th aspects, the operation circuit generates the change tendency information based on the multiple intensity signals received based on the measurement light from the trajectory of each welding, and stores it in advance in the storage device.

[0034] (Implementation method 1)

[0035] In the first embodiment, as an example using the laser welding monitoring device of the present disclosure, a processing system that performs welding processing using laser light and determines the processing state thereof will be described.

[0036] 1. Structure

[0037] Below, use Figure 1 as well as Figure 2 The structure of the processing system involved in Embodiment 1 is demonstrated.

[0038] 1-1. Structure of the processing system

[0039] Figure 1 1 is a schematic diagram of a processing system 100 in this embodiment. The processing system 100 includes: a laser processing machine 30 that performs laser welding processing by irradiating laser light; a spectrometer 13 that detects and splits light from a processing point 6 irradiated with laser light; and a monitoring device 14 that uses the detected light to monitor the processing state based on laser welding. The processing object, i.e., the workpiece to be processed, of the processing system 100 is, for example, a metal workpiece 5.

[0040] In the processing system 100 of this embodiment, for example, Figure 1As shown, the laser processing machine 30 includes a laser oscillator 1 , two fixed reflection mirrors 2 - 1 and 2 - 2 , a galvanometer mirror 3 , a scanning lens 4 , a focusing collimator 11 , and a photodetector 16 .

[0041] The laser oscillator 1 supplies light for generating the laser 10 in a pulsed state, for example. The laser oscillator 1 supplies light having a wavelength of 1070 nm, for example. The wavelength of the laser 10 is not limited thereto, and may be 532 nm or 450 nm, for example. The laser 10 from the laser oscillator 1 is reflected by, for example, fixed mirrors 2-1 and 2-2 and enters the galvanometer mirror 3. At this time, the photodetector 16 detects the leakage light 15 of the laser 10 transmitted through the mirror 2-1, and generates an electrical signal indicating the intensity of the detected light.

[0042] The galvanometer mirror 3 includes, for example, two mirrors rotatable about mutually orthogonal axes and a driving unit that rotates each mirror to a given angle. The galvanometer mirror 3 displaces the processing point 6 irradiated with the laser 10 on the workpiece 5 by, for example, changing the angle of each mirror, thereby enabling the laser 10 to scan two-dimensionally. The laser 10 reflected by the galvanometer mirror 3 is incident on the scanning lens 4.

[0043] The scanning lens 4 includes, for example, a plurality of lenses, and is configured so that the laser light 10 is at right angles to the imaging surface regardless of the incident angle. According to the scanning lens 4, for example, even if the laser light 10 is scanned on a plane on the workpiece 5 in an area that can be scanned by the galvanometer mirror 3, the laser light 10 can be focused to have a substantially constant spot diameter.

[0044] When the laser light 10 radiated from the laser processing machine 30 contacts the processing point 6 on the workpiece 5, laser welding of the workpiece 5 is performed. At this time, plasma light, which is the inherent light emission of the metal as a visible light component, and thermal radiation light in the near infrared region caused by the temperature rise are mainly generated at the processing point 6. In addition, a part of the laser light 10 that does not contribute to the processing is reflected from the processing point 6 as return light. In this way, reflected light, plasma light, and thermal radiation light are generated at the processing point 6 by irradiation of the laser light 10.

[0045] For example, Figure 1 As shown, the light generated at the processing point 6 is reflected by the galvanometer mirror 3 through the scanning lens 4 of the laser processing machine 30, and a part of it is transmitted through the reflector 2-2 and is incident on the focusing collimator 11. The focusing collimator 11 includes a focusing lens, and focuses the incident reflected light 7, plasma light 8, and thermal radiation light 9. The light focused by the focusing collimator 11 is transmitted to the spectrometer 13 through the optical fiber 12 connecting the laser processing machine 30 and the spectrometer 13.

[0046] The spectrometer 13 separates the transmitted light into reflected light 7, plasma light 8, and thermal radiation light 9 according to the wavelength. The spectrometer 13 includes, for example, three photodetectors that are highly sensitive to different wavelengths, and detects the reflected light 7, plasma light 8, and thermal radiation light 9 through each photodetector, and generates an electrical signal corresponding to the intensity of each light detected. For example, each photodetector detects the reflected light 7 having the same wavelength as the wavelength of the laser 10, the plasma light 8 having a wavelength in the visible region (for example, 400nm to 700nm), and the thermal radiation light 9 having an infrared wavelength (for example, 1300nm). The spectrometer 13 is not limited to the above example, and may also include a single photodetector that can detect the intensity of each wavelength.

[0047] The spectrometer 13 further includes, for example, an A / D converter, a CPU, and a communication circuit, etc., and converts the electrical signal from the photodetector into a digital signal (also referred to as a "signal") and sends it to the monitoring device 14. The monitoring device 14 can, for example, measure the intensity of the reflected light 7, plasma light 8, and thermal radiation light 9 corresponding to each signal based on the received signal. In the present disclosure, the reflected light 7, plasma light 8, and thermal radiation light 9 from the processing point 6 measured in this way are also collectively referred to as "measurement light".

[0048] When receiving a signal from the spectrometer 13, the monitoring device 14 determines the processing state at the processing point 6 based on the signal and outputs the determination result. The configuration of the monitoring device 14 and the operation of determining the processing state will be described later.

[0049] In addition, the electric signal generated by the photodetector 16 based on the leakage light 15 of the laser light 10 is converted into a digital signal by, for example, an A / D converter included in the photodetector 16, and is input to the monitoring device 14 through a signal line or the like as a laser output signal 17. The laser output signal 17 represents a signal corresponding to the output of the laser light 10 oscillated from the laser oscillator 1, and is used in the monitoring device 14 as a trigger signal for starting recording of the measurement light, for example.

[0050] The above-mentioned structure of the spectrometer 13 is an example, and the light may be split by a combination of dichroic mirrors or filters, or by a diffraction grating, etc. For example, a plurality of bandpass filters may be used to select the wavelengths to be passed.

[0051] 1-2. Structure of monitoring device

[0052] Figure 2This is a block diagram illustrating the structure of the monitoring device 14 in the processing system 100. The monitoring device 14 is composed of an information processing device such as a computer. The monitoring device 14 includes a CPU (Central Processing Unit) 51 for performing calculation processing, a communication circuit 52 for communicating with other devices, and a storage device 53 for storing data and computer programs.

[0053] The CPU 51 is an example of a calculation circuit in the monitoring device 14 of the present embodiment. The CPU 51 executes a control program 56 stored in the storage device 53, for example, to realize predetermined functions including determination of a machining state in the monitoring device 14 and preprocessing of a signal described later.

[0054] In addition, the arithmetic circuit configured as CPU51 in this embodiment can be implemented by various processors such as MPU or GPU, or can be configured by one or more processors. In addition, the arithmetic circuit can be a hardware circuit such as a dedicated electronic circuit or a reconfigurable electronic circuit designed to implement the above functions, or can be various semiconductor integrated circuits such as GPGPU, TPU, DSP, microcomputer, FPGA and ASIC.

[0055] The communication circuit 52 is, for example, a circuit that communicates in accordance with standards such as IEEE802.11, 4G or 5G. The communication circuit 52 can be connected to a communication network such as the Internet. The communication circuit 52 can also perform wired communication in accordance with standards such as Ethernet (registered trademark). In addition, the monitoring device 14 can communicate with other devices directly via the communication circuit 52 or communicate via an access point. The monitoring device 14 of this embodiment receives a signal from the spectrometer 13 via the communication circuit 52. In addition, the communication circuit 52 can also communicate with other devices without going through a communication network, and for example, it can also include connection terminals such as USB (registered trademark) terminals and / or HDMI (registered trademark) terminals.

[0056] The storage device 53 is a storage medium that stores computer programs and data required to realize the functions of the monitoring device 14 , and stores various data such as a control program 56 executed by the CPU 51 and a signal received from the spectrometer 13 .

[0057] The storage device 53 is composed of, for example, a magnetic storage device such as a hard disk drive (HDD), an optical storage device such as an optical disk drive, or a semiconductor storage device such as a solid state drive (SSD). The storage device 53 may include, for example, a temporary storage element composed of a RAM such as a DRAM or SRAM, or may function as an internal memory of the CPU 51.

[0058] 2. Action

[0059] Hereinafter, the operation of the processing system 100 in the present embodiment configured as described above will be described.

[0060] 2-1. About welding operation

[0061] Figure 1 The processing system 100 illustrated in the figure scans the focal point of the laser beam 10 on the workpiece 5 by using the optical system of the laser processing machine 30 including the galvanometer mirror 3 and the scanning lens 4, and performs linear welding, for example. Figure 3 , the operation of the machining system 100 of the present embodiment performing welding to form a linear welding mark, ie, a welding trajectory, on the workpiece 5 will be described.

[0062] Figure 3 Schematic diagram showing the trajectory of welding in this embodiment as viewed from the upper surface of the workpiece 5 (ie, the surface of the workpiece 5 on the laser processing machine 30 side). Figure 3 In the embodiment, four welding tracks 21 to 24 are formed in the area 20 where the laser beam 10 can be scanned by the optical system of the laser processing machine 30. The processing system 100 performs welding to form each track 21 to 24 for each pulse by, for example, repeatedly irradiating the pulsed laser beam 10. In the processing system 100 of the present embodiment, the optical system of the laser processing machine 30, for example, Figure 3 The laser beam 10 is scanned in a given welding pattern of a plurality of welding trajectories 21 to 24 as shown.

[0063] Processing system 100 Figure 3 Between the welding trajectories 21 to 24, welding is performed under the same welding conditions except for the position of the processing point 6 corresponding to the irradiation position of the laser 10 on the workpiece 5, that is, the processing position. The welding conditions can be set for the laser processing machine 30 as various parameters such as laser output, welding speed, and focal length.

[0064] exist Figure 3 Although four welding tracks 21 to 24 are illustrated in the figure, the number of welding tracks is not limited to four, and may be, for example, two or three, or may be five or more. The welding tracks may be formed at any position in the region 20 .

[0065] 2-2. Operation of monitoring device

[0066] Next, the operation of the monitoring device 14 for monitoring the machining state in the machining system 100 that performs welding in a given welding pattern as described above will be described.

[0067] Figure 4This is a flowchart illustrating the operation of the monitoring device 14 in this embodiment. For example, the processing shown in this flowchart is performed at a given time interval from the start of processing based on a given welding pattern in a workpiece 5 to the start of processing of the next workpiece 5, and each processing is executed by the CPU 51 of the monitoring device 14.

[0068] First, the CPU 51 acquires the signal of the measurement light sensed by the spectrometer 13 through the communication circuit 52 (S1). For example, the CPU 51 uses the reception of the laser output signal 17 input from the photodetector 16 of the laser processing machine 30 as a trigger to acquire the signals corresponding to the reflected light 7, the plasma light 8, and the thermal radiation light 9. In the monitoring device 14 of the present embodiment, each signal is taken in as a numerical string corresponding to the waveform data representing the time change of the intensity of the signal (hereinafter also referred to as "signal intensity"), and is stored in the internal memory of the CPU 51.

[0069] Next, the CPU 51 performs preprocessing on each acquired signal (S2). The CPU 51 applies preprocessing such as correction for each processing position described later to each signal of the measurement light. Figure 5 This is a diagram for explaining waveform data showing a signal before preprocessing, that is, before correction (hereinafter also referred to as “uncorrected waveform data”). Figure 5 The vertical axis represents the signal intensity corresponding to the intensity of the measurement light (also referred to as “light signal intensity”), and the horizontal axis represents the welding time which is the time for welding.

[0070] Figure 5 For example, in the formation Figure 3 The uncorrected waveform data of the plasma light 8 acquired during welding in the welding mode of the welding trajectories 21 to 24 are shown in FIG. Figure 5 In the figure, the time of receiving the trigger signal of each pulse corresponding to each welding track 21 to 24 is regarded as welding time "0", and the signal waveforms during the welding of each track 21 to 24 are overlapped and shown as "the first welding track", "the second welding track", "the third welding track" and "the fourth welding track" by different line types. Figure 5 In FIG. 1 , approximate straight lines obtained by linearly approximating the signal waveforms corresponding to the welding trajectories 21 to 24 are represented by the same line type as the signal waveforms as the “first approximate straight line”, “second approximate straight line”, “third approximate straight line” and “fourth approximate straight line” respectively.

[0071] like Figure 5As shown in the uncorrected waveform data, the signal intensity and the inclination of the signal waveform, that is, the inclination of the approximate straight line, vary for each welding trajectory 21 to 24, that is, for each processing position within the area 20 where the laser 10 can scan the workpiece 5. When the processing position varies due to the scanning of the laser 10, Figure 1 In the scanning lens 4 included in the optical system of the laser processing machine 30 shown, for example, light generated and returned from the processing point 6 is dispersed, emitted with a deviation from the optical axis, and enters the focusing collimator 11 at an angle. Therefore, the intensity of the measurement light transmitted to the monitoring device 14 via the optical fiber 12 varies depending on the processing position, so that, for example, Figure 5 As shown, the signal intensity changes over time in a different trend for each processing position.

[0072] Assuming that the processing state is determined using the uncorrected waveform data of the signal that changes for each processing position as described above, the change in signal strength caused by, for example, abnormality in welding is unlikely to appear in the signal, and there is a concern that it is difficult to determine the processing state with high accuracy. For this reason, in the monitoring device 14 of this embodiment, before determining the processing state, preprocessing (S2) such as correction to suppress the change for each processing position is performed on the acquired signal. The details of such signal preprocessing (S2) will be described later.

[0073] Return to Figure 4 , the CPU 51 performs a calculation process (S3) to determine the processing state of the processing point 6 based on the laser welding based on the signal of the measured light after preprocessing (S2). For example, the CPU 51 performs a calculation to compare the intensity of each signal of the reflected light 7, the plasma light 8, and the thermal radiation light 9 with a given threshold value, and determines whether the welding quality at the processing point 6 is qualified as the processing state.

[0074] After determining the processing state (S3), the CPU 51 outputs the determination result, for example, through the communication circuit 52 (S4). The determination result can be received and displayed, for example, by an external device that can perform data communication with the monitoring device 14. Alternatively, the monitoring device 14 may also include a display device such as a display that can be connected to the CPU 51 so as to communicate, and the display device may display the determination result. In addition, the CPU 51 may also notify the laser processing machine 30 of the determination result, for example, via a signal line that connects the monitoring device 14 and the laser processing machine 30.

[0075] When the determination result is output ( S4 ), the CPU 51 stores, for example, waveform data indicating the preprocessed signal ( S2 ) and the determination result of the processing state ( S3 ) in the storage device 53 , and ends the process of this flowchart.

[0076] According to the above processing, the monitoring device 14 obtains the signal of the measurement light from the spectrometer 13 (S1), performs preprocessing on the signal (S2), and determines the processing state based on the preprocessed signal (S3). Thus, for example, the processing state can be determined with high accuracy based on the signal corrected by the preprocessing.

[0077] In addition, the process of determining the machining state (S3) is not limited to the above-mentioned example, and may be performed based on a part of the measurement light signal, for example. In addition, for example, the machining state may be determined by inputting the measurement light signal to a learned model that has been learned through various machine learning to estimate the pass or fail of welding quality based on the signal intensity, and outputting the estimation result.

[0078] 2-2-1. Signal preprocessing

[0079] use Figure 6 Come to Figure 4 The details of step S2 are described below. Figure 6 This is a flowchart illustrating the pre-processing ( S2 ) of the signal in the monitoring device 14 according to the present embodiment.

[0080] In the monitoring device 14, for example, the acquired measurement light signal (S1) is stored in the internal memory of the CPU 51, and then the monitoring device 14 starts to monitor the light. Figure 6 The process of this flowchart is performed for each of the reflected light 7, the plasma light 8, and the thermal radiation light 9 in the measurement light, for example.

[0081] First, the CPU 51 performs a process of correcting the signal of the measurement light according to the processing position (S11). In such a correction process (S11) for each processing position, the CPU 51 performs a correction process based on the acquired signal, for example, Figure 3 In the welding pattern shown, the signal of the corresponding period is corrected for each welding trajectory 21 to 24. The correction process (S11) for each processing position will be described in detail later.

[0082] Next, the CPU 51 performs, for example, gain correction to correct the signal strength so as to absorb the variation in signal strength caused by the individual difference of the spectrometer 13 (S12). In addition, the CPU 51 applies, for example, a low-pass filter so as to reduce the noise contained in the signal (S13). Thereafter, the CPU 51 ends the processing of this flowchart and returns to Figure 5 Step S3.

[0083] According to the above processing, various preprocessing (S11 to S13) including correction (S11) for each processing position is applied to the signal of the measured light. Thus, the processing state (S3) can be determined with high accuracy using the preprocessed signal. In addition, the execution order of steps S11 to S13 is not limited to Figure 6 For example, you can also follow Figure 6 Execute in different order.

[0084] 2-2-2. Correction of each processing position

[0085] use Figure 7 to Figure 9 Come to Figure 7 The details of step S11 are described below.

[0086] Figure 7 This is a flowchart illustrating the correction process ( S11 ) for each machining position in the monitoring device 14 of the present embodiment. Figure 8 It is a diagram for explaining the change of waveform data before and after correction in this embodiment. Fig. 9 It is a diagram for explaining the waveform data after correction in the monitoring device 14 of the present embodiment.

[0087] In the monitoring device 14 of the present embodiment, for example, as described later, in advance in the same welding mode as that when determining the processing state, a correction table storing correction coefficients for correcting the signal is created based on the processing position on the workpiece 5 and the measurement result of the measurement light at the processing position. The created correction table is stored in the storage device 53, for example.

[0088] exist Figure 7 In the flowchart of FIG. 1 , first, CPU 51 reads out a correction table, for example, corresponding to Figure 3 The correction coefficients corresponding to the welding trajectories 21 to 24 are calculated ( S21 ).

[0089] The CPU 51 performs, for example, a comparison between the read correction coefficients and the correction coefficients obtained by welding each of the welding tracks 21 to 24. Figure 4 The CPU 51 performs correction of the acquired signal by multiplying the signal of the corresponding period in the signal acquired in step S1 ( S22 ). For example, the CPU 51 converts the waveform data of the signal so as to correct the slope and signal strength of the signal waveform.

[0090] Figure 8 Through Figure 5 Similarly, the vertical axis and the horizontal axis show the signal waveforms before and after correction of the signal corresponding to one track among the welding tracks 21 to 24. Figure 8 In FIG. 1 , the signal waveform before correction is shown by a dotted line, and the signal waveform after correction is shown by a solid line. Figure 8In FIG. 1 , the dashed line and the solid line respectively show the approximate straight line obtained by linearly approximating the signal waveforms before and after correction. Figure 8 In FIG. 1 , the waveform of the laser output signal is shown by a dotted line.

[0091] The CPU 51 converts the waveform data of the signal waveforms of the trajectories 21 to 24 of each welding based on the acquired signal by using a correction coefficient corresponding to the same welding trajectory as before correction, and outputs the waveform data so that the inclination of the corrected approximate straight line is changed from the inclination α of the approximate straight line before correction. n The CPU 51 converts the waveform data by using the correction coefficient, for example, so that the intra-track average representing the average signal strength of the waveform data based on each welding track 21 to 24 and the inter-track average representing the average signal strength between all welding tracks 21 to 24 are equal. average equal (S22). For the intra-trajectory average and inter-trajectory average I average The details will be described later.

[0092] In step S22, the CPU 51 corrects the inclination and signal strength of the signal waveform by, for example, multiplying the sampling position of each waveform data corresponding to each welding trajectory 21 to 24 by a pre-calculated correction coefficient as described later. The sampling position, for example, represents the sampling time based on the welding time "0" in the waveform data corresponding to each welding trajectory 21 to 24. By multiplying the correction coefficient in this way, for example, Figure 8 As shown, for example, waveform data after correction in which the slope of the signal waveform is changed from that before correction can be obtained.

[0093] According to the above processing, the correction coefficient corresponding to each welding track 21 to 24 is read from the correction table created in advance (S21), and the inclination and signal intensity of the signal waveform corresponding to the measurement light from each welding track 21 to 24 are corrected by the correction coefficient (S22). For example, in this embodiment, the inclination of the approximate straight line of each signal waveform is a given inclination β, and the intra-track average and inter-track average I based on the signal intensity of each signal waveform are further corrected. average Consistent.

[0094] Fig. 9 and Figure 6 Similarly, the corrected signal waveforms corresponding to the welding trajectories 21 to 24 are shown. Fig. 9 In the example, the average signal strength of each welding track 21 to 24 is Figure 5 Before correction, the average signal strength I averageSimilarly, the waveform data of the signal of the measurement light from the welding tracks 21 to 24 at different processing positions have almost the same profile. According to such correction processing (S11, S21 to S22) for each processing position, even when the processing position changes, for example, between the welding tracks 21 to 24, it is possible to obtain the same signal strength corresponding to the welding time, thereby suppressing the change of the signal strength caused by the change of the processing position.

[0095] 2-3. Creation of correction table

[0096] use Fig.10 as well as Fig.11 A process of creating in advance a correction table used in the correction process ( S11 ) for each machining position as described above will be described.

[0097] Fig.10 This is a flowchart illustrating the creation process of the correction table in the monitoring device 14 of this embodiment. Figure 4 Prior to the processing shown, CPU 51 executes Fig.10 The processing is shown in the flowchart shown. Fig.11 This is a diagram illustrating correction coefficients of waveform data in this embodiment.

[0098] exist Fig.10 In the flowchart, first, the CPU 51 obtains the uncorrected waveform data of the signal representing the measured light according to the trajectory 21 to 24 of each weld at the same processing position as when determining the processing state (S31). The CPU 51 obtains the signal of each welding trajectory 21 to 24 (S31) for each of the reflected light 7, the plasma light 8, and the thermal radiation light 9, and performs the processing after step S32. For example, for the plasma light 8, in step S31, the signal of the trajectory 21 to 24 of each weld is obtained. Figure 5 The same uncorrected waveform data.

[0099] In addition, in step S31, the CPU 51 may obtain uncorrected waveform data based on multiple weldings for each welding track 21 to 24, or may calculate average waveform data obtained by averaging multiple waveform data, for example. In this case, the average waveform data is used as the obtained uncorrected waveform data in subsequent processing.

[0100] Next, the CPU 51 calculates the average I / O value in the track for each of the welding tracks 21 to 24 based on the signal value of the uncorrected waveform data and the signal strength during a given period. n (S32). Average I within the trajectory nFor example, for the signals corresponding to the welding tracks 21 to 24, the time average of the integrated value of the signal intensity during a given period is represented. n The subscript n is a number (eg, n=1, 2, 3, 4) indicating each welding track 21 to 24. The predetermined period is set to correspond to, for example, one pulse of laser output, a period during which the signal intensity of each welding track 21 to 24 exceeds zero in the uncorrected waveform data.

[0101] In step S32, the CPU 51 calculates the intra-track average I of the signal strength of each of the welding tracks 21 to 24 using, for example, the following calculation formula: n .

[0102] I n = (integral value of signal intensity of uncorrected waveform data of the nth welding track) ÷ (welding time of the entire waveform data of the nth welding track - welding time when the laser output of the nth welding track is 0)

[0103] When the average I in each welding track 21 to 24 is calculated n When the CPU 51 calculates the average I in the track between the tracks 21 to 24 representing all the welding n The average of the inter-trajectory average I average (S33) The CPU 51 calculates the average I between tracks by, for example, the following calculation formula: average .exist Figure 3 In the example, the number of welding tracks N=4.

[0104] The average I between all welded tracks average =Σ(average I within each welding trajectory n ) ÷ the number of welding tracks N

[0105] The CPU 51 calculates the average I in each track for each welding track 21 to 24. n The average I between tracks 21 to 24 for all welding average The offset I offset，n (S34). The CPU 51 calculates each offset I by the following formula: offset，n . Each offset I offset，n In the correction process ( S11 ) for each processing position, the correction coefficient is used so that the average intensity in the corrected waveform data corresponding to the n-th welding trajectory coincides with the inter-trajectory average of all welding trajectories 21 to 24 .

[0106] I offset，n =I average -I n

[0107] Next, the CPU 51 performs linear approximation based on the uncorrected waveform data for each welding trajectory 21 to 24 to derive an approximate straight line of the signal waveform ( S35 ). Specifically, the CPU 51 calculates the inclination α of the approximate straight line for each welding trajectory 21 to 24 n and the intercept a n The approximate straight line is expressed by the following linear approximation formula.

[0108] The uncorrected linear approximation y n (x) = α n *x+a n

[0109] Here, y n (x) is the signal intensity of the uncorrected approximate straight line of each welding trajectory, and x is the sampling position of the signal waveform.

[0110] In the monitoring device 14 of the present embodiment, the waveform data is a digital signal sampled at a constant time interval, and is expressed as (welding time = sampling interval × sampling position). Since the sampling interval is constant, in the following processing performed by the monitoring device 14, the integer sampling position x is used for calculation instead of the welding time.

[0111] In addition, for example, in Figure 5 In the same uncorrected waveform data, the signal intensity in the period corresponding to the first 15% and the last 5% of the welding time of the entire waveform data varies greatly compared with other periods. Therefore, in this embodiment, for example, the start and end ends with such large variations may be excluded, and only the waveform data of the middle part corresponding to 15% to 95% of the welding time of the entire waveform data with relatively small variations may be used for the inclination α n Calculated.

[0112] When the approximate straight line before correction is derived (S35), the CPU 51 calculates the intercept b when the approximate straight line of the corrected waveform data has a given inclination β, for each of the welding trajectories 21 to 24, for example, using the following linear approximation formula: n , and derive a corrected approximate straight line (S36).

[0113] The corrected linear approximation y n (x)' = β*x + b n +I offset，n

[0114] Here, y n (x)' is the signal intensity of the corrected approximate straight line for each welding trajectory, b n In order to maintain the average I in each welding track 21 to 24 nThe intercept when the slope of the approximate straight line is converted into the corrected slope β in the state.

[0115] In deriving the corrected approximate straight line ( S36 ), the CPU 51 calculates the intercept b of the corrected approximate straight line using, for example, the following calculation formula: n .

[0116] b n =I n -{β / α n *(I n -a n )}

[0117] For a given inclination β in the corrected approximate straight line, for example, the inclination α of the approximate curve before correction calculated for each of the welding trajectories 21 to 24 is calculated using all the welding trajectories 21 to 24. n The average α average , so that the signal intensity has the same temporal variation tendency independent of the processing position.

[0118] Based on the derived approximate straight lines before and after correction ( S35 , S36 ), the CPU 51 calculates the correction coefficient γ for converting the uncorrected waveform data into the corrected waveform data as follows for each welding trajectory 21 to 24: n (x) (S37).

[0119] γ n (x) = y n (x)' / y n (x)

[0120] Fig.11 To the general Figure 3 The welding trajectories 21 to 24 of the welding are respectively used as the first to fourth welding trajectories and the correction coefficient γ calculated for the plasma light 8 n (x) is used as an example. Fig.11 The vertical axis represents the correction factor γ n (x), the horizontal axis represents the welding time corresponding to the sampling position. Fig.11 The correction coefficient γ calculated for each sampling position as a discrete time is shown. n (x) is interpolated. The correction factor γ n When (x) is 1, the signal intensity of the measurement light is not changed in the correction process ( S11 ) for each processing position, and when (x) is other than 1, the signal intensity of the measurement light is corrected to a value obtained by multiplying the correction coefficient.

[0121] When the correction factor γ is calculated n (x) (S37), the CPU 51 creates a correction table and records the correction coefficient γn (x) and stores it in the storage device 53 (S38). Then, the CPU 51 ends Fig.10 Flowchart of the process.

[0122] According to the above processing, the correction coefficient γ for correcting the uncorrected waveform data of each welding trajectory 21 to 24 is calculated. n (x), so that the welding tracks 21 to 24 have the same average I between tracks average , and each approximate straight line has a given inclination β (S31-S37). In addition, a correction coefficient γ is created and recorded. n In the correction process (S11) for each processing position, the correction table created in advance is used to correct the signal of the measurement light so as to suppress the variation of the intensity of the measurement light caused by the processing position, and it is possible to obtain Fig. 9 The corrected waveform data is shown.

[0123] The predetermined inclination β in the corrected approximate straight line is not limited to the above-mentioned α. average , for example, can also be based on Figure 8 Similarly, the waveform data representing the laser output signal is obtained from the inclination α before correction in step S35. n Similarly, the inclination α during the period of 15% to 95% of the welding time is calculated. laser The waveform data of the laser output signal is, for example, a monitoring value obtained by detecting the laser output provided by the photodetector 16 as a parameter of the welding condition when the correction table is created in the laser processing machine 30. The period corresponding to 15% to 95% of the welding time of the entire waveform data corresponds to, for example, a period when the laser output is in a stable state.

[0124] 3. Effect

[0125] As described above, the monitoring device 14 of this embodiment is an example of a laser welding monitoring device that monitors the processing state based on laser welding. The monitoring device 14 includes a communication circuit 52 as an example of an acquisition interface, a CPU 51 as an example of an operation circuit, and a storage device 53. The communication circuit 52 acquires a signal corresponding to the measurement light that is measured by irradiating the workpiece 5 (an example of the processed object) with the laser 10 (S1). The CPU 51 performs operation processing (S2, S3) to determine the processing state based on the signal acquired by the communication circuit 52. The storage device 53 pre-stores a correction table as an example of change tendency information corresponding to the tendency of the intensity of the measurement light to change according to the processing position, which is the position of the processing point 6 in the area 20 where the laser 10 can be scanned on the workpiece 5 by the optical system including the galvanometer mirror 3 and the scanning lens 4. The measurement light includes plasma light 8, thermal radiation light 9, and reflected light 7 of the laser 10. When the laser 10 is scanned by the optical system in a given welding mode, the measuring light is received through the optical system and split into plasma light, thermal radiation light and reflected light, the communication circuit 52 acquires a signal representing uncorrected waveform data as an example of an intensity signal representing the intensity of each of the plasma light 8, thermal radiation light 9 and reflected light 7 (S1). The CPU 51 corrects the acquired signal based on the correction table for each of the plasma light 8, thermal radiation light 9 and reflected light 7 so as to suppress the variation of the intensity caused by the processing position (S2, S11), and determines the processing state based on the corrected signal (S3).

[0126] According to the above monitoring device 14, even if the signal intensity of the measuring light varies due to the processing position, the intensity variation caused by the processing position can be corrected (S2, S11) along with the scanning of the laser 10 based on the optical system including the galvanometer mirror 3 and the scanning lens 4. As a result, the signal intensity can be suppressed from the influence of the varying processing position, and for example, the processing state can be determined with high accuracy based on the signal intensity of the three types of measuring light monitored (S3).

[0127] In the present embodiment, a given welding pattern includes multiple weldings that make the processing conditions consistent except for the processing position, and a correction table is generated based on multiple signals for each of the plasma light 8, the thermal radiation light 9, and the reflected light 7 according to the measurement light from the multiple welding tracks 21 to 24 formed by the multiple weldings in the area 20 where the laser 10 can be scanned (S31 to S38). Thus, for example, the correction table can be created by reflecting the change of the measurement light caused by the processing position between the welding tracks 21 to 24.

[0128] In this embodiment, the corrected signal has the same value as the average I between tracks.average The consistent intra-trajectory average, the inter-trajectory average I average The average I in the track is calculated based on the signal strength of each signal in a given period among the plurality of signals corresponding to the plurality of welding tracks 21 to 24. n The average between the tracks is the average (see S11 and S22). Thus, for example, the signal can be corrected so that changes in processing states other than the processing position under laser welding in each welding track 21 to 24 are easily reflected in the signal intensity.

[0129] In the present embodiment, an approximate straight line (an example of an approximate line) that approximates the time variation of the corrected signal has an average value α obtained by averaging the inclinations of the approximate straight lines of the intensities of the respective signals in the plurality of signals corresponding to the plurality of welding trajectories 21 to 24. average Alternatively, the approximate straight line has an inclination α which is similar to the inclination of the linear approximation that approximates the temporal variation of the intensity of the laser output of the oscillating laser 10. laser By correcting the inclination in this way, for example, it is possible to suppress the variation in the time change of the signal intensity caused by the processing position of each welding trajectory 21 to 24.

[0130] In the present embodiment, the correction table includes a correction coefficient as an example of a coefficient for correcting the signal corresponding to each of the plurality of welding trajectories 21 to 24 (S37, S38). Thus, in the correction process (S11) at each processing position, for example, a simple operation of multiplying the correction coefficient pre-recorded in the correction table by the acquired signal can be used to effectively correct the signal (S21 to S22).

[0131] In the present embodiment, the CPU 51 generates a correction table based on a plurality of signals received from the measurement light of each welding track 21 to 24, and stores it in advance in the storage device 53 (S31 to S38). In addition, the monitoring device 14 may not generate a correction table, for example, it may obtain a correction table generated by an external computing device in advance and store it in the storage device 53.

[0132] (Other Implementations)

[0133] As described above, the above-mentioned embodiment 1 is described as an example of the technology disclosed in this application. However, the technology in this disclosure is not limited thereto, and can also be applied to embodiments in which changes, substitutions, additions, omissions, etc. are appropriately made. Other embodiments are exemplified below.

[0134] In the first embodiment described above, an example is described in which the waveform data of the measurement light signal is used to calculate the corrected approximate straight line to have a constant inclination α that does not depend on time. average or α laser As a correction coefficient of the given inclination β (S36, S37), correction is performed for each processing position (S22). In the present embodiment, as the given inclination β, an inclination that changes with time may be used. For example, when the energy density of the laser 10 changes during welding as a parameter of the welding condition, an inclination determined according to the change in energy density may be used. In this case, for example, Figure 7 In step S22, the CPU 51 may also modify the acquired signal so that Figure 4 The approximate straight line of the signal intensity of the signal acquired in step S1 has a slope corresponding to the change in the energy density.

[0135] As described above, in this embodiment, when a parameter including the energy density of the laser 10 changes during welding, an approximate straight line (an example of an approximate line) that approximates the time variation of the intensity of the corrected signal (an example of a time variation of the intensity signal) has the same inclination as the inclination determined by the change in the energy density.

[0136] In the above-mentioned embodiments, the correction coefficient γ for each of the plurality of welding trajectories 21 to 24 is calculated. n (x)( Fig.10 ), and perform correction processing for each processing position ( Figure 7 In this embodiment, the correction coefficient may be calculated for a welding pattern that forms a single welding trajectory, and the correction coefficient may be used to correct each processing position. Fig.10 In the same process, the average I between the tracks can also be calculated. average , and the correction coefficient is calculated so that the slope of the corrected approximate curve becomes the slope α of the laser output waveform during the period when the laser output is in a stable state. laser .

[0137] In the above-mentioned embodiments, the monitoring device 14 is described for acquiring the signals of the reflected light 7, the plasma light 8, and the thermal radiation light 9. In this embodiment, for example, the corresponding signal may be acquired for a combination of one or two of the reflected light 7, the plasma light 8, and the thermal radiation light 9.

[0138] In each of the above-mentioned embodiments, an example is described in which the monitoring device 14 acquires a signal from the external spectrometer 13. However, the case where the spectrometer 13 and the monitoring device 14 are separate is only an example. The laser welding monitoring device of the present disclosure may also be a device in which the two are integrated. In this case, the acquisition interface in the laser welding monitoring device may be an internal input terminal or a connection terminal that receives a signal from a circuit that generates a signal by detecting light.

[0139] The present disclosure is not limited to the above-described embodiments, and various modifications are possible. That is, embodiments obtained by combining technical means appropriately modified by a person skilled in the art also fall within the scope of the present disclosure.

[0140] Industrial Applicability

[0141] The laser welding monitoring device disclosed in the present disclosure can be applied to a technology for monitoring a processing state of laser welding by scanning a laser beam on a workpiece based on measurement light from a processing point.

Claims

1. A laser welding monitoring device for monitoring the processing status based on laser welding, comprising: an acquisition interface for acquiring a signal corresponding to measurement light measured by irradiating the workpiece with laser light; an operation circuit that performs operation processing to determine the processing state based on the signal acquired by the acquisition interface; and a storage device that pre-stores information on a change tendency corresponding to a tendency of the intensity of the measurement light to change according to a position of a processing point, wherein the position of the processing point is a position of the processing point in an area where the laser light can be scanned on the workpiece by an optical system including a galvanometer mirror and a scanning lens, The measurement light includes plasma light and heat radiation light generated at the processing point by irradiation of the laser light, and reflected light of the laser light. The laser is scanned by the optical system in a given welding mode, and the measurement light is received through the optical system, wherein the measurement light is split into the plasma light, the thermal radiation light, and the reflected light. After the measurement light is split, the acquisition interface acquires intensity signals indicating the respective intensities of the plasma light, the thermal radiation light, and the reflected light. The operation circuit performs the following processing: For each of the plasma light, the thermal radiation light, and the reflected light, the intensity signal is corrected based on the variation tendency information so as to suppress the variation of the intensity caused by the position of the processing point. The processing state is determined based on the corrected intensity signal.

2. The laser welding monitoring device according to claim 1, wherein: The given welding mode includes a plurality of welding processes that make the processing conditions the same except for the position of the processing point, The multiple welding processes form multiple welding tracks in the scannable area. The variation tendency information is generated based on the plurality of intensity signals corresponding to the plurality of welding trajectories for each of the plasma light, the heat radiation light, and the reflected light of the measurement light respectively coming from the plurality of welding trajectories.

3. The laser welding monitoring device according to claim 2, wherein: The intensity signal subjected to the correction has an intra-track average that matches an inter-track average of an average of intra-track averages calculated based on the intensities of each of the plurality of intensity signals within a given period.

4. The laser welding monitoring device according to claim 2 or 3, wherein: An approximation line that approximates the time variation of the corrected intensity signal has the same inclination as an average value obtained by averaging the inclinations of the approximation lines of the plurality of intensity signals.

5. The laser welding monitoring device according to claim 2 or 3, wherein: The approximation line that approximates the time variation of the intensity signal that has been corrected has the same inclination as the inclination of the approximation line that approximates the time variation of the intensity of the laser output of the oscillating laser light.

6. The laser welding monitoring device according to claim 1, wherein: When parameters including the energy density of the laser light change during welding, the approximation line that approximates the time change of the corrected intensity signal has the same inclination as the inclination determined by the change in the energy density.

7. The laser welding monitoring device according to claim 2, wherein: The variation tendency information includes a coefficient for correcting the intensity signal corresponding to the welding trajectory for each of the plurality of welding trajectories.

8. The laser welding monitoring device according to claim 2, wherein: The calculation circuit generates the change tendency information based on the plurality of intensity signals and stores the information in advance in the storage device.

Images