TW202414928A - Laser device, laser processing device, learning device, inference device, laser processing system and laser processing method - Google Patents

Abstract

The laser device (500) comprises a Q-switched laser oscillator (100) for generating pulsed laser light, an amplifier (200) for amplifying the pulsed laser light, an optical switch element (26) provided on an optical path between the Q-switched laser oscillator (100) and the amplifier (200), and a control device (35) for modulating the transmittance of the optical switch element (26) based on a pulse characteristic time representing a time interval of the pulsed laser light generated by the Q-switched laser oscillator (100).

Description

Laser device, laser processing device, learning device, inference device, laser processing system and laser processing method

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.

In the case where a high-output pulsed laser light is required, a structure is adopted in which a pulsed laser light with a lower output than the required laser light is generated and then amplified by an amplifier. In this structure, when the pulse cycle (i.e., time interval) of the pulsed laser light is not constant, when the amplifier changes the time interval of the incident pulsed laser light, the gain of the amplifier varies for each pulsed laser light, and the pulse energy of the pulsed laser light output by the amplifier will deviate (be inconsistent).

Patent document 1 discloses a pulse laser system that suppresses the variation of the pulse laser light incident time interval in the amplifier and the gain variation of the amplifier by injecting a secondary signal having a different wavelength from the main signal into the amplifier in addition to the main signal when a pulse laser light having a non-constant pulse period is amplified by an amplifier. In this pulse laser system, the secondary signal is eliminated from the output of the amplifier to obtain the main signal as the laser output. [Prior art document] [Patent document]

[Patent Document 1] Japanese Patent Publication No. 2018-531524

[The problem that the invention is trying to solve]

However, according to the above-mentioned known technology, when the desired pulse cycle is long, more secondary signals are eliminated after amplification. Therefore, most of the input power is used for secondary signals that are not main signals, resulting in low energy efficiency.

In view of this, the purpose of the present disclosure is to obtain a laser device that can suppress the inefficiency of converting input power into laser output while suppressing the variation of amplified pulse energy even when the pulse cycle varies. [Means for solving the problem]

In order to solve the above-mentioned problems and achieve the purpose, the laser device disclosed in the present invention has a Q-switched laser oscillator for generating pulsed laser light, an amplifier for amplifying the pulsed laser light, an optical switch element arranged on the optical path between the Q-switched laser oscillator and the amplifier, and a control device for modulating the transmittance of the optical switch element based on the pulse characteristic time representing the time interval of the pulsed laser light generated by the Q-switched laser oscillator. [Effect of the invention]

The laser device obtained according to the present disclosure has the effect of suppressing the inefficiency of converting input power into laser output and suppressing the variation of amplified pulse energy even when the pulse cycle varies.

The laser device, laser processing device, learning device, inference device, laser processing system and laser processing method of the disclosed embodiments are described in detail below based on the drawings.

<Example 1> FIG. 1 shows the structure of the laser device 500 of Example 1. The laser device 500 has a Q-switch laser oscillator 100, an amplifier 200, an optical switch element 26, a control device 35, and an information processing device 36. The Q-switch laser oscillator 100 uses an element with a shutter function such as a Q switch to generate pulse laser light. The amplifier 200 amplifies the pulse laser light generated by the Q-switch laser oscillator 100. The optical switch element 26 is provided on the optical axis 2 between the Q-switch laser oscillator 100 and the amplifier 200.

The pulse laser light generated by the Q-switched laser oscillator 100 is incident on the optical switch element 26, passes through the optical switch element 26 with a set transmittance, and then is incident on the amplifier 200 along the optical axis 2. The transmittance of the optical switch element 26 is modulated according to the signal from the control device 35. In addition, "modulating the transmittance" means changing the transmittance over time. The optical switch element 26 is, for example, an acousto-optic element, an electro-optic element, etc.

The control device 35 controls the pulse cycle, pulse width, power input to the Q-switched laser oscillator 100 and the amplifier 200, etc. simultaneously with the transmittance control of the optical switch element 26. The pulse cycle is the time interval for the Q-switched laser oscillator 100 to oscillate the pulse laser light.

The information processing device 36 calculates the control signal of the optical switch element according to the oscillation interval of the pulse laser light based on the information obtained by the sensor (not shown) provided in the Q-switched laser oscillator 100 and the amplifier 200, and transmits the information including the calculation result to the control device 35.

The information obtained by the sensor is, for example, the pulse energy of the pulse laser light, the waveform of the pulse laser light, the temperature information of the laser device 500, etc. The control device 35 controls the operation of the laser device 500 by transmitting a control signal to the optical switch element 26, etc. based on the information from the information processing device 36.

The pulse laser light incident on the amplifier 200 through the optical switch element 26 is amplified by the amplifier 200 and then emitted along the optical axis 3. When the pulse laser light is continuously generated, the optical switch element 26 is controlled to reduce the variation of the pulse energy of the pulse laser light emitted from the amplifier 200. In addition, although the laser device 500 in FIG. 1 has one optical switch element 26, the laser device 500 may have a plurality of optical switch elements 26.

In addition, although FIG. 1 shows the case where the Q-switched laser oscillator 100 and the amplifier 200 are separate devices, as described below, the Q-switched laser oscillator 100 and the amplifier 200 may be a single device configured in a single housing.

In addition, the functions of the control device 35 and the information processing unit 36 can be realized by using processing circuits. These processing circuits can be realized by dedicated hardware or by using a control circuit of a CPU (central processing unit).

When the processing circuit is dedicated hardware, for example, a single circuit, a complex circuit, a programmed processor, a parallel programmed processor, an ASIC (application specific integrated circuit), an FPGA (field programmable gate array), or a combination of the foregoing is used.

When the above-mentioned processing circuit is implemented by a control circuit using a CPU, the control circuit has a processor and a memory. The processor is a CPU, which is also called a processing device, an arithmetic device, a microprocessor, a microcomputer, or a DSP (digital signal processor). The memory is, for example, a non-volatile or volatile semiconductor memory such as RAM (random access memory), ROM (read-only memory), flash memory, EPROM (erasable programmable read-only memory), EEPROM (registered trademark) (electronically erasable programmable read-only memory), a magnetic disk, a floppy disk, an optical disk, a compact disk, a DVD (digital versatile disk), etc.

When the above-mentioned processing circuit is realized by a control circuit using a CPU, the processor is realized by reading and executing a program stored in a memory corresponding to the processing of the control device 35 and the information processing device 36. In addition, the memory can be used as a temporary memory in each processing executed by the processor.

FIG2 shows an example of the detailed structure of the laser device 500 of Embodiment 1. In FIG2, the internal structure of the laser device 500 is shown in a three-dimensional diagram. The laser device 500 is, for example, a three-axis orthogonal carbon dioxide (CO2) laser. In addition, the laser 500 can be a carbon monoxide laser, an excimer laser, etc.

In the configuration of FIG. 2 , the Q-switched laser oscillator 100 and the amplifier 200 constituting the laser device 500 shown in FIG. 1 are accommodated in the same housing 300 .

Laser gas is sealed in the housing 300. The laser device 500 has blowers 40 and 41, a pair of electrodes 11 and a pair of electrodes 12, and heat exchangers 42 and 43 in the housing 300. The laser gas flows in the airflow directions 13 and 14 indicated by the arrows through the blowers 40 and 41, respectively. Then, the laser gas passes between the electrodes 11 and 12. According to the control of the discharge control device 44, high-frequency electricity is applied to the electrodes 11 and 12, and silent discharge is generated between the electrodes. This discharge excites the laser gas. Then, the laser gas flows along the airflow directions 15 and 16 indicated by the arrows and is cooled by the heat exchangers 42 and 43.

In addition, the laser device 500 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, and the total reflection mirror 21 and the partial reflection mirror 24 are arranged so that the optical axis 1 of the resonator is located in the discharge space of the excited laser gas. 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 further has a Q switch 22. The Q switch 22 is disposed on the optical axis 1 of the resonator, and generates pulse laser by pulse oscillation by controlling the Q value of the resonator. The Q switch 22 may be, for example, an acousto-optic element or an electro-optic element.

The total reflection mirror 21 constituting the resonator reflects most of the light but allows a small amount of light to pass through, and a photo sensor 50 is set at the position where the light that passes through is incident. The photo sensor 50 measures the pulse waveform, pulse energy, etc. of the pulsed laser light oscillating from the resonator. The laser device 500 further has a window 23 to block the laser gas in the housing 300 from the external air. In the example of FIG. 2 , the total reflection mirror 21, the Q switch 22, and the photo sensor 50 are set outside the housing 300, and the window 23 prevents the laser gas from leaking from the housing 300, while allowing the light from the partial reflection mirror 24 to pass through the housing 300 toward the total reflection mirror 21. The partial reflection mirror 24 is also responsible for shielding the laser gas in the housing 300 from the external air. The components arranged along the optical axis 1 from the photo sensor 50 to the partial reflection mirror 24 constitute a part of the Q-switch laser oscillator 100.

In the above description, the Q switch 22 is assumed to be set in the atmosphere, but the Q switch 22 can also be set in the housing 300, and the total reflection mirror 21 can replace the window 23 to block the laser gas in the housing 300 from the external air. In addition, more than one mirror can be set between the window 23 and the total reflection mirror 21.

The pulsed light emitted from the partial reflection mirror 24 of the oscillator enters the optical switch elements 26 and 27 through the mirror 25, and the pulsed laser light passing through the optical switch elements 26 and 27 enters the housing 300 again through the window 29 along the optical axis 2 through the mirror 28. At this time, the beam diameter of the pulsed laser light can be adjusted using a lens, a curvature mirror, etc. on the optical portion from the partial reflection mirror 24 of the resonator to the window 29. The pulsed laser light entering the housing 300 from the window 29 passes through the excited laser gas and is amplified, and then is reflected by the mirror 30 and enters the excited laser gas again. Afterwards, after passing through the excited laser gas and being amplified, it is reflected by the mirror 31 and enters the excited laser gas again. Afterwards, after passing through the excited laser gas and being amplified, it is reflected by the mirror 32 and enters the excited laser gas again. Afterwards, after passing through the excited laser gas and being amplified, the amplified pulse laser light is output from the window 33 to the outside of the housing 300. The amplified pulse laser light is reflected by the partial reflection mirror 34 along the optical axis 3 and output as the laser output. A portion of the light that passes through the partial reflection mirror 34 is incident on the photo sensor 51. The photo sensor 51 measures the pulse waveform, pulse energy, etc. of the pulse laser light output as the laser output.

The amplifier 200 is constituted by a portion including a portion where the pulsed laser light enters the housing 300 from the window 29 , passes through the laser gas excited in the housing 300 , and is output from the partial reflection mirror 34 , and the photo sensor 51 .

The laser device 500 shown in FIG2 comprises a Q-switched laser oscillator 100 and an amplifier 200 in a housing 300. The optical axes of the Q-switched laser oscillator 100 and the amplifier 200 are configured to pass through an integrated MOPA (master oscillator power amplifier) in at least one continuous discharge space. Here, an aperture (not shown) may be provided between the reflective surface of each mirror in the housing 300 and the discharge space to limit the range of light passing.

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

The control device 35 is connected to and controls the Q switch 22 , the photo-switching elements 26 , 27 , and the discharge control device 44 .

The information processing device 36 receives information from the photo sensors 50 and 51. In addition, the information processing device 36 can receive control signals of the photo switches 26 and 27 and a control signal of the discharge current from the control device 35. The information processing device 36 can calculate the pulse energy of the pulse laser light according to the pulse waveform of the pulse laser light.

In addition, the information processing device 36 can receive information indicating the state of the laser device 500 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. In addition, as described below, when the laser device 500 is used as a laser light source of a laser processing device, it can receive information indicating the timing of laser oscillation, the stabilization time, the static time, and the position determination time including the stabilization time and the static time of the galvanometer scanning mirror mounted on the laser processing device from the laser processing device.

The information processing device 36 calculates the information required for the Q switch 22, the photo-switch elements 26, 27, and the discharge control device 44 based on the received information, and transmits the information to the control device 35.

FIG3 shows the structure of a laser processing device 510 using the laser device 500 of Example 1 as a laser light source. The laser processing device 510 includes the laser device 500 and a drilling machine 400. The drilling machine 400 is a device that uses the pulsed laser light output by the laser device 500 to perform drilling, for example, drilling (opening) holes on an electronic substrate. The drilling machine 400 is positioned using a positioning mechanism such as a galvanometer scanner 403 so that the pulsed laser light is irradiated to a predetermined position on the electronic substrate.

The pulsed laser light outputted by the laser device 500 enters the drilling machine 400. The pulsed laser light is irradiated to the galvanometer scanner 403 via the mirrors 401 and 402 of the drilling machine 400. The galvanometer scanner 403 is an example of a deflection element, which adjusts the irradiation position of the pulsed laser light on the processing object 405 by deflecting the pulsed laser light. In addition, after the pulsed laser light enters the drilling machine 400, an optical element such as a lens or a spherical mirror (not shown) for adjusting the beam diameter may be arranged on the optical path from the pulsed laser light to the galvanometer scanner 403, and a mask for shaping the beam profile may be arranged. Although one galvanometer scanner 403 is shown in FIG. 3 , two galvanometer scanners 403 may be used as two axes to scan a plane.

The pulsed laser light irradiating the galvanometer scanner 403 is irradiated onto the processing object 405 through the lens 404 (i.e., the objective optical system). When the above-mentioned mask is used, it can be used as a transfer optical system to transfer the pulsed laser light shaped by the mask onto the processing object 405. The processing object 405 is set on a table movable in three orthogonal axis directions.

The position where the pulse laser light is irradiated on the processing object 405 is positioned by adjusting the angle of the galvanometer scanning mirror 403.

A pulsed laser light is irradiated once or multiple times at a position located by the galvanometer scanner 403, and then another pulsed laser light is irradiated at another position located by the galvanometer scanner 403. By repeatedly and continuously performing this process, a plurality of holes are processed in the object 405.

The time required for positioning the galvanometer scanning mirror 403 depends on the target shape of the hole to be processed on the electronic substrate, that is, the processing pattern of the hole. FIG4 shows an example of the target shape of the hole formed on the electronic substrate. The arrows in FIG4 indicate the processing path of the hole. In the following description, each processing path is specified by the number attached to the processing path. When processing is performed in the numerical order of processing path #1, processing path #2, processing path #3, etc., the time required for positioning the galvanometer scanner 403 becomes shorter in the portion where the intervals between holes of the electronic substrate are short, such as processing path #1, processing path #4, processing path #6, processing path #7, etc., and the time required for positioning the galvanometer scanner 403 becomes longer in the portion where the intervals between holes are long, such as processing path #3, processing path #5, processing path #8, etc. Therefore, when drilling multiple holes continuously on the electronic substrate, the positioning time of the galvanometer scanner 403 when processing each hole is not constant and varies over time.

Since the pulse laser light is irradiated at the time when the galvanometer scanning mirror 403 is positioned, the timing of emitting the pulse laser light from the laser device 500 also changes with time. In the Q-switch oscillation, since the oscillation timing of the pulse laser light changes with time and the accumulation time of each pulse laser light gain changes, the pulse energy of the oscillating pulse laser light also changes with time. Therefore, the pulse laser light oscillating by the Q-switch laser oscillator 100 is amplified by the amplifier 200, and the pulse energy of the pulse laser light also changes with time. In this case, the pulse energy of the pulse laser light for each processing hole in the drilling machine 400 will change. This situation is shown in FIG5 and FIG6.

FIG5 shows an example of a pattern of processing holes and a processing path processed on an electronic substrate. FIG6 shows the time change of the pulse laser light waveform when processing the processing holes of the pattern shown in FIG5. As shown in FIG5, when processing holes in the order of processing hole #1, processing hole #2, processing hole #3, and processing hole #4, the waveform of the pulse laser light caused by the oscillation of the Q switch becomes as shown in FIG6. When the horizontal axis of FIG6 is represented, the numbers attached to the horizontal axis correspond to the numbers attached to each processing hole in FIG5. Compared with processing hole #1 and processing hole #2, the interval between processing hole #3 and processing hole #4 is longer. Therefore, the pulse energy of the pulse laser light used to process the processing holes #3 and #4 becomes higher than that of the pulse laser light used to process the processing holes #1 and #2. In this case, the pulse energy of the pulse laser light used to process each processing hole will vary (inconsistent), and the processing quality may become unstable.

Therefore, in the laser device 500, in order to suppress the deviation of the pulse energy of each pulse laser light incident on the drilling machine 400, the transmittance of the pulse laser light emitted from the Q-switched laser oscillator 100 when passing through the optical switching element 26 is adjusted for each pulse laser light, so as to suppress the deviation of the pulse energy of the pulse laser light incident on the amplifier 200.

FIG7 is used to illustrate the effect of the electro-optical device 500. FIG7 (a) shows the waveform of the pulse laser light oscillated by the Q-switch laser oscillator 100. FIG7 (b) shows the transmittance of the optical switch element 26. FIG7 (c) shows the waveform of the pulse laser light input to the amplifier 200. FIG7 (d) shows the waveform of the pulse laser light output from the amplifier 200.

The oscillation timing of the Q-switched laser oscillator 100 is not a fixed period. When the pulse period changes, the pulse energy of the pulse laser light output from the Q-switched laser oscillator 100 increases as the time interval from the previous oscillation timing increases. For example, the interval length of the last pulse laser light is in the order of the longer one being the first, pulse laser light #3, pulse laser light #4, and pulse laser light #1. When pulse laser light #1 and pulse laser light #2 are the same, the pulse energy of the pulse laser light output from the Q-switch laser oscillator 100 is in the order of the higher one being the first, pulse laser light #3, pulse laser light #4, and pulse laser light #1. Pulse laser light #1 and pulse laser light #2 are the same.

Furthermore, in the amplifier 200, when pulsed laser light of the same pulse energy is input, the pulse energy of the output pulsed laser light increases as the time interval from the previous pulsed laser light increases. Therefore, the transmittance of the control optical switch element 26 decreases as the time interval of the pulsed laser light increases, thereby causing the pulse energy of the pulsed laser light incident on the amplifier 200 to decrease as the time interval of the pulsed laser light increases. In the example of FIG7 , the pulse energy of the pulse laser light incident on the amplifier 200 is in the order of the smaller one being the pulse laser light #3, the pulse laser light #4, and the pulse laser light #1, and the pulse laser light #1 and the pulse laser light #2 are the same; the transmittance is in the order of the smaller one being the pulse laser light #3, the pulse laser light #4, and the pulse laser light #1, and it is assumed that the pulse laser light #1 and the pulse laser light #2 are the same. In this way, the pulse energy of the pulse laser light emitted from the amplifier 200 is close to uniform, thereby suppressing deviation.

The time interval between pulsed laser lights is detected based on the pulse waveform, for example, and the detection value can be used to control the transmittance. Based on the measured voltage measured by a sensor that can measure the time dependency of the light intensity of pulsed laser lights from a photodetector, a phototube, etc., the occurrence time of the pulsed laser lights can be identified from the characteristic shape of the pulse waveform, and the time interval between the pulsed laser lights can be defined. For example, the peak time of the pulse can be set as the occurrence time, the time it takes for the light intensity to rise to a predetermined ratio relative to the pulse peak value can be set as the occurrence time, or the time it takes for the light intensity to decrease to a predetermined ratio relative to the peak value after the pulse rises to the peak value can be set as the occurrence time.

The laser device 500 modulates the transmittance of the optical switch element 26 for each pulse laser light according to the pulse characteristic time; the pulse characteristic time indicates the characteristics of the time interval of the pulse laser light generated by the Q-switched laser oscillator 100. The pulse characteristic time is, for example, the time interval between the plurality of pulse laser lights generated by the Q-switched laser oscillator 100, or the moving average of the time interval between the plurality of pulse laser lights generated by the Q-switched laser oscillator 100. The laser device 500 controls the transmittance of the optical switch element 26 so that the transmittance increases as the time interval indicated by the pulse characteristic time becomes shorter, and the energy variation between the pulse laser lights becomes smaller than a predetermined value.

In addition, as shown in FIG. 2 , when the laser device 500 has a plurality of optical switch elements 26 and 27 , the laser device 500 controls the combined transmittance of the optical switch elements 26 and 27 to satisfy the above-mentioned condition.

As described above, the laser device 500 of the first embodiment includes a Q-switched laser oscillator 100 for generating pulsed laser light, an amplifier 200 for amplifying the pulsed laser light, an optical switch element 26 disposed on an optical path between the Q-switched laser oscillator 100 and the amplifier 200, and a control device 3 for modulating the transmittance of the optical switch element 26 according to the pulse characteristic time; the pulse characteristic time represents the characteristics of the time interval at which the Q-switched laser oscillator 100 generates the pulsed laser light. Modulating the transmittance means changing the transmittance value with time. Specifically, in the first embodiment, the control device 35 changes the transmittance value for each pulsed laser light incident on the optical switch element 26. By this, the time interval at which the Q-switched laser oscillator 100 generates the pulse laser light is changed, and the pulse energy of the pulse laser light generated by the Q-switched laser oscillator 100 is different for each pulse laser light, and even when the amplification factor of the amplifier is changed, the variation of the pulse energy after amplification can be suppressed. In addition, at this moment, in the method of suppressing the time interval change of the pulse laser light incident by injecting a secondary signal other than the main signal into the amplifier, the reduction in the conversion efficiency from the input power to the laser output caused by eliminating the secondary signal and extracting the main signal as the laser output can be suppressed. Therefore, while suppressing the reduction in the conversion efficiency from the input power to the laser output, the variation of the pulse energy after amplification can be suppressed even if the pulse cycle changes.

Furthermore, the control device 35 obtains the time interval between the multiple pulse laser lights generated by the Q-switched laser oscillator 100 or the moving average of the time interval between the multiple pulse laser lights generated by the Q-switched laser oscillator 100 as the pulse time characteristic, and controls the transmittance of the optical switch element 26 so that the transmittance of the optical switch element 26 becomes higher as the time interval becomes shorter, and the energy variation between the multiple pulse laser lights emitted from the amplifier 200 is smaller than a predetermined value. Here, controlling so that the energy variation between the multiple pulse laser lights emitted from the amplifier 200 is smaller than a predetermined value means, for example, in the example of FIG. 7 , making the difference in pulse energy between the pulse laser lights #1 to #4 smaller than a predetermined value as shown in FIG. 7 (d). Therefore, while suppressing the decrease in conversion efficiency from input power to laser output, it is possible to suppress changes in pulse energy after amplification even when the pulse cycle changes.

In addition, in the above, although the pulsed laser light is irradiated at the time when the galvanometer scanner 403 completes the positioning, when the positioning time of the galvanometer scanner 403 is too short, the accumulated gain during the period when the Q-switched laser oscillator 100 stops oscillating will be too small, and the pulse energy of the pulsed laser light output from the amplifier 200 will be less than the desired value. In this case, the galvanometer scanner 403 can be stopped until the required pulse energy is obtained after amplification, and the pulsed laser light can be irradiated when the required pulse energy is obtained.

<Example 2> Example 2 describes the control process of the laser processing device 510 of the drilling machine 400 for using the laser output of the laser device 500 for electronic substrates.

FIG8 is a flow chart for explaining the control operation of the laser processing device 510 of Embodiment 2. The control device 35 of the laser device 500 first obtains data representing the drilling pattern of the electronic substrate processed by the drilling machine 400 (step S101). For example, the data representing the drilling pattern may include hole position information of the processed electronic substrate.

Next, the control device 35 calculates the drilling path of the electronic substrate according to the obtained data representing the drilling pattern (step S102). After that, the control device 35 calculates the oscillation interval of the pulsed laser light (step S103), and the oscillation interval refers to the time interval of the pulsed laser light generated by the Q-switched laser oscillator 100 to process each processing hole.

The control device 35 calculates the initial value of the transmittance of the optical switch element 26 for each pulsed laser light corresponding to the oscillation time interval of the pulsed laser light (step S104). The control device 35 stores the calculated data of the oscillation interval of the pulsed laser light and the transmittance of the optical switch element 26 in, for example, a memory in the control device 35, and sets the transmittance of the optical switch element 26 for each pulsed laser light (step S105).

The control device 35 uses the calculated oscillation interval of the pulsed laser light and the calculated transmittance of the optical switch element 26 to control the Q-switched laser oscillator 100 and the optical switch element 26, and causes the Q-switched laser oscillator 100 to oscillate (step S106).

In addition, the control device 35 obtains the pulse energy of the amplified pulse laser light obtained by the photo sensor 51 through the information processing device 36 (step S107). In addition, the number of times the laser oscillation is performed in step S106 can be the same as the number of holes in the next electronic substrate, or one electronic substrate can be divided into multiple times. The pulse laser light in this oscillation is not irradiated onto the electronic substrate, and no hole is drilled yet.

The control device 35 determines whether the value indicating the deviation of the pulse energy of the amplified multiple pulse laser light obtained in step S107 is below the threshold value (step S108). When the deviation of the pulse energy is not below the threshold value (step S108: No), the control device 35 corrects the transmittance of the optical switch element 26 when the deviation of the oscillation pulse energy is above the threshold value according to the deviation of the pulse energy of each pulse laser light after amplification (step S109), and returns to step S105. Therefore, the laser device 500 sets the corrected transmittance, controls the state of the optical switch element 26 using the corrected transmittance, performs laser oscillation again, and obtains the amplified pulse energy.

When the pulse energy deviation is below the threshold value (step S108: yes), the laser processing device 510 starts drilling holes in the electronic substrate (step S110). By performing such processing, the laser processing device 500 repeatedly corrects the transmittance of the optical switch element 26 until the deviation of the amplified pulse energy measured by the laser oscillation is below the threshold value.

The control device 35 may correct the transmittance using, for example, proportional integral differential (PID) control. Alternatively, the transmittance may be determined by machine learning as described below.

If it is known in advance that the deviation of the pulse energy will be within a predetermined range by using the set value as the initial value of the rate of the optical switch element 26, the processing of obtaining the pulse energy or correcting the deviation of the pulse energy can be omitted, and the processing can be performed as it is. In short, feedback control is not required during processing. In addition, obtaining the pulse energy or correcting the deviation of the pulse energy can be performed while the processing is performed, that is, feedback control can be performed.

In addition, the functional division between the control device 35 and the information processing device 36 is an example. In the above, the processing described as the function of the control device 35 can be executed by the information processing device 36, and the processing described as the function of the information processing device 36 can be executed by the control device 35. For example, the information processing device 36 can acquire the data of the drilling pattern of the electronic substrate, calculate the drilling path of the electronic substrate, calculate the oscillation interval of the pulsed laser light, calculate the initial value of the transmittance of the optical switch element 26, calculate the value representing the deviation of the pulse energy, calculate the correction amount of the transmittance of the optical switch element 26, etc.

As described above, the laser processing method of Example 2 includes the following steps: controlling the Q-switch laser oscillator 100 by the control device 35 to generate a pulse laser light; setting the transmittance of the pulse laser light through the optical switch element 26 by the control device 35; causing the pulse laser light to be incident on the optical switch element 26 with the set transmittance to change the pulse energy of the pulse laser light; and amplifying the pulse laser light by the amplifier 200. Step; a step of adjusting the position of the processing object 405 irradiated by the pulse laser light by deflecting the amplified pulse laser light by the galvanometer scanner 403; and a step of focusing or translating the deflected pulse laser light to irradiate the processing object 405; wherein, in the step of changing the pulse energy of the pulse laser light, the transmittance of the optical switch element 26 is modulated according to the pulse characteristic time representing the characteristic time interval of generating the pulse laser light.

<Example 3> In Example 2, although the deviation of the pulse energy of the pulsed laser light is controlled within a predetermined range, the deviation of the shape of the processed hole can also be controlled within a predetermined range.

Fig. 9 is a flow chart for explaining the control operation of the laser processing device 510 according to Embodiment 3. The processing from step S101 to step S106 is the same as that in Fig. 8, so the description thereof is omitted here.

When the laser is oscillating in step S106, the laser processing device 510 starts drilling holes in the electronic substrate using the pulsed laser light output by the laser device 500 (step S121). After the drilling is performed, the laser processing device 510 obtains the shape of the processed hole after drilling (step S122). The shape of the processed hole can be obtained from an image obtained using a camera, etc., or by scanning the processing surface with a probe laser. The number of times the pulsed laser light oscillates can be the same as the number of holes in one electronic substrate, or one electronic substrate can be divided into multiple times.

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

Here, the deviation of the processed hole shape is the fluctuation range of the value used to represent each characteristic of the plurality of processed holes. The value representing the characteristics of the processed hole can be a value used to represent the measurement result of each of the plurality of processed holes, or a value calculated from the measurement result. As an example of a value representing the characteristics of the processed hole, it can be 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, the amount of scattered materials around the hole, etc. In addition, the shape of the processed hole defines not only the shape of a single hole, but also the configuration of the hole, that is, where the hole is configured, how many holes are configured, etc., which can be described in the processing formula, instructions, etc.

When the deviation of the processed hole shape is not within the predetermined range (step S123: No), the control device 35 corrects the transmittance of the optical switch element 26 according to the deviation of the processed hole shape (step S124), and then returns to step S105. Here, although the transmittance is set corresponding to each pulsed laser light, the transmittance to be corrected is the transmittance corresponding to the pulsed laser light used to process the processing hole whose difference value is not within the predetermined range; the aforementioned difference value refers to the difference between the value representing the shape of the processed hole and the reference value.

When the deviation of the processed hole shape is within the predetermined range (step S123: yes), the electronic substrate is drilled at the set transmittance (step S125). The electronic substrate sample for adjustment can be used until the transmittance for drilling is determined.

The control device 35 can correct the transmittance using, for example, PID control. Alternatively, the transmittance can be determined by machine learning as described below.

In addition, in the third embodiment, the functional division between the control device 35 and the information processing device 36 is an example. In the above, the processing described as the function of the control device 35 can be executed by the information processing device 36, and the processing described as the function of the information processing device 36 can be executed by the control device 35. For example, the information processing device 36 can be used to obtain the data of the drilling pattern of the electronic substrate, calculate the drilling path of the electronic substrate, calculate the oscillation interval of the pulse laser light, calculate the initial value of the transmittance of the optical switch element 26, obtain the shape of the processed hole, calculate the value indicating the deviation of the pulse energy, calculate the correction amount of the transmittance of the optical switch element 26, etc.

As described above, the laser processing device 510 of the third embodiment includes: the electro-optical device 500; the galvanometer scanner 403, which is a deflection element for deflecting the pulse laser light output from the amplifier 200; the lens 404, which is an objective optical system for focusing or translating the pulse laser light from the galvanometer scanner 403 to irradiate the processing object 405; and a moving mechanism for moving the processing object 405. The control device 35 controls the transmittance based on the processing hole characteristic value, which represents the characteristics of the processing holes included in the processing path when drilling holes in the processing object 405 by the pulse laser light, so that the transmittance becomes higher as the interval of the processing holes becomes shorter and the value representing the deviation of the shapes of the plurality of processing holes is less than a predetermined value. When the laser device 500 is used for drilling holes, since the pulse characteristic time becomes longer as the interval between processed holes becomes longer, by controlling the transmittance based on the characteristic value of the processed holes, the transmittance based on the pulse characteristic time can be controlled, and the shapes of the processed holes can be homogenized.

In addition, the machined hole characteristic value indicates the interval between a plurality of machined holes on the machining path or the moving average value of the interval between a plurality of machined holes.

<Example 4> In Examples 2 and 3, although the drilling pattern of the electronic substrate is known in advance, when the number of holes in the electronic substrate is very large, it is necessary to save the transmittance corresponding to each hole as data, which puts pressure on memory resources. In addition, it is necessary to predetermine the oscillation timing of the pulse laser light.

The control method shown in the fourth embodiment can be applied to the case where the drilling pattern of the electronic substrate or the oscillation timing of the pulsed laser light is not known in advance.

FIG10 illustrates the structure of the learning device 60 of the fourth embodiment. The learning device 60 performs machine learning related to the laser device 500. The learning device 60 includes a learning data acquisition unit 61 for acquiring learning data for learning, and a model generation unit 62 for generating a learned model for inferring the transmittance of the optical switch element 26 using the learning data. The model generation unit 62 stores the generated learned model in the memory unit 70.

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

In addition, when the laser device 500 is a gas laser (i.e., the laser medium is gas), the state quantity obtained by the learning data acquisition unit 61 includes at least one of the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the gas temperature of the Q-switched laser oscillator 100 or 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 total discharge time after the optical component of the Q-switched laser oscillator 100 or the amplifier 200 is replaced.

The model generation unit 62 learns the transmittance of the optical switch element 26 based on the learning data generated by combining the state quantity and the transmittance of the optical switch element 26; the aforementioned state quantity includes the interval information output from the learning acquisition unit 61 and the pulse energy of the pulse laser light generated by the laser device 500 after amplification. Specifically, the model generation unit 62 generates a learned model for inferring the transmittance of the optical switch element 26 from the state quantity of the laser device 500, and the target value is the amplified pulse energy. Here, the learning data is data that correlates the state quantity and the transmittance of the optical switch element 26.

The learning algorithm used by the model generation unit 62 may be a well-known algorithm such as supervised learning, unsupervised learning, or reinforcement learning. As an example, the case where the algorithm is applied to a neural network will be described.

The model generation unit 62 learns the transmittance of the optical switch element 26 by so-called supervised learning based on, for example, a neural network model. Here, so-called supervised learning is to give a data set of input and result (label) to the learning device 60, thereby learning the characteristics of these learning data and inferring the result from the input.

A neural network includes an input layer composed of multiple neurons, an intermediate layer composed of multiple neurons, i.e., a storage layer, and an output layer composed of multiple neurons. The intermediate layer may be one layer or two or more layers.

FIG11 shows an example of a 3-layer neural network. For example, if it is a 3-layer neural network as shown in FIG11, after a plurality of inputs are input to the input layer (X1~X3), their values are multiplied by weights W1 (w11~w16) and input to the middle layer (Y1~Y2), and their results are further multiplied by weights W2 (w21~w26) and output from the output layer (Z1~Z3). This output result changes according to the values of weights W1 and W2.

In the fourth embodiment, the neural network learns the transmittance through so-called supervised learning based on the learning data created by the combination of the state quantity and the transmittance acquired by the learning data acquisition unit 61.

That is, in the neural network, a state quantity including a target value of the amplified pulse energy is inputted into the input layer, and the result outputted from the output layer is learned by adjusting weights W1 and W2 so that the amplified pulse energy approaches the transmittance of the optical switch element 26 which is the target value.

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

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

Next, the learning process performed by the learning device 60 will be described using Fig. 12. Fig. 12 is a flowchart for describing the learning process of the learning device 60.

The learning data acquisition unit 61 acquires learning data including state quantities and transmittance (step S201). In addition, the state quantities and transmittance are acquired at the same time, but the state quantities and transmittance can be input in association with each other, and the state quantities and transmittance can be acquired at different times.

The model generation unit 62 generates a learning data based on the combination of state quantity and transmittance obtained by the learning data acquisition unit 61, and performs a learning process of the transmittance so that the pulse energy after learning amplification will be the target value through so-called supervised learning, thereby generating a learning-completed model (step S202).

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

13 is a diagram showing the configuration of the inference device 80 of the laser device 500. 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 quantities include the interval information and the target value of the amplified pulse energy. In addition, when the laser device 500 is a gas laser (i.e., the laser medium is gas), the state quantity obtained by the inferred data acquisition unit 81 includes at least one of the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the gas temperature of the Q-switched laser oscillator 100 or 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 total discharge time after the optical component of the Q-switched laser oscillator 100 or the amplifier 200 is replaced.

The inference unit 82 infers the transmittance obtained by the learned model. In other words, the inference unit 82 inputs the inference data obtained by the inference data acquisition unit 81 into the learned model stored in the learned memory unit 70, and can output the transmittance so that the amplified pulse energy inferred based on the state quantity becomes the target value.

In addition, the learned model using the inference device 80 can be a learned model learned using learning data obtained from the laser device 500 (i.e., the inference object of the inference device 80), or a learned model obtained from outside such as another laser device 500 that is different from the laser device 500 (i.e., the inference object of the inference device 80).

Next, the process of obtaining the transmittance using the inference device 80 will be described using Fig. 14. Fig. 14 is a flow chart for describing the process of obtaining the transmittance using the inference device 80.

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 memory unit 70 (step S302), and outputs the inference result of the transmittance to the laser device 500 (step S303).

The transmittance estimation result outputted from the laser device 500 is used to control the optical switching element 26 (step S304).

For example, in the drilling machine 400 shown in FIG. 3 , when drilling a hole on an electronic substrate, the control device 35 issues a pulse laser light output command to the Q switching oscillator 100, so that after the galvanometer scanner 403 completes positioning, the pulse laser light is irradiated on the processing object 405. In addition, in order to reduce the time loss caused by the delay of the command signal, the control device 35 predicts the time when the galvanometer scanner 403 completes positioning, and issues a pulse laser light output command to the Q switching oscillator 100 before the galvanometer scanner 403 completes positioning, so that the pulse laser light is immediately irradiated on the processing object 405 after the positioning is completed. In any case, the generation time of the pulsed laser light depends on the positioning time of the galvanometer scanning mirror 403. At this time, by using the learned model obtained by the learning device 60, the inference device 80 can determine the transmittance of the optical switch element 26 based on the interval information indicating the generation interval of the pulsed laser light and the state quantity including the target value of the amplified pulse energy. Therefore, by using the learned model, even if the generation interval of the pulsed laser light is not predetermined, the transmittance of the optical switch element 26 can be determined for each pulsed laser light, and the amplified pulse energy can be controlled.

In addition, in this embodiment, although the supervised learning is applied as the learning algorithm used by the model generation unit 62, the present invention is not limited to this. In addition to the supervised learning, the learning algorithm may use reinforcement learning, unsupervised learning, or semi-supervised learning.

In addition, the model generation unit 62 can learn the transmittance according to the learning data established for the plurality of laser devices 500. In addition, the model generation unit 62 can obtain the learning data from the plurality of laser devices 500 used in the same area, and can also learn the transmittance using the learning data collected from the plurality of laser devices 500 used in different areas and operating independently. In addition, the laser device 500 for which the learning data is collected can be added as an object or removed from the object during the process. In addition, the learning device 60 that has learned the transmittance of a certain laser device 500 can be applied to another laser device 500, and the transmittance of another laser device 500 can be relearned and updated.

In addition, as a learning algorithm used in the model generation unit 62, deep learning (Deep Learning) can be used to learn the feature quantity itself, or other well-known methods such as genetic programming, functional logic programming, support vector machine, etc. can be used to implement machine learning.

In addition, the learning device 60 and the inference device 80 are used to learn the transmittance of the optical switch element 26 of the laser device 500, and are connected to the laser device 500 through a network, for example, and can be devices separate from the laser device 500. In addition, the learning device 60 and the inference device 80 can be built into the laser device 500 or the laser processing device 510. For example, at least one of the learning device 60 and the inference device 80 can be part of the function of the control device 35 or the information processing device 36. In addition, the learning device 60 and the inference device 80 can be stored on a cloud server.

In addition, the above-mentioned method stabilizes the amplified pulse energy by controlling the transmittance of the optical switch element 26, but the amplified pulse energy may also be stabilized by controlling the transmittance of the optical switch element 26 and the discharge current or the amplified power.

In addition, as shown in Example 1, when the positioning time of the galvanometer scanner 403 is too short so that the pulse energy output by the amplifier 200 is less than the desired value, the galvanometer scanner 403 can be stopped until the required pulse energy after amplification is obtained, and control can be performed so that the pulse laser light is irradiated at the time point when the required pulse energy is obtained.

For example, in CO2 laser, the oscillation efficiency changes according to the gas temperature and gas pressure. Therefore, in order to correct the change of pulse energy or the transmittance of the optical switch element 26 caused by the gas temperature and gas pressure, these parameters can be used as the state quantity of machine learning. In addition, when performing continuous discharge, the pulse quantity may decrease due to the degradation of the laser gas. Therefore, by using the discharge time of the laser device 500 as the state quantity of machine learning, the dependence of the transmittance of the optical switch element 26 and the pulse energy is calculated, which can be used to correct the transmittance or pulse energy of the optical switch element 26. In addition, the total discharge time after replacing the optical component becomes an indicator of the degradation state of the optical component. When optical components deteriorate, the pulse energy changes due to a decrease in reflectivity or transmittance. Therefore, the total discharge time after the optical component is replaced can be used as a state quantity for machine learning.

As described above, the laser device 500 of the fourth embodiment may be further equipped with a learning device 60. The learning device 60 has a learning data acquisition unit 61 for acquiring learning data and a model generation unit 62 for generating a learned model. The learning data includes a state quantity and a transmittance of the optical switch element 26. The state quantity includes interval information indicating a time interval at which the laser device 500 generates a pulse laser light and a pulse energy after amplification of the pulse laser light. The learned model uses the learning data to infer the transmittance of the optical switch element 26 that makes the amplified pulse energy a cyclic value from the state quantity.

In addition, the laser device 500 of the fourth embodiment may be further equipped with an inference device 80. The inference device 80 includes: an inference data acquisition unit 81 that acquires a state quantity including interval information indicating a time interval for generating a pulse laser light and a target value of a pulse energy after amplification of the pulse laser light; and an inference unit 82 that infers transmittance from the state quantity acquired by the inference data acquisition unit 81 using a learned model for inferring the transmittance of the optical switch element 26 whose target value is the amplified pulse energy from the state quantity. The control device 35 controls the optical switch element 26 using the transmittance inferred by the inference unit 80.

In addition, the interval information used by the learning device 60 and the inference device 80 includes, for example, at least one of the pulse characteristic time of the pulse laser light generated by the Q-switched laser oscillator 100, the energy of the pulse laser light generated by the Q-switched laser oscillator 100, the waveform of the pulse laser light generated by the Q-switched laser oscillator 100, the waveform of the pulse laser light amplified by the amplifier 200, and the driving current or discharge force of the Q-switched laser oscillator 100 and the amplifier 200.

In addition, when the laser device 500 is a gas laser, the state quantities used by the learning state device 60 and the inference device 80 include at least one of the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the gas temperature of the Q-switched laser oscillator 100 or 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 total discharge time after the optical component of the Q-switched laser oscillator 100 or the amplifier 200 is replaced.

In addition, the learning device 60 and the inference device 80 may be devices separate from the laser device 500. A laser processing system may be constructed by being equipped with at least one of the learning device 60 and the inference device 80 separate from the laser device 500.

<Example 5> In Example 4, the transmittance of the optical switch element 26 whose target value is the amplified pulse energy is learned. When the pulse laser light output by the laser device 500 is used to drill holes in an electronic substrate, the stability of the processed shape is more desired.

Therefore, in Example 5, a method of learning the transmittance so that the shape of the processed hole after processing becomes the target shape is described.

FIG15 shows the structure of the learning device 60a of the fifth embodiment. The learning device 60a performs machine learning of the laser processing device 510 shown in FIG3, for example. The learning device 60a comprises: an acquisition unit 61a, which acquires learning data, i.e., data used for learning; and a model generation unit 62a, which uses the learning data to generate a learned model for inferring the transmittance of the optical switch element 26. The model generation unit 62a stores the generated learned model in the learned model storage unit 70a.

The learning data acquisition unit 61a acquires the state quantity including the interval information and the shape information and the transmittance of the optical switch element 26 as learning data. The interval information indicates the time interval of the pulse laser light generated by the laser device 500 of the laser processing device 510. The shape information indicates the shape of the processed hole processed by the laser processing device 510. When the interval information is the time interval of generating the pulse laser light, it can be any information. The interval information includes, for example, the pulse characteristic time, the energy of the pulse laser light generated by the Q-switching oscillator 100, the waveform of the pulse laser light generated by the Q-switching oscillator 100, the waveform of the pulse laser light amplified by the amplifier 200, and at least one of the drive current or discharge current of the Q-switching oscillator 100 and the amplifier 200. The learning device 60a, for example, obtains the shape of the processed hole obtained by processing while making various changes to the time interval of generating the pulse laser light and the transmittance of the optical switch element 26, and obtains the state quantity including the interval information and shape information at that time and the set transmittance as learning data.

In addition, when the laser device 500 is a gas laser whose laser medium is gas, the state quantities obtained by the learning data acquisition unit 61a include at least one of the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the gas temperature of the Q-switched laser oscillator 100 or 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 total discharge time after the optical components of the Q-switched laser oscillator 100 or the amplifier 200 are replaced.

The model generation unit 62a learns the transmittance of the optical switch element 26 based on the learning data generated based on the combination of the state quantity including the interval information output by the learning data acquisition unit 61a and the shape information indicating the shape of the processed hole processed by the laser processing device 510 and the transmittance of the optical switch element 26. Specifically, the model generation unit 62a generates a learned model for inferring the transmittance of the optical switch element 26 so that the shape of the processed hole after processing is a desired shape from the state quantity of the laser processing device 510.

The learning algorithm used by the model generation unit 62a may be a well-known algorithm such as supervised learning, unsupervised learning, and reinforcement learning. The following describes the case where the algorithm is applied to a neural network as an example.

The model generation unit 62a learns the transmittance of the optical switch element 26 by so-called supervised learning, for example, according to a neural network. Here, the so-called supervised learning method is to input result (label) data to the learning device 60a, learn the features in the learning data, and infer the result from the input.

The configuration of the neural network includes an input layer composed of multiple neurons, an intermediate layer composed of multiple neurons, i.e., a hidden layer, and an output layer composed of multiple neurons. The intermediate layer can be one layer or two or more layers. As in Embodiment 4, a neural network as shown in FIG. 11 can be used.

In the fifth embodiment, the neural network learns the transmittance that makes the processed hole shape a desired shape through so-called supervised learning based on the learning data created based on the combination of the state quantity and transmittance acquired by the learning data acquisition unit 61a.

In other words, the neural network learns by adjusting weights W1 and W2 so that the state quantity including the target value of the amplified pulse energy is input to the input layer and the result output by the output layer is close to the transmittance of the optical switch element 26 that makes the amplified pulse energy reach the target value.

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

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

The learning process of the learning device 60a is the same as that of the learning device 60 of Embodiment 4, and its description is omitted here.

Fig. 16 is a diagram showing the configuration of an inference device 80a of the laser processing device 510. 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 quantities include the above-mentioned interval information and the target shape of the processed hole after processing. In addition, when the laser device 500 is a gas laser whose laser medium is gas, the state quantity obtained by the inferred data acquisition unit 81a may further include at least one of the cooling water temperature of the Q-switched laser oscillator 100 or the amplifier 200, the gas temperature of the Q-switched laser oscillator 100 or 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 total discharge time after the optical component of the Q-switched laser oscillator 100 or the amplifier 200 is replaced.

The inference unit 82a infers the transmittance obtained by using the learned model. In other words, the inference unit 82a can infer and output the transmittance that makes the shape of the processed hole become the target shape based on the state quantity by inputting the inference data obtained by the inference data acquisition unit 81a into the learned model stored in the learned memory unit 70a.

Furthermore, the learned model used by the inference device 80a may be a learned model that has been learned using learning data obtained from the laser device 500 that is the inference target of the inference device 80a. The learned model may also be obtained from an external laser device 500 that is different from the laser device 500 that is the inference target of the inference device 80a.

The process of the inference action of the inference device 80a is the same as that of the inference device 80 of Example 4, and its description is omitted here.

For example, in the drilling machine 400 shown in FIG3 , when drilling (hole processing) an electronic substrate, the control device 35 sends a pulse laser light output command to the Q switching oscillator 100, so that after the galvanometer scanner 403 is positioned, the pulse laser light is irradiated on the processing object 405. In addition, in order to reduce the time loss caused by the delay of the command signal, the control device 30 presets the timing of the galvanometer scanner positioning, and sends a pulse laser light output command to the Q switching oscillator 100 before the galvanometer scanner 403 is positioned, so that the pulse laser light is irradiated on the processing object 405 immediately after the galvanometer scanner 403 is positioned. In any case, the timing of the pulse laser light generation is determined by the positioning time of the galvanometer scanner 403. At this time, by using the learned model obtained by the learning device 60a, the inference device 80a can determine the transmittance of the optical switch element 26 from the state quantity including the interval information indicating the time interval of the pulse laser light generation and the target shape of the processed hole. Therefore, by using the learned model, even if the interval of the pulse laser light generation is not predetermined, the transmittance of the optical switch element 26 can be determined for each pulse laser light, and the pulse quantity after amplification can be controlled.

In addition, in this embodiment, although the supervised learning is applied to the learning algorithm used by the model generation unit 62a, it is not limited to this. Regarding the learning algorithm, in addition to supervised learning, reinforcement learning, unsupervised learning, or semi-supervised learning may also be adopted.

In addition, the model generation unit 62a can learn the transmittance based on the learning data created for the plurality of laser devices 500. In addition, the model generation unit 62a can obtain the learning data from the plurality of laser devices 500 used in the same area, and can also learn the transmittance using the learning data collected from the plurality of laser devices 500 independently operating in different areas. In addition, the laser device 500 for which the learning data is collected can be added as an object or removed from the object during the process. In addition, the learning device 60a that has learned the transmittance of a certain laser device 500 can be applied to another laser device 500 to learn and update the transmittance of the other laser device 500.

In addition, as a learning algorithm used in the model generating device 62a, deep learning of learning and extracting feature quantities themselves can be used, or machine learning can be realized according to other well-known methods, such as genetic programming, functional logic programming, support vector machines, etc.

In addition, the learning device 60a and the inference device 80a are used to learn the transmittance of the optical switch element 26 of the laser device 500, and are connected to the laser device 500 through a network, for example, and can be devices separate from the laser device 500. In addition, the learning device 60a and the inference device 80a can be built into the laser device 500. In addition, the learning device 60a and the inference device 80a can be stored on a cloud server.

In addition, the above-mentioned method stabilizes the amplified pulse energy by controlling the transmittance of the optical switch element 26, but the amplified pulse energy may also be stabilized by controlling the transmittance of the optical switch element 26 and the discharge current or the amplified power.

In addition, as shown in Example 1, when the positioning time of the galvanometer scanner 403 is too short so that the pulse energy output by the amplifier 200 is less than the desired value, the galvanometer scanner 403 can be stopped until the required pulse energy after amplification is obtained, and control can be performed so that the pulse laser light is irradiated at the time point when the required pulse energy is obtained.

For example, in CO2 laser, the oscillation efficiency changes according to the gas temperature and gas pressure. Therefore, in order to correct the change of pulse energy or the transmittance of the optical switch element 26 caused by the gas temperature and gas pressure, these parameters can be used as the state quantity of machine learning. In addition, when performing continuous discharge, the pulse quantity may decrease due to the degradation of the laser gas. Therefore, by using the discharge time of the laser device 500 as the state quantity of machine learning, the dependence of the transmittance of the optical switch element 26 and the pulse energy is calculated, which can be used to correct the transmittance or pulse energy of the optical switch element 26. In addition, the total discharge time after replacing the optical component becomes an indicator of the degradation state of the optical component. When optical components deteriorate, the pulse energy changes due to a decrease in reflectivity or transmittance. Therefore, the total discharge time after the optical component is replaced can be used as a state quantity for machine learning.

As described above, the learning device 60a of Example 5 is a learning device 60a included in the laser device 500 of the laser processing device 510, and is used to learn the transmittance of the optical switch element 26. The learning device 60a has a learning data acquisition unit 61a that acquires learning data, and a model generation unit 62a that generates a learned model. The learning data includes a state quantity and the transmittance of the optical switch element 26. The state quantity includes interval information indicating the time interval of generating the pulsed laser light and shape information indicating the shape of the processed hole processed using the pulsed laser light. The learned model is to infer the transmittance of the optical switch element 26 that makes the shape of the processed hole become the target shape from the state quantity using the learning data. By having such an arrangement, it is possible to learn the relationship between the transmittance and the state quantity so as to make the shape of the processed hole uniform after processing.

In addition, the inference device 80a of the fifth embodiment is the inference device 80a of the laser device 500 of the laser processing device 510, and is used to infer the transmittance of the optical switch element 26. The inference device 80a includes: an inference data acquisition unit 81a that acquires a state quantity including interval information indicating the time interval of generating the pulsed laser light and shape information indicating the target shape of the processing hole processed by using the pulsed laser light; and an inference unit 82a that uses a learned model for inferring the transmittance of the optical switch element 26 from the state quantity and shape information to infer the transmittance from the state quantity and target shape acquired by the inference data acquisition unit 81a. By using such an inference device 80a, the transmittance that makes the shape of the processed hole after processing uniform can be inferred.

In addition, the learning device 60a and the inference device 80a may be built into the laser device 500 or the laser processing device 510, or may be devices separate from the laser device 500 and the laser processing device 510. In addition, a laser processing system equipped with at least one of the learning device 60a and the inference device 80a may be constructed.

<Example 6> In Example 1, it is assumed that the transmittance of the optical switch element 26 is determined for each pulse laser light, and the transmittance is constant when each pulse laser light is generated.

However, the pulse waveforms of pulse laser lights with different pulse intervals generated by the Q-switched laser oscillator 100 are different. FIG17 shows an example of the waveforms of pulse laser lights with different pulse intervals.

Fig. 17(a) shows the waveform of the pulse laser light oscillated by the Q-switched laser oscillator 100. Fig. 17(b) shows the transmittance of the optical switch element 26. Fig. 17(c) shows the waveform of the pulse laser light input to the amplifier 200. Fig. 17(d) shows the waveform of the pulse laser light output by the amplifier 200.

As shown in (a) of FIG. 17 , when pulse laser light with different pulse intervals is generated, the waveform of the pulse laser light output by the Q-switch laser oscillator 100 is not similar in shape, and the intensity ratio of the peak value at the top of the pulse to the stable value at the back of the pulse will be different. For example, relative to the increase in the peak value p2 of the pulse laser light #2 and the peak value p3 of the pulse laser light #3, the stable value c2 of the pulse laser light #2 and the stable value c3 of the pulse laser light #3 are approximately equal. In this case, using the same method as in Example 1, as shown in FIG17(b), when control is performed to set a certain transmittance for each pulse laser light, as shown in FIG17(d), even if the pulse energy is equal, the waveform of the amplified pulse laser light becomes different. Here, FIG17(c) is the waveform of the pulse laser light entering the amplifier 200.

In this way, when the waveforms of the multiple pulse laser lights generated by the Q-switched laser oscillator 100 are not similar, as shown in FIG. 18 , by varying the transmission of the optical switch element 26 within one pulse time, not only the pulse energy but also the waveform can be homogenized.

FIG18 illustrates an example of the change in transmittance over time during a pulse time. FIG18(a) shows the waveform of the pulse laser light oscillating from the Q-switched laser oscillator 100. FIG18(b) shows the transmittance of the optical switch element 26. FIG18(c) shows the waveform of the pulse laser light input to the amplifier 200. FIG18(d) shows the waveform of the pulse laser light output from the amplifier 200.

Furthermore, when such a transmittance configuration is implemented using machine learning, the state quantity may include the timing of the Q-switched laser oscillator 100 generating the pulse laser light, and at least one of the pulse energy of the pulse laser light, the pulse peak value at the top of the pulse, and the pulse stability value at the rear of the pulse. Furthermore, the transmittance, which is the learning result, is not a stable value for each pulse laser light, and may be a value that changes with time during the period when one pulse laser light is transmitted.

As described above, the pulse laser device 500 of the sixth embodiment changes the transmittance with time during the period when one pulse laser light passes through the optical switch element 26. Therefore, the waveform of the amplified pulse laser light can be homogenized. In addition, when the transmittance is set using machine learning, the model generation unit 62, 62a of the learning device 60, 60a can generate a learned model based on the state quantity for inferring the transmittance that changes with time during the period when one pulse laser light passes through the optical switch element 26.

<Example 7> When the interval of generating pulse laser light is short, not only the interval of the previous pulse laser light but also the history of multiple pulse laser lights may be affected. In this case, the interval information included in the state quantity shown in Examples 4 to 6 is not the time interval between the target pulse laser light and the previous pulse laser light, but may be information indicating multiple time intervals of multiple pulse laser lights included in a certain period of time.

FIG19 illustrates the transmittance control of Example 7. FIG19(a) shows the waveform of the pulse laser light oscillated by the Q-switch laser oscillator 100. FIG19(b) shows the transmittance of the optical switch element 26. FIG19(c) shows the waveform of the pulse laser light input to the amplifier 26. FIG19(d) shows the waveform of the pulse laser light output by the amplifier 200.

As shown in (a) of FIG. 19, even when pulse laser light #1 to #6 oscillates at a certain time interval, when the time interval is short, each pulse laser light #1 to #6 is affected not only by the previous pulse laser light but also by multiple pulse laser lights. In this case, even if pulse laser light #1 to #6 oscillates at a certain time interval, assuming that the duration of the response is TL, for pulse laser light #1, one pulse laser light is included in the TL period before the occurrence of pulse laser light #1, and for pulse laser light #3, two pulse laser lights #1 to #2 are included in the TL period before the occurrence of pulse laser light #3. In addition, for pulse laser light #4, three pulse laser lights #1 to #3 are included in the TL period before the occurrence of pulse laser light #1. Pulse laser lights #5 and #6 are also the same as pulse laser light #4, and three pulse laser lights are included in the TL period before each pulse laser light occurs. Therefore, the pulse energy in pulse laser lights #1 to #4 is variable, and the pulse energy in pulse laser lights #4 to #6 is constant. In this case, by controlling the transmittance as shown in FIG. 19(b) based on the total value of the time interval between the pulse laser lights within the TL period on which the pulse laser lights are affected, that is, the moving average value, the deviation (inconsistency) of the pulse energy after amplification can be suppressed as shown in FIG. 19(d).

In addition, since the value of TL varies depending on conditions such as pulse output, as shown in Examples 4 and 5, when machine learning is used, the value of TL is varied under various conditions to obtain data, and the value of TL is included in the state quantity to perform machine learning, thereby obtaining the most appropriate range.

<Example 8> When the interval between pulse laser light generation is very long, the discharge power can be changed. For example, when the laser device 500 is a three-axis orthogonal CO2 laser, it takes a certain amount of time from the start of discharge to the gain rise.

Fig. 20 shows a partial structure between electrodes 11 of the laser device 500 of Embodiment 8. For simplicity, Fig. 20 shows only a portion of the electrode 11, and does not show the electrode 12.

Fig. 20 shows the discharge space between the electrodes 11 of the laser device 500 shown in Fig. 2 as viewed from the optical axis direction. In Fig. 20, Vg is the gas flow rate, and Dwd is the electrode width of the electrode 11 in the gas flow direction 13.

FIG21 shows the gain distribution of the stable state between the electrodes 11 of the laser device 500 configured as shown in FIG20. From the start of discharge to the gain distribution in the stable state, it takes time for the laser gas to flow from the end to the end of the electrode 11 in the gas flow direction 13, which is the width direction of the electrode 11. In other words, when the time from the start of discharge to the gain distribution becoming a stable state is set to τ, τ=Dwd/Vg. For example, when the gas flow rate Vg=80m/s and the electrode width Dwd=40mm, τ=0.5msec. In other words, when the time interval of the pulse laser light is longer than the above τ, the discharge is stopped to reduce the power consumption, and the discharge can be started τ before the next pulse laser light is generated. In addition, even in the case where the time interval of the pulse laser light is shorter than the above τ, when the pulse energy is too large, the discharge power can be controlled to be reduced. In this way, even in the case of low pulse frequency, the energy efficiency can be improved. In addition, for a time longer than τ, due to the influence of the previous pulse laser light, τ can be made into the TL value described in Example 7.

As described above, in the laser device 500 of the eighth embodiment, the pulse characteristic time is less than or equal to the value of the electrode width Dwg of the discharge electrode divided by the gas flow rate Vg; wherein the electrode width Dwg of the discharge electrode is the length of the electrodes 11 and 12 of the Q-switch laser oscillator 100 in the direction of laser gas flow. Therefore, the pulse energy of the pulse laser light output from the amplifier 200 can be stabilized, and the processed shapes processed using the laser device 500 can be homogenized.

<Example 9> Figure 22 shows the internal structure of the laser device 500a of Example 9. In the laser device 500a, a part of the configuration of the part constituting the resonator is different from that of the laser device 500. Specifically, in the laser device 500a, there is a total reflection mirror 54 to replace the total reflection mirror 21. The total reflection mirror 54 is composed of two plane mirrors whose normals are perpendicular to each other. At this time, the two normals are parallel to the electrode plane, that is, parallel to the XZ plane. Therefore, the total reflection mirror 54 becomes a recursive reflection mirror in the XZ plane direction.

When the discharge space between the electrodes 11 of the laser device 500a shown in Fig. 22 is observed from the optical axis direction, the state is the same as that shown in Fig. 20. In addition, Fig. 23 shows the temperature distribution between the electrodes 11 of the laser device 500a in a stable state.

The laser gas is supplied with electricity by continuous discharge while flowing from the upstream side of the gas flow, i.e., the positive direction of the X-axis, to the downstream side, i.e., the negative direction of the X-axis. Therefore, a temperature distribution is generated in the gas flow direction 13, and the temperature of the laser gas on the downstream side becomes higher than that on the upstream side. Therefore, a refractive index distribution appears in the laser gas between the electrodes 11, and the optical axis of the light propagating through this part is slightly bent. In a resonator using a general concave mirror pair, in this case, the optical axis is moved on the upstream side of the gas flow, and the oscillation is stopped according to the situation. In Embodiment 9, since the total reflection mirror 54, i.e., the recursive reflector, is used to form the resonator, the resonant condition is that the optical axis position is limited to the tangent line of the two mirrors of the recursive reflector. In other words, when the recursive reflector is used, even if the temperature distribution appears in the airflow direction, the movement of the optical axis is limited and can oscillate stably. In this case, the pulse energy and pulse waveform of the resonator, the mirror after the output resonator can use the partial reflection mirror 55 instead of the mirror 25 of the laser device 500, and the light sensor 50 can observe part of the transmitted light. In FIG. 22 , although the partial reflection mirror 24 and the total reflection mirror 54 , which is the recursive reflection mirror, are arranged on a straight line, a reflection mirror may be arranged between the partial reflection mirror 24 and the total reflection mirror 54 .

As described above, the Q-switch laser oscillator 100 of the laser device 500a of the ninth embodiment has a recursive mirror as the total reflection mirror 54 constituting the resonator. Therefore, even when there is a temperature distribution in the air flow direction, stable oscillation can be achieved.

<Example 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 inferred from the pulse energy of the Q-switched laser oscillator 100. For example, when the pulse energy of the pulsed laser light in the oscillation section decreases due to the degradation of the laser gas, the gain of the amplifier section will also decrease. The relationship between the gain state in the amplifier 200 and the pulse energy in the Q-switched laser oscillator 100 is modeled. By correcting the transmittance of the optical switch element 26 using this model, the amplified pulse energy can be stabilized, so that the processed shape after laser processing using this pulse laser light can be homogenized.

Fig. 24 is a flow chart for explaining the operation of the laser processing device 510 of Embodiment 10. Since steps S101 to S104 are the same as those in Fig. 8 and the like, their description is omitted here.

The control device 35 stores the calculated transmittance data of the optical switch element 26 in, for example, a memory within the control device 35, and sets the transmittance of the optical switch element 26 for each pulsed laser light (step S131).

The control device 35 uses the oscillation interval of the pulse laser light calculated in step S103 and the transmittance of the optical switch element 26 set in step S131 to control the Q-switched laser oscillator 100 and the optical switch element 26, and causes the Q-switched laser oscillator 100 to oscillate only one pulse (step S132).

Furthermore, the control device 35 obtains the pulse energy of the pulse laser light of the Q-switched laser oscillator 100 measured by the photo sensor 50 through 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 (process) only one hole on the electronic substrate (step S134), and then returns to step S132. In addition, while the hole drilling is being performed, the control device 35 calculates the pulse energy of the next pulsed laser light based on the model (step S135). As described above, the model used here is a model that represents the relationship between the state of gain in the amplifier 200 and the pulse energy in the Q-switch laser oscillator 100. The control device 35 corrects the transmittance of the optical switch element 26 according to the pulse energy of the next pulsed laser light calculated based on the model (step S136), and then returns to step S131.

Through the above-mentioned processing, the pulse energy of the oscillation segment can be obtained every time a pulse oscillates, and the transmittance of the optical switch element 26 can be corrected based on the model before the next pulse laser light oscillates.

At this time, the model is created through machine learning.

The configuration shown in the above embodiment is an example. The configuration shown in the above embodiment is only an example. Without departing from the scope of the subject matter of the present invention, it can be combined with other known technologies, or it can be combined with other embodiments, and part of the configuration can be omitted or changed.

For example, in the above-mentioned embodiments 1 to 10, although the laser devices 500 and 500a are assumed to be gas lasers whose laser medium is gas, when the oscillation interval of the pulse laser light changes, the change of the pulse energy when the Q-switch oscillator 100 oscillates will have the same influence on solid lasers, etc. Therefore, in addition to the unique parts of the gas laser shown above, the above-mentioned configuration can also be applied to solid lasers, etc., and the same effect can be expected.

In addition, although the above mainly describes an example for controlling the transmittance of the optical switch element 26, as shown in FIG. 2, when the laser device 500 has a plurality of optical switch elements 26, 27, the transmittance of the plurality of optical switch elements 26, 27 can be permanently controlled to meet the above conditions.

1,2,3: optical axis 11,12: electrode 13,14,15,16: airflow direction 21,54: total reflector 22: Q switch 23,29,33: window 24,34,55: partial reflector 25,28,30,31,401,402: mirror 26,27: optical switch element 35: control device 36: information processing device 40,41: blower 42,43: heat exchanger 44: discharge control device 50,51: photo sensor 60,60a: learning device 61,61a: learning data acquisition unit 62,62a: model generation unit 70,70a: Learning model memory unit 80,80a: Inference device 81,81a: Inference data acquisition unit 82,82a: Inference unit 100: Q-switch laser oscillator 200: Amplifier 300: Housing 400: Drilling machine 403: Galvanometer scanner 404: Lens 405: Processing object 500,500a: Laser device 510: Laser processing device

FIG. 1 illustrates the structure of the laser device of Example 1. FIG. 2 illustrates an example of the detailed structure of the laser device of Example 1. FIG. 3 illustrates the structure of a laser processing device using the laser device of Example 1 as a laser light source. FIG. 4 illustrates an example of the target shape of a hole formed on an electronic substrate. FIG. 5 illustrates an example of the pattern of a processing hole processed on an electronic substrate and a processing path. FIG. 6 illustrates the time variation of the pulse laser light waveform when processing the processing hole of the pattern shown in FIG. 5. FIG. 7 is used to illustrate the effect of the electro-optical device. FIG. 8 is a flowchart for illustrating the control action of the laser processing device of Example 2. FIG. 9 is a flowchart for illustrating the control action of the laser processing device of Example 3. FIG. 10 illustrates the structure of the learning device of Example 4. FIG11 shows an example of a three-layer neural network. FIG12 is a flowchart for explaining the learning process of the learning device. FIG13 is a diagram showing the structure of the inference device of the laser device. FIG14 is a diagram showing the process of using the inference device to obtain the transmittance. FIG15 shows the structure of the learning device of Example 5. FIG16 is a diagram showing the structure of the inference device of the laser processing device. FIG17 shows an example of a pulse laser light waveform with different pulse intervals. FIG18 shows an example of the change of transmittance over time within the pulse time. FIG19 shows the transmittance control of Example 7. FIG20 shows the partial structure between the electrodes of the laser device of Example 8. FIG. 21 shows the gain distribution of the stable state between the electrodes of the laser device configured as shown in FIG. 20. FIG. 22 shows the internal structure of the laser device of Example 9. FIG. 23 shows the temperature distribution of the stable state between the electrodes of the laser device. FIG. 24 is a flow chart for explaining the operation of the laser processing device of Example 10.

2,3: optical axis

26: Optical switch components

35: Control device

36: Information processing device

100:Q switch laser oscillator

200:Amplifier

500:Laser device

Claims (23)

A laser device, comprising: a Q-switched laser oscillator, generating a pulse laser light; an amplifier, amplifying the pulse laser light; an optical switch element, disposed on an optical path between the Q-switched laser oscillator and the amplifier; and a control device, modulating the transmittance of the optical switch element based on a pulse characteristic time characteristic of a time interval at which the Q-switched laser oscillator generates the pulse laser light. A laser device as claimed in claim 1, wherein the control device obtains the time interval between the plurality of pulse laser lights generated by the Q-switched laser oscillator or the moving average of the time interval between the plurality of pulse laser lights generated by the Q-switched laser oscillator as the pulse characteristic time, and controls the transmittance so that the transmittance increases as the time interval becomes shorter, and the energy variation between the plurality of pulse laser lights output by the amplifier becomes smaller than a predetermined value. The laser device of claim 1 further includes a learning device, wherein the learning device has a learning data acquisition unit for acquiring learning data; the learning data includes a state quantity and a transmittance of the optical switch element; the state quantity includes interval information for indicating the time interval of the pulse laser light generated by the laser device and the pulse energy of the pulse laser light after amplification; and a model generation unit for generating a learning model based on the state quantity using the learning data; the learning model is used to infer the transmittance of the optical switch element with the amplified pulse energy being the target value. The laser device of claim 1 further includes an inference device for inferring the transmittance of the optical switch element, wherein the inference device has an inference data acquisition unit for acquiring a state quantity; the state quantity includes interval information indicating the time interval for generating the pulse laser light and a target value of the pulse energy of the pulse laser light after amplification; and the inference unit uses a learned model to infer the transmittance based on the state quantity obtained by the inference data acquisition unit; the learned model infers the transmittance of the optical switch element based on the state quantity with the amplified pulse quantity as the target value; the control device uses the transmittance inferred by the inference device to control the optical switch element. The laser device of claim 3 further includes an inference device for inferring the transmittance of the optical switch element, wherein the inference device has an inference data acquisition unit for acquiring a state quantity; the state quantity includes interval information indicating the time interval for generating the pulse laser light and a target value of the pulse energy of the pulse laser light after amplification; and the inference unit uses a learned model to infer the transmittance based on the state quantity obtained by the inference data acquisition unit; the learned model infers the transmittance of the optical switch element based on the state quantity with the amplified pulse quantity as the target value; the control device uses the transmittance inferred by the inference device to control the optical switch element. A laser device as claimed in any one of claims 3 to 5, wherein the interval information includes at least one of the pulse characteristic time of the pulse laser light generated by the Q-switched laser oscillator, the energy of the pulse laser light generated by the Q-switched laser oscillator, the waveform of the pulse laser light generated by the Q-switched laser oscillator, the waveform of the pulse laser light amplified by the amplifier, and the driving current or discharge power of the Q-switched laser oscillator and the amplifier. A laser device as claimed in any one of claims 3 to 5, wherein the laser device is a gas laser whose laser medium is gas; the state quantity includes at least one of the cooling water temperature of the Q-switched laser oscillator or the amplifier, the gas temperature of the Q-switched laser oscillator or the amplifier, the gas pressure of the Q-switched laser oscillator or the amplifier, the continuous discharge time of the Q-switched laser oscillator or the amplifier, the electrode temperature of the Q-switched laser oscillator or the amplifier, and the total discharge time after the optical component of the Q-switched laser oscillator or the amplifier is replaced. A laser device as claimed in any one of claims 1 to 5, wherein the aforementioned pulse characteristic time is less than or equal to the value obtained by dividing the electrode width by the gas flow rate; the aforementioned electrode width is the length of the discharge electrode of the aforementioned Q-switch laser oscillator in the direction of gas flow. A laser device as claimed in any one of claims 1 to 5, wherein the control device changes the transmittance over time during the period in which one of the pulsed laser light passes through the switch element. As in claim 3, the laser device, wherein the model generation device generates a learned model based on the state quantity; the learned model is used to infer the transmittance that varies with time during the period when one of the pulsed laser light passes through the switch element. A laser processing device, comprising: a laser device, having a Q-switched laser oscillator, generating a pulse laser light; an amplifier, amplifying the pulse laser light; an optical switch element, disposed on an optical path between the Q-switched laser oscillator and the amplifier; and a control device, modulating the transmittance of the optical switch element based on a pulse characteristic time characteristic of the time interval at which the Q-switched laser oscillator generates the pulse laser light; a deflection element, deflecting the pulse laser light outputted from the amplifier; and an objective optical system, collecting or transcribing the pulse laser light from the deflection element, and irradiating it onto a processing object; the control device, The transmittance is controlled according to the processing hole characteristic value, and the transmittance is controlled so that the transmittance becomes higher as the interval of the processing holes becomes shorter, and the value representing the deviation of the shapes of the plurality of processing holes is less than a predetermined value; The processing hole characteristic value represents the processing hole characteristic included in the processing path when the processing object is subjected to the hole opening processing by the pulse laser light. A laser processing device as claimed in claim 11, wherein the processing hole characteristic value is a value representing the interval between a plurality of processing holes on the processing path, or a moving average of the interval between a plurality of processing holes. A laser processing device as claimed in claim 11 or 12, wherein the aforementioned Q-switch laser oscillator has a recursive reflector as a three-total reflector constituting a resonator. A learning device for learning the transmittance of the optical switch element of the laser device of claim 1, comprising: a learning data acquisition unit for acquiring learning data; the learning data includes a state quantity and the transmittance of the optical switch element; the state quantity includes interval information for indicating the time interval of the pulse laser light generated by the laser device, and the pulse energy of the amplified pulse laser light; and a model generation unit for generating a learning model based on the state quantity using the learning data; the learning model is used to infer the transmittance of the optical switch element with the pulse energy being the target value. A learning device for learning the transmittance of the optical switch element of the laser processing device of claim 11, comprising: a learning data acquisition unit for acquiring learning data; the learning data includes a state quantity and the transmittance of the optical switch element; the state quantity includes interval information for indicating the time interval of the pulse laser light generated by the laser device, and shape information for indicating the shape of the processed hole processed by the pulse laser light; and a model generation unit for generating a learning model based on the state quantity using the learning data; the learning model is used to infer the transmittance of the optical switch element with the shape of the processed hole as the target shape. A learning device as claimed in claim 14 or 15, wherein the interval information includes at least one of the pulse characteristic time of the pulse laser light generated by the Q-switched laser oscillator, the energy of the pulse laser light generated by the Q-switched laser oscillator, the waveform of the pulse laser light generated by the Q-switched laser oscillator, the waveform of the pulse laser light amplified by the amplifier, and the driving current or discharge power of the Q-switched laser oscillator and the amplifier. A learning device as claimed in claim 14 or 15, wherein the laser device is a gas laser whose laser medium is gas; the state quantity includes at least one of the cooling water temperature of the Q-switched laser oscillator or the amplifier, the gas temperature of the Q-switched laser oscillator or the amplifier, the gas pressure of the Q-switched laser oscillator or the amplifier, the continuous discharge time of the Q-switched laser oscillator or the amplifier, the electrode temperature of the Q-switched laser oscillator or the amplifier, and the total discharge time after the optical component of the Q-switched laser oscillator or the amplifier is replaced. An inference device for inferring the transmittance of the optical switch element of the laser device of claim 1 or 2, comprising: an inference data acquisition unit, which acquires interval information including a time interval for generating the pulse laser light and a target value of the pulse energy of the amplified pulse laser light; and an inference unit, which uses a learned model to infer the transmittance based on the state quantity and the target value obtained by the inference data acquisition unit; the pre-learned model infers the transmittance of the optical switch element based on the state quantity and the target value. An inference device for inferring the transmittance of the optical switch element of the laser device of the laser processing device of claim 11 or 12, comprising: an inference data acquisition unit, which acquires a state quantity including interval information for indicating the time interval of generating the pulse laser light, and shape information for indicating the target shape of a processing hole processed using the pulse laser light; and an inference unit, which uses a learned model to infer the transmittance based on the state quantity and the target shape obtained by the inference data acquisition unit; the pre-learned model infers the transmittance of the optical switch element based on the state quantity and the shape information. A laser processing system includes: A Q-switch laser oscillator that generates pulsed laser light; An amplifier that amplifies the pulsed laser light; An optical switch element that is disposed on an optical path between the Q-switch laser oscillator and the amplifier; A deflection element that deflects the pulsed laser light output by the amplifier; and An objective optical system that collects or transcribes the pulsed laser light from the deflection element and irradiates the object to be processed; and A control device that controls the transmittance of the optical switch element according to a processing hole characteristic value, and controls the transmittance so that the transmittance increases as the interval between the processing holes becomes shorter, and the value representing the deviation of the shapes of the plurality of processing holes is less than a predetermined value, thereby modulating the transmittance; The aforementioned processing hole characteristic value represents the processing hole characteristic included in the processing path when the aforementioned pulse laser light is used to perform hole opening processing on the aforementioned processing object. The laser processing system of claim 20 further includes a learning device, wherein the learning device includes, a learning data acquisition unit, which acquires learning data; the learning data includes a state quantity and shape information for indicating the shape of a processing hole processed using the pulsed laser light; the state quantity includes interval information for indicating the time interval at which the laser device generates the pulsed laser light; and a model generation unit, which uses the learning data to generate a learning model based on the state quantity; the learning model is used to infer the transmittance of the optical switch element whose shape of the processing hole is a target shape. The laser processing system of claim 20 or 21 further includes an inference device, wherein the pre-inference device includes, an inference data acquisition unit, which acquires a state quantity including interval information for indicating the time interval for generating the aforementioned pulse laser light, and shape information for indicating the target shape of the processing hole processed using the aforementioned pulse laser light; and an inference unit, which uses a learned model to infer the aforementioned transmittance based on the aforementioned state quantity and the aforementioned target shape obtained by the aforementioned inference data acquisition unit; the pre-learned model infers the transmittance of the aforementioned optical switch element based on the aforementioned state quantity and the aforementioned shape information. A laser processing method, comprising: A step of generating pulsed laser light; A step of setting the transmittance of the pulsed laser light through an optical switch element; A step of changing the pulse energy of the pulsed laser light by causing the pulsed laser light to enter the switch element with the set transmittance; A step of igniting the pulsed laser light; A step of adjusting the position of the pulsed laser light irradiating the processing object by deflecting the amplified pulsed laser light; and A step of collecting or transcribing the deflected pulsed laser light to irradiate the processing object; The step of changing the pulse energy of the pulse laser light is to adjust the transmittance of the optical switch element based on the pulse characteristic time that represents the time interval of generating the pulse laser light.

Images