JP7250231B1 - Laser device, laser processing device, learning device, reasoning device, laser processing system, and laser processing method - Google Patents

Abstract

The laser device (500) includes a Q-switched laser oscillator (100) that generates pulsed laser light, an amplifier (200) that amplifies the pulsed laser light, and a Q-switched laser oscillator (100) and amplifier (200). Transmittance of the optical switching element (26) based on the optical switching element (26) arranged on the optical path and the pulse characteristic time that indicates the characteristics of the time interval at which the Q-switched laser oscillator (100) generates the pulsed laser light. a control device (35) for modulating the

Description

The present disclosure relates to a laser device for laser processing, a laser processing device, a learning device, an inference device, a laser processing system, and a laser processing method.

When a high-power pulsed laser beam is required, a configuration is employed in which a pulsed laser beam with a lower output than the required laser output is generated and then amplified by an amplifier. In such a configuration, if the pulse period, which is the time interval for generating the pulsed laser light, is not constant, the time interval at which the pulsed laser light is incident on the amplifier changes, and the gain of the amplifier changes for each pulsed laser light. Variation occurs in the pulse energy of the pulsed laser light output from the amplifier.

In Patent Document 1, when a pulsed laser beam with an irregular pulse period is amplified by an amplifier, in addition to a main signal, a secondary signal having a wavelength different from that of the main signal is input to the amplifier, thereby increasing the pulsed laser beam to the amplifier. A pulsed laser system is disclosed that suppresses variations in the time interval between incident beams and suppresses amplifier gain fluctuations. In this pulsed laser system, the secondary signal is removed from the output of the amplifier and the main signal is extracted as the laser output.

Japanese Patent Publication No. 2018-531524

However, according to the above-described conventional technique, when the desired pulse period is long, more secondary signals are removed after amplification. As a result, most of the input power is used for the secondary signal rather than the main signal, resulting in a decrease in energy efficiency.

The present disclosure has been made in view of the above, and suppresses fluctuations in post-amplification pulse energy even when the pulse period changes while suppressing a decrease in conversion efficiency from input power to laser output. It is an object of the present invention to obtain a laser device capable of

In order to solve the above-described problems and achieve the object, the laser device of the present disclosure includes a Q-switched laser oscillator that generates pulsed laser light, an amplifier that amplifies the pulsed laser light, and a Q-switched laser oscillator and the amplifier. a control device that modulates the transmittance of the optical switching element based on the optical switching element disposed on the optical path between the Q-switched laser oscillator and the pulse characteristic time that indicates the characteristics of the time interval at which the Q-switched laser oscillator generates the pulsed laser light; characterized by comprising

According to the present disclosure, it is possible to obtain a laser device capable of suppressing fluctuations in post-amplification pulse energy even when the pulse period changes while suppressing deterioration in conversion efficiency from input power to laser output. It has the effect of being able to

1 shows a configuration of a laser device according to a first embodiment; FIG. 1 is a diagram showing an example of a detailed configuration of a laser device according to a first embodiment; FIG. 1 is a diagram showing the configuration of a laser processing apparatus using the laser apparatus according to Embodiment 1 as a laser light source; FIG. A diagram showing an example of a target shape of a hole formed in an electronic substrate A diagram showing an example of a pattern of processing holes to be processed on an electronic substrate and a processing route. FIG. 6 is a diagram showing the time change of the waveform of the pulsed laser beam when machining the machined holes of the pattern shown in FIG. 5 ; Explanatory diagram of the effect of the laser device Flowchart for explaining the control operation of the laser processing apparatus according to the second embodiment Flowchart for explaining the control operation of the laser processing apparatus according to the third embodiment FIG. 11 shows a configuration of a learning device according to a fourth embodiment; Diagram showing an example of a three-layer neural network Flowchart for explaining learning processing of the learning device Configuration diagram of an inference device for a laser device Flowchart for explaining processing for obtaining transmittance using an inference device FIG. 11 shows a configuration of a learning device according to a fifth embodiment; Configuration diagram of inference device for laser processing equipment FIG. 4 shows an example of waveforms of pulsed laser beams with different pulse intervals; Explanatory diagram of an example of changing the transmittance over time within the pulse time Explanatory diagram of transmittance control according to the seventh embodiment FIG. 12 is a diagram showing a partial configuration between electrodes of a laser device according to an eighth embodiment; FIG. 20 is a graph showing steady-state gain distribution between electrodes of the laser device having the configuration shown in FIG. FIG. 11 shows an internal configuration of a laser device according to a ninth embodiment; Diagram showing the steady-state temperature distribution between the electrodes of the laser device Flowchart for explaining the operation of the laser processing apparatus according to the tenth embodiment

A laser device, a laser processing device, a learning device, an inference device, a laser processing system, and a laser processing method according to embodiments of the present disclosure will be described below in detail with reference to the drawings.

Embodiment 1.
FIG. 1 is a diagram showing the configuration of a laser device 500 according to the first embodiment. The laser device 500 includes a Q-switched laser oscillator 100 that generates pulsed laser light using an element having a shutter function such as a Q-switch, an amplifier 200 that amplifies the pulsed laser light generated by the Q-switched laser oscillator 100, It has an optical switching element 26 arranged on the optical axis 2 between the Q-switched laser oscillator 100 and the amplifier 200, a control device 35, and an information processing device .

A pulsed laser beam generated by the Q-switched laser oscillator 100 is incident on the optical switching element 26 , passes through the optical switching element 26 with a set transmittance, and is incident on the amplifier 200 along the optical axis 2 . The transmittance of optical switching element 26 is modulated according to a signal from controller 35 . Note that "modulating the transmittance" refers to changing the transmittance over time. The optical switching element 26 is, for example, an acousto-optic element, an electro-optic element, or the like.

In parallel with the control of the transmittance of the optical switching element 26, the control device 35 controls the pulse cycle, which is the time interval for the Q-switched laser oscillator 100 to oscillate the pulsed laser light, the pulse width, the Q-switched laser oscillator 100 and the amplifier 200. It controls the input power to each.

The information processing device 36 calculates a control signal for the optical switching element corresponding to the oscillation interval of the pulsed laser light based on information acquired by sensors (not shown) installed in the Q-switched laser oscillator 100 and the amplifier 200, Information including the calculation result is sent to the control device 35 .

Information acquired by the sensor includes, for example, the pulse energy of the pulsed laser light, the waveform of the pulsed laser light, and the temperature information of the laser device 500 . The control device 35 controls the operation of the laser device 500 by sending control signals to the optical switching element 26 and the like based on the information from the information processing device 36 .

The pulsed laser light that has passed through the optical switching element 26 and entered the amplifier 200 is amplified by the amplifier 200 and then emitted along the optical axis 3 . The optical switching element 26 is controlled so that fluctuations in the pulse energy of the pulsed laser light after exiting the amplifier 200 are reduced when the pulsed laser light is continuously generated. Although the laser device 500 has one optical switching element 26 in FIG. 1 , the laser device 500 may have a plurality of optical switching elements 26 .

Further, FIG. 1 shows the case where the Q-switched laser oscillator 100 and the amplifier 200 are separate devices. It may also be a single device with

The functions of the control device 35 and the information processing device 36 are realized using processing circuits. These processing circuits may be implemented by dedicated hardware, or may be control circuits using a CPU (Central Processing Unit).

If the processing circuit is dedicated hardware, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or any of these A combination is used.

When the processing circuit described above is realized by a control circuit using a CPU, the control circuit includes a processor and a memory. A processor is a CPU, and is also called a processing device, an arithmetic device, a microprocessor, a microcomputer, a DSP (Digital Signal Processor), or the like. Memory includes, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable ROM), EEPROM (registered trademark) (Electrically EPROM), magnetic They include discs, flexible discs, optical discs, compact discs, mini discs, and DVDs (Digital Versatile Discs).

When the above processing circuit is realized by a control circuit using a CPU, the processor reads out and executes a program corresponding to the processing of the control device 35 and the information processing device 36 stored in the memory. The memory is also used as temporary memory in each process performed by the processor.

FIG. 2 is a diagram showing an example of a detailed configuration of the laser device 500 according to the first embodiment. FIG. 2 shows a perspective view of the internal configuration of the laser device 500 . The laser device 500 is, for example, a triaxial orthogonal carbon dioxide laser. Also, the laser device 500 may be a carbon monoxide laser, an excimer laser, or the like.

In the configuration shown in FIG. 2, the parts forming the Q-switched laser oscillator 100 and the amplifier 200 of the laser device 500 shown in FIG.

A laser gas is sealed in the housing 300 . The laser device 500 has fans 40 and 41 , a pair of electrodes 11 and 12 , and heat exchangers 42 and 43 in a housing 300 . The laser gas is caused to flow by blowers 40 and 41 in gas flow directions 13 and 14 indicated by arrows, respectively. The laser gas then passes between electrodes 11 and 12 . A high-frequency power is applied to the electrodes 11 and 12 under the control of the discharge controller 44 to generate silent discharge between the electrodes. This discharge excites the laser gas. The laser gas then flows along gas flow directions 15,16 indicated by arrows and is cooled by heat exchangers 42,43.

The laser device 500 also has a total reflection mirror 21 and a partial reflection mirror 24 . The total reflection mirror 21 and the partial reflection mirror 24 constitute a resonator. are placed. However, since the laser gas remains excited for a certain period of time even after passing through the discharge space, the optical axis 1 may pass through a position outside the discharge space.

The laser device 500 also has a Q switch 22 . The Q switch 22 is arranged on the optical axis 1 of the resonator and controls the Q value of the resonator to cause pulse oscillation and generate pulsed laser light. The Q switch 22 may be, for example, an acoustooptic device or an electrooptic device.

The total reflection mirror 21 constituting the resonator reflects most of the light but allows a small amount of light to pass therethrough. The optical sensor 50 measures the pulse waveform, pulse energy, etc. of the pulsed laser light oscillated by the resonator. The laser device 500 further has a window 23 that blocks the laser gas in the housing 300 from the outside air. In the example shown in FIG. 2, the total reflection mirror 21, the Q switch 22, and the optical sensor 50 are arranged outside the housing 300, and the window 23 prevents laser gas from leaking out of the housing 300 and partially reflects light. Light from the mirror 24 is passed outside the housing 300 toward the total reflection mirror 21 . The partial reflection mirror 24 also plays a role of blocking the laser gas inside the housing 300 from the outside air. The element arranged between the optical sensor 50 and the partially reflecting mirror 24 along the optical axis 1 constitutes the Q-switched laser oscillator 100 .

In the above description, the Q switch 22 is installed in the atmosphere, but the Q switch 22 is installed in the housing 300 and the total reflection mirror 21 instead of the window 23 is used to separate the laser gas in the housing 300 from the outside air. and may be blocked. Also, one or more mirrors may be installed between the window 23 and the total reflection mirror 21 .

The pulsed laser light emitted from the partially reflecting mirror 24 of the oscillator is incident on the optical switching elements 26 and 27 via the mirror 25, and the pulsed laser light transmitted through the optical switching elements 26 and 27 is transmitted via the mirror 28 to the window. 29 and enters the housing 300 again along the optical axis 2 . At this time, the beam diameter of the pulsed laser light may be adjusted by using a lens, a curvature mirror, or the like on the optical path from the partially reflecting mirror 24 of the resonator to the window 29 . The pulsed laser beam entering the housing from the window 29 passes through the excited laser gas, is amplified, is reflected by the mirror 30, and enters the excited laser gas again. Then, after passing through the excited laser gas and being amplified, the light is reflected by the mirror 31 and enters the excited laser gas again. Then, after passing through the pumped laser gas and being amplified, the light is reflected by the mirror 32 and enters the pumped laser gas again. After that, after passing through the pumped laser gas and being amplified, the amplified pulsed laser light is output to the outside of the housing 300 from the window 33 . The pulsed laser beam after amplification is reflected by the partially reflecting mirror 34 along the optical axis 3 and output as laser output. Light partially transmitted by the partially reflecting mirror 34 enters the optical sensor 51 . The optical sensor 51 measures the pulse waveform, pulse energy, etc. of the pulsed laser light output as the laser output.

A portion including the optical sensor 51 and the pulsed laser beam entering the housing through the window 29 passes through the excited laser gas in the housing 300 and is output from the partially reflecting mirror 34, and the amplifier 200. This is the part to do.

The laser apparatus 500 shown in FIG. 2 includes a Q-switched laser oscillator 100 and an amplifier 200 in one housing 300, and the optical axes of the Q-switched laser oscillator 100 and the amplifier 200 extend through at least one continuous discharge space. It has an integrated MOPA (Master Oscillator Power Amplifier) configuration. Here, an aperture (not shown) may be arranged between the reflecting surface of each mirror in the housing 300 and the discharge space to limit the range through which light passes.

The laser device 500 can perform pulse oscillation by continuously discharging and rapidly changing the Q value of the resonator with the Q switch 22 .

The controller 35 is connected to the Q switch 22, the optical switching elements 26 and 27, and the discharge controller 44 and controls them.

The information processing device 36 receives information from the optical sensors 50 and 51 . The information processing device 36 may also receive control signals for the optical switching elements 26 and 27 and control signals for the discharge current from the control device 35 . The information processing device 36 may calculate the pulse energy of the pulsed laser light from the pulse waveform of the pulsed laser light.

The information processing device 36 also receives information from sensors (not shown) such as the temperature of the laser gas, the gas pressure, the temperature of the laser device 500, the temperature of the cooling water, and other information that changes with time. may receive information indicating Further, as described below, when the laser device 500 is used as a laser light source for a laser processing device, information indicating the timing of laser oscillation and the setting of the galvanomirror mounted on the laser processing device are transmitted from the laser processing device. Information such as positioning time, including time, rest time, settling time and rest time may be received.

The information processing device 36 calculates information necessary for controlling the Q switch 22 , the optical switching elements 26 and 27 and the discharge control device 44 based on the received information, and sends the information to the control device 35 .

FIG. 3 is a diagram showing the configuration of a laser processing device 510 using the laser device 500 according to the first embodiment as a laser light source. The laser processing device 510 has a laser device 500 and a drilling machine 400 . The drilling machine 400 is a device that drills holes using the pulsed laser beam output from the laser device 500, and can drill holes in electronic substrates, for example. The drilling machine 400 is positioned using a positioning mechanism such as a galvanomirror 403 so that the pulsed laser beam is applied to a predetermined position on the electronic substrate.

A pulsed laser beam output from the laser device 500 is incident on the drilling machine 400 . The pulsed laser beam is applied to the galvanomirror 403 via the mirrors 401 and 402 of the drilling machine 400 . The galvanomirror 403 is an example of a deflection element that adjusts the irradiation position of the pulsed laser beam on the workpiece 405 by deflecting the pulsed laser beam. Optical elements such as a lens and a spherical mirror (not shown) for adjusting the beam diameter are arranged on the optical path from when the pulsed laser beam is incident on the drilling machine 400 to when it is irradiated onto the galvanomirror 403. or a mask may be placed to shape the beam profile. In addition, although one galvanomirror 403 is shown in FIG. 3, two galvanomirrors 403 may be used for two axes so that a plane can be scanned.

The pulsed laser beam irradiated to the galvanomirror 403 is irradiated onto the processing object 405 via the lens 404 which is an objective optical system. When the above mask is used, a transfer optical system may be used to transfer the pulsed laser beam shaped by the mask onto the object 405 to be processed. The workpiece 405 may be placed on a table movable in three orthogonal axial directions.

The position of the object 405 to be irradiated with the pulsed laser light is positioned by adjusting the angle of the galvanomirror 403 .

A position positioned by the galvanomirror 403 is irradiated with one shot or multiple shots of the pulsed laser beam, then another position is positioned by the galvanomirror 403, and then the next pulsed laser beam is irradiated. A large number of holes are machined on the workpiece 405 by continuously repeating such processing.

The time required for the positioning of the galvanomirror 403 depends on the target shape of the hole to be processed in the electronic substrate, that is, the hole processing pattern. FIG. 4 is a diagram showing an example of a target shape of a hole to be formed in an electronic substrate. The arrows in FIG. 4 indicate the machining paths of the holes, and in the following description, each machining path is specified by the number attached to the machining path. When the holes are machined in numerical order such as machining path #1, machining path #2, machining path #3 . , Machining path #6, Machining path #7, etc., the time required for the positioning of the galvanomirror 403 is short, and the portions where the holes are far apart, for example, Machining path #3, Machining path #5, Machining path #8 For example, the time required for the positioning of the galvanomirror 403 is long. Therefore, when a plurality of holes are continuously drilled in an electronic substrate, the positioning time of the galvanomirror when processing each hole is not constant and fluctuates with time.

Since the pulsed laser light is emitted at the timing when the positioning of the galvanomirror 403 is completed, the timing at which the pulsed laser light is emitted from the laser device 500 also fluctuates with time. In the Q-switched oscillation, if the oscillation timing of the pulsed laser light changes with time, the gain accumulation time changes for each pulsed laser light, so the pulse energy of the oscillating pulsed laser light also changes with time. Therefore, if the pulsed laser light oscillated by the Q-switched laser oscillator 100 is amplified by the amplifier 200 as it is, the pulse energy of the pulsed laser light also changes with time. In this case, in the drilling machine 400, the pulse energy of the pulse laser light changes for each hole to be processed. This state is shown in FIGS. 5 and 6. FIG.

FIG. 5 is a diagram showing an example of a pattern of machining holes to be machined on an electronic substrate and a machining path. FIG. 6 is a diagram showing temporal changes in the waveform of the pulsed laser beam when machining holes having the pattern shown in FIG. As shown in FIG. 5, when drilling holes #1, #2, #3, and #4 in order, the waveform of pulsed laser light by Q-switch oscillation is shown in FIG. become. The horizontal axis in FIG. 6 indicates time, and the numbers attached to the horizontal axis correspond to the numbers attached to the machined holes in FIG. Compared to the machined hole #1 and the machined hole #2, the machined hole #3 and the machined hole #4 have a longer hole interval. For this reason, the pulse energy of the machined hole #3 and the machined hole #4 is higher than that of the pulsed laser light irradiated when machining the machined hole #1 and the machined hole #2. In this case, since the pulse energy of the pulsed laser beam used for machining varies for each hole to be machined, the machining quality may not be stable.

Therefore, in the laser device 500, the pulsed laser light emitted from the Q-switched laser oscillator 100 is transmitted through the optical switching element 26 in order to suppress variation in pulse energy for each pulsed laser light incident on the drilling machine 400. The actual transmittance is adjusted for each pulsed laser beam to suppress variation in the pulse energy of the pulsed laser beam incident on the amplifier 200 .

FIG. 7 is an explanatory diagram of the effect of the laser device 500. FIG. FIG. 7A shows the waveform of pulsed laser light oscillated by the Q-switched laser oscillator 100. FIG. FIG. 7B shows the transmittance of the optical switching element 26. FIG. FIG. 7(c) shows the waveform of the pulsed laser light input to the amplifier 200. FIG. FIG. 7(d) shows the waveform of the pulsed laser light output from the amplifier 200. FIG.

When the oscillation timing of the Q-switched laser oscillator 100 is not a constant period but the pulse period changes, the pulse energy of the pulsed laser light output from the Q-switched laser oscillator 100 increases as the time interval from the previous oscillation timing increases. growing. For example, the lengths of intervals from the previous pulsed laser beam are, in order from the longest, pulsed laser beam #3, pulsed laser beam #4, pulsed laser beam #1, pulsed laser beam #1 and pulsed laser beam #1. 2 are the same, the pulse energy of the pulse laser light output from the Q-switched laser oscillator 100 is increased in order of pulse laser light #3, pulse laser light #4, and pulse laser light #1. Yes, the pulsed laser beam #1 and the pulsed laser beam #2 are the same.

In the amplifier 200, when pulsed laser light with the same pulse energy is input, the longer the time interval from the previous pulsed laser light, the higher the pulse energy of the output pulsed laser light. Therefore, the transmittance of the optical switching element 26 is controlled to be smaller as the time interval of the pulsed laser light becomes longer so that the pulse energy of the pulsed laser light incident on the amplifier 200 becomes smaller as the time interval of the pulsed laser light becomes longer. . In the example of FIG. 7, the order of the pulse energy of the pulsed laser beams incident on the amplifier 200 is, from the smaller one, the pulsed laser beam #3, the pulsed laser beam #4, and the pulsed laser beam #1. Pulsed laser beam #3, pulsed laser beam #4, and pulsed laser beam #1 have the same transmittance so that #1 and pulsed laser beam #2 are the same. 1 and pulsed laser beam #2 are the same. As a result, the pulse energy of the pulsed laser light emitted from the amplifier 200 approaches uniformity, and variations are suppressed.

The time interval between the pulsed laser beams can be detected, for example, based on the pulse waveform, and the detected value can be used to control the transmittance. Identifying the generation time of the pulsed laser light from the characteristic shape of the pulse waveform based on the measured voltage of a sensor capable of measuring the time dependence of the light intensity of the pulsed laser light, such as a photodetector or phototube, A time interval between pulsed laser light can be defined. For example, the peak time of the pulse may be set as the generation time, the time when the light intensity rises to a predetermined ratio with respect to the pulse peak may be set as the generation time, or the pulse may reach the peak after rising to the peak. On the other hand, the time when the light intensity has decreased to a predetermined ratio may be set as the occurrence time.

The laser device 500 modulates the transmittance of the optical switching element 26 for each pulsed laser beam based on the pulse characteristic time that indicates the characteristics of the time interval at which the Q-switched laser oscillator 100 generates the pulsed laser beam. The pulse characteristic time is, for example, a time interval between a plurality of pulsed laser beams generated by the Q-switched laser oscillator 100, or a moving average of time intervals between a plurality of pulsed laser beams generated by the Q-switched laser oscillator 100. It's okay. The laser device 500 has a higher transmittance as the time interval indicated by the pulse characteristic time becomes shorter, and the transmittance of the optical switching element 26 is adjusted so that the variation in energy between the pulsed laser beams becomes smaller than a predetermined value. to control.

As shown in FIG. 2, when the laser device 500 has a plurality of optical switching elements 26 and 27, the combined transmittance of the plurality of optical switching elements 26 and 27 in the laser device 500 satisfies the above conditions. to control.

As described above, the laser device 500 according to the first embodiment includes the Q-switched laser oscillator 100 that generates pulsed laser light, the amplifier 200 that amplifies the pulsed laser light, and the Q-switched laser oscillator 100 and the amplifier 200. Control for modulating the transmittance of the optical switching element 26 based on the optical switching element 26 arranged on the optical path between the Q-switched laser oscillator 100 and the pulse characteristic time that indicates the characteristics of the time interval at which the Q-switched laser oscillator 100 generates the pulsed laser light. a device 35; Modulating the transmittance means changing the transmittance value over time. Specifically, in Embodiment 1, the control device 35 changes the transmittance value for each pulsed laser beam incident on the optical switching element 26 . As a result, the time interval at which the Q-switched laser oscillator 100 generates pulsed laser light changes, the pulse energy of the pulsed laser light generated by the Q-switched laser oscillator 100 differs for each pulsed laser light, and the amplification factor of the amplifier 200 fluctuates. Even if the pulse energy is increased, it is possible to suppress variations in the pulse energy after amplification. At this time, in the method of suppressing the change in the time interval at which the pulsed laser light is incident on the amplifier by inputting the secondary signal into the amplifier in addition to the main signal, the secondary signal is removed and the main signal is output as the laser signal. A reduction in conversion efficiency from input power to laser output, which is caused by extraction as output, is suppressed. Therefore, it is possible to suppress fluctuations in post-amplification pulse energy even when the pulse period changes, while suppressing a decrease in conversion efficiency from input power to laser output.

In addition, the control device 35 calculates the time intervals between the plurality of pulsed laser beams generated by the Q-switched laser oscillator 100, or the moving average of the time intervals between the plurality of pulsed laser beams generated by the Q-switched laser oscillator 100. Obtained as pulse characteristic time, the shorter the time interval, the higher the transmittance of the optical switching element 26 and the smaller the variation in energy among the plurality of pulsed laser beams emitted from the amplifier 200 than a predetermined value. The transmittance of the optical switching element 26 is controlled as follows. Here, for example, in the example of FIG. It refers to control so that the difference in pulse energy between the pulsed laser beams #1 to #4 shown is smaller than a predetermined value. This makes it possible to suppress variations in post-amplification pulse energy even when the pulse period changes, while suppressing a decrease in conversion efficiency from input power to laser output.

In the above description, the pulsed laser beam is emitted at the timing when the positioning of the galvanometer mirror 403 is completed. It is conceivable that the gain accumulated in between is too small and the pulse energy of the pulsed laser light output from the amplifier 200 becomes smaller than the desired value. In this case, control may be performed such that the galvanomirror 403 is stopped until the required pulse energy is obtained after amplification, and the pulse laser beam is irradiated at the time when the required pulse energy is obtained.

Embodiment 2.
In Embodiment 2, a control flow of a laser processing device 510 that uses the laser output of the laser device 500 for the drilling machine 400 for electronic substrates will be described.

FIG. 8 is a flow chart for explaining the control operation of the laser processing device 510 according to the second embodiment. The control device 35 of the laser device 500 first acquires data indicating the drilling pattern of the electronic substrate to be processed by the drilling machine 400 (step S101). For example, the data indicating the drilling pattern may include positional information of the holes in the electronic substrate to be processed.

Subsequently, the control device 35 calculates the drilling path of the electronic board from the acquired data indicating the drilling pattern (step S102). After that, the control device 35 calculates an oscillation interval of pulsed laser light, which is a time interval at which the Q-switched laser oscillator 100 generates pulsed laser light for machining each hole (step S103).

The control device 35 calculates the initial value of the transmittance of the optical switching element 26 for each pulsed laser beam according to the oscillation interval of the pulsed laser beam (step S104). The controller 35 stores the calculated oscillation interval of the pulsed laser beam and the transmittance of the optical switching element 26 in, for example, a memory in the controller 35, and calculates the transmittance of the optical switching element 26 for each pulsed laser beam. is set (step S105).

The control device 35 controls the Q-switched laser oscillator 100 and the optical switching element 26 using the calculated oscillation interval of the pulsed laser light and the calculated transmittance of the optical switching element 26 , and the Q-switched laser oscillator 100 Laser oscillation is performed (step S106).

In addition, the control device 35 acquires the pulse energy of the amplified pulsed laser light acquired by the optical sensor 51 via the information processing device 36 (step S107). The number of times laser oscillation is performed in step S106 may be, for example, the same number as the number of holes in one electronic board, or one electronic board may be divided into a plurality of times. The pulsed laser light oscillated here is not applied to the electronic substrate, and the hole drilling process is not yet performed.

The control device 35 determines whether or not the value indicating variations in pulse energy of the amplified pulsed laser beams obtained in step S107 is equal to or less than a threshold (step S108). If the variation in pulse energy is not equal to or less than the threshold (step S108: No), the control device 35 oscillates a pulsed laser beam whose variation in pulse energy is equal to or greater than the threshold based on the variation in pulse energy for each pulsed laser beam after amplification. Then, the transmittance of the optical switching element 26 is corrected (step S109), and the process returns to step S105. Therefore, the laser device 500 sets the post-correction transmittance, controls the optical switching element 26 using the post-correction transmittance, causes laser oscillation again, and obtains pulse energy after amplification. It will be.

If the variation in pulse energy is equal to or less than the threshold (step S108: Yes), the laser processing device 510 starts drilling the electronic substrate (step S110). By performing such processing, the laser device 500 repeats the correction of the transmittance of the optical switching element 26 until the variation in pulse energy after amplification measured by laser oscillation becomes equal to or less than the threshold.

The controller 35 can correct the transmittance using, for example, PID (Proportional Integral Differential) control. Alternatively, the transmittance may be determined by machine learning, which will be described later.

If it is known in advance that the variation in pulse energy is within a predetermined range at the value set as the initial value of the transmittance of the optical switching element 26, the acquisition of the pulse energy and the correction of the variation in pulse energy are performed. Processing may be omitted and processing may be performed as it is. In other words, feedback control may be unnecessary during machining. Also, the acquisition of pulse energy and the correction of variations in pulse energy may be performed while processing is being performed, that is, feedback control may be performed.

The division of functions between the control device 35 and the information processing device 36 is an example, and the processing described above as the functions of the control device 35 may be performed by the information processing device 36, or the functions of the information processing device 36 may be performed by the information processing device 36. may be performed by the control device 35 . For example, acquisition of drilling pattern data for electronic boards, calculation of drilling paths for electronic boards, calculation of oscillation intervals of pulsed laser light, calculation of initial values of transmittance of the optical switching element 26, and calculation of values indicating variations in pulse energy. The information processing device 36 may perform the calculation, the calculation of the correction amount of the transmittance of the optical switching element 26, and the like.

As described above, the laser processing method according to the second embodiment includes the step of generating pulsed laser light by the control device 35 controlling the Q-switched laser oscillator 100, and the control device 35 transmitting the pulsed laser light. a step of setting the transmittance of the optical switching element 26 that allows the light to be transmitted; a step of changing the pulse energy of the pulsed laser light by making the pulsed laser beam incident on the optical switching element 26 with the set transmittance; a step of amplifying the light; a step of deflecting the amplified pulsed laser light by the galvanomirror 403 to adjust a position where the pulsed laser light is irradiated onto the workpiece 405; and irradiating the workpiece 405 with light or transfer, and in the step of changing the pulse energy of the pulsed laser light, the light is The transmittance of the switching element 26 is modulated.

Embodiment 3.
In the second embodiment, the variation in the pulse energy of the pulsed laser light is controlled to be within a predetermined range. good.

FIG. 9 is a flow chart for explaining the control operation of the laser processing device 510 according to the third embodiment. The processing from step S101 to step S106 is the same as that in FIG. 8, and thus the description is omitted here.

When the laser is oscillated in step S106, the laser processing device 510 uses the pulsed laser beam output from the laser device 500 to start drilling the electronic substrate (step S121). After performing the drilling process, the laser processing device 510 acquires the shape of the processed hole after processing (step S122). The shape of the machined hole may be obtained, for example, from an image obtained using a camera or the like, or may be obtained by scanning the machined surface with a distance sensor using a probe laser. The number of times the pulsed laser light is oscillated may be the same as the number of holes in one electronic board, or one electronic board may be divided into a plurality of times.

The control device 35 calculates a value indicating variation in the shape of the machined hole, and determines whether the variation in the shape of the machined hole is within a predetermined range (step S123).

Here, the variation in the shape of the machined holes is the range of variation in the values indicating the characteristics of each of the plurality of machined holes. The value indicating the feature of the machined hole may be a value indicating a measurement result obtained by measuring each of a plurality of machined holes, or may be a value calculated from the measurement result. Examples of values that indicate features of a machined hole include the hole diameter, the aspect ratio of the hole, the area of the hole, the depth of the hole, the unevenness of the bottom surface of the hole, and the amount of scattered matter around the hole. In addition, the shape of a machined hole is not only specified for individual hole shapes, but also the arrangement of holes, i.e., where and how many holes are to be arranged, is described in machining programs, commands, etc. may Also, the mutual positional relationship between a plurality of machined holes may be specified.

If the variation in the shape of the machined hole is not within the predetermined range (step S123: No), the control device 35 corrects the transmittance of the optical switching element 26 based on the variation in the shape of the machined hole (step S124), It returns to the process of step S105. Here, the transmittance is set corresponding to each pulsed laser beam, but the transmittance to be corrected is the one whose difference from the reference value of the value indicating the shape of the processed hole is not within a predetermined range. This is the transmittance corresponding to the pulsed laser beam for machining holes.

If the variation in the shape of the processed hole is within the predetermined range (step S123: Yes), the electronic substrate is drilled with the set transmittance (step S125). A preparatory electronic substrate sample may be used pending the determination of the transmittance used to perform the drilling process.

The controller 35 can correct the transmittance using, for example, PID control. Alternatively, the transmittance may be determined by machine learning, which will be described later.

Also in Embodiment 3, the division of functions between the control device 35 and the information processing device 36 is an example, and the processing described above as the functions of the control device 35 may be performed by the information processing device 36. , the processing described as the function of the information processing device 36 may be performed by the control device 35 . For example, acquisition of drilling pattern data for an electronic board, calculation of a drilling path for an electronic board, calculation of an oscillation interval of a pulsed laser beam, calculation of an initial value of the transmittance of the optical switching element 26, acquisition of a machined hole shape, machined hole The information processing device 36 may perform the calculation of the value indicating the variation in shape, the calculation of the correction amount of the transmittance of the optical switching element 26, and the like.

As described above, the laser processing apparatus 510 according to the third embodiment includes the laser apparatus 500, the galvanomirror 403 which is a deflecting element that deflects the pulsed laser beam output from the amplifier 200, and the pulse from the galvanomirror 403. It has a lens 404 which is an objective optical system for condensing or transferring laser light and irradiating it on the object 405 to be processed, and a moving mechanism for moving the object 405 to be processed. The control device 35 controls the transmittance according to the processing hole characteristic value indicating the characteristics of the processing holes included on the processing path when the object 405 is drilled by the pulsed laser beam, and the intervals between the processing holes are controlled. The transmittance is controlled so that the shorter the hole, the higher the transmittance, and the value indicating variations in the shape of the plurality of machined holes is smaller than a predetermined value. When the laser device 500 is used for drilling, the longer the distance between the holes to be machined, the longer the pulse characteristic time. The transmittance can be controlled, and the shape of the processed hole can be homogenized.

The machined hole characteristic value is a value that indicates the interval between a plurality of machined holes on the machining path or the moving average of the intervals between the plurality of machined holes.

Embodiment 4.
In Embodiments 2 and 3, the case where the drilling pattern of the electronic board is known in advance was shown, but when the number of holes in the electronic board is very large, it is necessary to store the transmittance for each hole as data. Yes, memory resources are under pressure. Also, it is necessary to determine the oscillation timing of the pulsed laser light in advance.

Embodiment 4 shows a control method that can be applied even when the drilling pattern of the electronic substrate and the oscillation timing of the pulsed laser light are not known in advance.

FIG. 10 is a diagram showing the configuration of a learning device 60 according to the fourth embodiment. The learning device 60 performs machine learning on the laser device 500 . The learning device 60 generates a learned model for inferring the transmittance of the optical switching element 26 using a learning data acquisition unit 61 that acquires learning data, which is data used for learning, and the learning data. and a model generating unit 62 for generating the model. The model generating unit 62 stores the generated trained model in the trained model storage unit 70 .

The learning data acquisition unit 61 acquires state quantities including interval information indicating the time intervals at which the laser device 500 generates pulsed laser light and the pulse energy after amplification of the pulsed laser light, the transmittance of the optical switching element 26, is acquired as training data. The interval information may be any information as long as it indicates the time interval for generating the pulsed laser light. The interval information includes, for example, the pulse characteristic time described above, the energy of the pulsed laser light generated by the Q-switched laser oscillator 100, the waveform of the pulsed laser light generated by the Q-switched laser oscillator 100, and the pulse amplified by the amplifier 200. The information includes at least one of the waveform of the laser light and the drive current or discharge current of the Q-switched laser oscillator 100 and the amplifier 200 . For example, the learning device 60 obtains the pulse energy of the amplified pulsed laser light when the time interval for generating the pulsed laser light and the transmittance of the optical switching element 26 are changed to various values. A state quantity including time interval information and pulse energy after amplification, and a set transmittance are acquired as learning data.

When the laser device 500 is a gas laser whose laser medium is a gas, the state quantities acquired by the learning data acquisition unit 61 are the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200 and the Q-switched laser oscillator 100 or The gas temperature of the amplifier 200, the gas pressure of the Q-switched laser oscillator 100 or the amplifier 200, the continuous discharge time of the Q-switched laser oscillator 100 or the amplifier 200, the electrode temperature of the Q-switched laser oscillator 100 or the amplifier 200, and the Q-switched laser and the total discharge time from when the optical component of the oscillator 100 or the amplifier 200 is replaced.

The model generation unit 62 combines the state quantity including the interval information output from the learning data acquisition unit 61 and the amplified pulse energy of the pulsed laser light generated by the laser device 500, and the transmittance of the optical switching element 26. The transmittance of the optical switching element 26 is learned based on the learning data created based on. Specifically, the model generator 62 generates a learned model for inferring the transmittance of the optical switching element 26 whose target value is the pulse energy after amplification from the state quantity of the laser device 500 . Here, the learning data is data in which the state quantity and the transmittance of the optical switching element 26 are associated with each other.

Known algorithms such as supervised learning, unsupervised learning, and reinforcement learning can be used as the learning algorithm used by the model generator 62 . As an example, a case where a neural network is applied will be described.

The model generator 62 learns the transmittance of the optical switching element 26 by so-called supervised learning according to, for example, a neural network model. Here, supervised learning refers to a method of inferring a result from an input by giving a set of input and result (label) data to the learning device 60 to learn features in the data for learning. .

A neural network is composed of an input layer consisting of a plurality of neurons, an intermediate layer which is a hidden layer consisting of a plurality of neurons, and an output layer consisting of a plurality of neurons. The intermediate layer may be one layer, or two or more layers.

FIG. 11 is a diagram showing an example of a three-layer neural network. For example, in a three-layer neural network as shown in FIG. 11, when multiple inputs are input to the input layer (X1-X3), the value is multiplied by the weight W1 (w11-w16) and the intermediate layer ( Y1-Y2), and the result is further multiplied by weight W2 (w21-w26) and output from the output layer (Z1-Z3). This output result varies depending on the values of weight W1 and weight W2.

In Embodiment 4, the neural network learns transmittance by so-called supervised learning according to learning data created based on combinations of state quantities and transmittances acquired by learning data acquisition section 61 .

That is, the neural network has an optical switching element 26 such that the state quantity including the target value of the pulse energy after amplification is input to the input layer, and the result output from the output layer is the target value of the pulse energy after amplification. is learned by adjusting the weight W1 and the weight W2 so as to approach the transmittance of .

The model generation unit 62 generates a learned model by executing the learning as described above, and outputs the learned model to the learned model storage unit 70 .

The learned model storage unit 70 stores the learned model output from the model generation unit 62 .

Next, a process of learning by the learning device 60 will be described with reference to FIG. 12 . FIG. 12 is a flowchart for explaining the learning process of the learning device 60. As shown in FIG.

The learning data acquisition unit 61 acquires learning data including the state quantity and the transmittance (step S201). Here, the state quantity and the transmittance are acquired at the same time.

The model generation unit 62 performs so-called supervised learning according to learning data created based on the combination of the state quantity and the transmittance acquired by the learning data acquisition unit 61 so that the amplified pulse energy becomes the target value. A learned model is generated by performing such a learning process for learning the transmittance (step S202).

The learned model storage unit 70 stores the learned model generated by the model generation unit 62 (step S203).

FIG. 13 is a block diagram of the reasoning device 80 for the laser device 500. As shown in FIG. The inference device 80 includes an inference data acquisition unit 81 and an inference unit 82 .

The inference data acquisition unit 81 acquires state quantities as inference data. The state quantity includes the interval information described above and the target value of the pulse energy after amplification. When the laser device 500 is a gas laser whose laser medium is a gas, the state quantities acquired by the inference data acquisition unit 81 are the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the Q-switched laser oscillator 100 or The gas temperature of the amplifier 200, the gas pressure of the Q-switched laser oscillator 100 or the amplifier 200, the continuous discharge time of the Q-switched laser oscillator 100 or the amplifier 200, the electrode temperature of the Q-switched laser oscillator 100 or the amplifier 200, and the Q-switched laser and the total discharge time from when the optical component of the oscillator 100 or the amplifier 200 is replaced.

The inference unit 82 infers the transmittance obtained using the trained model. That is, the inference unit 82 inputs the inference data acquired by the inference data acquisition unit 81 to the learned model stored in the learned model storage unit 70, so that the amplified pulse is inferred from the state quantity. It is possible to output a transmittance that brings the energy to the target value.

The trained model used by the inference device 80 may be a trained model trained with learning data acquired from the laser device 500, which is the inference target of the inference device 80, or the inference target of the inference device 80. A trained model may be acquired from outside such as another laser device 500 different from the laser device 500 and used.

Next, the processing for obtaining the transmittance using the inference device 80 will be described with reference to FIG. FIG. 14 is a flow chart for explaining the processing for obtaining the transmittance using the inference device 80. As shown in FIG.

The inference data acquisition unit 81 acquires inference data (step S301).

The inference unit 82 inputs the state quantity to the learned model stored in the learned model storage unit 70 (step S302), and outputs the inference result of the transmittance to the laser device 500 (step S303).

The laser device 500 controls the optical switching element 26 using the output transmittance estimation result (step S304).

For example, in the drilling machine 400 as shown in FIG. , the Q-switched laser oscillator 100 is instructed to output a pulsed laser beam. Alternatively, in order to reduce the time loss due to the delay of the command signal, the control device 35 predicts the positioning time of the galvanomirror 403 so that the pulsed laser beam is applied to the workpiece 405 immediately after the positioning is completed. First, before the positioning of the galvanomirror 403 is completed, the Q-switched laser oscillator 100 is instructed to output a pulsed laser beam. In any case, the generation time of the pulsed laser light depends on the positioning time of the galvanomirror 403 . At this time, by using the learned model obtained by the learning device 60, the inference device 80 obtains the state quantity including interval information indicating the time interval for generating the pulsed laser light and the target value of the pulse energy after amplification. , the transmittance of the optical switching element 26 can be determined. Therefore, by using the trained model, even if the generation interval of the pulsed laser light is not predetermined, the transmittance of the optical switching element 26 can be determined for each pulsed laser light. Pulse energy can be controlled.

In the present embodiment, a case where supervised learning is applied to the learning algorithm used by the model generation unit 62 has been described, but the present invention is not limited to this. In addition to supervised learning, it is also possible to apply reinforcement learning, unsupervised learning, semi-supervised learning, and the like as learning algorithms.

Further, the model generator 62 may learn transmittance according to learning data created for a plurality of laser devices 500 . Note that the model generating unit 62 may acquire learning data from a plurality of laser devices 500 used in the same area, or learning data collected from a plurality of laser devices 500 operating independently in different areas. Transmittance may be learned using data for It is also possible to add or remove the laser device 500 that collects learning data from the target during the process. Furthermore, the learning device 60 that has learned the transmittance of a certain laser device 500 may be applied to another laser device 500 to re-learn and update the transmittance of the other laser device 500. .

In addition, as the learning algorithm used in the model generation unit 62, deep learning that learns to extract the feature amount itself can be used, and other known methods such as genetic programming, functional logic programming, Machine learning may be performed according to support vector machines and the like.

The learning device 60 and the inference device 80 are used to learn the transmittance of the optical switching element 26 of the laser device 500. may be a separate device. Also, the learning device 60 and the reasoning device 80 may be built in the laser device 500 or the laser processing device 510 . For example, at least one of the learning device 60 and the reasoning device 80 may be part of the functions of the control device 35 or the information processing device 36 . Furthermore, learning device 60 and reasoning device 80 may reside on a cloud server.

In the above description, the pulse energy after amplification is stabilized by controlling the transmittance of the optical switching element 26. pulse energy may be stabilized.

Also, as shown in Embodiment 1, if the positioning time of the galvanomirror 403 is too short and the pulse energy output from the amplifier 200 is smaller than the desired value, the necessary pulse energy can be obtained after amplification. The galvanomirror 403 may be stopped until the required pulse energy is obtained, and then the pulsed laser beam may be irradiated.

For example, in a CO2 laser, the oscillation efficiency changes depending on gas temperature and gas pressure. Therefore, in order to correct changes in pulse energy due to gas temperature and gas pressure, and the transmittance of the optical switching element 26, these parameters may be used as state quantities for machine learning. In addition, if the discharge is continuously performed, the pulse energy may decrease due to deterioration of the laser gas. Therefore, by using the continuous discharge time of the laser device 500 as a state quantity for machine learning, the dependence of the transmittance of the optical switching element 26 and the pulse energy is calculated, and the transmittance of the optical switching element 26 or the pulse energy is calculated. It can be used for correction. Furthermore, the total discharge time from the time of optical component replacement is an index of the deterioration state of the optical components. When the optical components deteriorate, the pulse energy may change due to the decrease in reflectance and transmittance. Therefore, the total discharge time from the time of optical component replacement may be used as the state quantity for machine learning.

As described above, the laser device 500 according to the fourth embodiment can further include the learning device 60 . The learning device 60 performs learning including state quantities including interval information indicating the time intervals at which the laser device 500 generates pulsed laser light, pulse energy after amplification of the pulsed laser light, and transmittance of the optical switching element 26. and a trained model for inferring the transmittance of the optical switching element 26 whose pulse energy after amplification is the target value from the state quantity using the learning data acquisition unit 61 that acquires the data for learning. and a model generator 62 for generating.

Also, the laser device 500 according to the fourth embodiment can further include an inference device 80 . The inference device 80 includes an inference data acquisition unit 81 that acquires state quantities including interval information indicating a time interval for generating the pulsed laser light and a target value of the pulse energy after amplification of the pulsed laser light; an inference unit 82 that infers the transmittance from the state quantity acquired by the inference data acquisition unit 81 using a trained model for inferring the transmittance of the optical switching element 26 whose pulse energy after amplification is a target value; , has The control device 35 controls the optical switching element 26 using the transmittance inferred by the inference device 80 .

The interval information used by each of the learning device 60 and the reasoning device 80 is, for example, the pulse characteristic time of the pulsed laser light generated by the Q-switched laser oscillator 100 and the energy of the pulsed laser light generated by the Q-switched laser oscillator 100. , the waveform of the pulsed laser light generated by the Q-switched laser oscillator 100, the waveform of the pulsed laser light amplified by the amplifier 200, and the drive current or discharge power of the Q-switched laser oscillator 100 and the amplifier 200. It may be information including

When the laser device 500 is a gas laser, the state quantities used by the learning device 60 and the reasoning device 80 are the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200 and the gas temperature of the Q-switched laser oscillator 100 or the amplifier 200. Temperature, gas pressure of Q-switched laser oscillator 100 or amplifier 200, continuous discharge time of Q-switched laser oscillator 100 or amplifier 200, electrode temperature of Q-switched laser oscillator 100 or amplifier 200, Q-switched laser oscillator 100 or amplifier The information may further include at least one of 200 total discharge time since the optical component was replaced.

Also, the learning device 60 and the reasoning device 80 may be separate devices from the laser device 500 . It is also possible to configure a laser processing system that includes at least one of the learning device 60 and the reasoning device 80 that are separate devices from the laser device 500 .

Embodiment 5.
In the fourth embodiment, the transmittance of the optical switching element 26 is learned so that the pulse energy after amplification becomes the target value. It is more desirable that the processed shape is stable when it is processed.

Therefore, in the fifth embodiment, a method of learning the transmittance so that the shape of the machined hole after machining becomes the target shape will be described.

FIG. 15 is a diagram showing the configuration of a learning device 60a according to the fifth embodiment. The learning device 60a performs machine learning on a laser processing device 510 as shown in FIG. 3, for example. The learning device 60a generates a learned model for inferring the transmittance of the optical switching element 26 using a learning data acquisition unit 61a that acquires learning data, which is data used for learning, and the learning data. and a model generator 62a. The model generation unit 62a stores the generated learned model in the learned model storage unit 70a.

The learning data acquisition unit 61a includes interval information indicating the time interval at which the laser device 500 of the laser processing device 510 generates the pulsed laser light and shape information indicating the shape of the hole processed by the laser processing device 510. The amount and the transmittance of the optical switching element 26 are acquired as learning data. The interval information may be any information as long as it indicates the time interval for generating the pulsed laser light. The interval information includes, for example, the pulse characteristic time described above, the energy of the pulsed laser light generated by the Q-switched laser oscillator 100, the waveform of the pulsed laser light generated by the Q-switched laser oscillator 100, and the pulse amplified by the amplifier 200. The information includes at least one of the waveform of the laser light and the drive current or discharge current of the Q-switched laser oscillator 100 and the amplifier 200 . For example, the learning device 60a acquires shape information indicating the shape of the machined hole when the time interval for generating the pulsed laser light and the transmittance of the optical switching element 26 are changed to various values. , the state quantity including the interval information and the shape information at that time, and the set transmittance are acquired as learning data.

When the laser device 500 is a gas laser whose laser medium is a gas, the state quantities acquired by the learning data acquisition unit 61a are the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the Q-switched laser oscillator 100 or The gas temperature of the amplifier 200, the gas pressure of the Q-switched laser oscillator 100 or the amplifier 200, the continuous discharge time of the Q-switched laser oscillator 100 or the amplifier 200, the electrode temperature of the Q-switched laser oscillator 100 or the amplifier 200, and the Q-switched laser and the total discharge time from when the optical component of the oscillator 100 or the amplifier 200 is replaced.

The model generation unit 62a is a state quantity including the interval information output from the learning data acquisition unit 61a and the shape information indicating the shape of the machined hole machined by the laser processing device 510, and the transmittance of the optical switching element 26. The transmittance of the optical switching element 26 is learned based on the learning data created based on the combination. Specifically, the model generation unit 62a generates a learned model for inferring the transmittance of the optical switching element 26 that makes the shape of the processed hole after processing a desired shape from the state quantity of the laser processing device 510. . Here, the learning data is data in which the state quantity and the transmittance of the optical switching element 26 are associated with each other.

Known algorithms such as supervised learning, unsupervised learning, and reinforcement learning can be used as the learning algorithm used by the model generator 62a. As an example, a case where a neural network is applied will be described.

The model generator 62a learns the transmittance of the optical switching element 26 by so-called supervised learning according to, for example, a neural network model. Here, supervised learning refers to a method of inferring a result from an input by giving a set of input and result (label) data to the learning device 60a to learn features in the learning data.

A neural network is composed of an input layer consisting of a plurality of neurons, an intermediate layer which is a hidden layer consisting of a plurality of neurons, and an output layer consisting of a plurality of neurons. The intermediate layer may be one layer, or two or more layers. As in the fourth embodiment, a neural network as shown in FIG. 11 can be used.

In the fifth embodiment, the neural network obtains the desired shape of the machined hole by so-called supervised learning according to the learning data created based on the combination of the state quantity and the transmittance acquired by the learning data acquisition unit 61a. Learn the transmittance that gives the shape of

That is, the neural network has an optical switching element 26 such that the state quantity including the target value of the pulse energy after amplification is input to the input layer, and the result output from the output layer is the target value of the pulse energy after amplification. is learned by adjusting the weight W1 and the weight W2 so as to approach the transmittance of .

The model generation unit 62a generates a learned model by executing the learning as described above, and outputs the learned model to the learned model storage unit 70a.

The learned model storage unit 70a stores the learned model output from the model generation unit 62a.

Since the flow of processing for learning by the learning device 60a is the same as that of the learning device 60 according to the fourth embodiment, description thereof is omitted here.

FIG. 16 is a configuration diagram of an inference device 80a related to the laser processing device 510. As shown in FIG. The inference device 80a includes an inference data acquisition unit 81a and an inference unit 82a.

The inference data acquisition unit 81a acquires state quantities as inference data. The state quantity includes the above-described interval information and the target shape of the machined hole after machining. When the laser device 500 is a gas laser whose laser medium is a gas, the state quantities acquired by the inference data acquisition unit 81a are the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the Q-switched laser oscillator 100 or The gas temperature of the amplifier 200, the gas pressure of the Q-switched laser oscillator 100 or the amplifier 200, the continuous discharge time of the Q-switched laser oscillator 100 or the amplifier 200, the electrode temperature of the Q-switched laser oscillator 100 or the amplifier 200, and the Q-switched laser and the total discharge time from when the optical component of the oscillator 100 or the amplifier 200 is replaced.

The inference unit 82a infers the transmittance obtained using the trained model. That is, the inference unit 82a inputs the inference data acquired by the inference data acquisition unit 81a to the learned model stored in the learned model storage unit 70a, thereby inferring the shape of the machined hole from the state quantity. can output a transmittance that gives a target shape.

The trained model used by the inference device 80a may be a trained model trained with learning data acquired from the laser device 500, which is the inference target of the inference device 80a, or the inference target of the inference device 80a. A trained model may be acquired from outside such as another laser device 500 different from the laser device 500 and used.

Since the flow of the inference operation of the inference device 80a is the same as that of the inference device 80 according to the fourth embodiment, the explanation is omitted here.

For example, in the drilling machine 400 as shown in FIG. , the Q-switched laser oscillator 100 is instructed to output a pulsed laser beam. Alternatively, in order to reduce the time loss due to the delay of the command signal, the control device 35 predicts the positioning time of the galvanomirror 403 so that the pulsed laser beam is applied to the workpiece 405 immediately after the positioning is completed. First, before the positioning of the galvanomirror 403 is completed, the Q-switched laser oscillator 100 is instructed to output a pulsed laser beam. In any case, the generation time of the pulsed laser light depends on the positioning time of the galvanomirror 403 . At this time, by using the learned model obtained by the learning device 60a, the inference device 80a can determine the light The transmittance of switching element 26 can be determined. Therefore, by using the trained model, even if the generation interval of the pulsed laser light is not predetermined, the transmittance of the optical switching element 26 can be determined for each pulsed laser light. Pulse energy can be controlled.

In this embodiment, the case where supervised learning is applied to the learning algorithm used by model generation unit 62a has been described, but the present invention is not limited to this. In addition to supervised learning, it is also possible to apply reinforcement learning, unsupervised learning, semi-supervised learning, and the like as learning algorithms.

Further, the model generator 62a may learn transmittance according to learning data created for a plurality of laser devices 500. FIG. Note that the model generation unit 62a may acquire learning data from a plurality of laser devices 500 used in the same area, or learning data collected from a plurality of laser devices 500 operating independently in different areas. Transmittance may be learned using data for It is also possible to add or remove the laser device 500 that collects learning data from the target during the process. Furthermore, the learning device 60a that has learned the transmittance of a certain laser device 500 may be applied to another laser device 500, and the transmittance of the other laser device 500 may be re-learned and updated. .

In addition, as a learning algorithm used in the model generating unit 62a, deep learning that learns to extract the feature amount itself can be used, and other known methods such as genetic programming, functional logic programming, support vector machine, etc. Machine learning may be performed according to

The learning device 60a and the inference device 80a are used to learn the transmittance of the optical switching element 26 of the laser device 500. may be a separate device. Also, the learning device 60 a and the reasoning device 80 a may be built in the laser device 500 . Furthermore, learning device 60a and reasoning device 80a may reside on a cloud server.

In the above description, the pulse energy after amplification is stabilized by controlling the transmittance of the optical switching element 26. pulse energy may be stabilized.

Also, as shown in Embodiment 1, if the positioning time of the galvanomirror 403 is too short and the pulse energy output from the amplifier 200 is smaller than the desired value, the necessary pulse energy can be obtained after amplification. The galvanomirror 403 may be stopped until the required pulse energy is obtained, and then the pulsed laser beam may be irradiated.

For example, in a CO2 laser, the oscillation efficiency changes depending on gas temperature and gas pressure. Therefore, in order to correct changes in pulse energy due to gas temperature and gas pressure, and the transmittance of the optical switching element 26, these parameters may be used as state quantities for machine learning. In addition, if the discharge is continuously performed, the pulse energy may decrease due to deterioration of the laser gas. Therefore, by using the continuous discharge time of the laser device 500 as a state quantity for machine learning, the dependence of the transmittance of the optical switching element 26 and the pulse energy is calculated, and the transmittance of the optical switching element 26 or the pulse energy is calculated. It can be used for correction. Furthermore, the total discharge time from the time of optical component replacement is an index of the deterioration state of the optical components. When the optical components deteriorate, the pulse energy may change due to the decrease in reflectance and transmittance. Therefore, the total discharge time from the time of optical component replacement may be used as the state quantity for machine learning.

As described above, the learning device 60a according to the fifth embodiment is a learning device 60a that learns the transmittance of the optical switching element 26 of the laser device 500 of the laser processing device 510, and generates a pulsed laser beam. Learning data for acquiring learning data including a state quantity including interval information indicating a time interval and shape information indicating the shape of a machined hole machined using a pulsed laser beam, and the transmittance of the optical switching element 26 An acquisition unit 61a, and a model generation unit 62a for generating a learned model for inferring the transmittance of an optical switching element having a machined hole having a target shape from state quantities using learning data. . With such a configuration, it becomes possible to learn the relationship between the transmittance and the state quantity that homogenizes the shape of the machined hole after machining.

An inference device 80a according to the fifth embodiment is an inference device 80a that infers the transmittance of the optical switching element 26 of the laser device 500 of the laser processing device 510, and indicates the time interval for generating the pulsed laser light. An inference data acquisition unit 81a for acquiring a state quantity including interval information and shape information indicating a target shape of a hole to be machined using a pulsed laser beam; and an inference unit 82a that infers the transmittance from the state quantity and the target shape acquired by the inference data acquisition unit 81a using a trained model for inference. By using such an inference device 80a, it becomes possible to infer the transmittance that homogenizes the shape of the machined hole after machining.

The learning device 60a and the reasoning device 80a may also be built in the laser device 500 or the laser processing device 510, or may be separate devices from the laser device 500 and the laser processing device 510. It is also possible to configure a laser processing system that includes at least one of the learning device 60a and the reasoning device 80a.

Embodiment 6.
In Embodiment 1, the transmittance of the optical switching element 26 is determined for each pulsed laser beam, and the transmittance is constant when each pulsed laser beam is generated.

However, when pulsed laser light with different pulse intervals is generated by the Q-switched laser oscillator 100, the pulse waveforms may not have similar shapes. FIG. 17 is a diagram showing an example of waveforms of pulsed laser light with different pulse intervals.

FIG. 17A shows the waveform of pulsed laser light oscillated by the Q-switched laser oscillator 100. FIG. FIG. 17(b) shows the transmittance of the optical switching element 26. FIG. FIG. 17(c) shows the waveform of the pulsed laser light input to the amplifier 200. FIG. FIG. 17(d) shows the waveform of the pulsed laser light output from the amplifier 200. FIG.

As shown in FIG. 17( a ), the waveform of the pulsed laser light output from the Q-switched laser oscillator 100 is not similar when the pulsed laser light with different pulse intervals is generated. , the intensity ratio of the pulse trailing portion to the steady-state value may be different. For example, while the peak value p2 of the pulsed laser beam #2 and the peak value p3 of the pulsed laser beam #3 are significantly different, the steady-state value c2 of the pulsed laser beam #2 and the peak value p3 of the pulsed laser beam #3 It has almost the same value as the steady-state value c3. In such a case, as shown in FIG. 17(b), control is performed so that the transmittance is constant for each pulsed laser beam in the same manner as in the first embodiment, resulting in the following as shown in FIG. 17(d): , the amplified pulsed laser light has different waveforms even if the pulse energy is the same. Here, FIG. 17C shows the waveform of the laser pulse light incident on the amplifier 200. FIG.

In this way, when the waveforms of a plurality of pulsed laser beams generated by the Q-switched laser oscillator 100 are not similar, the transmittance of the optical switching element 26 is changed within one pulse time as shown in FIG. Thereby, not only the pulse energy but also the waveform can be homogenized.

FIG. 18 is an explanatory diagram of an example in which the transmittance is changed over time within the pulse time. FIG. 18A shows the waveform of pulsed laser light oscillated by the Q-switched laser oscillator 100. FIG. FIG. 18(b) shows the transmittance of the optical switching element 26. FIG. FIG. 18(c) shows the waveform of the pulsed laser light input to the amplifier 200. FIG. FIG. 18(d) shows the waveform of the pulsed laser light output from the amplifier 200. FIG.

When the transmittance is set using machine learning, the timing at which the Q-switched laser oscillator 100 generates the pulsed laser beam, the pulse energy of the pulsed laser beam, and the pulse peak at the beginning of the pulse are used as state quantities. and at least one of a pulse steady-state value of the pulse trailing portion. Also, the transmittance, which is the learning result, may be a value that changes with time while one pulse laser beam is transmitted, rather than a constant value for each pulse laser beam.

As described above, the laser device 500 according to the sixth embodiment changes the transmittance over time while one pulsed laser beam is transmitted through the optical switching element 26 . This makes it possible to homogenize the waveform of the amplified laser pulse light. When the transmittance is set using machine learning, the model generators 62 and 62a of the learning devices 60 and 60a change with time while one pulse laser beam is transmitted through the optical switching element 26 from the state quantity. It is also possible to generate a trained model for inferring the transmittance.

Embodiment 7.
When the interval of generating the pulsed laser light is short, not only the interval of the previous pulsed laser light but also the history of a plurality of laser pulsed lights may affect. In this case, the interval information included in the state quantities shown in Embodiments 4 to 6 is not only the time interval between the target pulsed laser beam and the previous pulsed laser beam, but also a plurality of intervals included in the fixed time. It may be information indicating a plurality of time intervals for the pulsed laser beam.

FIG. 19 is an explanatory diagram of transmittance control according to the seventh embodiment. FIG. 19A shows the waveform of pulsed laser light oscillated by the Q-switched laser oscillator 100. FIG. FIG. 19(b) shows the transmittance of the optical switching element 26. FIG. FIG. 19(c) shows the waveform of the pulsed laser light input to the amplifier 200. FIG. FIG. 19(d) shows the waveform of the pulsed laser light output from the amplifier 200. FIG.

As shown in FIG. 19A, even when the pulsed laser beams #1 to #6 are oscillated at regular time intervals, if the time intervals are short, each of the pulsed laser beams #1 to #6 is oscillated immediately before. In some cases, it is affected not only by one pulsed laser beam, but also by a plurality of pulsed laser beams. In this case, even if the pulsed laser beams #1 to #6 are oscillated at regular intervals, the affected period length is TL. While one pulsed laser beam is included in the previous TL period, the pulsed laser beam #3 includes two pulsed laser beams #1 to #2 in the TL period before the generation time of the pulsed laser beam #3. . Further, the pulsed laser beam #4 includes three pulsed laser beams #1 to #3 during the period TL before the generation time of the pulsed laser beam #4. Similar to the pulse laser beam #4, the pulse laser beams #5 and #6 also include three pulse laser beams in the period TL before the generation time of each pulse laser beam. Therefore, the pulse energy of the pulse laser beams #1 to #4 varies, and the pulse energy of the pulse laser beams #4 to #6 is constant. In this case, if the transmittance is controlled as shown in FIG. As shown in 19(d), variations in pulse energy after amplification can be suppressed.

Since the value of TL varies depending on conditions such as pulse output, when machine learning is used, the value of TL is also changed to acquire data under various conditions, as shown in Embodiments 4 and 5. , TL may be included in the state quantity and machine learning may be performed to obtain the optimum range.

Embodiment 8.
If the interval at which the pulsed laser light is generated is long enough, the discharge power may be changed. For example, if the laser device 500 is a triaxial orthogonal CO2 laser, it takes a certain amount of time from the start of discharge until the gain rises.

FIG. 20 is a diagram showing a partial configuration between the electrodes 11 of the laser device 500 according to the eighth embodiment. For simplicity, only the electrode 11 portion is shown in FIG. 20, and the electrode 12 is not shown.

FIG. 20 is a view of the discharge space between the electrodes 11 of the laser device 500 shown in FIG. 2 as seen from the optical axis direction. In FIG. 20, Vg is the gas flow velocity, and Dwd is the electrode width of the electrode 11 in the gas flow direction 13 .

FIG. 21 is a diagram showing a steady-state gain distribution between the electrodes 11 of the laser device 500 having the configuration shown in FIG. It takes time for the laser gas to flow from end to end of the electrode 11 in the gas flow direction 13 , which is the width direction of the electrode 11 , until the steady-state gain distribution is obtained from the start of discharge. That is, when the time from the start of discharge until the gain distribution reaches a steady state is τ, τ=Dwd/Vg. For example, when gas flow velocity Vg=80 m/s and electrode width Dwd=40 mm, τ=0.5 msec. In other words, when the time interval for generating the pulsed laser light is longer than the above τ, discharge is stopped in order to suppress power consumption, and discharge is started τ before the time at which the next pulsed laser light is generated. can be Even when the time interval for generating the pulsed laser light is shorter than τ, the discharge power may be reduced when the pulse energy is too large. By doing so, energy efficiency can be improved even when the pulse frequency is low. Further, since the time longer than τ is not affected by the previous pulsed laser beam, the value of TL described in Embodiment 7 may be set to τ.

As described above, in the laser device 500 according to the eighth embodiment, the pulse characteristic time is the length of the discharge electrodes 11 and 12, which are the discharge electrodes of the Q-switched laser oscillator 100, in the direction in which the laser gas flows. It is equal to or less than the value obtained by dividing the electrode width Dwd by the gas flow velocity Vg. As a result, the pulse energy of the pulsed laser light output from the amplifier 200 can be stabilized, and the processed shape of the processing performed using this laser device 500 can be homogenized.

Embodiment 9.
FIG. 22 is a diagram showing the internal configuration of a laser device 500a according to the ninth embodiment. The laser device 500a is different from the laser device 500 in part of the configuration of the part that constitutes the resonator. Specifically, the laser device 500a has a total reflection mirror 54 instead of the total reflection mirror 21, and the total reflection mirror 54 is a combination of two plane mirrors whose normals are perpendicular to each other. At this time, the two normals are parallel to the electrode surface, that is, parallel to the XZ plane. Therefore, the total reflection mirror 54 becomes a retroreflection mirror in the XZ plane direction.

When the discharge space between the electrodes 11 of the laser device 500a shown in FIG. 22 is viewed from the optical axis direction, the state is similar to that shown in FIG. FIG. 23 is a diagram showing the temperature distribution in the steady state between the electrodes 11 of the laser device 500a.

While the laser gas flows from the upstream side of the gas flow, ie, the positive direction of the X-axis, to the downstream side, ie, the negative direction of the X-axis, power is continuously applied by the discharge. Therefore, a temperature distribution occurs in the gas flow direction 13, and the temperature of the laser gas is higher on the downstream side than on the upstream side. As a result, a refractive index distribution occurs in the laser gas between the electrodes 11, and the optical axis of the light propagating through this portion is slightly bent. In a resonator using a normal pair of concave mirrors, in this case, the optical axis moves to the upstream side of the gas flow, and oscillation stops in some cases. In the ninth embodiment, since the resonator is constructed using the total reflection mirror 54 which is a retroreflection mirror, the resonance condition is that the optical axis position is limited to the tangential line of the two retroreflection mirrors. become. That is, when a retroreflecting mirror is used, even if a temperature distribution occurs in the direction of gas flow, the movement of the optical axis is restricted, and stable oscillation can be achieved. In this case, the pulse energy and pulse waveform of the oscillator are obtained by using a partially reflecting mirror 55 instead of the mirror 25 in the laser device 500 as the mirror after outputting the resonator, and observing a part of the transmitted light with the optical sensor 50. good too. In FIG. 22, the partially reflecting mirror 24 and the total reflecting mirror 54, which is a retroreflecting mirror, are arranged on a straight line. good.

As described above, the Q-switched laser oscillator 100 of the laser device 500a according to the ninth embodiment has a retroreflection mirror as the total reflection mirror 54 forming the resonator. As a result, stable oscillation can be achieved even when temperature distribution occurs in the direction of gas flow.

Embodiment 10.
In the laser device 500 shown in FIG. 2, the Q-switched laser oscillator 100 and the amplifier 200 are housed in the same housing 300 and share the laser medium. Therefore, the gain state of the amplifier 200 can be estimated from the pulse energy of the Q-switched laser oscillator 100 . For example, when the laser gas deteriorates and the pulse energy of the pulsed laser light in the oscillation stage decreases, the gain in the amplification stage also decreases. The relationship between the state of the gain in the amplifier 200 and the pulse energy in the Q-switched laser oscillator 100 is modeled, and by correcting the transmittance of the optical switching element 26 from the model, the pulse energy after amplification is stabilized. As a result, it is possible to homogenize the processed shape of laser processing using this pulsed laser beam.

FIG. 24 is a flow chart for explaining the operation of the laser processing device 510 according to the tenth embodiment. Since steps S101 to S104 are the same as in FIG.

The controller 35 stores the calculated transmittance data of the optical switching element 26 in, for example, a memory in the controller 35, and sets the transmittance of the optical switching element 26 for each pulsed laser beam (step S131).

The control device 35 controls the Q-switched laser oscillator 100 and the optical switching element 26 using the oscillation interval of the pulsed laser light calculated in step S103 and the transmittance of the optical switching element 26 set in step S131. The Q-switched laser oscillator 100 is caused to oscillate for one pulse (step S132).

Also, the control device 35 acquires the pulse energy of the pulsed laser light of the Q-switched laser oscillator 100 acquired by the optical sensor 50 via the information processing device 36 (step S133).

The laser processing device 510 uses the pulsed laser light oscillated by the laser device 500 to drill one hole in the electronic substrate (step S134), and then returns to step S132. In parallel with the execution of this drilling process, the controller 35 calculates the pulse energy of the next pulsed laser light based on the model (step S135). The model used here is the model that shows the relationship between the state of gain in the amplifier 200 and the pulse energy in the Q-switched laser oscillator 100 described above. The controller 35 corrects the transmittance of the optical switching element 26 based on the pulse energy of the next pulsed laser beam calculated based on the model (step S136), and returns to the process of step S131.

By performing the above processing, the pulse energy of the oscillation stage is acquired each time one pulse is oscillated, and the transmittance of the optical switching element is corrected based on the model before the next pulsed laser beam is oscillated. It can be performed.

At this time, the model may be created by machine learning.

The configurations shown in the above embodiments are only examples, and can be combined with other known techniques, or can be combined with other embodiments, without departing from the scope of the invention. It is also possible to omit or change part of the configuration.

For example, in Embodiments 1 to 10 described above, the laser devices 500 and 500a are gas lasers in which the laser medium is a gas. Fluctuations in pulse energy during oscillation have similar effects on solid-state lasers and the like. Therefore, except for the portion specific to the gas laser shown above, the above configuration can also be applied to solid-state lasers, etc., and similar effects can be expected.

In the above description, an example of controlling the transmittance of the optical switching element 26 was mainly described. However, as shown in FIG. Control may be performed so that the combined transmittance of the switching elements 26 and 27 satisfies the above conditions.

1,2,3 optical axis, 11,12 electrode, 13,14,15,16 gas flow direction, 21,54 total reflection mirror, 22 Q switch, 23,29,33 window, 24,34,55 partial reflection mirror , 25, 28, 30, 31, 32, 401, 402 mirror, 26, 27 optical switching element, 35 control device, 36 information processing device, 40, 41 fan, 42, 43 heat exchanger, 44 discharge control device, 50 , 51 optical sensor 60, 60a learning device 61, 61a learning data acquisition unit 62, 62a model generation unit 70, 70a learned model storage unit 80, 80a inference device 81, 81a inference data acquisition unit , 82, 82a reasoning unit, 100 Q-switch laser oscillator, 200 amplifier, 300 housing, 400 drilling machine, 403 galvanomirror, 404 lens, 405 object to be processed, 500, 500a laser device, 510 laser processing device.

Claims (23)

a Q-switched laser oscillator that generates pulsed laser light;
an amplifier that amplifies the pulsed laser light;
an optical switching element arranged on an optical path between the Q-switched laser oscillator and the amplifier;
a control device that modulates the transmittance of the optical switching element based on a pulse characteristic time that indicates the characteristics of the time interval at which the Q-switched laser oscillator generates the pulsed laser light;
A laser device comprising: The control device is
The time interval between the plurality of pulsed laser beams generated by the Q-switched laser oscillator, or the moving average of the time intervals between the plurality of pulsed laser beams generated by the Q-switched laser oscillator, is defined as the pulse characteristic time. seek,
The transmittance is controlled such that the shorter the time interval, the higher the transmittance, and the change in energy among the plurality of pulsed laser beams emitted from the amplifier is smaller than a predetermined value. 2. The laser device according to claim 1, wherein: further equipped with a learning device,
The learning device
learning data including state quantity including interval information indicating the time interval at which the laser device generates the pulsed laser beam and pulse energy after amplification of the pulsed laser beam; and transmittance of the optical switching element. a learning data acquisition unit to acquire;
a model generation unit that generates a trained model for inferring the transmittance of the optical switching element whose target value is the pulse energy after amplification from the state quantity, using the learning data;
The laser device according to claim 1 , characterized by comprising: further comprising an inference device for inferring the transmittance of the optical switching element;
The reasoning device is
an inference data acquisition unit for acquiring state quantities including interval information indicating a time interval for generating the pulsed laser light and a target value of pulse energy after amplification of the pulsed laser light;
The transmittance from the state quantity acquired by the data acquisition unit for inference using a trained model for inferring the transmittance of the optical switching element whose target value is the pulse energy after amplification from the state quantity an inference unit that infers
has
2. The laser device according to claim 1 , wherein the control device controls the optical switching element using the transmittance inferred by the inference device. further comprising an inference device for inferring the transmittance of the optical switching element;
The reasoning device is
an inference data acquisition unit for acquiring state quantities including interval information indicating a time interval for generating the pulsed laser light and a target value of pulse energy after amplification of the pulsed laser light;
The transmittance from the state quantity acquired by the data acquisition unit for inference using a trained model for inferring the transmittance of the optical switching element whose target value is the pulse energy after amplification from the state quantity an inference unit that infers
has
The control device controls the optical switching element using the transmittance inferred by the inference device.
4. The laser device according to claim 3, characterized in that: The interval information includes the pulse characteristic time of the pulsed laser light generated by the Q-switched laser oscillator, the energy of the pulsed laser light generated by the Q-switched laser oscillator, and the pulse generated by the Q-switched laser oscillator. 4. The waveform includes at least one of a waveform of a laser beam, a waveform of the pulsed laser beam amplified by the amplifier, and drive current or discharge power of the Q-switched laser oscillator and the amplifier. 6. The laser device according to any one of 3 to 5 . The laser device is a gas laser in which the laser medium is a gas,
The state quantity includes a cooling water temperature of the Q-switched laser oscillator or the amplifier, a gas temperature of the Q-switched laser oscillator or the amplifier, a gas pressure of the Q-switched laser oscillator or the amplifier, and the Q-switched laser oscillator. or at least one of a continuous discharge time of the amplifier, an electrode temperature of the Q-switched laser oscillator or the amplifier, and a total discharge time from the time of optical component replacement of the Q-switched laser oscillator or the amplifier. 6. The laser device according to any one of claims 3 to 5, characterized in that: 6. The pulse characteristic time is equal to or less than a value obtained by dividing an electrode width, which is a length of a discharge electrode of the Q-switched laser oscillator in a gas flow direction, by a gas flow velocity. 2. The laser device according to item 1. 6. The laser device according to any one of claims 1 to 5 , wherein the control device changes the transmittance with time while one of the pulsed laser beams is transmitted through the optical switching element. The model generation unit generates a learned model for inferring the transmittance that changes with time while one of the pulsed laser beams is transmitted through the optical switching element, from the state quantity. Item 4. The laser device according to item 3. A Q-switched laser oscillator that generates a pulsed laser beam, an amplifier that amplifies the pulsed laser beam, an optical switching element arranged on an optical path between the Q-switched laser oscillator and the amplifier, and the Q-switched laser oscillator. a control device that modulates the transmittance of the optical switching element based on a pulse characteristic time that characterizes the time interval for generating the pulsed laser light;
a deflection element that deflects the pulsed laser light output from the amplifier;
an objective optical system for condensing or transferring the pulsed laser beam from the deflecting element and irradiating the object to be processed;
with
The control device controls the transmittance according to a processing hole characteristic value indicating characteristics of a processing hole included on a processing path when the processing object is drilled by the pulsed laser beam, and controls the processing hole. The transmittance is controlled so that the transmittance increases as the interval between the laser processing equipment. 12. The laser according to claim 11 , wherein the machined hole characteristic value is a value that indicates the interval between the plurality of machined holes on the machining path or a moving average of the intervals between the plurality of machined holes. processing equipment. 13. The laser processing apparatus according to claim 11 , wherein the Q-switched laser oscillator has a retroreflection mirror as a total reflection mirror forming a resonator. A learning device for learning the transmittance of the optical switching element of the laser device according to claim 1 ,
For learning to acquire learning data including a state quantity including interval information indicating a time interval for generating the pulsed laser beam and pulse energy after amplification of the pulsed laser beam, and the transmittance of the optical switching element. a data acquisition unit;
a model generating unit that generates a trained model for inferring the transmittance of the optical switching element whose pulse energy is a target value from the state quantity, using the learning data;
A learning device comprising: A learning device for learning the transmittance of the optical switching element included in the laser device of the laser processing device according to claim 11 ,
Learning including a state quantity including interval information indicating a time interval for generating the pulsed laser beam and shape information indicating a shape of a machined hole machined using the pulsed laser beam, and the transmittance of the optical switching element. a learning data acquisition unit that acquires data for
a model generation unit that generates a trained model for inferring the transmittance of the optical switching element having the shape of the machined hole as a target shape from the state quantity, using the learning data;
A learning device comprising: The interval information includes the pulse characteristic time of the pulsed laser light generated by the Q-switched laser oscillator, the energy of the pulsed laser light generated by the Q-switched laser oscillator, and the pulse generated by the Q-switched laser oscillator. 4. The waveform includes at least one of a waveform of a laser beam, a waveform of the pulsed laser beam amplified by the amplifier, and drive current or discharge power of the Q-switched laser oscillator and the amplifier. 16. The learning device according to 14 or 15 . The laser device is a gas laser in which the laser medium is a gas,
The state quantity includes a cooling water temperature of the Q-switched laser oscillator or the amplifier, a gas temperature of the Q-switched laser oscillator or the amplifier, a gas pressure of the Q-switched laser oscillator or the amplifier, and the Q-switched laser oscillator. or at least one of a continuous discharge time of the amplifier, an electrode temperature of the Q-switched laser oscillator or the amplifier, and a total discharge time from the time of optical component replacement of the Q-switched laser oscillator or the amplifier. 16. The learning device according to claim 14 or 15, characterized in that: An inference device for inferring the transmittance of the optical switching element of the laser device according to claim 1 or 2,
an inference data acquisition unit that acquires a state quantity including interval information indicating a time interval for generating the pulsed laser light and a target value of pulse energy after amplification of the pulsed laser light;
inferring the transmittance from the state quantity and the target value obtained by the data obtaining unit for inference using a trained model for inferring the transmittance of the optical switching element from the state quantity and the target value; an inference unit;
A reasoning device characterized by having: An inference device for inferring the transmittance of the optical switching element of the laser device of the laser processing device according to claim 11 or 12 ,
an inference data acquisition unit for acquiring a state quantity including interval information indicating a time interval for generating the pulsed laser beam and shape information indicating a target shape of a hole to be machined using the pulsed laser beam;
inferring the transmittance from the state quantity and the target shape obtained by the data obtaining unit for inference using a trained model for inferring the transmittance of the optical switching element from the state quantity and the shape information; an inference unit;
A reasoning device characterized by having: a Q-switched laser oscillator that generates pulsed laser light;
an amplifier that amplifies the pulsed laser light;
an optical switching element arranged on an optical path between the Q-switched laser oscillator and the amplifier;
a deflection element that deflects the pulsed laser light output from the amplifier;
an objective optical system for condensing or transferring the pulsed laser beam from the deflecting element and irradiating the object to be processed;
The transmittance of the optical switching element is controlled in accordance with a machined hole characteristic value indicating the characteristics of the machined hole included on the machined path when the machined object is machined by the pulsed laser beam. The transmittance is modulated by controlling the transmittance so that the transmittance increases as the distance becomes shorter, and a value indicating variation in shape of the plurality of processed holes becomes smaller than a predetermined value. a control device that causes
A laser processing system comprising: Learning data acquisition for acquiring learning data including a state quantity including interval information indicating a time interval for generating the pulsed laser beam, and shape information indicating a shape of a machined hole machined using the pulsed laser beam. Department and
a model generation unit that generates a trained model for inferring the transmittance of the optical switching element having the shape of the machined hole as a target shape from the state quantity, using the learning data;
a learning device having
21. The laser processing system of claim 20 , further comprising: an inference data acquisition unit for acquiring a state quantity including interval information indicating a time interval for generating the pulsed laser beam and shape information indicating a target shape of a hole to be machined using the pulsed laser beam;
inferring the transmittance from the state quantity and the target shape obtained by the data obtaining unit for inference using a trained model for inferring the transmittance of the optical switching element from the state quantity and the shape information; an inference unit;
A reasoning device having
22. The laser processing system of claim 20 or 21 , further comprising: generating pulsed laser light;
setting the transmittance of an optical switching element that transmits the pulsed laser light;
changing the pulse energy of the pulsed laser light by making the pulsed laser light incident on the optical switching element having a set transmittance;
Amplifying the pulsed laser light;
a step of adjusting a position at which the pulsed laser beam is irradiated onto the workpiece by deflecting the amplified pulsed laser beam;
condensing or transferring the deflected pulsed laser beam and irradiating it onto the object to be processed;
including
The step of changing the pulse energy of the pulsed laser beam modulates the transmittance of the optical switching element based on a pulse characteristic time that indicates the characteristics of the time interval for generating the pulsed laser beam. Method.

Images