CONTROL DEVICE, LASER PROCESSING SYSTEM, AND LASER PROCESSING METHOD

Abstract

The control device (3) includes a learned model holding unit (26) that holds a plurality of learned models that are the result of learning the relationship between features indicating the state of machining by the machining machine and the state of machining by the machining machine; a machining state evaluation unit (23) that evaluates the state of machining by the machining machine by inputting features to each of the plurality of learned models or by inputting features to a combination of the plurality of learned models; a monitoring model determination unit (24) that determines a learned model or a combination of a plurality of learned models, which is a monitoring model to be used for monitoring the state of machining by the machining machine, based on the result of evaluating the state of machining by observing a workpiece machined by the machining machine and the evaluation result by the machining state evaluation unit (23); and a machining control unit (28) that controls the machining machine based on the result of monitoring the state of machining by the machining machine using the monitoring model.

Description

[Technical field]

[0001] The present disclosure relates to a control device for controlling a laser processing machine, a laser processing system, and a laser processing method. [Background technology]

[0002] The control device that controls the laser processing machine reads out preset processing conditions and controls the laser processing machine according to the read out processing conditions. The control device has preset multiple processing conditions associated with the material of the workpiece, the thickness of the workpiece, the type of processing gas, etc., and selects processing conditions suitable for processing from the multiple processing conditions.

[0003] The processing conditions that enable good processing by the laser processing machine may vary from the preset processing conditions due to individual differences in the workpieces depending on the workpiece manufacturer or individual differences in the workpieces depending on the workpiece lot. Therefore, the control device may monitor the state of processing by the laser processing machine and adjust the processing parameters used for processing based on the monitoring results. The laser processing machine can perform good processing by appropriately adjusting the processing parameters.

[0004] Patent Document 1 discloses a numerical control device that detects light emitted from near the machining point of a workpiece and evaluates the quality of the machining state of the workpiece based on the detected light intensity distribution. The numerical control device disclosed in Patent Document 1 generates a trained model by learning the relationship between the detected light intensity distribution and the quality of the machining state, and evaluates the quality of the machining state of the workpiece using the trained model. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2022-118774 A Summary of the Invention [Problem to be solved by the invention]

[0006] The state of light emitted from the vicinity of the processing point of the workpiece when the processing state is good may change due to the state of the laser processing machine or the variation in the components contained in the workpiece. While the state of light may change in this way, in the case of the conventional technology disclosed in Patent Document 1, a pre-generated trained model is used to evaluate the quality of the processing state, so that the quality of the processing state may not be accurately evaluated. According to the conventional technology disclosed in Patent Document 1, there is a problem that the quality of the processing state may not be accurately evaluated, and therefore good processing by the laser processing machine may not be able to be continued.

[0007] The present disclosure has been made in consideration of the above, and has an object to provide a control device that can continue to perform good processing using a laser processing machine. [Means for solving the problem]

[0008] In order to solve the above-mentioned problems and achieve the object, the control device disclosed herein comprises a learned model holding unit that holds multiple learned models that are the result of learning the relationship between features indicating the state of machining by the machining machine and the state of machining by the machining machine; a machining state evaluation unit that evaluates the state of machining by the machining machine by inputting features to each of the multiple learned models or by inputting features to a combination of multiple learned models; a monitoring model determination unit that determines a learned model or a combination of multiple learned models, which is a monitoring model to be used for monitoring the state of machining by the machining machine, based on the result of evaluating the state of machining by observing a workpiece machined by the machining machine and the evaluation result by the machining state evaluation unit; and a machining control unit that controls the machining machine based on the result of monitoring the state of machining by the machining machine using the monitoring model. Effect of the Invention

[0009] The control device according to the present disclosure has an effect of enabling good processing to be continued by the laser processing machine. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing a configuration example of a laser processing system according to a first embodiment. [Diagram 2] FIG. 1 is a diagram showing a configuration example of a control device included in a laser processing system according to a first embodiment; [Diagram 3] FIG. 1 is a diagram showing an example of a configuration of a trained model in the first embodiment. [Figure 4] 1 is a flowchart showing a first example of a procedure of a process executed by the laser processing system according to the first embodiment; [Diagram 5] 11 is a flowchart showing a second example of a procedure of a process executed by the laser processing system according to the first embodiment. [Figure 6] 11 is a flowchart showing a third example of a procedure of a process executed by the laser processing system according to the first embodiment. [Figure 7] FIG. 13 is a diagram illustrating a configuration example of a monitoring model determination unit included in the control device according to the second embodiment. [Figure 8] FIG. 13 is a diagram showing a configuration example of a control device included in a laser processing system according to a third embodiment. [Figure 9] FIG. 13 is a diagram illustrating a configuration example of a monitoring model determination unit included in the control device according to the third embodiment. [Figure 10] FIG. 1 shows an example of the configuration of a control circuit according to first to third embodiments. [Figure 11] FIG. 1 is a diagram showing an example of a configuration of a dedicated hardware circuit according to the first to third embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A control device, a laser processing system, and a laser processing method according to embodiments will be described in detail below with reference to the drawings.

[0012] Embodiment 1 FIG. 1 is a diagram showing a configuration example of a laser processing system 1 according to a first embodiment. The laser processing system 1 includes a laser processing machine 2 and a control device 3. The laser processing machine 2 is a processing machine that processes a workpiece 10 by laser light. The laser processing machine 2 processes the workpiece 10 by locally melting the workpiece 10 by irradiating the workpiece with laser light. The laser processing machine 2 performs laser processing such as cutting, welding, lamination processing, or heat treatment. In the example shown in FIG. 1, the workpiece 10 is a metal plate. Also, the laser processing machine 2 performs laser processing to cut the workpiece 10. The control device 3 controls the laser processing machine 2.

[0013] The laser processing machine 2 includes a laser oscillator 4 which is a light source that outputs laser light, a processing head 6, a cable 5 which is a transmission path of the laser light from the laser oscillator 4 to the processing head 6, and a table on which a workpiece 10 is placed. The table is not shown. The cable 5 includes, for example, an optical fiber. The emission end of the cable 5 is placed inside the processing head 6.

[0014] FIG. 1 shows a schematic diagram of the internal configuration of the processing head 6. Inside the processing head 6, a collimating optical system 11, an imaging optical system 12, and a protective glass 13 are provided. The collimating optical system 11 receives laser light diverging from the output end of the cable 5. The collimating optical system 11 emits parallel light. The imaging optical system 12 forms an image of the output end of the cable 5. The protective glass 13 is a flat glass plate. The protective glass 13 is an optical component for protecting the internal configuration of the processing head 6. Note that FIG. 1 shows the collimating optical system 11, which is a single lens, and the imaging optical system 12, which is a single lens. At least one of the collimating optical system 11 and the imaging optical system 12 may be composed of a plurality of lenses.

[0015] The processing head 6 has a processing nozzle 7 through which the laser light to be irradiated onto the workpiece 10 and the processing gas to be sprayed onto the workpiece 10 pass. The laser light that passes through the collimating optical system 11, the imaging optical system 12, and the protective glass 13 passes through the processing nozzle 7. The processing gas is supplied from a gas supply source outside the processing head 6 to the inside of the processing head 6. The laser processing machine 2 sprays the processing gas from the processing head 6 through the processing nozzle 7 onto the workpiece 10. An illustration of the gas supply source is omitted.

[0016] The laser processing machine 2 includes a processing head drive unit 14 that drives the processing head 6. The laser processing machine 2 moves the laser light and the workpiece 10 relatively by moving the processing head 6 relative to the table using the processing head drive unit 14. The laser processing machine 2 moves the laser light and the workpiece 10 relatively by controlling the incident position of the laser light on the workpiece 10. Note that the laser processing machine 2 may move the laser light and the workpiece 10 relatively by moving the table relative to the processing head 6 without moving the processing head 6.

[0017] Inside the processing head 6, a moving mechanism 15 for moving the collimating optical system 11 and a moving mechanism 16 for moving the imaging optical system 12 are provided. The moving mechanism 16 changes the imaging position by moving the imaging optical system 12 in the optical axis direction. The laser processing machine 2 changes the positional relationship between the imaging position and the workpiece 10 without changing the positional relationship between the processing nozzle 7 and the workpiece 10 by changing the imaging position. The moving mechanism 15 adjusts the position of the collimating optical system 11 by moving the collimating optical system 11 in the optical axis direction in accordance with the movement of the imaging optical system 12. Alternatively, the moving mechanism 15 may adjust the divergence angle of the laser light incident on the imaging optical system 12 by moving the collimating optical system 11 in the optical axis direction. Note that the optical axis direction in the above description refers to the direction of the optical axis of each of the collimating optical system 11 and the imaging optical system 12. The optical axis of the collimating optical system 11 and the optical axis of the imaging optical system 12 coincide with each other. These optical axes coincide with the central axis of the laser light incident on the workpiece 10.

[0018] An optical system other than the collimating optical system 11 and the imaging optical system 12 may be provided inside the processing head 6. A zoom optical system that changes the size of the image by moving in the optical axis direction may be provided inside the processing head 6. The laser processing machine 2 moves at least one of the multiple optical systems provided inside the processing head 6 in the optical axis direction. The imaging position may be on the beam waist or may be shifted from the beam waist. A focusing optical system may be provided inside the processing head 6 instead of the imaging optical system 12. In the following description, the optical systems provided inside the processing head 6 may be collectively referred to as the processing optical system.

[0019] An example of the laser oscillator 4 is a fiber laser oscillator. The laser oscillator 4 may be a direct diode laser, a carbon dioxide gas laser, a copper vapor laser, or various ion lasers. Alternatively, the laser oscillator 4 may be a solid-state laser having an excitation medium such as YAG (Yttrium Aluminum Garnet) crystal. The laser processing machine 2 may include a wavelength conversion unit that performs wavelength conversion of the laser light output from the laser oscillator 4.

[0020] The control device 3 controls each of the laser oscillator 4, the processing head drive unit 14, the moving mechanism 15, and the moving mechanism 16 by outputting control signals to each of the laser oscillator 4, the processing head drive unit 14, the moving mechanism 15, and the moving mechanism 16.

[0021] The laser processing system 1 has a sound sensor 8 and an optical sensor 9. Each of the sound sensor 8 and the optical sensor 9 is a sensor that detects the state of processing by the laser processing machine 2. The sound sensor 8 detects sound generated near the processing point, and outputs a signal indicating the volume of the detected sound to the control device 3. The optical sensor 9 detects light emitted from near the processing point, and outputs a signal indicating the intensity of the detected light to the control device 3. The processing point is the position on the workpiece 10 where processing is being performed, and is the position where the laser light is incident.

[0022] In the following description, the data obtained by each of the sound sensor 8 and the light sensor 9 will be referred to as sensor data. The sensor data obtained by the sound sensor 8 is data indicating the volume of sound generated during processing. The sensor data obtained by the light sensor 9 is data indicating the intensity of light emitted during processing. Each of the sensor data obtained by the sound sensor 8 and the sensor data obtained by the light sensor 9 is data indicating the state of processing by the laser processing machine 2.

[0023] The laser processing system 1 may have sensors other than the sound sensor 8 and the light sensor 9. The laser processing system 1 may have, for example, a vibration sensor that detects vibrations generated near the processing point, or a camera that captures the vicinity of the processing point. The sensor data obtained by the vibration sensor is data indicating the magnitude of vibrations generated during processing. The control device 3 processes the image captured by the camera to obtain data indicating the dimensions of the processed portion, or data indicating the melting state of the workpiece 10. The sensor data obtained by the camera is data obtained by processing an image captured near the processing point. The laser processing system 1 has at least one of the sound sensor 8, the light sensor 9, the vibration sensor, and the camera. The laser processing system 1 may have two or more of the sound sensor 8, the light sensor 9, the vibration sensor, and the camera. The sensor of the laser processing system 1 may be a sensor that can detect the state of processing by the laser processing machine 2, and is not limited to the sensor described in the first embodiment.

[0024] 1, the optical sensor 9 is disposed outside the processing head 6. The optical sensor 9 may be disposed inside the processing head 6. Since the optical sensor 9 has a short response time, the laser processing system 1 can quickly detect the processing state by providing the optical sensor 9. The sound sensor 8 has the advantage that the installation position can be easily determined since it can detect sound over a wide range.

[0025] Next, the configuration of the control device 3 will be described. Fig. 2 is a diagram showing an example of the configuration of the control device 3 included in the laser processing system 1 according to the first embodiment. The control device 3 includes a processing monitor unit 20 that monitors the state of processing by the laser processing machine 2, and a processing control unit 28 that controls the laser processing machine 2.

[0026] When performing continuous machining, the laser machining machine 2 may cause machining defects due to heat accumulation in the components constituting the machining head 6 or heat accumulation in the workpiece 10. The machining monitoring unit 20 evaluates the machining state of the laser machining machine 2 to monitor machining defects.

[0027] In the laser processing system 1 shown in FIG. 1, the control device 3 is an external device of the laser processing machine 2. The laser processing machine 2 and the control device 3 are connected to each other so that they can communicate with each other, and information is transmitted and received between the laser processing machine 2 and the control device 3. The control device 3 may be a device included in the laser processing machine 2. Alternatively, the processing control unit 28 of the control device 3 may be realized by a device included in the laser processing machine 2, and the processing monitoring unit 20 of the control device 3 may be realized by a device external to the laser processing machine 2. When the processing monitoring unit 20 is realized by a device external to the laser processing machine 2, the device that realizes the processing monitoring unit 20 and the device that realizes the processing control unit 28 are connected to each other so that they can communicate with each other.

[0028] The machining monitoring unit 20 includes a data acquisition unit 21, a feature calculation unit 22, a machining state evaluation unit 23, a monitoring model determination unit 24, a machining parameter correction unit 25, a learned model holding unit 26, and a monitoring model holding unit 27.

[0029] The data acquisition unit 21 acquires sensor data output from each of the sound sensor 8 and the light sensor 9. That is, the data acquisition unit 21 acquires sensor data that is the detection result of light generated during processing by the laser processing machine 2 and sensor data that is the detection result of sound generated during processing by the laser processing machine 2. By continuously inputting sensor data from each of the sound sensor 8 and the light sensor 9, the data acquisition unit 21 acquires sensor data that is time-series data. The data acquisition unit 21 outputs the acquired sensor data to the feature calculation unit 22.

[0030] The feature amount calculation unit 22 calculates the feature amount from the sensor data acquired by the data acquisition unit 21. In the first embodiment, the feature amount indicates the state of processing by the laser processing machine 2. The feature amount calculation unit 22 outputs the calculated feature amount to the processing state evaluation unit 23.

[0031] The feature amount is a value obtained by processing the sensor data, and is, for example, a statistical amount such as an average value or a standard deviation. The feature amount may be a value obtained by processing such as frequency analysis, a filter bank, or a wavelet transform. The feature amount calculation unit 22 may calculate a combination of multiple values ​​as the feature amount.

[0032] The feature amount calculated by the feature amount calculation unit 22 is not limited to the one exemplified here. The feature amount may be a feature amount that can be calculated by any method that is a general analysis method for time series data. The feature amount calculation unit 22 may output the value of the sensor data as it is as the feature amount. The feature amount calculation unit 22 may store the position in the feature space of a feature vector indicating the feature amount at the start of processing, and output the amount of change in the position of the feature vector after the start of processing as the feature amount. When the data acquisition unit 21 acquires sensor data from multiple sensors, the feature amount calculation unit 22 may calculate a feature amount reflecting the output of each sensor, or may calculate a feature amount reflecting a combination of the outputs of multiple sensors.

[0033] The trained model holding unit 26 holds a plurality of trained models. Each of the trained models is a result of learning the relationship between the feature amount indicating the state of machining by the laser machining machine 2 and the state of machining by the laser machining machine 2. Each of the trained models held in the trained model holding unit 26 is a result of learning the relationship between the feature amount and the state of machining in each of a plurality of cases in which at least one of the thickness of the plate-shaped workpiece 10 and the type of machining gas used during machining by the laser machining machine 2 is different from each other. As a result, the control device 3 holds trained models for a plurality of cases in which the thickness of the workpiece 10 or the type of machining gas is different from each other in the trained model holding unit 26.

[0034] The machining state evaluation unit 23 reads out a plurality of trained models from the trained model storage unit 26. The machining state evaluation unit 23 inputs the feature amount calculated by the feature amount calculation unit 22 into each of the plurality of trained models to evaluate the state of machining by the laser processing machine 2. The machining state evaluation unit 23 evaluates the state of machining by determining whether the state of machining is good or bad. In other words, the machining state evaluation unit 23 determines whether the state of machining by the laser processing machine 2 is good or bad. The machining state evaluation unit 23 outputs the evaluation result of the state of machining to the monitoring model determination unit 24.

[0035] The monitoring model determination unit 24 receives an evaluation result by the machining state evaluation unit 23. In addition to the evaluation result by the machining state evaluation unit 23, an evaluation result by observing the workpiece 10 machined by the laser processing machine 2 is input to the monitoring model determination unit 24. The monitoring model determination unit 24 determines a trained model, which is a monitoring model, based on the result of evaluating the machining state by observing the workpiece 10 machined by the laser processing machine 2 and the evaluation result by the machining state evaluation unit 23. In the example described here, the user evaluates the machining state by observing the workpiece 10 machined by the laser processing machine 2. The monitoring model determination unit 24 determines, as a monitoring model, a trained model in which the evaluation result output by the machining state evaluation unit 23 matches the evaluation result by the user, among the multiple trained models held in the trained model holding unit 26.

[0036] The trained model determined as the monitoring model may be selected by the user or may be selected by the monitoring model determination unit 24. When the trained model determined as the monitoring model is selected by the user, the user checks the evaluation results output by the processing state evaluation unit 23 for each of the multiple trained models, and selects one trained model whose evaluation result output by the processing state evaluation unit 23 matches the evaluation result by the user. Alternatively, the user selects one trained model whose evaluation result output by the processing state evaluation unit 23 is closest to the evaluation result by the user.

[0037] The user inputs information indicating the selected trained model to the monitoring model determination unit 24, for example, by operating an input device provided in the control device 3. In FIG. 2, the input device is not shown. The monitoring model determination unit 24 determines the trained model selected by the user as the monitoring model. The laser processing system 1 may be provided with a display device that displays the evaluation result by the processing state evaluation unit 23. The user can select the trained model by checking the evaluation result displayed on the display device. In FIG. 2, the display device is not shown.

[0038] When the trained model to be determined as the monitoring model is selected by the monitoring model determination unit 24, the user inputs the result of evaluating the state of processing by the laser processing machine 2 to the monitoring model determination unit 24. The user inputs the result of evaluating the state of processing to the monitoring model determination unit 24, for example, by operating an input device provided in the control device 3. The monitoring model determination unit 24 selects one trained model whose evaluation result output by the processing state evaluation unit 23 matches the evaluation result by the user from among the multiple trained models. Alternatively, the monitoring model determination unit 24 selects one trained model whose evaluation result output by the processing state evaluation unit 23 is closest to the evaluation result by the user from among the multiple trained models. The monitoring model determination unit 24 determines the trained model selected by the monitoring model determination unit 24 as the monitoring model.

[0039] The monitoring model determination unit 24 stores the determined monitoring model in the monitoring model storage unit 27. The monitoring model storage unit 27 stores the monitoring model. The control device 3 determines the monitoring model by comparing the evaluation result by the user with the evaluation result output by the machining state evaluation unit 23, thereby obtaining a monitoring model that can perform an evaluation similar to that by the user.

[0040] The processing monitoring unit 20 monitors the state of processing by the laser processing machine 2 using the monitoring model determined by the monitoring model determination unit 24. When the processing monitoring unit 20 monitors the state of processing by the laser processing machine 2, the data acquisition unit 21 acquires sensor data output from each of the sound sensor 8 and the light sensor 9. The feature calculation unit 22 calculates feature amounts from the sensor data acquired by the data acquisition unit 21.

[0041] When the processing monitoring unit 20 monitors the state of processing by the laser processing machine 2, the processing state evaluation unit 23 reads out the monitoring model from the monitoring model storage unit 27. The processing state evaluation unit 23 inputs the feature amount calculated by the feature amount calculation unit 22 to the monitoring model, thereby evaluating the state of processing by the laser processing machine 2. The processing state evaluation unit 23 outputs the evaluation result of the processing state to the processing parameter correction unit 25.

[0042] The processing parameter correction unit 25 calculates the correction amount of the processing parameter based on the evaluation result by the processing state evaluation unit 23. In the first embodiment, the processing parameter is a parameter used for processing by the laser processing machine 2. The processing parameter correction unit 25 outputs information on the calculated correction amount to the processing control unit 28.

[0043] The machining control unit 28 adjusts the machining parameters according to the machining conditions. When the machining control unit 28 acquires the information on the correction amount, the machining control unit 28 corrects the machining parameters based on the correction amount. The machining control unit 28 controls the laser oscillator 4, the machining head driving unit 14, the moving mechanism 15, and the moving mechanism 16 according to the corrected machining parameters.

[0044] In the above description, the processing state evaluation unit 23 judges whether the processing state by the laser processing machine 2 corresponds to good or bad. In this case, the processing state evaluation unit 23 outputs information indicating whether the processing state corresponds to good or bad. Whether the processing state corresponds to good or bad is expressed by two values. The processing state evaluation unit 23 outputs a value indicating the judgment result. Note that the evaluation by the processing state evaluation unit 23 is not limited to judging whether the processing state corresponds to good or bad. The processing state evaluation unit 23 may evaluate the processing state by calculating an evaluation value indicating the degree to which the processing state is good. An example of the evaluation value is a value in the range from 0% to 100%. For example, the evaluation value is a higher value as the processing state is better. The evaluation value may be a value in the range from 0 to 10. The processing state evaluation unit 23 outputs the calculated evaluation value.

[0045] The processing state evaluation unit 23 may set a plurality of items regarding the processing state and output the evaluation result for each item. For example, when determining whether the processing state corresponds to good or bad, the processing state evaluation unit 23 determines whether each item corresponds to good or bad. The processing state evaluation unit 23 outputs the judgment result for each item.

[0046] Examples of items related to the processing condition include the occurrence of dross and roughness of the cut surface. Dross is molten material that adheres to the workpiece 10 when the workpiece 10 is cut. Dross adheres to the lower end of the cut surface of the workpiece 10. The less dross there is, the better the processing condition can be said to be. Roughness of the cut surface is periodic unevenness that occurs on the upper part of the cut surface. When roughness occurs on the cut surface, the depth of the scratches that occur on the cut surface becomes deeper compared to when roughness does not occur. The less roughness of the cut surface is, the better the processing condition can be said to be.

[0047] When the processing gas used in laser processing is oxygen, an oxide film is generated on the cut surface. When the processing gas used in laser processing is oxygen, the peeling of the oxide film generated on the cut surface may be included in the items regarding the processing condition. The less the peeling of the oxide film generated on the cut surface, the better the processing condition can be said to be.

[0048] The items regarding the machining state are not limited to the above items. The items regarding the machining state may include items such as discoloration of the workpiece 10 or the presence or absence of vibration on the surface of the workpiece 10. The machining state evaluation unit 23 can output an evaluation result regarding any one or more items among the multiple items described in the first embodiment. The machining state evaluation unit 23 may change the items to be evaluated depending on the machining conditions, etc. For example, the machining state evaluation unit 23 may change the items to be evaluated depending on the combination of the laser output, the machining speed, and the thickness of the workpiece 10. The thickness of the workpiece 10 is defined as the thickness in the direction of the central axis of the laser light incident on the workpiece 10.

[0049] Alternatively, the processing state evaluation unit 23 may change the items to be evaluated depending on the type of processing gas. For example, when the processing gas used in the laser processing is oxygen, the processing state evaluation unit 23 includes peeling of an oxide film in the items to be evaluated. On the other hand, when the processing gas used in the laser processing is nitrogen, since no oxide film is generated on the cut surface, the processing state evaluation unit 23 excludes peeling of an oxide film from the items to be evaluated. For the laser processing machine 2 that performs welding processing, the items regarding the processing state may include items such as the occurrence of spatter. For the laser processing machine 2 that performs layered processing, the items regarding the processing state may include items such as the height of the molded object or excessive melting of the molded object.

[0050] The processing state evaluation unit 23 may determine the processing state for two or more items and output a result of summarizing the determination for each item as a processing state determination result. For example, when the processing state is determined to be poor for a preset number of items or more, the processing state evaluation unit 23 may output a determination result indicating that the processing state is poor. Alternatively, when the processing state is determined to be poor, the processing state evaluation unit 23 may analyze the pass / fail of each item.

[0051] As described above, the laser processing system 1 may have a display device that displays the evaluation result by the processing state evaluation unit 23. The display device may be provided in the laser processing machine 2, or may be provided in a device external to the laser processing machine 2. For example, the display device displays information indicating whether the processing state is good or bad. The display device may display the evaluation result by the processing state evaluation unit 23 only when the processing state evaluation unit 23 judges that the processing state is poor. In this case, the display device displays information indicating that the processing state is poor, but does not display information indicating that the processing state is good.

[0052] In the above description, the processing state evaluation unit 23 evaluates the processing state of the laser processing machine 2 by inputting the feature amount calculated by the feature amount calculation unit 22 into each of the multiple trained models. The processing state evaluation unit 23 may evaluate the processing state of the laser processing machine 2 by inputting information other than the feature amount calculated by the feature amount calculation unit 22 into each of the multiple trained models. Examples of information input to each of the multiple trained models include the value of the processing parameter at the time of evaluation, the temperature of the processing optical system at the time of evaluation, the temperature change of the processing optical system, the thickness of the workpiece 10, or the material of the workpiece 10. In this case, each of the multiple trained models is the result of learning the relationship between such information, the feature amount, and the processing state. In addition, the processing state evaluation unit 23 evaluates the processing state of the laser processing machine 2 by inputting such information and the feature amount into the monitoring model.

[0053] The processing parameter correction unit 25 calculates the correction amount of the processing parameter based on the evaluation result by the processing state evaluation unit 23. For example, when a judgment result indicating that the processing state of the laser processing machine 2 is poor is input to the processing parameter correction unit 25, the processing parameter correction unit 25 calculates the correction amount of the processing parameter. The processing control unit 28 corrects the processing parameter using the correction amount calculated by the processing parameter correction unit 25. The processing monitoring unit 20 repeats the calculation of the correction amount by the processing parameter correction unit 25 and the correction of the processing parameter by the processing control unit 28 until a judgment result indicating that the processing state is good is obtained.

[0054] The processing parameter correction unit 25 may calculate the correction amount based on the value of the processing parameter set in the processing control unit 28 and the judgment result by the processing state evaluation unit 23. In this case, the processing parameter correction unit 25 acquires the value of the processing parameter set in the processing control unit 28 from the processing control unit 28.

[0055] The machining parameter correcting unit 25 may determine the presence or absence of a sign of a machining defect based on the evaluation result by the machining state evaluating unit 23, and may calculate a correction amount when a sign of a machining defect is confirmed. The machining monitoring unit 20 may repeat the calculation of the correction amount by the machining parameter correcting unit 25 and the correction of the machining parameters by the machining control unit 28 until the sign of a machining defect is no longer confirmed.

[0056] The processing parameters corrected by the correction amount calculated by the processing parameter correction unit 25 are, for example, parameters indicating the laser output, the beam quality of the laser light, the pressure of the processing gas, the processing speed, the focal length of the imaging optical system 12, the diameter of the image formed by the imaging optical system 12, the pulse frequency of the laser oscillator 4, the duty ratio of the pulse of the laser oscillator 4, the magnification of the imaging optical system 12, the nozzle diameter, the distance between the workpiece 10 and the processing nozzle 7, or the type of mode of the laser light. The nozzle diameter is the diameter of the hole in the processing nozzle 7 through which the laser light passes. The processing parameter corrected by the correction amount calculated by the processing parameter correction unit 25 may be a parameter indicating the positional relationship between the center position of the hole in the processing nozzle 7 through which the laser light passes and the central axis of the laser light.

[0057] When the judgment results for each of a plurality of items are input to the processing parameter correction unit 25, the processing parameter correction unit 25 may determine the processing parameters to be corrected and the correction amounts of the processing parameters to be corrected based on a combination of the judgment results for each item.

[0058] Here, the information indicating that the machining state is good is "1", and the information indicating that the machining state is bad is "0". The machining state evaluation unit 23 outputs the judgment results for each of the items of dross generation, roughness of the cut surface, and peeling of the oxide film generated on the cut surface. When "1" for dross generation, "0" for roughness of the cut surface, and "0" for peeling of the oxide film generated on the cut surface are input to the machining parameter correction unit 25, the machining parameter correction unit 25 determines the machining parameters to be corrected and the correction amount based on the combination of the input values ​​"1, 0, 0". Based on such a combination, the machining parameter correction unit 25 determines, for example, each parameter of the laser output and the pressure of the machining gas as the machining parameters to be corrected. Based on such a combination, the machining parameter correction unit 25 also determines the correction amount for each of the laser output and the pressure of the machining gas. For example, the machining parameter correction unit 25 calculates the correction amount for increasing the laser output and the correction amount for decreasing the pressure of the machining gas.

[0059] The relationship between the processing parameters to be corrected and the correction amount for each combination of a plurality of combinations of the judgment results for each item is previously determined in the processing parameter correction unit 25. The processing parameter correction unit 25 can determine the processing parameters to be corrected and the correction amount based on the previously determined relationship and the combination of the judgment results for each item.

[0060] In the above description, the machining parameter correction unit 25 calculates the correction amount based on the evaluation result by the machining state evaluation unit 23, but the correction amount may be calculated based on the feature amount. The machining parameter correction unit 25 may calculate the correction amount by inputting the feature amount calculated by the feature amount calculation unit 22 to the monitoring model. Each of the multiple trained models held by the trained model holding unit 26 may be a trained model that outputs at least one of the evaluation result of the machining state and the correction amount by inputting the feature amount. When the trained model outputs the correction amount, the machining state is evaluated by the trained model, and then the correction amount is calculated.

[0061] In the above description, the processing monitoring unit 20 repeats the calculation of the correction amount by the processing parameter correction unit 25 and the correction of the processing parameters by the processing control unit 28 until a judgment result indicating that the processing state is good is obtained. If the laser processing system 1 continues to be unable to obtain a judgment result indicating that the processing state is good, it may stop laser processing by the laser processing machine 2.

[0062] As described above, the monitoring model determination unit 24 determines the monitoring model based on the result of the user's evaluation of the state of processing by the laser processing machine 2 and the evaluation result by the processing state evaluation unit 23. The user may, for example, visually check the cut surface of the workpiece 10 to determine whether the processing is good or bad. The result of the user's evaluation of the state of processing may be the result of measuring the roughness of the cut surface using a measuring device such as a three-dimensional measuring device or a roughness measuring device. The monitoring model determination unit 24 may accept information indicating the result of the user's evaluation of the state of processing. The user inputs the result of the evaluation of the state of processing to the monitoring model determination unit 24, for example, by operating an input device provided in the control device 3.

[0063] In the above description, the result of evaluation by the user is input to the monitoring model determination unit 24 as a result of evaluating the state of processing by observing the workpiece 10 processed by the laser processing machine 2. The evaluation result input to the monitoring model determination unit 24 is not limited to the evaluation result by the user. In other words, the result of evaluating the state of processing by observing the workpiece 10 is not limited to the evaluation result by the user. The result of judging the quality of processing from a judgment device that judges the state of processing by observing the workpiece 10 may be input to the monitoring model determination unit 24. The judgment device automatically judges the state of processing based on an image taken by a camera, for example. In this way, either the result of evaluation of the state of processing by the user or the result of evaluation of the state of processing by the judgment device may be input to the monitoring model determination unit 24. The monitoring model determination unit 24 can select one learned model among the multiple learned models whose evaluation result output by the processing state evaluation unit 23 is closest to the result of evaluating the state of processing by observing the workpiece 10. Note that the judgment device is not illustrated.

[0064] When the evaluation result by the user or the judgment device is input to the monitoring model determination unit 24, and the machining state evaluation unit 23 judges whether the machining state corresponds to good or bad, the result of the judgment by the user or the judgment device as to whether the machining state corresponds to good or bad is input to the monitoring model determination unit 24. When the judgment result of the machining state evaluation unit 23 matches the judgment result of the user or the judgment device, the monitoring model determination unit 24 determines the learned model that has output the judgment result of the machining state evaluation unit 23 as the monitoring model.

[0065] In the case where the evaluation result by the user or the judgment device is input to the monitoring model determination unit 24, and the processing state evaluation unit 23 outputs the evaluation result for each of the multiple items, the result of the user or the judgment device judging the processing state for each of the multiple items is input to the monitoring model determination unit 24. In the case where the judgment result for each item by the processing state evaluation unit 23 matches the judgment result for each item by the user or the judgment device, the monitoring model determination unit 24 decides that the learned model that output the judgment result of the processing state evaluation unit 23 is the monitoring model.

[0066] When the evaluation result by the user or the judgment device is input to the monitoring model determination unit 24, and the processing state evaluation unit 23 outputs an evaluation value for the processing state, the evaluation value determined by the user or the judgment device for the processing state is input to the monitoring model determination unit 24. When the evaluation value output by the processing state evaluation unit 23 is included in an allowable range that includes the evaluation value determined by the user or the judgment device, the monitoring model determination unit 24 determines the trained model that output the evaluation value of the processing state evaluation unit 23 as the monitoring model. The allowable range is a range based on the evaluation value determined by the user or the judgment device, and is a range of a preset numerical width.

[0067] In the above description, the monitoring model determination unit 24 determines, as a monitoring model, a learned model whose evaluation result output by the machining state evaluation unit 23 matches the result of evaluating the machining state by observing the workpiece 10, among the multiple learned models. When each of the multiple learned models is a learned model that outputs a correction amount, the monitoring model determination unit 24 may determine the learned model to be the monitoring model based on the correction amount output from each of the multiple learned models. For example, the monitoring model determination unit 24 compares the machining parameters corrected based on the correction amount output from the learned model with the machining parameters based on the machining conditions adjusted by the user. The monitoring model determination unit 24 may determine the learned model to be the monitoring model by such a comparison.

[0068] The trained model holding unit 26 holds, for example, a plurality of trained models generated by a manufacturer of the laser processing machine 2. The plurality of trained models held in the trained model holding unit 26 may include trained models generated by users in addition to trained models generated by manufacturers.

[0069] Here, an example of the trained model in the first embodiment will be described. The learning algorithm used by the laser processing system 1 can be a known algorithm such as supervised learning, unsupervised learning, or reinforcement learning. As an example, a case in which a neural network is applied will be described.

[0070] The trained model in the first embodiment is the result of learning the relationship between features and processing states by supervised learning. Here, supervised learning is a method of learning features in training data and inferring results from the input by providing a set of input and result data to a learning device. The training data includes an input and a label which is a result corresponding to the input. The features correspond to the input. The processing state is training data and corresponds to the label. The neural network is composed of an input layer consisting of a plurality of neurons, a hidden layer which is an intermediate layer consisting of a plurality of neurons, and an output layer consisting of a plurality of neurons. The intermediate layer may be one layer, or two or more layers.

[0071] FIG. 3 is a diagram showing a configuration example of a trained model in the first embodiment. FIG. 3 shows a configuration example of a neural network. The neural network shown in FIG. 3 is a three-layer neural network. The input layer includes neurons X1, X2, and X3. The intermediate layer includes neurons Y1 and Y2. The output layer includes neurons Z1, Z2, and Z3. The number of neurons in each layer is arbitrary. A plurality of values ​​input to the input layer are multiplied by weights W1, w12, w13, w14, w15, and w16, and input to the intermediate layer. A plurality of values ​​input to the intermediate layer are multiplied by weights W2, w21, w22, w23, w24, w25, and w26, and output from the output layer. The output result output from the output layer changes according to the values ​​of weights W1 and W2. The neural network is generated by inputting features to the input layer and adjusting weights W1 and W2 so that the result output from the output layer approaches the processed state.

[0072] Next, a description will be given of the procedure of the process executed by the laser processing system 1. Here, three examples of the procedure of the process when determining a monitoring model will be described.

[0073] 4 is a flowchart showing a first example of a processing procedure executed by the laser processing system 1 according to the first embodiment. In the first example, the laser processing system 1 determines a monitoring model by having the laser processing machine 2 perform trial processing. In the first example, the monitoring model determination unit 24 determines the monitoring model based on the result of the user's evaluation of the processing state and the evaluation result by the processing state evaluation unit 23.

[0074] In step S1, the laser processing machine 2 starts trial processing. When the trial processing starts, the sound sensor 8 and the optical sensor 9 each detect the processing state and output sensor data. In step S2, the data acquisition unit 21 acquires the sensor data output from the sound sensor 8 and the optical sensor 9. The data acquisition unit 21 outputs the acquired sensor data to the feature calculation unit 22.

[0075] In step S3, the feature amount calculation unit 22 calculates the feature amount based on the sensor data. The feature amount calculation unit 22 outputs the calculated feature amount to the machining state evaluation unit 23. In step S4, the machining state evaluation unit 23 evaluates the machining state based on the trained model. The machining state evaluation unit 23 inputs the feature amount to one of the multiple trained models held in the trained model holding unit 26 to obtain an evaluation result of the machining state. The machining state evaluation unit 23 outputs the obtained evaluation result to the monitoring model determination unit 24.

[0076] In step S5, the monitoring model determination unit 24 judges whether or not the evaluation result of the processing state in step S4 coincides with the evaluation result by the user. In the first example shown in Fig. 4, it is assumed that the evaluation result by the user is input to the monitoring model determination unit 24. The monitoring model determination unit 24 compares the evaluation result of the processing state in step S4 with the evaluation result by the user to judge whether or not an evaluation result that coincides with the evaluation result by the user has been obtained.

[0077] If the evaluation result of the machining state in step S4 matches the evaluation result by the user (step S5, Yes), the monitoring model determination unit 24 determines a monitoring model in step S6. The monitoring model determination unit 24 determines the trained model used in the evaluation of the machining state in step S4 as the monitoring model.

[0078] On the other hand, if the evaluation result of the processing state in step S4 does not match the evaluation result by the user (step S5, No), in step S7, the processing state evaluation unit 23 selects a trained model to be used for evaluating the processing state. The processing state evaluation unit 23 selects one of the trained models stored in the trained model storage unit 26 other than the trained model used for evaluating the processing state in step S4. After that, the laser processing system 1 returns the procedure to step S4 and performs the processes of steps S4 and S5 for the trained model selected in step S7.

[0079] By completing step S6, the laser processing system 1 ends the processing according to the procedure shown in Fig. 4. The laser processing system 1 monitors the processing state when the laser processing machine 2 processes a product by using a monitoring model. When the judgment device judges the processing state by observing the workpiece 10, the monitoring model determination unit 24 determines the monitoring model based on the result of evaluation of the processing state by the judgment device and the evaluation result by the processing state evaluation unit 23 in step S5.

[0080] The user may determine whether the evaluation result of the processing state by the processing state evaluation unit 23 matches the evaluation result by the user. The user checks the evaluation result output by the processing state evaluation unit 23 for each of the multiple trained models, and selects one trained model whose evaluation result output by the processing state evaluation unit 23 matches the evaluation result by the user. In this case, the monitoring model determination unit 24 determines the trained model selected by the user as the monitoring model. In the second example, the monitoring model determination unit 24 determines the monitoring model based on the result of the user's evaluation of the processing state and the evaluation result by the processing state evaluation unit 23.

[0081] FIG. 5 is a flowchart showing a second example of the procedure of the process executed by the laser processing system 1 according to the first embodiment. In the above first example, the laser processing system 1 obtains an evaluation result from each of the multiple trained models, and compares the obtained evaluation result with the evaluation result by the user. In the second example, the laser processing system 1 obtains evaluation results from all of the multiple trained models, and compares each obtained evaluation result with the evaluation result by the user. Also in the second example, the laser processing system 1 determines the monitoring model by having the laser processing machine 2 perform trial processing.

[0082] In step S11, the laser processing machine 2 starts test processing. In step S12, the data acquisition unit 21 acquires sensor data. In step S13, the feature calculation unit 22 calculates feature amounts based on the sensor data. Steps S11 to S13 are similar to steps S1 to S3 shown in FIG. 4.

[0083] In step S14, the machining state evaluation unit 23 evaluates the machining state based on each of the multiple trained models. The machining state evaluation unit 23 obtains an evaluation result based on each of the multiple trained models by inputting feature amounts to all of the multiple trained models held in the trained model holding unit 26. The machining state evaluation unit 23 outputs the obtained evaluation result to the monitoring model determination unit 24.

[0084] In step S15, the monitoring model determination unit 24 selects, from among the multiple trained models, a trained model whose evaluation result of the processing state is closest to the evaluation result by the user. In the second example shown in Fig. 5, it is assumed that the evaluation result by the user is input to the monitoring model determination unit 24. The monitoring model determination unit 24 compares the evaluation results of each of the multiple trained models with the evaluation result by the user, and selects one trained model that has obtained an evaluation result closest to the evaluation result by the user.

[0085] In step S16, the monitoring model determination unit 24 determines a monitoring model. The monitoring model determination unit 24 determines the trained model selected in step S15 as the monitoring model. By completing step S16, the laser processing system 1 ends the processing according to the procedure shown in Fig. 5. The laser processing system 1 uses the monitoring model to monitor the processing state when the laser processing machine 2 processes the product.

[0086] The trained model whose evaluation result of the machining state is closest to the evaluation result by the user may be selected by the user. The user checks the evaluation results of each of the multiple trained models and selects one trained model whose evaluation result is closest to the evaluation result by the user. In this case, the monitoring model determination unit 24 determines the trained model selected by the user as the monitoring model. Note that, when the determination of the machining state by observing the workpiece 10 is performed by a determination device, the monitoring model determination unit 24 selects, in step S15, from the multiple trained models, the trained model whose evaluation result of the machining state is closest to the evaluation result by the determination device.

[0087] 6 is a flowchart showing a third example of a procedure of a process executed by the laser processing system 1 according to the first embodiment. In the third example, the laser processing system 1 determines a monitoring model while the laser processing machine 2 is processing a product. The laser processing system 1 may determine a monitoring model by trial processing as in the first or second example, or may determine a monitoring model while the product is being processed as in the third example. In the third example, the monitoring model determination unit 24 determines a monitoring model based on a result of a user's evaluation of the processing state and an evaluation result by the processing state evaluation unit 23.

[0088] In step S21, the laser processing machine 2 starts processing the product. In step S22, the data acquisition unit 21 acquires sensor data. In step S23, the feature calculation unit 22 calculates feature values ​​based on the sensor data. In step S24, the processing state evaluation unit 23 evaluates the processing state based on each of the multiple trained models. In step S25, the monitoring model determination unit 24 selects, from the multiple trained models, a trained model whose evaluation result of the processing state is closest to the evaluation result by the user. In step S26, the monitoring model determination unit 24 determines the monitoring model. Steps S22 to S26 are similar to steps S12 to S16 shown in FIG. 5.

[0089] By completing step S26, the laser processing system 1 ends the processing according to the procedure shown in FIG. 6. The laser processing system 1 continues product processing by the laser processing machine 2 and monitors the processing state using the monitoring model. When processing defects are tolerable to a certain extent in product processing, the laser processing system 1 may determine the monitoring model while the product is being processed, as in the third example. Note that, when the judgment device judges the processing state by observing the workpiece 10, the monitoring model determination unit 24 selects, in step S25, from among the multiple learned models, a learned model whose evaluation result of the processing state is closest to the evaluation result by the judgment device.

[0090] When a product is being processed, the timing for executing the process for determining the monitoring model is arbitrary. For example, the laser processing system 1 executes the process for determining the monitoring model when the operation of the laser processing machine 2 for product processing is terminated. The laser processing system 1 may execute the process for determining the monitoring model after processing of a certain number of products is completed. The laser processing system 1 may execute the process for determining the monitoring model every time one product is processed. The laser processing system 1 may execute the process for determining the monitoring model when a processing defect occurs. The laser processing system 1 may execute the process for determining the monitoring model when the operation of the laser processing machine 2 for product processing is started.

[0091] The laser processing system 1 can determine the learned model to be used as the monitoring model based on a large amount of data by determining the monitoring model while the product is being processed. The laser processing system 1 can evaluate the processing state using the monitoring model that reflects the aging of the laser processing machine 2 by determining the monitoring model at any time while the product is being processed.

[0092] According to the first embodiment, the control device 3 determines a learned model, which is a monitoring model used to monitor the state of processing by the laser processing machine 2, based on the result of evaluating the state of processing by observing the workpiece 10 processed by the laser processing machine 2 and the evaluation result by the processing state evaluation unit 23. The control device 3 can obtain a monitoring model that enables accurate evaluation of the processing state even when the processing state measured by the sensor changes due to the state of the laser processing machine 2 or the variation in the components contained in the workpiece 10. The control device 3 can accurately evaluate the quality of the processing state, thereby allowing the laser processing machine 2 to continue good processing. As a result, the control device 3 has the effect of allowing the laser processing machine 2 to continue good processing.

[0093] Embodiment 2 In the first embodiment, the monitoring model determination unit 24 determines, as the monitoring model, a trained model in which the evaluation result output by the machining state evaluation unit 23 matches the evaluation result obtained by observing the workpiece 10, from among a plurality of trained models. In the second embodiment, an example in which a combination of a plurality of trained models is determined as the monitoring model will be described. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and configurations different from those in the first embodiment will be mainly described.

[0094] Fig. 7 is a diagram showing a configuration example of the monitoring model determination unit 30 included in the control device 3 according to the second embodiment. The control device 3 according to the second embodiment includes the monitoring model determination unit 30 shown in Fig. 7 instead of the monitoring model determination unit 24 shown in Fig. 2. The monitoring model determination unit 30 adjusts the weight for the output value of each learned model for a combination of multiple learned models held in the learned model holding unit 26, and determines the combination of multiple learned models in which the weight for each learned model has been adjusted as a monitoring model.

[0095] The monitoring model determination unit 30 includes a weight determination unit 31. The weight determination unit 31 adjusts the weight for each trained model and determines the weight for each trained model.

[0096] The monitoring model determination unit 30 reads out a plurality of learned models from the learned model holding unit 26. In the example shown in Fig. 7, it is assumed that three learned models are held in the learned model holding unit 26, and the monitoring model determination unit 30 reads out the three learned models.

[0097] v1, v2, and v3 shown in Fig. 7 represent weights for the output values ​​of each trained model. The weight determination unit 31 adjusts each of v1, v2, and v3. The dashed arrows shown in Fig. 7 indicate that each of v1, v2, and v3 is adjusted by the weight determination unit 31. Each trained model is, for example, the neural network shown in Fig. 3.

[0098] The machining state evaluation unit 23 evaluates the state of machining by the laser processing machine 2 by a combination of multiple trained models in which the weights of the trained models are adjusted. The monitoring model determination unit 30 adjusts each of v1, v2, and v3 so that the difference between the evaluation result output by the machining state evaluation unit 23 and the evaluation result by observation of the workpiece 10 is small. As a result, the monitoring model determination unit 30 adjusts each of v1, v2, and v3 so that the evaluation result output by the machining state evaluation unit 23 matches the evaluation result by observation of the workpiece 10. The evaluation result by observation of the workpiece 10 is the result of a user evaluating the machining state by observing the machined workpiece 10, or the evaluation result by a judgment device that judges the machining state by observing the workpiece 10.

[0099] The monitoring model determination unit 30 determines, as the monitoring model, a combination of three trained models whose weights have been adjusted so that the evaluation result output by the machining state evaluation unit 23 coincides with the evaluation result obtained by observing the workpiece 10. Alternatively, the monitoring model determination unit 30 determines, as the monitoring model, a combination of three trained models whose weights have been adjusted so that the evaluation result output by the machining state evaluation unit 23 is closest to the evaluation result obtained by observing the workpiece 10. In this manner, the monitoring model determination unit 30 determines, as the monitoring model, a combination of multiple trained models whose weights have been adjusted for each trained model. The monitoring model determination unit 30 stores the determined monitoring model in the monitoring model holding unit 27. The monitoring model holding unit 27 holds the monitoring model.

[0100] The processing monitoring unit 20 monitors the state of processing by the laser processing machine 2 using the monitoring model determined by the monitoring model determination unit 30. The monitoring model receives the feature values ​​and outputs the sum of the weighted output values ​​of each trained model. The monitoring model may receive the feature values ​​and output a weighted average of the output values ​​of each trained model. The monitoring model determination unit 30 may determine the weight of each trained model by learning the relationship between the weight, the feature values, and the processing state.

[0101] When each trained model is a neural network, the monitoring model determination unit 30 does not change the weights set in the neural network, i.e., w11-w16, w21-w26 shown in FIG. 3. The monitoring model determination unit 30 determines the monitoring model by adjusting the weights for the output values ​​of each trained model without changing the weights set in the neural network. The processing monitoring unit 20 can use a combination of multiple trained models as a monitoring model by adjusting the weights for the output values ​​of each trained model with the monitoring model determination unit 30.

[0102] The control device 3 obtains a monitoring model for evaluating the state of processing by the laser processing machine 2 by adjusting the weight for each learned model. Each of the multiple learned models held in the learned model holding unit 26 can be a model that is used as the basis of the monitoring model. The model that is used as the basis of the monitoring model is, for example, a model set for each manufacturer of the workpiece 10, or a model set for each material of the workpiece 10. The control device 3 can reduce the number of learned models that are set in advance, compared to a case where a learned model to be used as a monitoring model is selected from multiple learned models. The control device 3 can determine the monitoring model based on a relatively small amount of data.

[0103] According to the second embodiment, the control device 3 determines a combination of multiple trained models in which the weights of the trained models are adjusted as a monitoring model. The control device 3 can obtain a monitoring model that enables accurate evaluation of the processing state, and can continue good processing by the laser processing machine 2. As a result, the control device 3 has an effect of allowing good processing by the laser processing machine 2 to be continued. The laser processing system 1 can improve productivity by determining a monitoring model based on a relatively small amount of data.

[0104] Embodiment 3 In the first and second embodiments, an example in which a plurality of trained models are set in advance is described. In the third embodiment, an example in which a trained model is added by training in the laser processing system 1 is described. In the third embodiment, the same components as those in the first or second embodiment are denoted by the same reference numerals, and configurations different from those in the first or second embodiment are mainly described.

[0105] 8 is a diagram showing a configuration example of a control device 40 included in the laser processing system 1 according to the third embodiment. The laser processing system 1 according to the third embodiment includes a laser processing machine 2 and a control device 40. The control device 40 includes a processing control unit 28 and a processing monitoring unit 41. The control device 40 may be a device external to the laser processing machine 2, or may be a device included in the laser processing machine 2. Alternatively, the processing control unit 28 of the control device 40 may be realized by a device included in the laser processing machine 2, and the processing monitoring unit 41 of the control device 40 may be realized by a device external to the laser processing machine 2.

[0106] The processing monitoring unit 41 includes a data acquisition unit 21, a feature calculation unit 22, a processing state evaluation unit 23, a processing parameter correction unit 25, a learned model holding unit 26, a monitoring model holding unit 27, a monitoring model determination unit 42, and a learning unit 43.

[0107] The learning unit 43 acquires the feature amounts calculated by the feature amount calculation unit 22 from the feature amount calculation unit 22. The learning unit 43 acquires information indicating the state of processing by the laser processing machine 2. The learning unit 43 generates a trained model by learning the relationship between the feature amounts and the state of processing by the laser processing machine 2, and adds the generated trained model to the trained model holding unit 26.

[0108] The monitoring model determination unit 42 adjusts the weight for the output value of each of the multiple trained models for the combination of the multiple trained models held in the trained model holding unit 26, and determines the combination of the multiple trained models in which the weight for each trained model has been adjusted as the monitoring model. The multiple trained models held in the trained model holding unit 26 include the trained models added by the learning unit 43.

[0109] Each of the multiple trained models is, for example, a neural network as shown in Fig. 3. The learning unit 43 learns the relationship between the feature amount and the processing state by supervised learning, and generates a trained model.

[0110] 9 is a diagram illustrating a configuration example of the monitoring model determination unit 42 included in the control device 40 according to the third embodiment. The monitoring model determination unit 42 includes a weight determination unit 44. The weight determination unit 44 adjusts the weight for each trained model to determine the weight for each trained model.

[0111] The monitoring model determination unit 42 reads out a plurality of learned models from the learned model holding unit 26. In the example shown in Fig. 9, it is assumed that five learned models are held in the learned model holding unit 26, and the monitoring model determination unit 42 reads out the five learned models. Two of the five learned models are learned models added by the learning unit 43.

[0112] v1, v2, v3, v4, and v5 shown in Fig. 9 represent weights for the output values ​​of each trained model. The weight determination unit 44 adjusts each of v1, v2, v3, v4, and v5. The dashed arrows shown in Fig. 9 indicate that each of v1, v2, v3, v4, and v5 is adjusted by the weight determination unit 44.

[0113] The machining state evaluation unit 23 evaluates the state of machining by the laser processing machine 2 by a combination of multiple trained models in which the weights for each trained model have been adjusted. The monitoring model determination unit 42 adjusts each of v1, v2, v3, v4, and v5 so as to reduce the difference between the evaluation result output by the machining state evaluation unit 23 and the evaluation result obtained by observing the workpiece 10. As a result, the monitoring model determination unit 42 adjusts each of v1, v2, v3, v4, and v5 so that the evaluation result output by the machining state evaluation unit 23 matches the evaluation result obtained by observing the workpiece 10.

[0114] The monitoring model determination unit 42 determines, as the monitoring model, a combination of five trained models whose weights have been adjusted so that the evaluation result output by the machining state evaluation unit 23 matches the evaluation result obtained by observing the workpiece 10. Alternatively, the monitoring model determination unit 42 determines, as the monitoring model, a combination of five trained models whose weights have been adjusted so that the evaluation result output by the machining state evaluation unit 23 is closest to the evaluation result obtained by observing the workpiece 10. In this manner, the monitoring model determination unit 42 determines, as the monitoring model, a combination of multiple trained models whose weights have been adjusted for each trained model. The monitoring model determination unit 42 stores the determined monitoring model in the monitoring model holding unit 27. The monitoring model holding unit 27 holds the monitoring model.

[0115] The processing monitoring unit 20 monitors the state of processing by the laser processing machine 2 using the monitoring model determined by the monitoring model determination unit 42. The monitoring model receives the feature values ​​and outputs the sum of the weighted output values ​​of each trained model. The monitoring model may receive the feature values ​​and output a weighted average of the output values ​​of each trained model. The monitoring model determination unit 42 may determine the weight of each trained model by learning the relationship between the weight, the feature values, and the processing state.

[0116] The learning unit 43 generates a trained model using feature amounts calculated based on sensor data actually obtained from the laser processing machine 2. The control device 40 creates a monitoring model based on such a trained model, thereby obtaining a highly accurate monitoring model based on the actual state of the laser processing machine 2.

[0117] The learning unit 43 generates a new trained model using the feature amount and information indicating the processing state. Alternatively, the learning unit 43 may perform additional learning of a trained model generated in the past. The learning unit 43 may add a trained model to the trained model holding unit 26, whereby the added trained model may be accumulated in the trained model holding unit 26. The trained model added to the trained model holding unit 26 may be deleted from the trained model holding unit 26 as appropriate.

[0118] In the control device 40, when the material of the workpiece 10 processed by the laser processing machine 2 is changed, the learned model generated by the learning unit 43 may be added to the learned model holding unit 26, or the learned model generated by the learning unit 43 and added to the learned model holding unit 26 may be deleted from the learned model holding unit 26. When the material of the workpiece 10 is changed, the control device 40 can obtain a highly accurate monitoring model according to the material of the workpiece 10 by updating the learned model held in the learned model holding unit 26.

[0119] In the control device 40, when the state of processing by the laser processing machine 2 changes, the learned model generated by the learning unit 43 may be added to the learned model holding unit 26, or the learned model generated by the learning unit 43 and added to the learned model holding unit 26 may be deleted from the learned model holding unit 26. A change in the processing state is, for example, a change in the processing state from a good state to a poor state. When the processing state changes, the control device 40 can obtain a highly accurate monitoring model according to the processing state by updating the learned model held in the learned model holding unit 26.

[0120] The trained model, which is the result of learning by the learning unit 43, may be a trained model that outputs at least one of an evaluation result of the machining state and a correction amount by inputting the feature amount. The machining parameter correction unit 25 may calculate the correction amount by inputting the feature amount calculated by the feature amount calculation unit 22 to the monitoring model.

[0121] The machining state evaluation unit 23 may evaluate the state of machining by the laser processing machine 2 by inputting information other than the feature amount calculated by the feature amount calculation unit 22 to each of the multiple trained models. Examples of information input to each of the multiple trained models include the value of the machining parameter at the time of evaluation, the temperature of the machining optical system at the time of evaluation, the temperature change of the machining optical system, the thickness of the workpiece 10, or the material of the workpiece 10. In this case, the learning unit 43 generates a trained model by learning the relationship between such information, the feature amount, and the machining state. In addition, the machining state evaluation unit 23 evaluates the state of machining by the laser processing machine 2 by inputting such information and the feature amount to the monitoring model.

[0122] In the above description, the learning unit 43 is provided inside the control device 40. The learning unit 43 may be realized by a device external to the control device 40. A learning device that realizes the learning unit 43 may be a device that can be connected to the control device 40 via a network. The learning device may be a device that exists on a cloud server.

[0123] According to the third embodiment, the control device 40 determines a combination of multiple learned models in which the weights of the learned models are adjusted as the monitoring model. The control device 40 can obtain a monitoring model that enables accurate evaluation of the processing state, and can continue good processing by the laser processing machine 2. As described above, the control device 40 has an effect of allowing good processing by the laser processing machine 2 to be continued. The laser processing system 1 can improve productivity by determining a monitoring model based on a relatively small amount of data. The control device 40 can obtain a highly accurate monitoring model by generating a learned model using a feature amount calculated based on sensor data actually obtained from the laser processing machine 2.

[0124] Next, a description will be given of hardware for realizing the control devices 3 and 40 according to the first to third embodiments. The control devices 3 and 40 are realized by a processing circuit. The processing circuit may be a circuit in which a processor executes software, or may be a dedicated circuit.

[0125] When the processing circuit is realized by software, the processing circuit is, for example, a control circuit 50 shown in Fig. 10. Fig. 10 is a diagram showing a configuration example of the control circuit 50 according to the first to third embodiments. The control circuit 50 includes an input unit 51, a processor 52, a memory 53, and an output unit 54. The input unit 51 is an interface circuit that receives data input from outside the control circuit 50 and provides the data to the processor 52. The output unit 54 is an interface circuit that sends data from the processor 52 or the memory 53 to outside the control circuit 50.

[0126] When the processing circuit is the control circuit 50 shown in FIG. 10, the control device 3, 40 is realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in the memory 53. The processing circuit realizes each function of the control device 3, 40 by the processor 52 reading and executing the program stored in the memory 53. That is, the processing circuit includes the memory 53 for storing the program that will result in the processing of the control device 3, 40. It can also be said that these programs cause a computer to execute the procedures and methods of the control device 3, 40.

[0127] The processor 52 is a CPU (Central Processing Unit). The processor 52 may be a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a processor, or a DSP (Digital Signal Processor). The memory 53 may be, for example, a random access memory (RAM), a read only memory (ROM), a flash memory, an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM (registered trademark)), or other non-volatile or volatile semiconductor memory, a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a digital versatile disk (DVD).

[0128] Fig. 10 shows an example of hardware in which the control devices 3 and 40 are realized by a general-purpose processor 52 and a memory 53, but the control devices 3 and 40 may be realized by a dedicated hardware circuit. Fig. 11 shows an example of the configuration of a dedicated hardware circuit 55 according to the first to third embodiments.

[0129] The dedicated hardware circuit 55 includes an input unit 51, an output unit 54, and a processing circuit 56. The processing circuit 56 is a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a circuit that is a combination of these. Each function of the control device 3, 40 may be realized by the processing circuit 56 on a function-by-function basis, or each function may be realized collectively by the processing circuit 56. The control device 3, 40 may be realized by combining the control circuit 50 and the hardware circuit 55.

[0130] When the learning unit 43 is an external device to the control device 40, the learning device is realized by a processing circuit, similar to the control devices 3 and 40. The processing circuit that realizes the learning device is the control circuit 50 shown in FIG. 10 or a dedicated hardware circuit 55 shown in FIG. 11.

[0131] The configurations shown in the above embodiments are examples of the contents of the present disclosure. The configurations of each embodiment can be combined with other known technologies. The configurations of each embodiment can be appropriately combined. A part of the configuration of each embodiment can be omitted or modified without departing from the gist of the present disclosure. [Explanation of symbols]

[0132] 1 laser processing system, 2 laser processing machine, 3,40 control device, 4 laser oscillator, 5 cable, 6 processing head, 7 processing nozzle, 8 sound sensor, 9 optical sensor, 10 workpiece, 11 collimating optical system, 12 imaging optical system, 13 protective glass, 14 processing head drive unit, 15,16 moving mechanism, 20,41 processing monitoring unit, 21 data acquisition unit, 22 feature calculation unit, 23 processing state evaluation unit, 24,30,42 monitoring model determination unit, 25 processing parameter correction unit, 26 learned model storage unit, 27 monitoring model storage unit, 28 processing control unit, 31,44 weight determination unit, 43 learning unit, 50 control circuit, 51 input unit, 52 processor, 53 memory, 54 output unit, 55 hardware circuit, 56 processing circuit.

Claims

1. A learned model holding unit that holds a plurality of learned models, which are the results of learning the relationship between the feature quantity indicating the processing state by the processing machine and the processing state by the processing machine; A processing state evaluation unit that evaluates the processing state by the processing machine by inputting the feature quantity to each of the plurality of learned models or by inputting the feature quantity to a combination of the plurality of learned models; A monitoring model determination unit that determines the learned model or a combination of the plurality of learned models, which is a monitoring model used for monitoring the processing state by the processing machine, based on the result of evaluating the processing state by observing the workpiece processed by the processing machine and the evaluation result by the processing state evaluation unit; A processing control unit that controls the processing machine based on the result of monitoring the processing state by the processing machine using the monitoring model, comprising A control device characterized by the above.

2. The monitoring model determination unit determines one of the plurality of learned models held by the learned model holding unit as the monitoring model The control device according to claim 1, characterized by the above.

3. The monitoring model determination unit adjusts the weights for the output values of each of the learned models for a combination of the plurality of learned models held by the learned model holding unit, and determines a combination of the plurality of learned models with the weights adjusted for each of the learned models as the monitoring model The control device according to claim 1, characterized by the above.

4. The processing machine is a laser processing machine that processes the workpiece with laser light, A data acquisition unit that acquires data indicating the detection result of light generated during the processing by the laser processing machine and data indicating the detection result of sound generated during the processing by the laser processing machine; A feature quantity calculation unit that calculates the feature quantity from the data acquired by the data acquisition unit, comprising The control device according to any one of claims 1 to 3, characterized by the above.

5. The processing machine is a laser processing machine that processes the workpiece with laser light, Each of the plurality of learned models held by the learned model holding unit is the result of learning the relationship between the feature quantity and the processing state in each of a plurality of cases where at least one of the thickness of the plate-shaped workpiece and the type of processing gas during the processing by the laser processing machine is different from each other The control device according to any one of claims 1 to 3, characterized by the above.

6. A learning unit that generates the learned model by learning the relationship between the feature amount and the machining state by the machine tool, and adds the generated learned model to the learned model holding unit. The control device according to any one of claims 1 to 3, characterized in that.

7. When the material of the workpiece machined by the machine tool is changed, the learned model generated by the learning unit is added to the learned model holding unit, or the learned model generated by the learning unit and added to the learned model holding unit is deleted from the learned model holding unit. The control device according to claim 6, characterized in that.

8. When the machining state by the machine tool changes, the learned model generated by the learning unit is added to the learned model holding unit, or the learned model generated by the learning unit and added to the learned model holding unit is deleted from the learned model holding unit. The control device according to claim 6, characterized in that.

9. A laser processing machine that processes a workpiece with laser light, A control device that controls the laser processing machine, comprising: The control device is A learned model holding unit that holds a plurality of learned models that are the results of learning the relationship between the feature amount indicating the machining state by the laser processing machine and the machining state by the laser processing machine; A machining state evaluation unit that evaluates the machining state by the laser processing machine by inputting the feature amount to each of the plurality of learned models or by inputting the feature amount to a combination of the plurality of learned models; A monitoring model determination unit that determines the learned model or a combination of the plurality of learned models, which is a monitoring model used for monitoring the machining state by the laser processing machine, based on the result of evaluating the machining state by observing the workpiece machined by the laser processing machine and the evaluation result by the machining state evaluation unit; A machining control unit that controls the laser processing machine based on the result of monitoring the machining state by the laser processing machine using the monitoring model. A laser processing system, characterized in that.

10. A step of obtaining a feature amount indicating the machining state by a laser processing machine that processes a workpiece with laser light. By inputting the feature amount obtained for each of a plurality of learned models, which is the result of learning the relationship between the processing state by the laser processing machine and the feature amount, or by inputting the feature amount obtained for a combination of the plurality of learned models, a step of evaluating the processing state by the laser processing machine; Based on the result of evaluating the processing state by observing the workpiece processed by the laser processing machine and the evaluation result by the step of evaluating the processing state, determining the learned model or a combination of the plurality of learned models, which is a monitoring model used for monitoring the processing state by the laser processing machine; Controlling the laser processing machine based on the result of monitoring the processing state by the laser processing machine using the monitoring model, and including A laser processing method characterized by the above.

Images