Data generation method, computer program, recording medium, and machining device

Abstract

This data generation method includes: acquiring test machining condition information including the pulse count for which pulsed beams should be emitted respectively to a plurality of irradiation positions on a test workpiece in order to perform machining so as to form the test workpiece in a target shape; measuring a test workpiece shape that is the shape of the test workpiece after machining the test workpiece on the basis of the test machining condition information; and calculating a goal pulse count for which the pulsed beams should be emitted to the plurality of irradiation positions on the machining object workpiece, on the basis of the test machining condition information, the test workpiece shape, and prediction information of a shape of the portion in which machining is to be performed through the emission of the pulsed beams in the unit pulse count.

Description

Data generation method, computer program, recording medium and processing device

[0001] The present invention relates to a data generation method, a computer program, a recording medium, and a processing device that generate control data for controlling a processing device that can remove and process a workpiece by irradiating the surface of the workpiece with a pulse energy beam.

[0002] Patent Document 1 describes a processing device that processes an object by irradiating the object with laser light. This type of processing device is required to process the object appropriately.

[0003] US Patent Application Publication No. 2002 / 0017509

[0004] According to a first aspect, there is provided a data generation method for generating control data for controlling a processing device capable of removing and processing a workpiece to be processed by irradiating a pulse energy beam onto the surface of the workpiece, the data generation method including: acquiring test processing condition information including the number of pulses at which the pulse energy beam should be irradiated to each of multiple irradiation positions of the test workpiece in order to process the test workpiece into a target shape; measuring a test workpiece shape, which is the shape of the test workpiece after processing the test workpiece based on the test processing condition information; and calculating a target number of pulses at which the pulse energy beam should be irradiated to each of multiple irradiation positions of the workpiece to be processed based on the test processing condition information, the test workpiece shape, and predicted information on the shape of the part to be processed by irradiating the pulse energy beam with a unit number of pulses.

[0005] According to a second aspect, there is provided a data generation method for generating control data for controlling a processing device capable of removing and processing a workpiece to be processed by irradiating a pulse energy beam onto the surface of the workpiece, the data generation method including: inputting test processing condition information indicating the number of pulses with which the pulse energy beam should be irradiated at each of a plurality of irradiation positions of the test workpiece in order to process the test workpiece into a target shape; measuring a test workpiece shape related to the shape of the test workpiece processed by the pulse energy beam based on the test processing condition information; performing machine learning to generate model information including prediction information of the shape to be processed by the pulse energy beam with a unit number of pulses based on the test processing condition information and the test workpiece shape; and calculating a target number of pulses with which the pulse energy beam should be irradiated at a plurality of irradiation positions of the workpiece to be processed based on the model information generated by the machine learning and the target shape.

[0006] According to a third aspect, there is provided a data generation method for generating control data for controlling a processing device capable of removing and processing a workpiece to be processed by irradiating a pulse energy beam onto the surface of the workpiece to be processed, the data generation method including: acquiring target pulse number information indicating a target number of pulses with which the pulse energy beam should be irradiated to each of a plurality of irradiation positions on the workpiece to be processed, and target shape information regarding the target shape of the workpiece to be processed after removal and processing; and generating the target pulse number information as the control data based on the model information and the target shape information.

[0007] According to a fourth aspect, there is provided a data generation method for generating control data for controlling a processing device capable of removing and processing a workpiece to be processed by irradiating a pulse energy beam onto the surface of the workpiece to be processed, the data generation method including: acquiring model information that can be used to predict a predicted value of the processing amount of the workpiece to be processed when the pulse energy beam is irradiated at a first irradiation position among a plurality of irradiation positions on the workpiece to be processed by a unit pulse number, the predicted value of the processing amount at the first irradiation position and the predicted value of the processing amount at a position other than the first irradiation position; and generating, as the control data, target pulse number information indicating a target number of pulses with which the pulse energy beam should be irradiated at each of a plurality of irradiation positions on the workpiece to be processed based on the model information and target shape information regarding a target shape of the workpiece to be processed after removal processing.

[0008] According to a fifth aspect, there is provided a data generation method for generating control data for controlling a processing device capable of removing and processing a workpiece to be processed by irradiating a pulse energy beam onto the surface of the workpiece, the data generation method including: acquiring test processing condition information for processing a test workpiece to a target shape; measuring a test workpiece shape, which is the shape of the test workpiece after processing the test workpiece based on the test processing condition information; and calculating processing conditions for the workpiece to be processed based on the test processing condition information, the test workpiece shape, and predicted information about the shape to be processed by the pulse energy beam.

[0009] According to a sixth aspect, there is provided a data generation method for generating control data for controlling a machining device capable of removing a workpiece to be machined by irradiating a surface of the workpiece with a pulse energy beam, the data generation method including: generating first test machining condition information indicating the number of pulses with which the pulse energy beam should be irradiated to each of a plurality of irradiation positions of a first test workpiece; acquiring first test shape information regarding the shape of the first test workpiece that has been removed by irradiating the first test workpiece with the pulse energy beam based on the first test machining condition information; generating second test machining condition information indicating the number of pulses with which the pulse energy beam should be irradiated to each of a plurality of irradiation positions of a second test workpiece in order to machine a second test workpiece into a target shape; acquiring second test shape information regarding the shape of the second test workpiece that has been removed by irradiating the second test workpiece with the pulse energy beam based on the second test machining condition information; and calculating a target number of pulses with which the pulse energy beam should be irradiated to each of the irradiation positions of the workpiece to be machined based on the first test shape information, the second test shape information, the second test machining conditions, and the target shape.

[0010] According to a seventh aspect, there is provided a computer program causing a computer to execute the data generation method provided by any one of the first to sixth aspects.

[0011] According to an eighth aspect, there is provided a recording medium on which the computer program provided by the seventh aspect is recorded.

[0012] According to a ninth aspect, there is provided a processing device that removes and processes the workpiece to be processed using the control data generated by the data generation method provided by any one of the first to sixth aspects.

[0013] FIG. 1 is a block diagram showing the overall configuration of a processing system according to this embodiment. FIG. 2 is a perspective view schematically showing the exterior of the processing apparatus according to this embodiment. FIG. 3 is a system configuration diagram showing the system configuration of the processing apparatus according to this embodiment. FIG. 4 is a perspective view showing the configuration of an irradiation optical system. FIGS. 5(a) to 5(c) are cross-sectional views showing the state of removal processing performed on a workpiece. FIG. 6 shows the workpiece before and after removal processing. FIG. 7 shows a removal target portion to be removed from the workpiece. FIG. 8 is a cross-sectional view showing multiple removal layers to be removed from the workpiece. FIGS. 9(a) to 9(d) are plan views showing areas on the surface of the workpiece W where removal processing is performed in the process of removing each removal layer. FIGS. 10(a) and 10(b) are plan views showing irradiation target positions on the surface of the workpiece. FIG. 11 is a block diagram showing the configuration of a data generation server. FIG. 12 is a flowchart showing the flow of a data generation operation. FIG. 13(a) is a perspective view showing an example of a riblet structure, and FIG. 13(b) is a cross-sectional view showing an example of a riblet structure. Fig. 14 conceptually illustrates machine learning for generating model information. Fig. 15 conceptually illustrates an example of model information. Fig. 16 conceptually illustrates an example of prediction information. Fig. 17 conceptually illustrates calculation of a target pulse number based on model information.

[0014] Hereinafter, embodiments of a data generation method, a computer program, a recording medium, and a processing device will be described with reference to the drawings. Hereinafter, embodiments of a data generation method, a computer program, a recording medium, and a processing device will be described using a processing system SYS that processes a workpiece W using processing light EL. However, the present invention is not limited to the embodiments described below.

[0015] In the following description, the positional relationships of the various components constituting the machining system SYS will be described using an XYZ Cartesian coordinate system defined by mutually orthogonal X, Y, and Z axes. For convenience of explanation, the X-axis and Y-axis directions are each assumed to be horizontal (i.e., a predetermined direction within a horizontal plane), and the Z-axis direction is assumed to be vertical (i.e., a direction perpendicular to the horizontal plane, essentially an up-down direction). Furthermore, the rotation directions around the X-axis, Y-axis, and Z-axis (in other words, tilt directions) are referred to as the θX direction, θY direction, and θZ direction, respectively. Here, the Z-axis direction may be the direction of gravity. Furthermore, the XY plane may be assumed to be horizontal.

[0016] (1) Overall Configuration of Machining System SYS First, the overall configuration of the machining system SYS will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the overall configuration of the machining system SYS.

[0017] As shown in FIG. 1 , the processing system SYS includes a processing device 1 and a data generating server 2. Furthermore, the processing system SYS includes a client terminal device 3. However, the processing system SYS does not necessarily include the client terminal device 3. The processing device 1, the data generating server 2, and the client terminal device 3 can communicate with each other via a communication network 4 including at least one of a wired communication network and a wireless communication network. In this case, the data generating server 2 may function as a cloud server for the processing device 1 and the client terminal device 3. If the data generating server 2 can function as a cloud server, the data generating server 2 may be referred to as a cloud server or a cloud system. However, at least one of the processing device 1, the data generating server 2, and the client terminal device 3 does not necessarily have to be able to communicate with at least one other of the processing device 1, the data generating server 2, and the client terminal device 3.

[0018] The processing apparatus 1 is capable of processing a workpiece W (see FIG. 2 ), which is an object to be processed (i.e., a workpiece to be processed). The workpiece W may also be referred to as a workpiece to be processed. The workpiece W may be, for example, a metal, an alloy (e.g., duralumin, etc.), a semiconductor (e.g., silicon), a resin, a composite material such as CFRP (Carbon Fiber Reinforced Plastic), glass, ceramics, or any other material. Examples of such materials include at least one of gypsum, rubber such as polyurethane, and elastomers. Furthermore, a portion of the workpiece W may be made of one material, and another portion may be made of a different material.

[0019] In this embodiment, an example will be described in which the processing device 1 performs removal processing on the workpiece W. In particular, in this embodiment, an example will be described in which the processing device 1 performs removal processing by irradiating the workpiece W with processing light EL. However, the processing device 1 may perform processing on the workpiece W that is different from removal processing. For example, the processing device 1 may perform additional processing on the workpiece W.

[0020] The processing device 1 may form a riblet structure on the workpiece W by processing the workpiece W. The riblet structure may be a structure that can reduce the resistance of the surface of the workpiece W to the fluid (particularly, at least one of frictional resistance and turbulent frictional resistance). The riblet structure may include a structure that can reduce noise generated when the fluid and the surface of the workpiece W move relative to each other. The riblet structure may include, for example, a structure in which grooves extending in a first direction (e.g., the Y-axis direction) along the surface of the workpiece W are arranged in a plurality of rows along a second direction (e.g., the X-axis direction) that is along the surface of the workpiece W and intersects the first direction.

[0021] The processing apparatus 1 may process the workpiece W to form an arbitrary structure having an arbitrary shape on the surface of the workpiece W. One example of the arbitrary structure is a structure that generates vortices in the flow of a fluid on the surface of the workpiece W. Another example of the arbitrary structure is a structure that imparts hydrophobic properties to the surface of the workpiece W. Another example of the arbitrary structure is a regularly or irregularly formed fine texture structure (typically an uneven structure) on the order of micrometers or nanometers. Such a fine texture structure may include at least one of a shark skin structure and a dimple structure that have the function of reducing resistance due to a fluid (gas and / or liquid). The fine texture structure may also include a lotus leaf surface structure that has at least one of a liquid-repellent function and a self-cleaning function (e.g., has the lotus effect). The fine texture structure may include at least one of a micro-protrusion structure having a liquid transport function (see U.S. Patent Publication No. 2017 / 0044002), a concave-convex structure having a lyophilic function, a concave-convex structure having an anti-fouling function, a moth-eye structure having at least one of a reflectance reducing function and a liquid repellent function, a concave-convex structure that exhibits a structural color by intensifying only light of a specific wavelength through interference, a pillar array structure having an adhesive function utilizing van der Waals forces, a concave-convex structure having an aerodynamic noise reducing function, a honeycomb structure having a droplet collecting function, and a concave-convex structure that improves adhesion with a layer formed on the surface.

[0022] The data generating server 2 is capable of generating control data for controlling the processing device 1. The control data may be any data as long as it can be used to control the processing device 1. For example, the control data may include data that directly or indirectly specifies the processing conditions of the processing device 1. For example, the control data may include data that directly or indirectly specifies the operation details of the processing device 1. For example, the control data may include data that can directly control the processing device 1 (e.g., command data, etc.). For example, the control data may include data that can be used to generate data that is actually used to control the processing device 1 (e.g., slice data, etc., described below).

[0023] To generate the control data, the data generating server 2 may acquire (i.e., receive) information referenced by the data generating server 2 to generate the control data (hereinafter referred to as "reference information") from at least one of the processing apparatus 1 and the client terminal device 3 via the communication network 4. The reference information may include, for example, information about the processing light EL used by the processing apparatus 1 to process the workpiece W. The information about the processing light EL may include at least one of information about the intensity of the processing light EL, information about the shape of the processing light EL in a plane intersecting the traveling direction of the processing light EL (in other words, the irradiation direction), information about the light intensity distribution of the processing light EL in a plane intersecting the traveling direction of the processing light EL, information about the fluence of the processing light EL, and information about the fluence distribution of the processing light EL. The reference information may include, for example, device identification information for identifying the processing apparatus 1. The reference information may include, for example, workpiece information for identifying the workpiece W to be processed by the processing apparatus 1. The reference information may include, for example, information about the material of the workpiece W to be processed by the processing apparatus 1. The information about the material of the workpiece W may include information about the type of material constituting the workpiece W (e.g., the type of metal material). The reference information may include, for example, information about the shape of the workpiece W before the processing device 1 performs removal processing. The reference information may include, for example, information about the shape of the workpiece W after the processing device 1 performs removal processing. In the following description, the shape of the workpiece W after the processing device 1 performs removal processing is referred to as the target shape (target shape). The reference information may include information about the quality of processing by the processing device 1 (processing quality information). The processing quality information may include at least one of information about the resolution of processing by the processing device 1, information about the surface roughness of the workpiece W processed by the processing device 1, and information about the accuracy of processing by the processing device 1. The reference information may include information about the throughput of processing by the processing device 1 (processing throughput information).

[0024] The data generating server 2 may acquire all of the necessary reference information from the client terminal device 3. The data generating server 2 may acquire all of the necessary reference information from the processing device 1. The data generating server 2 may acquire part of the necessary reference information from the client terminal device 3, and acquire another part of the necessary reference information from the processing device 1. The data generating server 2 may acquire the reference information from the processing device 1 via the client terminal device 3. The data generating server 2 may acquire the reference information from the client terminal device 3 via the processing device 1.

[0025] The data generating server 2 may be capable of storing the acquired reference information. Note that the information about the processed light EL and the identification information for identifying the processing device 1 are both information about the processing device 1. In this case, the data generating server 2 may be capable of storing the information about the processed light EL and the identification information in a state in which the information about the processed light EL and the identification information are associated with each other. Alternatively, a server different from the data generating server 2 (e.g., a cloud server) may acquire the reference information from at least one of the processing device 1 and the client terminal device 3 via the communication network 4 and store the acquired reference information. In this case, the data generating server 2 may acquire the reference information from the cloud server that stores the reference information.

[0026] The data generating server 2 may generate control data based on the acquired reference information. For example, the data generating server 2 may generate control data by performing a calculation based on the acquired reference information. As an example, the data generating server 2 may determine processing conditions by performing a calculation based on the acquired reference information, and generate control data for controlling the processing device 1 so that the processing device 1 operates according to the determined processing conditions. An example of the calculation is a deconvolution calculation. Alternatively, for example, if multiple control data candidates are prepared, the data generating server 2 may select one control data based on the acquired reference information. As an example, the data generating server 2 may select one processing condition from multiple pre-prepared processing condition candidates based on the acquired reference information, and select (generate) control data for controlling the processing device 1 so that the processing device 1 operates according to the selected processing condition. In other words, in this embodiment, the operation of generating control data may include at least one of an operation of generating new control data by a calculation and an operation of selecting pre-prepared control data. The processing conditions may also be referred to as a processing recipe.

[0027] The data generating server 2 may be installed at the location where the processing device 1 is installed. The data generating server 2 may be installed at a location different from the location where the processing device 1 is installed. The data generating server 2 may be installed at a location where the client terminal device 3 is installed. The data generating server 2 may be installed at a location different from the location where the client terminal device 3 is installed. As an example, the data generating server 2 may be installed at a business establishment different from the business establishment where at least one of the processing device 1 and the client terminal device 3 is installed. As another example, the data generating server 2 may be installed in a country different from the country where at least one of the processing device 1 and the client terminal device 3 is installed.

[0028] The client terminal device 3 is a terminal device that can be used by a user of the processing device 1. The client terminal device 3 may include, for example, at least one of a personal computer, a smartphone, and a tablet terminal.

[0029] (2) Processing Device 1 Next, the processing device 1 included in the processing system SYS will be described.

[0030] (2-1) Configuration of Processing Apparatus 1 First, the configuration of the processing apparatus 1 will be described with reference to Fig. 2 and Fig. 3. Fig. 2 is a perspective view showing the configuration of the processing apparatus 1. Fig. 2 is a block diagram showing the system configuration of the processing apparatus 1.

[0031] 2 and 3 , the processing apparatus 1 includes a processing unit 11, a measurement unit 12, a stage unit 13, and a control unit 14. The processing unit 11, the measurement unit 12, the stage unit 13, and the control unit 14 may be referred to as a processing apparatus, a measurement apparatus, a stage apparatus, and a control apparatus, respectively. The processing unit 11, the measurement unit 12, and the stage unit 13 are housed in a housing 15. However, at least a portion of the processing unit 11, the measurement unit 12, and the stage unit 13 may not be housed in the housing 15. The processing apparatus 1 may not include a housing 15 that houses the processing unit 11, the measurement unit 12, and the stage unit 13.

[0032] The machining unit 11 is capable of removing and machining the workpiece W by irradiating the workpiece W with machining light EL under the control of the control unit 14. Specifically, the machining unit 11 is capable of removing and machining the workpiece W by irradiating the surface of the workpiece W with machining light EL under the control of the control unit 14. In order to machine the workpiece W, the machining unit 11 is equipped with a machining light source 111, a machining head 112, and a head drive system 113.

[0033] The processing light source 111 emits, as processing light EL, at least one of infrared light, visible light, ultraviolet light, and extreme ultraviolet light, under the control of the control unit 14. However, other types of light may be used as the processing light EL.

[0034] The processing light EL includes pulsed light (i.e., multiple pulse beams). Since light is an example of an energy beam, the pulsed light may also be referred to as a pulsed energy beam. In this case, the processing light source 111 may emit pulsed light having a pulse width on the order of femtoseconds, picoseconds, or nanoseconds as the processing light EL. However, the processing light EL does not have to include pulsed light. For example, the processing light EL may be continuous light.

[0035] The processing light EL may be a laser beam. In this case, the processing light source 111 may include a laser light source (for example, a semiconductor laser such as a laser diode (LD)). The laser light source may be a fiber laser, a CO ２ The processing light source 111 may include at least one of a laser, a YAG laser, an excimer laser, etc. However, the processing light EL does not have to be a laser beam. The processing light source 111 may include any light source (for example, at least one of an LED (Light Emitting Diode), a discharge lamp, etc.).

[0036] The machining head 112, under the control of the control unit 14, processes the workpiece W by irradiating the workpiece W with the machining light EL emitted by the machining light source 111. The machining head 112 performs removal processing on the workpiece W placed on a stage 132 (described later). That is, the removal processing is performed on the stage 132. In this case, the machining head 112 may be disposed above the stage 132 on which the workpiece W is placed. For example, the machining head 112 may be attached to a gate-shaped support frame 16 disposed on a base plate 131 provided in the stage unit 13. The support frame 16 may include a pair of leg members 161 protruding from the base plate 131 along the Z-axis direction and a beam member 162 connecting the pair of leg members 161 via the upper ends of the pair of leg members 161. The beam member 162 may be disposed above the stage 132. The machining head 112 may be attached to this beam member 162. 2, the machining head 112 is attached to the beam member 162 via a head drive system 113, which will be described later. When the machining head 112 is disposed above the stage 132, the machining head 112 may irradiate the workpiece W with the machining light EL by emitting the machining light EL downward from the machining head 112. In other words, the machining head 112 may irradiate the workpiece W with the machining light EL traveling along the Z-axis direction by emitting the machining light EL traveling along the Z-axis direction.

[0037] The processing head 112 is equipped with an irradiation optical system 1120 in order to irradiate the workpiece W with the processing light EL. The irradiation optical system 1120 will now be described with reference to Fig. 4. Fig. 4 is a perspective view showing the configuration of the irradiation optical system 1120.

[0038] As shown in FIG. 4, the irradiation optical system 1120 may include, for example, a focus changing optical system 1121, a galvanometer mirror 1122, and an fθ lens 1123.

[0039] The focus-changing optical system 1121 is an optical element that can change the focus position of the processing light EL (i.e., the convergence position of the processing light EL) along the traveling direction of the processing light EL. The focus-changing optical system 1121 may include, for example, multiple lenses aligned along the traveling direction of the processing light EL. In this case, the focus position of the processing light EL may be changed by moving at least one of the multiple lenses along its optical axis direction.

[0040] The processing light EL that passes through the focus change optical system 1121 is incident on the galvanometer mirror 1122. The galvanometer mirror 1122 deflects the processing light EL (i.e., changes the emission angle of the processing light EL), thereby changing the emission direction of the processing light EL from the galvanometer mirror 1122. When the emission direction of the processing light EL from the galvanometer mirror 1122 is changed, the position from which the processing light EL is emitted from the processing head 112 is changed. When the position from which the processing light EL is emitted from the processing head 112 is changed, the position of the target irradiation area EA on the surface of the workpiece W onto which the processing light EL is irradiated is changed.

[0041] The galvanometer mirror 1122 includes, for example, an X-scanning mirror 1122X and a Y-scanning mirror 1122Y. Each of the X-scanning mirror 1122X and the Y-scanning mirror 1122Y is a tilt-angle variable mirror whose angle relative to the optical path of the processing light EL incident on each mirror can be changed. The X-scanning mirror 1122X reflects the processing light EL toward the Y-scanning mirror 1122Y. The X-scanning mirror 1122X can swing or rotate about a rotation axis along the Y-axis. By swinging or rotating the X-scanning mirror 1122X, the processing light EL scans the surface of the workpiece W along the X-axis direction. By swinging or rotating the X-scanning mirror 1122X, the target irradiation area EA moves on the surface of the workpiece W along the X-axis direction. The Y-scanning mirror 1122Y reflects the processing light EL toward the fθ lens 1123. The Y-scanning mirror 1122Y can swing or rotate about a rotation axis along the X-axis. By swinging or rotating the Y scanning mirror 1122Y, the processing light EL scans along the Y-axis direction on the surface of the workpiece W. By swinging or rotating the Y scanning mirror 1122Y, the target irradiation area EA moves on the surface of the workpiece W along the Y-axis direction.

[0042] The galvanometer mirror 1122 allows the processing light EL to scan or sweep a processing area PSA defined relative to the processing head 112. In other words, the galvanometer mirror 1122 allows the target irradiation area EA to move within the processing area PSA defined relative to the processing head 112. The processing area PSA indicates the area (in other words, the range) in which removal processing is performed by the processing head 112 while the positional relationship between the processing head 112 and the workpiece W is fixed (i.e., without change). Typically, the processing area PSA is set to coincide with or be narrower than the scanning range of the processing light EL deflected by the galvanometer mirror 1122 while the positional relationship between the processing head 112 and the workpiece W is fixed. Furthermore, the processing area PSA (target irradiation area EA) can be moved relatively on the surface of the workpiece W by moving the processing head 112 using the head drive system 113 (described later) and / or moving the stage 132 using the stage drive system 133 (described later).

[0043] Depending on the height of the surface of the workpiece W, the processing head 112 may be moved in the Z-axis direction (a direction intersecting with the surface of the workpiece W) by the head drive system 113, the stage 132 may be moved in the Z-axis direction by the stage drive system 133, or the focus position may be changed using a focus changing optical system 1121. At least two of these three methods may be used in combination.

[0044] The fθ lens 1123 is an optical system for emitting the processing light EL from the galvanometer mirror 1122 toward the workpiece W. In particular, the fθ lens 1123 is an optical element that can focus the processing light EL from the galvanometer mirror 1122 on a focusing surface. Therefore, the fθ lens 1123 may also be referred to as a focusing optical system or an objective optical system. The focusing surface of the fθ lens 1123 may be set, for example, on the surface of the workpiece W. Alternatively, the focusing surface of the fθ lens 1123 may be set on a plane away from the surface of the workpiece W in a direction along the optical axis AX of the fθ lens 1123. Note that the focusing surface of the fθ lens 1123 may be set on a plane that includes the rear focal position of the fθ lens 1123. In this case, the galvanometer mirror 1122 may be disposed at the front focal position of the fθ lens 1123. When the galvanometer mirror 1122 includes a plurality of scanning mirrors (for example, an X scanning mirror 1122X and a Y scanning mirror 1122Y), the front focal position of the fθ lens 1123 may be set between the plurality of scanning mirrors).

[0045] 2 and 3 , the head drive system 113, under the control of the control unit 14, moves the machining head 112 (particularly, the irradiation optical system 1120) along at least one of the X-axis direction, Y-axis direction, Z-axis direction, θX direction, θY direction, and θZ direction. Therefore, the head drive system 113 may also be referred to as a moving device. FIG. 2 shows an example in which the head drive system 113 moves the machining head 112 along the Z-axis direction. In this case, the head drive system 113 may include, for example, a Z slider member 1131 extending along the Z-axis direction. The Z slider member 1131 is disposed on a support frame 16 disposed on a surface plate 131 via a vibration isolation device. The Z slider member 1131 is disposed on a beam member 162 via a support member 163 extending along the Z-axis direction, for example. The machining head 112 is connected to the Z slider member 1131 so as to be movable along the Z slider member 1131.

[0046] When the machining head 112 moves, the positional relationship between the machining head 112 and the stage 132, which will be described later, changes. Furthermore, when the machining head 112 moves, the positional relationship between the machining head 112 and the workpiece W placed on the stage 132 changes. Therefore, moving the machining head 112 may be considered equivalent to changing the positional relationships between the machining head 112 and the stage 132 and between the machining head 112 and the workpiece W. Furthermore, when the machining head 112 moves, the target irradiation area EA and the processing area PSA, onto which the processing light EL is irradiated on the surface of the workpiece W, move relative to the surface of the workpiece W.

[0047] The measurement unit 12 is capable of measuring the measurement object under the control of the control unit 14. In order to measure the measurement object, the measurement unit 12 includes a measurement head 121 and a head drive system 122.

[0048] The measurement head 121 is capable of measuring (in other words, measuring) the measurement object under the control of the control unit 14. Specifically, the measurement head 121 is capable of measuring any characteristic of the measurement object. One example of the characteristic of the measurement object is the position of the measurement object. Another example of the characteristic of the measurement object is the shape (e.g., two-dimensional shape or three-dimensional shape) of the measurement object. Another example of the characteristic of the measurement object is at least one of the reflectance of the measurement object, the transmittance of the measurement object, and the surface roughness of the measurement object.

[0049] The measurement object may include a workpiece W. Specifically, the measurement object may include at least one of a workpiece W that has not yet been subjected to removal processing by the processing unit 11, a workpiece W that is in the middle of being subjected to removal processing by the processing unit 11, and a workpiece W that has finished being subjected to removal processing by the processing unit 11. The measurement object may include a stage 132 on which the workpiece W can be placed.

[0050] The measurement head 121 may measure the measurement object using any measurement method. For example, the measurement head 121 may measure the measurement object optically, electrically, magnetically, physically, chemically, or thermally. The measurement head 121 may measure the measurement object without contacting the measurement object. The measurement head 121 may measure the measurement object by contacting the measurement object. In this embodiment, an example will be described in which the measurement head 121 optically measures the measurement object by irradiating the measurement object with measurement light ML without contacting the measurement object. For example, the measurement head 121 may measure the measurement object using a light-section method in which the measurement light ML, which is a slit light, is projected onto the surface of the measurement object and the shape of the projected slit light is measured. For example, the measurement head 121 may measure the measurement object using white light interferometry in which the interference pattern between the measurement light ML, which is white light that passes through the measurement object, and white light that does not pass through the measurement object is measured. For example, the measurement head 121 may measure the measurement object using a pattern projection method in which measurement light ML that draws a light pattern on the surface of the measurement object is projected and the shape of the projected pattern is measured. For example, the measurement head 121 may measure the measurement object using a time-of-flight method in which measurement light ML is projected onto the surface of the measurement object and the distance to the measurement object is measured from the time it takes for the projected measurement light ML to return, and this operation is performed at multiple positions on the measurement object. For example, the measurement head 121 may measure the measurement object using at least one of moire topography (specifically, grating illumination method or grating projection method), holographic interferometry, autocollimation method, stereo method, astigmatism method, critical angle method, knife-edge method, interferometry method, and confocal method.

[0051] The measurement head 121 may be disposed above the stage 132 on which the workpiece W is placed. Specifically, the measurement head 121 may be attached to the beam member 162, similar to the processing head 112. In the example shown in FIG. 2, the measurement head 121 is attached to the beam member 162 via the head drive system 122. When the measurement head 121 is disposed above the stage 132, the measurement head 121 may measure the workpiece W from above the workpiece W. The measurement head 121 may measure the stage 132 from above the stage 132. When the measurement head 121 is disposed above the stage 132, the measurement head 121 may irradiate at least one of the workpiece W and the stage 132 with the measurement light ML by emitting the measurement light ML downward from the measurement head 121. In other words, the measurement head 121 may irradiate at least one of the workpiece W and the stage 132 with the measurement light ML traveling along the Z-axis direction by emitting the measurement light ML traveling along the Z-axis direction.

[0052] The measurement head 121 may include a plurality of measuring instruments each capable of measuring a measurement target. The plurality of measuring instruments may include at least two measuring instruments with different measurement resolutions (in other words, different measurement accuracies). The plurality of measuring instruments may include at least two measuring instruments with different measurement area sizes.

[0053] The head drive system 122 moves the measurement head 121 along at least one of the X-axis direction, the Y-axis direction, the Z-axis direction, the θX direction, the θY direction, and the θZ direction under the control of the control unit 14. For this reason, the head drive system 122 may also be referred to as a moving device. FIG. 2 shows an example in which the head drive system 122 moves the measurement head 121 along the Z-axis direction. In this case, the head drive system 122 may include, for example, a Z slider member 1221 extending along the Z-axis direction. The Z slider member 1221 is disposed on a support frame 16 that is disposed on a surface plate 131 via a vibration isolation device. The Z slider member 1221 is disposed on a beam member 162 via a support member 164 that extends along the Z-axis direction, for example. The measurement head 121 is connected to the Z slider member 1221 so as to be movable along the Z slider member 1221.

[0054] When the measurement head 121 moves, the positional relationship between the measurement head 121 and the stage 132 (described later) changes. Furthermore, when the measurement head 121 moves, the positional relationship between the measurement head 121 and the workpiece W placed on the stage 132 changes. Therefore, moving the measurement head 121 may be considered equivalent to changing the positional relationships between the measurement head 121 and the stage 132 and between the measurement head 121 and the workpiece W.

[0055] The stage unit 13 includes a surface plate 131 , a stage 132 , and a stage drive system 133 .

[0056] The surface plate 131 is placed on the bottom surface of the housing 15 (or on a support surface such as a floor on which the housing 15 is placed). A stage 132 is placed on the surface plate 131. A vibration-isolating device (not shown) for reducing transmission of vibrations from the surface plate 131 to the stage 132 may be installed between the surface plate 131 and the bottom surface of the housing 15 or a support surface such as a floor on which the housing 15 is placed. Furthermore, the support frame 16 described above may be installed on the surface plate 131. Note that leg members may be provided between the surface plate 131 and the bottom surface of the housing 15 (or a support surface such as a floor on which the housing 15 is placed). In this case, vibration-isolating devices may be installed between the leg members and the surface plate 131 and / or between the leg members and the bottom surface (or support surface).

[0057] The stage 132 is a mounting device on which the workpiece W is placed. The stage 132 may be capable of holding the workpiece W placed on the stage 132. Alternatively, the stage 132 may not be capable of holding the workpiece W placed on the stage 132. In this case, the workpiece W may be placed on the stage 132 in a clampless manner. When the stage 132 is capable of holding the workpiece W, the stage 132 may be equipped with at least one of a mechanical chuck, an electrostatic chuck, a vacuum chuck, or the like to hold the workpiece W.

[0058] The stage drive system 133 moves the stage 132 under the control of the control unit 14. For example, the stage drive system 133 may move the stage 132 along at least one of the X axis, Y axis, Z axis, θX direction, θY direction, and θZ direction under the control of the control unit 14. The stage drive system 133 may also be referred to as a moving device.

[0059] In the example shown in FIG. 2 , the stage drive system 133 moves the stage 132 along each of the X-axis and the Y-axis. That is, in the example shown in FIG. 2 , the stage drive system 133 moves the stage 132 along a direction along an XY plane that intersects with the propagation directions of the processing light EL and the measurement light ML. In this case, the stage drive system 133 may include, for example, an X-slide member 1331 extending along the X-axis direction (two X-slide members 1331 arranged parallel to each other in the example shown in FIG. 2 ) and a Y-slide member 1332 extending along the Y-axis direction (one Y-slide member 1332 in the example shown in FIG. 2 ). The two X-slide members 1331 are arranged on the base 131 so as to be aligned along the Y-axis direction. The Y-slide member 1332 is connected to the two X-slide members 1331 so as to be movable along the two X-slide members 1331. The stage 132 is connected to the Y-slide member 1332 so as to be movable along the Y-slide member 1332. 2, a plurality of X slide members 1331 are provided, but there may be only one X slide member 1331. Furthermore, the stage 132 may be supported by floating above the base 131 by an air bearing.

[0060] When the stage drive system 133 moves the stage 132, the positional relationships between the processing head 112 and the measurement head 121, respectively, and the stage 132 and workpiece W, respectively, change. For this reason, the stage drive system 133 may be considered to be capable of functioning as a position changing device that can change the positional relationships between the processing head 112 and the measurement head 121, respectively, and the stage 132 and workpiece W, respectively. Furthermore, when the stage drive system 133 moves the stage 132, the stage 132 and workpiece W each move relative to the processing area PSA where the processing head 112 performs removal processing and the measurement area where the measurement head 121 performs measurement, respectively.

[0061] The control unit 14 controls the operation of the machining apparatus 1. For example, the control unit 14 may generate machining control information (e.g., machining path information) for machining the workpiece W, and may control the machining unit 11 and the stage unit 13 based on the machining control information so that the workpiece W is machined in accordance with the generated machining control information. In other words, the control unit 14 may control the machining of the workpiece W. For example, the control unit 14 may generate measurement control information for measuring a measurement object, and may control the machining unit 11 and the stage unit 13 based on the measurement control information so that the measurement object is measured in accordance with the generated measurement control information. In other words, the control unit 14 may control the measurement of the measurement object.

[0062] The control unit 14 may generate at least one of the processing control information and the measurement control information based on the control data generated by the data generating server 2. In this case, the control unit 14 may acquire the control data from the data generating server 2 via the communication network 4. Alternatively, the control unit 14 may acquire the control data via the communication network 4 from the client terminal device 3 that has acquired the control data from the data generating server 2.

[0063] The control unit 14 may include, for example, an arithmetic device 141 and a storage device 142. The arithmetic device 141 may include, for example, at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The control unit 14 functions as a device that controls the operation of the machining device 1 by the arithmetic device 141 executing a computer program. This computer program is a computer program for causing the control unit 14 (e.g., the arithmetic device 141) to perform (i.e., execute) the operations to be performed by the control unit 14, which will be described later. In other words, this computer program is a computer program for causing the control unit 14 to function so as to cause the machining device 1 to perform the operations to be performed later. The computer program executed by the arithmetic device 141 may be recorded in a storage device 142 (i.e., a recording medium) included in the control unit 14, or may be recorded in any storage medium (e.g., a hard disk or semiconductor memory) built into the control unit 14 or externally attachable to the control unit 14. Alternatively, the computing device 141 may download the computer program to be executed from a device external to the control unit 14 via a network interface.

[0064] The control unit 14 does not have to be provided inside the processing apparatus 1. For example, the control unit 14 may be provided as a server or the like outside the processing apparatus 1. For example, the control unit 14 may be provided as a computer (e.g., a laptop computer) connectable to the processing apparatus 1. For example, the control unit 14 may be provided as a computer (e.g., a laptop computer) installed near the processing apparatus 1. In this case, the control unit 14 and the processing apparatus 1 may be connected via a wired and / or wireless network (or a data bus and / or a communication line). As the wired network, for example, a network using a serial bus interface represented by at least one of IEEE 1394, RS-232x, RS-422, RS-423, RS-485, and USB may be used. As the wired network, a network using a parallel bus interface may be used. The wired network may use a network using an interface compliant with Ethernet (registered trademark), such as at least one of 10BASE-T, 100BASE-TX, and 1000BASE-T. The wireless network may use a network using radio waves. An example of a network using radio waves is a network compliant with IEEE 802.1x (e.g., at least one of wireless LAN and Bluetooth (registered trademark)). The wireless network may use a network using infrared rays. The wireless network may use a network using optical communication. In this case, the control unit 14 and the processing device 1 may be configured to be able to send and receive various information via the network. Furthermore, the control unit 14 may be capable of sending information such as commands and control parameters to the processing device 1 via the network. The processing device 1 may be equipped with a receiving device that receives information such as commands and control parameters from the control unit 14 via the network. Alternatively, a first control device that performs part of the processing performed by the control unit 14 may be provided inside the processing device 1, while a second control device that performs another part of the processing performed by the control unit 14 may be provided outside the processing device 1.

[0065] The control unit 14 may be capable of functioning as the client terminal device 3. For example, a computer may be used as both the control unit 14 and the client terminal device 3. In other words, the control unit 14 and the client terminal device 3 may be an integrated device (or an integrated system). However, typically, two different computers may be used as the control unit 14 and the client terminal device 3, respectively.

[0066] A computational model that can be constructed by machine learning may be implemented in the control unit 14 by the arithmetic device 141 executing a computer program. An example of a computational model that can be constructed by machine learning is a computational model including a neural network (so-called artificial intelligence (AI)). In this case, learning of the computational model may include learning of parameters of the neural network (e.g., at least one of a weight and a bias). The control unit 14 may control the operation of the processing device 1 using the computational model. In other words, the operation of controlling the operation of the processing device 1 may include the operation of controlling the operation of the processing device 1 using the computational model. Note that the control unit 14 may be implemented with a computational model that has been constructed by offline machine learning using training data. Furthermore, the computational model implemented in the control unit 14 may be updated on the control unit 14 by online machine learning. Alternatively, the control unit 14 may control the operation of the processing device 1 using a calculation model implemented in a device external to the control unit 14 (i.e., a device provided outside the processing device 1) in addition to or instead of the calculation model implemented in the control unit 14.

[0067] The recording medium for recording the computer program executed by the arithmetic device 141 may be at least one of the following: a CD-ROM, CD-R, CD-RW, a flexible disk, an MO, a DVD-ROM, a DVD-RAM, a DVD-R, a DVD+R, a DVD-RW, a DVD+RW, and an optical disk such as Blu-ray (registered trademark), a magnetic medium such as a magnetic tape, a magneto-optical disk, a semiconductor memory such as a USB memory, and any other medium capable of storing a program. The recording medium may include a device capable of recording a computer program (for example, a general-purpose device or a dedicated device in which a computer program is implemented in an executable state in at least one form such as software or firmware). Furthermore, each process or function included in the computer program may be realized by a logical processing block realized within the control unit 14 when the control unit 14 (i.e., the computer) executes the computer program, or may be realized by hardware such as a predetermined gate array (FPGA, Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit) provided in the control unit 14, or may be realized in a form that combines logical processing blocks and partial hardware modules that realize some elements of the hardware.

[0068] (2-2) Removal Processing Performed by Processing Apparatus 1 Next, an example of removal processing using processing light EL will be described with reference to Fig. 5(a) to Fig. 5(c), each of which is a cross-sectional view showing the state of removal processing performed on the workpiece W.

[0069] As shown in FIG. 5A, the processing device 1 irradiates a target irradiation area EA set (i.e., formed) on the surface of the workpiece W with the processing light EL. When the target irradiation area EA is irradiated with the processing light EL, the energy of the processing light EL is transmitted to the target irradiation area EA and a portion of the workpiece W adjacent to the target irradiation area EA. When heat resulting from the energy of the processing light EL is transmitted, the heat resulting from the energy of the processing light EL melts the material constituting the target irradiation area EA and a portion of the workpiece W adjacent to the target irradiation area EA. The molten material scatters as droplets. Alternatively, the molten material evaporates due to the heat resulting from the energy of the processing light EL. As a result, the portion of the workpiece W adjacent to the target irradiation area EA is removed. That is, as shown in FIG. 5B, a recess (i.e., a groove) having a depth corresponding to the reference processing amount Δz resulting from the processing light EL is formed on the surface of the workpiece W. Here, if the processing light EL is pulsed light, the reference processing amount Δz may be the depth of removal of the workpiece W when the processing light EL is irradiated onto the workpiece W by a unit number of pulses. Here, the unit pulse number may be 1 or a number greater than 1. Furthermore, since the reference processing amount Δz is the amount of processing performed by irradiating the processing light EL with a unit number of pulses, it may also be referred to as a unit processing amount. In this case, the processing device 1 may be considered to be performing removal processing of the workpiece W using the principle of so-called thermal processing. Note that, when thermal processing is performed, the processing light EL may include pulsed light or continuous light having an emission time of milliseconds or more. Note that, if the processing light EL is continuous light, the reference processing amount Δz may be the amount of processing of the workpiece W (depth of removal) when the processing light EL of a unit energy is irradiated onto the workpiece W for a unit time.

[0070] On the other hand, depending on the characteristics of the processing light EL, the processing apparatus 1 may also process the workpiece W using the principle of non-thermal processing (e.g., ablation processing). That is, the processing apparatus 1 may perform non-thermal processing (e.g., ablation processing) on ​​the workpiece W. For example, when light with a high photon density (in other words, fluence) is used as the processing light EL, the material constituting the target irradiation area EA and the portion adjacent to the target irradiation area EA of the workpiece W instantaneously evaporates and disperses. That is, the material constituting the target irradiation area EA and the portion adjacent to the target irradiation area EA of the workpiece W evaporates and disperses within a time sufficiently shorter than the thermal diffusion time of the workpiece W. In this case, the material constituting the target irradiation area EA and the portion adjacent to the target irradiation area EA of the workpiece W may be released from the workpiece W as at least one of ions, atoms, radicals, molecules, clusters, and solid fragments. When non-thermal processing is performed, the processing light EL may include pulsed light having an emission time of picoseconds or less (or, in some cases, nanoseconds or femtoseconds or less). When pulsed light having an emission time of picoseconds or less (or, in some cases, nanoseconds or femtoseconds or less) is used as the processing light EL, the material constituting the target irradiation area EA and the portion of the workpiece W adjacent to the target irradiation area EA may sublimate without passing through a molten state. Therefore, it is possible to process the workpiece W while minimizing the effect on the workpiece W of heat caused by the energy of the processing light EL.

[0071] The processing apparatus 1 uses the above-described galvanometer mirror 1122 to move the target irradiation area EA on the surface of the workpiece W. That is, the processing apparatus 1 scans the surface of the workpiece W with the processing light EL. As a result, as shown in FIG. 5C, at least a portion of the surface of the workpiece W is removed along the scanning trajectory of the processing light EL (i.e., the movement trajectory of the target irradiation area EA). Therefore, the processing apparatus 1 can appropriately remove the portion of the workpiece W that is to be removed by scanning the surface of the workpiece W with the processing light EL along the desired scanning trajectory corresponding to the area to be removed. The processing apparatus 1 can appropriately remove the portion of the workpiece W that is to be removed by scanning the surface of the workpiece W with the processing light EL along the desired scanning trajectory corresponding to the area to be removed. The processing apparatus 1 can remove the removal layer SL, which has a thickness corresponding to the reference processing amount Δz and is the portion to be removed, from the workpiece W by scanning the surface of the workpiece W with the processing light EL along the desired scanning trajectory corresponding to the area to be removed.

[0072] The processing apparatus 1 may process the workpiece W so that the shape of the workpiece W becomes a desired shape by repeating the operation of removing the removal layer SL having a thickness corresponding to the reference processing amount Δz. In the following explanation, for convenience of explanation, an example will be described in which the shape of the workpiece W, which has a rectangular parallelepiped shape shown on the left side of Fig. 6, becomes the shape of a cone protruding from a plate-like member shown on the right side of Fig. 6. In this case, the processing apparatus 1 performs removal processing to remove the removal target portion W_rmv to be removed from the workpiece W, as shown in Fig. 7.

[0073] 6 shows an example in which the surface of the workpiece W before removal processing (particularly, the surface of the workpiece W on which removal processing is performed) is a plane along the XY plane. However, the surface of the workpiece W (particularly, the surface of the workpiece W on which removal processing is performed) may include a plane inclined with respect to the XY plane. The surface of the workpiece W (particularly, the surface of the workpiece W on which removal processing is performed) may include a curved surface.

[0074] To perform the removal process to remove the removal target portion W_rmv, as shown in FIG. 8 , the processing apparatus 1 sequentially removes multiple removal layers SL obtained by slicing the removal target portion W_rmv along the Z-axis direction. Each removal layer SL may be considered to correspond to a removal portion removed by scanning the processing light EL while the positional relationship between the workpiece W and the processing head 112 in the Z-axis direction (particularly, the positional relationship between the workpiece W and the focus position of the processing light EL) is fixed. In this case, the processing apparatus 1 first performs the removal process to remove the top removal layer SL#1 from the workpiece W. Then, the processing apparatus 1 moves the stage 132 and / or the processing head 112 so that the processing head 112 approaches the workpiece W by a distance corresponding to the reference processing amount Δz. Alternatively, the processing apparatus 1 controls the irradiation optical system 1120 (particularly, the focus change optical system 1121) so that the focus position of the processing light EL approaches the workpiece W by a distance corresponding to the reference processing amount Δz. Thereafter, the processing device 1 performs a removal process to remove the second removal layer SL#2 from the workpiece W from which the first removal layer SL#1 has been removed. Thereafter, the processing device 1 repeats the same operation until all removal layers SL (for example, in the example shown in FIG. 8 , Q (where Q is a constant representing an integer of 2 or greater) removal layers SL#1 to SL#Q) are removed. As a result, the shape of the workpiece W, which had a rectangular parallelepiped shape shown on the left side of FIG. 6 , changes to the shape of a cone protruding from a plate-like member shown on the right side of FIG. 6 .

[0075] The processing apparatus 1 may sequentially remove multiple removal layers SL based on slice data indicating the area on the surface of the workpiece W where removal processing is performed during the process of removing each removal layer SL. The slice data may be an example of processing control information for processing the workpiece W. For example, the processing apparatus 1 may remove removal layer SL#1 by irradiating the area on the surface of the workpiece W indicated by the first slice data (see FIG. 9( a)) indicating the area on the surface of the workpiece W where removal processing is performed during the process of removing removal layer SL#1 with the processing light EL. Then, the processing apparatus 1 may remove removal layer SL#2 by irradiating the area on the surface of the workpiece W indicated by the second slice data (see FIG. 9( b)) indicating the area on the surface of the workpiece W where removal processing is performed during the process of removing removal layer SL#2 with the processing light EL. Thereafter, the processing device 1 may remove the removal layer SL#3 by irradiating the area on the surface of the workpiece W indicated by the third slice data (see FIG. 9(c)), which indicates the area on the surface of the workpiece W where removal processing is performed in the process of removing the removal layer SL#3, with the processing light EL. The processing device 1 then repeats the same operation. Finally, the processing device 1 may remove the removal layer SL#Q by irradiating the area on the surface of the workpiece W indicated by the Qth slice data (see FIG. 9(d)), which indicates the area on the surface of the workpiece W where removal processing is performed in the process of removing the removal layer SL#Q. Note that in each of FIGS. 9(a) to 9(d), the area on the surface of the workpiece W where removal processing is performed (i.e., the area irradiated with the processing light EL) is indicated by hatching.

[0076] The processing device 1 scans the processing area PSA set on the surface of the workpiece W along the XY plane with the processing light EL. In this case, as shown in FIG. 10(a), it can be considered that the number of irradiation target positions (irradiation positions) C that can be irradiated with the processing light EL is set within the processing area PSA according to the scanning speed and pulse frequency of the processing light EL. Furthermore, the processing device 1 moves at least one of the processing head 112 and the stage 132 to move the processing area PSA along the XY plane on the surface of the workpiece W. Therefore, as shown in FIG. 10(b), in the process of removing each removal layer SL, it can be considered that the number of irradiation target positions C that can be irradiated with the processing light EL is set within the processing area PSA, which is a plane along the scanning direction of the processing light EL and the movement direction of the processing area PSA within the processing area PSA and is within the processing surface PL (typically a plane along the XY plane) on which the processing device 1 performs removal processing, according to the number of irradiation target positions C set within the processing area PSA and the number of times the processing area PSA moves on the surface of the workpiece W (i.e., the number of times the processing area PSA is set on the surface of the workpiece W). In this case, in the process of removing each removal layer SL, it may be considered that a plurality of irradiation target positions C are set on the surface of the workpiece W. In the following description, for convenience of explanation, as shown in FIG. 10(b), N (where N is a constant indicating an integer of 2 or more) irradiation target positions C (specifically, irradiation target positions C １ From C Ｎ ) is set will be described.

[0077] 10(a) and 10(b) schematically represent each irradiation target position C using a beam spot (i.e., a circular beam spot) of the processing light EL irradiated on the irradiation target position C. The beam spot of the processing light EL may refer to an area irradiated with the processing light EL whose intensity exceeds a predetermined intensity threshold. In this case, when the processing light EL is irradiated on each irradiation target position C, the area outside the circle schematically showing each irradiation target position C may be irradiated with the processing light EL whose intensity does not exceed a predetermined intensity threshold. Alternatively, the beam spot of the processing light EL may refer to an area corresponding to the full width at half maximum of the intensity distribution (e.g., Gaussian distribution) of the processing light EL.

[0078] 10(a) and 10(b) show an example in which one irradiation target position C does not overlap with another irradiation target position C adjacent to the one irradiation target position C, for ease of viewing the drawings. However, in reality, one irradiation target position C may at least partially overlap with another irradiation target position C adjacent to the one irradiation target position C. In other words, the beam spot of the processing light EL irradiated on one irradiation target position C may at least partially overlap with the beam spot of the processing light EL irradiated on another irradiation target position C adjacent to the one irradiation target position C.

[0079] The slice data is the irradiation target position C １ From C Ｎ 9(d) may be regarded as equivalent to information indicating whether or not the processing light EL is irradiated to each of the irradiation target positions C １ From C Ｎ At least one irradiation target position C included in the hatched area shown in FIG. 9(d) should be irradiated with the processing light EL, while the irradiation target position C １ From C Ｎ 9D. The information may be regarded as equivalent to information indicating that the processing light EL should not be irradiated to at least one other irradiation target position C that is not included in the hatched area shown in FIG. 9D.

[0080] The data generating server 2 may generate control data for generating slice data shown in Fig. 9(a) to Fig. 9(d). For example, each of the Q slice data is １ From C Ｎ In other words, each of the Q slice data indicates whether or not the processing light EL is irradiated to each of the Q slice data. １ From C Ｎ It can be said that the number of times to irradiate each of the Q slice data with the processing light EL is 1 or 0 (or some other number). Therefore, the Q slice data as a whole substantially corresponds to the irradiation target position C １ From C ＮIn this case, the data generating server 2 determines the number of times that the processing light EL should be irradiated to each of the irradiation target positions C １ From C Ｎ In this case, the control unit 14 may generate control data indicating the number of times to irradiate each of the Q removal layers SL with the processing light EL based on the control data, and sequentially remove the Q removal layers SL based on the Q slice data.

[0081] As described above, since the processing light EL is pulsed light, the number of times the processing light EL should be irradiated onto the irradiation target position C may refer to the number of times the unit pulse number of the processing light EL should be irradiated onto the irradiation target position C. In other words, the number of times the processing light EL should be irradiated onto the irradiation target position C may refer to the number of times the pulses of the pulsed light contained in the processing light EL should be irradiated onto the irradiation target position C. In other words, the number of times the processing light EL should be irradiated onto the irradiation target position C may refer to the number of pulses to be irradiated onto the irradiation target position C. In the following description, taking into account that the processing light EL is pulsed light, for convenience of explanation, the number of times the processing light EL should be irradiated onto the irradiation target position C will be referred to as the target pulse number. The target pulse number is a specific example of the processing conditions of the above-mentioned processing device 1.

[0082] (3) Data Generation Server 2 Next, the data generation server 2 included in the machining system SYS will be described. The data generation server 2 that generates control data indicating the target pulse number, which is an example of the machining conditions, will be described below. That is, the irradiation target position C １ the target pulse number, which is the number of times the processing light EL is irradiated to the target position C ２ a target pulse number, which is the number of times the processing light EL is irradiated to the target position C Ｎ The data generating server 2 generates, as at least a part of the control data, target pulse number information indicating the target pulse number, which is the number of times the processing light EL is irradiated onto the target object.

[0083] (3-1) Configuration of Data Generating Server 2 First, the configuration of the data generating server 2 will be described with reference to Fig. 11. Fig. 11 is a block diagram showing the configuration of the data generating server 2.

[0084] 11 , the data generating server 2 includes a calculation device 21, a storage device 22, and a communication device 23. The data generating server 2 may further include an input device 24 and an output device 25. However, the data generating server 2 does not necessarily have to include at least one of the input device 24 and the output device 25. The calculation device 21, the storage device 22, the communication device 23, the input device 24, and the output device 25 may be connected via a data bus 26.

[0085] The arithmetic device 21 includes, for example, at least one of a CPU and a GPU. The arithmetic device 21 loads a computer program. For example, the arithmetic device 21 may load a computer program stored in the storage device 22. For example, the arithmetic device 21 may load a computer program stored in a computer-readable, non-transitory storage medium using a storage medium reading device (not shown). The arithmetic device 21 may acquire (i.e., download or read) a computer program from a device (not shown) located outside the data generating server 2 via the communication device 23. That is, the arithmetic device 21 may acquire (i.e., download or read) a computer program stored in a storage device of a device (not shown) located outside the data generating server 2 via the communication device 23. The arithmetic device 21 executes the loaded computer program. As a result, logical functional blocks for executing operations to be performed by the data generating server 2 (e.g., data generation operations for generating control data) are realized within the arithmetic device 21. That is, the arithmetic device 21 can function as a controller for realizing logical functional blocks for executing the operations to be performed by the data generating server 2. In this case, any device (typically, a computer) that executes a computer program can function as the data generating server 2.

[0086] 11 shows an example of a logical functional block realized in the arithmetic device 21 for performing an operation of generating control data indicating a target number of pulses. As shown in FIG. 11, a model generation unit 211 and a control data generation unit 212 are realized in the arithmetic device 21. The operations of the model generation unit 211 and the control data generation unit 212 will be described in detail later with reference to FIG. 12 etc., but an outline thereof will be briefly explained below. The model generation unit 211 generates model information LM. The model information LM is used to generate an irradiation target position C for machining the workpiece W. １ From C Ｎ The control data generating unit 212 calculates the target number of pulses based on the model information LM. Specifically, the control data generating unit 212 calculates the target number of pulses for processing the workpiece W so that the shape of the workpiece W becomes the desired target shape, based on the model information LM. In other words, the control data generating unit 212 generates control data indicating the target number of pulses.

[0087] A computational model that can be constructed by machine learning may be implemented in the computational device 21 by the computational device 21 executing a computer program. An example of a computational model that can be constructed by machine learning is a computational model including a neural network (so-called artificial intelligence (AI)). In this case, learning of the computational model may include learning of parameters of the neural network (e.g., at least one of a weight and a bias). The computational device 21 may perform a data generation operation using the computational model. That is, the data generation operation may include an operation of generating control data using the computational model. That is, at least one of the model generation unit 211 and the control data generation unit 212 may be realized using the computational model. In other words, the operation performed by at least one of the model generation unit 211 and the control data generation unit 212 may be performed by the computational model. Note that a computational model that has been constructed by offline machine learning using teacher data may be implemented in the computational device 21. Furthermore, the computational model implemented in the computational device 21 may be updated by online machine learning on the computational device 21. Alternatively, the computational device 21 may generate control data using a computational model implemented in a device external to the computational device 21 (that is, a device provided outside the data generating server 2), in addition to or instead of the computational model implemented in the computational device 21.

[0088] Note that at least some of the functional blocks (i.e., the model generation unit 211 and the control data generation unit 212) in the arithmetic device 21 of the data generating server 2 may not be included in the arithmetic device 21 (i.e., the data generating server 2). For example, the client terminal device 3 may include at least some of the functional blocks (i.e., the model generation unit 211 and the control data generation unit 212) in the arithmetic device 21. For example, the processing device 1 (e.g., the control unit 14) may include at least some of the functional blocks (i.e., the model generation unit 211 and the control data generation unit 212) in the arithmetic device 21.

[0089] The storage device 22 can store desired data. For example, the storage device 22 may temporarily store a computer program executed by the arithmetic device 21. The storage device 22 may temporarily store data temporarily used by the arithmetic device 21 when the arithmetic device 21 is executing a computer program. The storage device 22 may also store data that the data generation server 2 stores long-term. In this embodiment, the storage device 22 may store model information LM generated by the model generation unit 211. The storage device 22 may include at least one of a RAM (Random Access Memory), a ROM (Read Only Memory), a hard disk device, a magneto-optical disk device, an SSD (Solid State Drive), and a disk array device. In other words, the storage device 22 may include a non-temporary recording medium.

[0090] The communication device 23 is capable of communicating with at least one of the processing device 1 and the client terminal device 3 via the communication network 4. In this embodiment, the communication device 23 is capable of receiving, from at least one of the processing device 1 and the client terminal device 3 via the communication network 4, reference information that the data generating server 2 refers to in order to generate control data. Furthermore, the communication device 23 is capable of transmitting the generated control data to at least one of the processing device 1 and the client terminal device 3 via the communication network 4.

[0091] The input device 24 is a device that accepts information input to the data generating server 2 from outside the data generating server 2. For example, the input device 24 may include an operation device (e.g., at least one of a keyboard, a mouse, and a touch panel) that can be operated by a server user. For example, the input device 24 may include a reading device that can read information recorded as data on a recording medium that can be attached externally to the data generating server 2.

[0092] The output device 25 is a device that outputs information to the outside of the data generating server 2. For example, the output device 25 may output information as an image. That is, the output device 25 may include a display device (a so-called display) that can display an image showing the information to be output. For example, the output device 25 may output information as sound. That is, the output device 25 may include an audio device (a so-called speaker) that can output sound. For example, the output device 25 may output information on paper. That is, the output device 25 may include a printing device (a so-called printer) that can print desired information on paper.

[0093] The data generating server 2 may be capable of functioning as the client terminal device 3. For example, a computer may be used both as the data generating server 2 and as the client terminal device 3. In other words, the data generating server 2 and the client terminal device 3 may be an integrated device (or an integrated system). However, typically, two different computers may be used as the data generating server 2 and the client terminal device 3, respectively.

[0094] (3-2) Data Generation Operation Performed by Data Generating Server 2 Next, the data generation operation (i.e., the operation of generating control data) performed by the data generating server 2 will be described with reference to Fig. 12. Fig. 12 is a flowchart showing the flow of the data generation operation (i.e., the operation of generating control data) performed by the data generating server 2.

[0095] 12 , the data generating server 2 first generates model information LM to be used for generating control data (steps S11 to S15). In this embodiment, the data generating server 2 may generate the model information LM by performing machine learning using training data. In this case, the data generating server 2 first acquires, as training data, information on the results of test machining actually performed by the machining device 1 on a test workpiece W.

[0096] In the following description, the workpiece W for test machining will be referred to as test workpiece W_test, and the workpiece W that will actually be machined using the control data generated by the data generating server 2 will be referred to as the workpiece W_target to be machined, to distinguish between the two. The test workpiece W_test may be a workpiece W of the same specifications as the workpiece W_target to be machined. The test workpiece W_test may be a workpiece W of the same size as the workpiece W_target to be machined. The test workpiece W_test may be a workpiece W of the same shape as the workpiece W_target to be machined. The test workpiece W_test may be a workpiece W made from the same type of material as the workpiece W_target to be machined.

[0097] In order to obtain the results of the test machining performed by the machining device 1 on the workpiece W for test machining as training data, the model generation unit 211 of the data generation server 2 obtains test machining condition information indicating the machining conditions of the machining device 1 performing the test machining (step S11). That is, the model generation unit 211 obtains test machining condition information regarding the machining conditions of the machining device 1 performing the test machining (step S11). The test machining condition information indicates the machining conditions for machining the test workpiece W_test so that the shape of the test workpiece W_test becomes the desired test target shape. That is, the test machining condition information indicates the machining conditions of the test machining performed on the test workpiece W_test so that the shape of the test workpiece W_test becomes the desired test target shape.

[0098] The test target shape is the same as the machining target shape, which is the shape that the workpiece W_target to be machined should have after machining. In other words, the test target shape is the same as the machining target shape, which is the target shape of the workpiece W_target to be machined after machining. However, the test target shape may be a shape different from the machining target shape. When test machining is performed multiple times using multiple different test machining condition information as described below, at least one of test machining for machining the test workpiece W_test so that the shape of the test workpiece W_test becomes a test target shape that is the same as the machining target shape, and test machining for machining the test workpiece W_test so that the shape of the test workpiece W_test becomes a test target shape that is different from the machining target shape may be performed.

[0099] As described above, the machining apparatus 1 may form a riblet structure on the workpiece W by machining the workpiece W to be machined. An example of a riblet structure is shown in FIGS. 13( a) and 13(b). As shown in FIGS. 13(a) and 13(b), the riblet structure may include a structure in which grooves GV extending in a first direction (e.g., the X-axis direction) along the surface of the workpiece W to be machined are arranged in a plurality of rows along a second direction (e.g., the Y-axis direction) along the surface of the workpiece W to be machined and intersecting the first direction. In other words, the riblet structure may include a structure in which convex structures LD extending in a first direction (e.g., the X-axis direction) along the surface of the workpiece W to be machined are arranged in a plurality of rows along a second direction (e.g., the Y-axis direction) along the surface of the workpiece W to be machined and intersecting the first direction. In this case, at least one of the test target shape and the machining target shape may be a shape corresponding to a riblet structure. For example, the three-dimensional shape of the riblet structure may be used as at least one of the test target shape and the processing target shape. For example, a three-dimensional profile indicating the three-dimensional shape of the riblet structure (e.g., a three-dimensional profile indicating the three-dimensional shape shown in FIG. 13( a)) may be used as at least one of the test target shape and the processing target shape.

[0100] At least one of the test target shape and the machining target shape may have a shape whose height changes in at least one direction. Specifically, the test target shape may have a shape whose height of the test workpiece W_test (e.g., the height of the surface of the test workpiece W_test) changes in at least one direction, and the machining target shape may have a shape whose height of the workpiece W_target (e.g., the height of the surface of the workpiece W_target) changes in at least one direction. For example, as shown in FIGS. 13( a) and 13(b), the shape of the above-described riblet structure is a shape whose height (i.e., the size in a third direction intersecting the first and second directions, for example, the size in the Z-axis direction) changes along the second direction in which the multiple grooves GV (multiple convex structures LD) are arranged. In this case, at least one of the test target shape and the machining target shape may have a shape whose height changes along the second direction in which the multiple grooves GV (multiple convex structures LD) are arranged. In other words, the test target shape may be a cross-sectional profile (in other words, a height profile or two-dimensional profile, for example, a two-dimensional profile showing the two-dimensional shape shown in Figure 13(b)) that shows the height that changes along the second direction in which the multiple grooves GV (multiple convex structures LD) are arranged.

[0101] In this embodiment, the test machining condition information includes at least test pulse number information indicating the number of test pulses, which is the number of times that the processing light EL should be irradiated onto the irradiation target position C for test machining. Specifically, as described above, on the processing surface PL of the workpiece W, there are N irradiation target positions C. １ From C Ｎ Therefore, N irradiation target positions C are also set on the processing surface PL of the test workpiece W_test, which is the workpiece W. １ From C Ｎ In this case, the test processing condition information is set as the irradiation target position C １ the number of test pulses, which is the number of times the processing light EL is irradiated to the target position C ２ the number of test pulses, which is the number of times the processing light EL is irradiated to the target irradiation position C ＮThe test pulse number may include test pulse number information indicating the number of times the processing light EL is irradiated to the target object.

[0102] The model generation unit 211 may acquire test processing condition information input to the data generation server 2 using the input device 24. For example, the model generation unit 211 may acquire test processing condition information input to the data generation server 2 by a user of the processing system SYS using the input device 24.

[0103] The model generation unit 211 may acquire the test processing condition information transmitted to the data generation server 2 using the communication device 23. For example, the model generation unit 211 may acquire the test processing condition information by receiving the test processing condition information stored in a server external to the data generation server 2 using the communication device 23.

[0104] The model generation unit 211 may acquire the test machining condition information by generating test machining condition information. For example, the model generation unit 211 may generate test machining condition information indicating the set machining conditions by setting machining conditions for machining the test workpiece W_test so that the shape of the test workpiece W_test becomes a desired test target shape.

[0105] Referring again to FIG. 12 , after the test machining condition information is acquired in step S11, the test workpiece W_test is machined based on the test machining condition information acquired in step S11 (step S12). Specifically, the model generation unit 211 controls the machining device 1 to machine the test workpiece W_test based on the test machining condition information. For example, the model generation unit 211 may transmit the test machining condition information to the machining device 1 as control data. The machining device 1 (particularly, the control unit 14) may generate machining control information for controlling the machining device 1 to perform test machining based on the test machining condition information transmitted from the data generation server 2. For example, the control unit 14 may generate machining control information for controlling the machining device 1 to machine the test workpiece W_test so that the shape of the test workpiece W_test becomes the desired test target shape. Thereafter, the machining device 1 may machine the test workpiece W_test based on the generated machining control information. In other words, the machining device 1 may perform test machining of the test workpiece W_test based on the generated machining control information.

[0106] After test machining of the test workpiece W_test is performed in step S12, the test workpiece shape, which is the shape of the test workpiece W_test after test machining, is measured (step S13). That is, the test workpiece shape, which is the shape of the test workpiece W_test after machining based on the test machining condition information acquired in step S11, is measured (step S13). For example, the machining device 1 may measure the test workpiece shape using the measurement unit 12. Then, test workpiece shape information (test shape information) indicating the measurement results of the test workpiece shape by the measurement unit 12 may be transmitted from the machining device 1 to the data generation server 2 via the communication network 4. The model generation unit 211 of the data generation server 2 may acquire the test workpiece shape information (test shape information) indicating the measurement results of the test workpiece shape by the measurement unit 12.

[0107] The acquired test work shape information is information regarding the results of test machining performed on the test work W_test by the machining device 1. In other words, the model generation unit 211 acquires the test work shape information as training data for generating model information LM. More specifically, the model generation unit 211 acquires training data for generating model information LM, which is data that associates the test machining condition information acquired in step S11 with test work shape information regarding the results of test machining performed based on the test machining condition information.

[0108] Thereafter, the model generation unit 211 determines whether to end the test processing (step S14). For example, the model generation unit 211 may determine to end the test processing when the number of acquired teacher data exceeds a predetermined number of samples indicating the number of teacher data necessary to generate the model information LM. On the other hand, the model generation unit 211 may determine not to end the test processing when the number of acquired teacher data is less than the predetermined number of samples. Typically, it is preferable to use multiple teacher data to generate the model information LM. Therefore, it is preferable that the predetermined number of samples is two or more.

[0109] If it is determined in step S14 that the test machining should not be terminated (step S14: No), the operations from step S11 to step S13 are performed again. That is, the model generation unit 211 acquires new test machining condition information (step S11), performs test machining of the test workpiece W_test based on the newly acquired test machining condition information (step S12), and measures the test workpiece shape, which is the shape of the test workpiece W_test that has been test machined based on the newly acquired test machining condition information (step S13).

[0110] Thus, in this embodiment, test machining may be performed multiple times to obtain multiple training data. For example, first test machining condition information may be acquired (step S11), test machining of a first test workpiece W_test based on the first test machining condition information may be performed (step S12), and a first test workpiece shape, which is the shape of the first test workpiece W_test after test machining based on the first test machining condition information, may be measured (step S13). As a result, first training data may be acquired in which the first test machining condition information and test workpiece shape information indicating the first test workpiece shape are associated. Then, second test machining condition information different from the first test machining condition information may be acquired (step S11), test machining of a second test workpiece W_test based on the second test machining condition information may be performed (step S12), and a second test workpiece shape, which is the shape of the second test workpiece W_test after test machining based on the second test machining condition information, may be measured (step S13). The second test workpiece W_test may be different from the first test workpiece W_test. As a result, second teacher data is acquired in which the second test machining condition information and test workpiece shape information indicating the second test workpiece shape are associated. This operation is repeated until the required number of teacher data have been acquired.

[0111] On the other hand, if the result of the judgment in step S14 is that the test processing is to be terminated (step S14: Yes), the model generation unit 211 generates model information LM by performing machine learning using a teacher data set including the acquired multiple teacher data.

[0112] As described above, the model information LM is the irradiation target position C １ From C Ｎ , and the shape of the workpiece W after machining. In this case, the model generation unit 211 calculates the model information LM based on the irradiation target position C １ From C ＮMachine learning may be performed using training data in which the test machining condition information (the number of test pulses) and the test work shape information are associated so that the model information LM becomes information that appropriately indicates the relationship between the number of times the machining light EL is irradiated to each of the test workpieces (i.e., the number of test pulses) and the test workpiece shape, which is the shape of the test workpiece W_test after machining. In other words, the model generation unit 211 may perform machine learning using training data in which the test machining condition information and the test workpiece shape information are associated so that the model information LM becomes information that appropriately indicates the relationship between the test machining condition information and the test workpiece shape information.

[0113] Here, if the model information LM is model information indicating the relationship between the number of test pulses and the test work shape, the model generation unit 211 can calculate (in other words, predict) either the number of test pulses or the test work shape based on the model information LM and either one of the number of test pulses or the test work shape. In this case, to perform machine learning to generate the model information LM, the model generation unit 211 may predict the test work shape based on test pulse number information indicating the number of test pulses and the model information LM, as shown in FIG. 14. Note that if the model information LM has not yet been generated (i.e., the operation of step S15 in FIG. 12 has not been performed even once), the model generation unit 211 may predict the test work shape using default model information LM (i.e., model information LM in its initial state). If the model information LM has already been generated (i.e., the operation of step S15 in FIG. 12 has been performed at least once), the model generation unit 211 may predict the test work shape using the generated model information LM.

[0114] Here, the closer the model information LM is to the ideal model information LM, the smaller the test work shape difference Δs_test, which is the difference between the test work shape prediction result based on the model information LM and the actual measurement result of the test work shape (i.e., the test work shape information). Therefore, the model generation unit 211 may generate the model information LM by performing machine learning so that the test work shape difference Δs_test is small (or minimized). As an example, when multiple pieces of teacher data are acquired, the model generation unit 211 may generate multiple test work shape prediction results based on multiple pieces of test pulse number information included in each piece of teacher data and the model information LM. For example, the model generation unit 211 may generate a first test work shape prediction result included in the first teacher data based on the first test pulse number information included in the first teacher data and the model information LM, and generate a second test work shape prediction result included in the second teacher data based on the second test pulse number information included in the second teacher data and the model information LM. The model generation unit 211 may then calculate multiple test work shape differences Δs_test, which are the differences between multiple prediction results of the test work shape using the model information LM and multiple measurement results of the test work shape included in each of the multiple teacher data sets (i.e., multiple test work shape information sets). For example, the model generation unit 211 may calculate a first test work shape difference Δs_test, which is the difference between a first test work shape actually included in the first teacher data set and the prediction result of the first test work shape, and calculate a second test work shape difference Δs_test, which is the difference between a second test work shape actually included in the second teacher data set and the prediction result of the second test work shape. The model generation unit 211 may then generate the model information LM by performing machine learning so that the loss determined based on the multiple test work shape differences Δs_test is small (or minimized). The loss determined based on the multiple test work shape differences Δs_test may be a loss whose value decreases as the multiple test work shape differences Δs_test decrease.Examples of losses include the mean squared error calculated by dividing the sum of the squares of multiple test work shape differences Δs_test by the total number of test work shape differences Δs_test, and the mean absolute error calculated by dividing the sum of the absolute values ​​of multiple test work shape differences Δs_test by the total number of test work shape differences Δs_test.

[0115] The model information LM may be any information as long as it indicates the relationship between the number of test pulses and the test work shape. For example, the model information LM may include model information that can predict the test work shape from the number of test pulses. For example, the model information LM may include information that can predict the number of test pulses from the test work shape. For example, the model information LM may include model information that can output a predicted result of the test work shape when the number of test pulses is input. For example, the model information LM may include information that can output a predicted result of the number of test pulses when the test work shape is input.

[0116] An example of the model information LM is shown in Fig. 15. As shown in Fig. 15, the model information LM may include prediction information PI indicating a predicted value of the processing amount of the test workpiece W_test when the processing light EL is irradiated a unit number of times onto the irradiation target position C. Specifically, as shown in Fig. 15, the model information LM may include prediction information PI indicating a predicted value of the processing amount of the test workpiece W_test when the irradiation target position C is irradiated a unit number of times with the processing light EL. １ Prediction information PI indicating a predicted value of the processing amount of the test workpiece W_test when the processing light EL is irradiated a unit number of times. １ and irradiation target position C ２ Prediction information PI indicating a predicted value of the processing amount of the test workpiece W_test when the processing light EL is irradiated a unit number of times. ２ , and irradiation target position C ｉ (Note that i is a variable indicating an integer greater than or equal to 1 and less than or equal to n) ｉ , and irradiation target position C ｎ Prediction information PI indicating a predicted value of the processing amount of the test workpiece W_test when the processing light EL is irradiated a unit number of times. ｎIn this embodiment, an example in which the unit number of times is one will be described.

[0117] The "machining amount of the test workpiece W_test" indicated by the prediction information PI may refer to the distribution of the machining amount (specifically, the removal amount) of the test workpiece W_test in a direction along the surface of the test workpiece W_test (e.g., at least one of the X-axis direction and the Y-axis direction). In other words, the "machining amount of the test workpiece W_test" indicated by the prediction information PI may refer to a one-dimensional distribution of the machining amount of the test workpiece W_test. Alternatively, the "machining amount of the test workpiece W_test" indicated by the prediction information PI may refer to the distribution of the machining amount of the test workpiece W_test within a plane along the surface of the test workpiece W_test (e.g., a plane along the XY plane). In other words, the "machining amount of the test workpiece W_test" indicated by the prediction information PI may refer to a two-dimensional distribution of the machining amount of the test workpiece W_test.

[0118] As an example, the prediction information PI ｉ As shown in FIG. 16, which conceptually shows the irradiation target position C ｉ Prediction information PI indicating a predicted value of the processing amount of the test workpiece W_test when the processing light EL is irradiated a unit number of times. ｉ is the irradiation target position C ｉ Irradiation target position C when processing light EL is irradiated a unit number of times ｉ Not only the predicted value of the processing amount at the irradiation target position C ｉ Irradiation target position C when processing light EL is irradiated a unit number of times ｉ Irradiation target position C different from ｊ (Note that j is a variable indicating an integer that is equal to or greater than 1 and equal to or less than n, and is different from the variable i.) In other words, the prediction information PI ｉ is the irradiation target position C ｉ Irradiation target position C when processing light EL is irradiated a unit number of times ｉ The predicted value of the processing amount at the irradiation target position C ｉ Irradiation target position C when processing light EL is irradiated a unit number of times ｊ The information may include information that can be used to predict the predicted value of the amount of processing in the target area.

[0119] For example, the prediction information PI １ is the irradiation target position C １ Irradiation target position C when processing light EL is irradiated a unit number of times １ Not only the predicted value of the processing amount at the irradiation target position C １ Irradiation target position C when processing light EL is irradiated a unit number of times ２ From C ｎ For example, the prediction information PI ２ is the irradiation target position C ２ Irradiation target position C when processing light EL is irradiated a unit number of times ２ Not only the predicted value of the processing amount at the irradiation target position C ２ Irradiation target position C when processing light EL is irradiated a unit number of times １ and C ３ From C ｎ For example, the prediction information PI ｉ is the irradiation target position C ｉ Irradiation target position C when processing light EL is irradiated a unit number of times ｉ Not only the predicted value of the processing amount at the irradiation target position C ｉ Irradiation target position C when processing light EL is irradiated a unit number of times １ From C ｉ－１ and C ｉ＋１ From C ｎ For example, the prediction information PI ｎ is the irradiation target position C ｎ Irradiation target position C when processing light EL is irradiated a unit number of times ｎ Not only the predicted value of the processing amount at the irradiation target position C ｎ Irradiation target position C when processing light EL is irradiated a unit number of times １ From C ｎ－１ The predicted value of the amount of processing for each of the above may be shown.

[0120] In this way, the prediction information PI ｉ One of the reasons why the distribution of the processing amount of the test workpiece W_test is ｉThe processing light EL irradiated to the irradiation target position C ｉ Not only a part of the test workpiece W_test located at the irradiation target position C ｊ Specifically, there is a possibility that another part of the test workpiece W_test located at the irradiation target position C ｉ When the processing light EL is irradiated to the irradiation target position C ｊ When the fluence of the processing light EL at the irradiation target position C exceeds the lower limit threshold of the fluence capable of processing the test workpiece W_test, ｉ The processing light EL irradiated to the irradiation target position C ｊ This is because another part of the test workpiece W_test located at is processed.

[0121] Furthermore, the irradiation target position C ｉ The state in which the processing light EL is irradiated to the target position C is such that the center of the beam spot of the processing light EL is aligned with the irradiation target position C ｉ or irradiation target position C ｉ The beam spot of the processing light EL may mean a region irradiated with the processing light EL whose intensity exceeds a predetermined intensity threshold. The beam spot of the processing light EL may mean a region corresponding to the full width at half maximum of the intensity distribution (e.g., Gaussian distribution) of the processing light EL.

[0122] Forecast Information PI ｉ indicates the distribution of the machining amount of the test workpiece W_test, the prediction information PI ｉ is the irradiation target position C ｉ It may be considered that the prediction information PI indicates at least one of the width and depth of the removal target portion W_rmv removed from the test workpiece W_test when the processing light EL is irradiated a unit number of times. ｉ is the irradiation target position C ｉIt may be considered that the data includes at least one of information regarding the width and depth of the removal target portion W_rmv removed from the test workpiece W_test when the processing light EL is irradiated a unit number of times. Note that the width of the removal target portion W_rmv may mean the size of the removal target portion W_rmv in at least one of the X-axis direction and the Y-axis direction. The depth of the removal target portion W_rmv may mean the size of the removal target portion W_rmv in the Z-axis direction. The depth of the removal target portion W_rmv may be synonymous with the processing amount.

[0123] When the model information LM includes prediction information PI, the machine learning for generating (updating) the model information LM may include machine learning for generating (updating) the prediction information PI included in the model information LM. That is, the model generation unit 211 may generate the model information LM by performing machine learning for generating the prediction information PI included in the model information LM. For example, the model generation unit 211 may generate the model information LM by performing machine learning for generating the prediction information PI so that the loss determined based on the above-described multiple test work shape differences Δs_test is small (or minimized).

[0124] Forecast Information PI ｉ may include a model parameter A for defining the distribution of the machining amount of the test workpiece W_test. For example, as shown in FIG. ｉ is m (where m is a constant indicating an integer of 1 or more) basis functions PM (specifically, basis functions PM １ to PM ｍ ) are multiplied by m model parameters A (specifically, model parameters A ｉ，１ From A ｉ，ｍ ) may be included. １ to PM ｍ may represent the unit processing amount of the test workpiece W_test. １ to PM ｍ may represent m different unit processing amounts. In the example shown in FIG. 16, １ to PM ｍindicates m unit processing amounts having different sizes (i.e., processing amounts) in the width direction (i.e., at least one of the X-axis direction and the Y-axis direction). １ to PM ｍ may indicate m unit processing amounts having different sizes (i.e., processing amounts) in the depth direction (i.e., Z-axis direction). In this case, the prediction information PI ｉ is the basis function PM １ to PM ｍ and model parameter A ｉ，１ From A ｉ，ｍ The distribution of the machining amounts obtained by adding the multiplied values ​​of the prediction information PI and the test workpiece W_test may be shown as the distribution of the machining amounts of the test workpiece W_test. ｉ is the basis function PM １ and model parameter A ｉ，１ and the basis function PM ２ and model parameter A ｉ，２ and the multiplication value of the basis function PM ｍ and model parameter A ｉ，ｍ The distribution of the machining amounts obtained by adding the multiplied values ​​of and may be shown as the distribution of the machining amounts of the test workpiece W_test.

[0125] When the prediction information PI includes a model parameter A, the machine learning for generating (updating) the model information LM may include machine learning for generating (updating) the model parameter A included in the prediction information PI. That is, the model generation unit 211 may generate the model information LM by performing machine learning for generating the model parameter A included in the prediction information PI. For example, the model generation unit 211 may generate the model information LM by performing machine learning for generating the model parameter A so that the loss determined based on the above-described multiple test work shape differences Δs_test is small (or minimized).

[0126] The model information LM is the predicted information PI (i.e., the predicted information PI １ From PI ｎ15, the relationship between the number of test pulses and the predicted result of the test work shape is essentially established as follows: Number of test pulses × predicted information PI = predicted result of the test work shape. Specifically, as shown in FIG. 15, the model generating unit 211 ｉ Irradiation target position C ｉ is multiplied by the number of test pulses, which is the number of times the processing light EL should be irradiated, to obtain the irradiation target position C ｉ The model generating unit 211 can predict the machining amount of the test workpiece W_test when the machining light EL is irradiated onto the target position C the same number of times as the number of test pulses. １ From C ｎ The model generating unit 211 can predict the machining amount of the test workpiece W_test when the machining light EL is irradiated onto each of the irradiation target positions C １ From C ｎ The amount of machining of the test work W_test (i.e., the test work shape) can be predicted by adding up the n predicted values ​​of the amount of machining of the test work W_test corresponding to each of the n values. As a result, the model generation unit 211 calculates a test work shape difference Δs_test, which is the difference between the predicted result of the test work shape and the actual measurement result of the test work shape (i.e., the test work shape information), and performs machine learning so that the loss determined based on the calculated test work shape difference Δs_test is small (or minimized), thereby generating model information LM.

[0127] Through the machine learning described above, the model generation unit 211 can generate model information LM that appropriately indicates the relationship between the number of test pulses and the shape of the test workpiece. In other words, the model generation unit 211 can acquire model information LM that appropriately indicates the relationship between the number of test pulses and the shape of the test workpiece.

[0128] 12 again, after the model information LM is generated in step S15, the control data generation unit 212 of the data generation server 2 calculates (in other words, generates) the target pulse number based on the model information LM generated in step S15 (step S21). That is, the control data generation unit 212 generates target pulse number information indicating the target pulse number based on the model information LM generated in step S15 (step S21).

[0129] Specifically, the number of test pulses is set to the irradiation target position C １ From C Ｎ The model information LM indicates the number of times the machining light EL is irradiated onto each of the test pulses. Therefore, it is assumed that the model information LM that appropriately indicates the relationship between the number of test pulses and the test work shape indicates the relationship between the target pulse number and the shape of the workpiece W_target machined based on the target pulse number. Therefore, by generating model information LM that appropriately indicates the relationship between the number of test pulses and the test work shape, the model generation unit 211 may be considered to have generated model information LM that appropriately indicates the relationship between the target pulse number and the shape of the workpiece W_target machined based on the target pulse number. In other words, it is assumed that the model information LM indicates the relationship between the target pulse number and the machining target shape. Therefore, it is assumed that the model generation unit 211 generates model information LM that appropriately indicates the relationship between the number of test pulses and the test work shape,

[0130] It is assumed that the prediction information PI included in the model information LM generated by machine learning also indicates a predicted value of the machining amount of the workpiece W_target to be machined when the processing light EL is irradiated a unit number of times to the irradiation target position C. In other words, the model generation unit 211 generates model information LM including prediction information PI indicating a predicted value of the machining amount of the test workpiece W_test when the processing light EL is irradiated a unit number of times to the irradiation target position C, thereby making it possible to generate model information LM including prediction information PI indicating a predicted value of the machining amount of the workpiece W_target to be machined when the processing light EL is irradiated a unit number of times to the irradiation target position C.

[0131] Because the model information LM showing the relationship between the number of test pulses and the test workpiece shape can be considered to show the relationship between the target number of pulses and the machining target shape, the model information LM can be used to predict the target number of pulses recommended for machining the workpiece W_target so that the shape of the workpiece W_target becomes the desired machining target shape, based on the machining target shape, which is the target value for the shape of the workpiece W_target after machining. In other words, the model information LM can be used to predict the target number of pulses recommended for machining the workpiece W_target so that the shape of the workpiece W_target becomes the desired machining target shape, based on the machining target shape information, under circumstances where machining target shape information regarding the machining target shape is known (i.e., the machining target shape has already been set). For this reason, the model information LM can be considered to be prediction information that can be used to generate unknown target pulse number information (i.e., predict the target pulse number) from known machining target shape information. Therefore, in step S21, the control data generation unit 212 calculates the target pulse number based on the machining target shape information indicating the machining target shape and the model information LM. That is, the control data generation unit 212 predicts the target pulse number recommended for machining the workpiece W_target so that the shape of the workpiece W_target becomes the desired machining target shape, using the machining target shape information and the model information LM. In other words, the control data generation unit 212 calculates the target pulse number based on the machining target shape information indicating the machining target shape and the model information LM, by assuming that the model information LM indicating the relationship between the test pulse number and the test workpiece shape indicates the relationship between the target pulse number and the machining target shape. For example, as shown in FIG. 17 , the control data generation unit 212 may input the machining target shape information indicating the machining target shape into the model information LM, and calculate the target pulse number as the output (prediction result) of the model information LM.

[0132] Here, it is unlikely that there is only one type of target pulse number recommended for machining the workpiece W_target so that the shape of the workpiece W_target becomes the desired machining target shape. In other words, it is possible that the shape of the workpiece W_target machined using one target pulse number will become the desired machining target shape, and that the shape of the workpiece W_target machined using a different target pulse number will also become the desired machining target shape. For this reason, in step S21, the control data generation unit 212 generates multiple pieces of target pulse number information, each indicating a different target pulse number. However, in step S21, the control data generation unit 212 does not have to generate multiple pieces of target pulse number information. In step S21, the control data generation unit 212 may generate a single piece of target pulse number information.

[0133] 12, the control data generation unit 212 then predicts the shape of the workpiece W_target after machining, assuming that the workpiece W_target is machined using the target pulse number indicated by the target pulse number information, based on the target pulse number information generated in step S21 and the model information LM (step S22). The shape of the workpiece W_target predicted in step S22 is referred to as the predicted workpiece shape. In particular, because multiple pieces of target pulse number information are generated in step S21, the control data generation unit 212 predicts multiple predicted workpiece shapes corresponding to the multiple pieces of target pulse number information in step S22.

[0134] Specifically, as described above, since the model information LM can be considered to indicate the relationship between the target pulse number and the shape of the machining target, the model information LM can be used to predict, from the target pulse number, the shape of the workpiece W_target after machining when it is assumed that the workpiece W_target is machined using the target pulse number (in this case, the machining target shape corresponding to the target pulse number). In other words, the model information LM can be used to predict the shape of the workpiece W_target after machining when it is assumed that the workpiece W_target is machined using the target pulse number.１ From C Ｎ The target pulse number information can be used to predict the shape of the workpiece W_target that would result from the assumption that the machining light EL is irradiated onto each of the workpieces W_target a number of times equal to the target pulse number. Therefore, the model information LM may be considered to be prediction information that can be used to predict the post-machining shape of the workpiece W_target, which is unknown information, from the target pulse number information, which is known information. Therefore, in step S22, the control data generation unit 212 can predict the predicted workpiece shape based on the target pulse number information and the model information LM generated in step S21. In other words, in step S22, the control data generation unit 212 can predict the predicted workpiece shape based on the target pulse number information and the model information LM by considering the model information LM, which indicates the relationship between the test pulse number and the test workpiece shape, to indicate the relationship between the target pulse number and the machining target shape. For example, as shown in FIG. 17 , the control data generation unit 212 may input target pulse number information indicating the target pulse number into the model information LM, and calculate the predicted workpiece shape as the output (prediction result) of the model information LM.

[0135] Referring again to FIG. 12 , the control data generation unit 212 then selects one target pulse number information from the plurality of target pulse number information generated in step S21 that satisfies a predetermined selection criterion based on the predicted workpiece shape predicted in step S22 (step S23). For example, as shown in FIG. 17 , the control data generation unit 212 may select one target pulse number information that satisfies a predetermined selection criterion based on the target workpiece shape difference Δ_target, which is the difference between the predicted workpiece shape and the machining target shape. As an example, the control data generation unit 212 may select one target pulse number information corresponding to the predicted workpiece shape that satisfies the condition that the target workpiece shape difference Δ_target is the smallest, as one target pulse number information that satisfies the predetermined selection criterion. In other words, the control data generation unit 212 may select, from the plurality of target pulse number information, one target pulse number information that is predicted to result in the shape of the target workpiece W_target after machining that is closest to the machining target shape if the target workpiece W_target is machined using the target pulse number indicated by the target pulse number information, as one target pulse number information that satisfies the predetermined selection criterion. In this case, the operation of selecting one target pulse number information from multiple target pulse number information may be considered equivalent to the operation of generating one target pulse number information so that the predicted work shape approaches the machining target shape (i.e., so that the machining target work shape difference Δ_target, which is the difference between the predicted work shape and the machining target shape, becomes smaller).

[0136] The control data generation unit 212 may select, from among the multiple pieces of target pulse number information, one piece of target pulse number information that satisfies selection criteria other than the selection criteria based on the target workpiece shape difference Δ_target. For example, as described above, the machining device 1 forms a riblet structure in the target workpiece W_target by machining the target workpiece W_target. In this case, the control data generation unit 212 may select, from among the multiple pieces of target pulse number information, one piece of target pulse number information that satisfies selection criteria based on the characteristics of the riblet structure that would be formed if the target workpiece W_target were machined using the target pulse number indicated by the target pulse number information. In this case, the control data generation unit 212 may calculate the characteristics of the riblet structure from the predicted workpiece shape predicted in step S22 (i.e., the predicted result of the shape of the riblet structure that would be formed if the target workpiece W_target were machined using the target pulse number indicated by the target pulse number information), and then select, based on the calculated characteristics of the riblet structure, one piece of target pulse number information that satisfies the selection criteria based on the characteristics of the riblet structure. For example, as described above, the riblet structure has the effect of reducing the resistance of the surface of the workpiece W_target to a fluid. In this case, the characteristics of the riblet structure may include the resistance reduction effect of the riblet structure. In this case, the control data generating unit 212 may select, from among the multiple pieces of target pulse number information, one piece of target pulse number information that satisfies the condition that the resistance reduction effect of the riblet structure formed when the workpiece W_target is machined using the target pulse number indicated by the target pulse number information is the highest.

[0137] If the control data generating unit 212 does not generate multiple pieces of target pulse number information in step S21 (i.e., generates a single piece of target pulse number information), the control data generating unit 212 does not need to perform steps S22 to S23. In this case, the single piece of target pulse number information generated in step S21 may be used as the single piece of target pulse number information selected in step S23.

[0138] Thereafter, the target workpiece W_target is machined based on the target pulse number information selected in step S23 (step S24). Specifically, the control data generation unit 212 controls the machining device 1 to machine the test workpiece W_test based on the target pulse number information. For example, the control data generation unit 212 may transmit the target pulse number information to the machining device 1 as control data. The machining device 1 (particularly, the control unit 14) may generate machining control information for controlling the machining device 1 based on the target pulse number information transmitted from the data generation server 2. For example, the control unit 14 may generate machining control information for controlling the machining device 1 to machine the machining mode workpiece W_target so that the shape of the target workpiece W_target becomes the desired machining target shape. Thereafter, the machining device 1 may machine the target workpiece W_target based on the generated machining control information.

[0139] After the workpiece W_target is machined, a shape of the workpiece W_target after machining is measured (step S24). That is, a shape of the workpiece W_target after machining is measured based on the target pulse number information selected in step S23 (step S24). For example, the machining device 1 may measure the shape of the workpiece W_target using the measurement unit 12. Thereafter, machined workpiece shape information (machined shape information) indicating the measurement results of the shape of the workpiece W_target by the measurement unit 12 may be transmitted from the machining device 1 to the data generating server 2 via the communication network 4. The model generating unit 211 of the data generating server 2 may acquire the shape of the workpiece W_target indicating the measurement results of the shape of the workpiece W_target by the measurement unit 12.

[0140] Thereafter, the control data generation unit 212 determines whether the shape of the workpiece to be machined satisfies a predetermined standard (step S25). For example, if the difference between the shape of the workpiece to be machined and the shape of the machining target is less than the allowable amount, it is highly likely that the workpiece to be machined W_target has been properly machined based on the target pulse number information. Therefore, in this case, the control data generation unit 212 may determine that the shape of the workpiece to be machined satisfies the predetermined standard. On the other hand, for example, if the difference between the shape of the workpiece to be machined and the shape of the machining target is greater than the allowable amount, it is highly likely that the workpiece to be machined W_target has not been properly machined based on the target pulse number information. Therefore, in this case, the control data generation unit 212 may determine that the shape of the workpiece to be machined does not satisfy the predetermined standard.

[0141] If it is determined in step S25 that the shape of the workpiece to be machined satisfies the predetermined standard (step S25: Yes), it is assumed that the target pulse number information selected in step S25 is appropriate as the target pulse number information to be used for machining the workpiece to be machined W_target so that the shape of the workpiece to be machined W_target becomes the predetermined machining target shape. Therefore, in this case, the control data generation unit 212 may generate control data indicating the target pulse number information selected in step S25 and output the generated control data to the machining device 1.

[0142] If the result of the judgment in step S25 is that the shape of the workpiece to be machined does not meet the predetermined standard (step S25: No), it is assumed that the target pulse number information selected in step S25 is not appropriate as the target pulse number information used to machine the workpiece to be machined W_target so that the shape of the workpiece to be machined W_target becomes the predetermined machining target shape. In this case, the model information LM used to generate the target pulse number information may not have been sufficiently learned. Therefore, the data generation server 2 performs the operation of step S15 again to perform machine learning again to generate (in this case, update) the model information LM. In particular, when performing machine learning again, the model generation unit 211 may use, as new training data, data associated with the target pulse number information selected in step S23 and the workpiece shape information indicating the shape of the workpiece to be machined measured in step S24, in addition to the training data acquired by the operations of steps S11 to S13 (i.e., training data associated with test machining condition information and test workpiece shape information).

[0143] After machine learning is performed again to generate (in this case, update) the model information LM, the data generating server 2 performs the operations of steps S21 to S25 again. That is, the control data generating unit 212 generates new target pulse number information based on the updated model information LM. Note that generating new target pulse number information may be considered equivalent to updating the target pulse number information (i.e., updating the target pulse number indicated by the target pulse number information).

[0144] Here, because the model information LM has been updated, it is expected that the target pulse number information generated based on the updated model information LM will be more likely to be appropriate as target pulse number information used to machine the workpiece W_target to be machined so that the shape of the workpiece W_target becomes a predetermined machining target shape, compared to the target pulse number information generated based on the model information LM before the update. In other words, it is expected that the target pulse number information generated based on the updated model information LM will be more likely to be appropriate as target pulse number information used to machine the workpiece W_target to be machined so that the shape of the workpiece W_target becomes a predetermined machining target shape, compared to the target pulse number information generated based on the model information LM before the update.

[0145] In this case, when target pulse number information is generated based on the updated model information LM, it is expected that the workpiece shape difference Δ_target corresponding to the one target pulse number information selected in step S23 will be smaller than when target pulse number information is generated based on the model information LM before the update. For this reason, updating the model information LM and generating (in other words, updating) target pulse number information based on the updated model information LM may be considered equivalent to generating (in other words, updating) target pulse number information so as to reduce the workpiece shape difference Δ_target.

[0146] Furthermore, when target pulse number information is generated based on the updated model information LM, it is expected that the shape of the workpiece to be machined will be determined to meet the standard in step S25 more likely than when target pulse number information is generated based on the model information LM before the update. In other words, it is expected that the difference between the shape of the workpiece to be machined and the shape of the machining target will be smaller. For this reason, updating the model information LM and generating (in other words, updating) target pulse number information based on the updated model information LM may be considered equivalent to generating (in other words, updating) target pulse number information so that the shape of the workpiece to be machined will be determined to meet the standard more likely (for example, so that the difference between the shape of the workpiece to be machined and the shape of the machining target will be smaller).

[0147] (4) Technical Effects As described above, in this embodiment, the data generating server 2 determines the irradiation target position C １ From C Ｎ The target number of pulses can be calculated using model information LM that indicates the relationship between the number of times the processing light EL is irradiated onto each of the workpieces and the shape of the workpiece W after processing. Therefore, the time required to calculate the target number of pulses is reduced compared to when the model information LM is not used.

[0148] Specifically, if the model information LM is not used, the operator of the machining system SYS must manually set the target pulse number. Furthermore, the workpiece W must be machined based on the set target pulse number, the shape of the machined workpiece W must be measured, and based on the measurement results of the shape of the workpiece W, it must be evaluated whether the target pulse number manually set by the operator is appropriate. Furthermore, if it is determined that the target pulse number manually set by the operator is inappropriate (i.e., the shape of the machined workpiece W differs from the target shape), the operator must manually reset the target pulse number and re-evaluate whether the target pulse number manually set by the operator is appropriate. Therefore, it takes a significant amount of time to set an appropriate target pulse number.

[0149] On the other hand, in this embodiment, the target pulse number is calculated using the model information LM, eliminating the need for the operator to manually set the target pulse number. Furthermore, since the test workpiece W_test is actually machined and the model information LM is generated using the measurement results of the shape of the actually machined test workpiece W_test, the model information LM is likely to indicate the relationship between the target pulse number and the target shape of the workpiece W. Therefore, the target pulse number calculated using the model information LM is likely to be an appropriate target pulse number. Therefore, compared to when the model information LM is not used, the likelihood of having to reset the target pulse number is reduced. Furthermore, to obtain the training data used to generate the model information LM, the test workpiece W_test is machined, and the shape of the machined test workpiece W_test is measured. However, since the machining and measurement of the test workpiece W_test are performed to obtain training data, the number of machining and measurement of the test workpiece W_test is reduced compared to the number of machining and measurement performed to evaluate whether the target pulse number is appropriate. For these reasons, the time required to set the target number of pulses is reduced.

[0150] (5) Other Modifications In the above description, the data generation server 2 generates the model information LM by performing machine learning. However, the data generation server 2 may generate the model information LM without performing machine learning. For example, the data generation server 2 may generate model information LM that indicates the relationship between the test machining condition information and the test work shape information based on the test machining condition information and the test work shape information.

[0151] In the above description, the data generating server 2 generates model information LM based on the test machining condition information and the test workpiece shape information, and then generates target pulse number information based on the generated model information LM and the machining target shape information. However, the data generating server 2 may generate target pulse number information without generating model information LM. For example, the data generating server 2 may generate target pulse number information based on the test machining condition information, the test workpiece shape information, and the machining target shape information without generating model information LM.

[0152] In the above description, the prediction information PI included in the model information LM indicates a predicted value of the machining amount of the workpiece W when the processing light EL is irradiated one time onto the irradiation target position C as a predicted value of the machining amount of the workpiece W when the processing light EL is irradiated a unit number of times onto the irradiation target position C. That is, in the above description, an example in which the unit number of times is one time is described. However, the unit number of times may be two or more times. In this case, the target pulse number may indicate the number of times to perform a unit irradiation operation in which the processing light EL is irradiated a unit number of times onto the irradiation target position C. Specifically, the target pulse number information is １ a target pulse number, which is the number of times that a first unit irradiation operation should be performed to irradiate the processing light EL a unit number of times, and an irradiation target position C ２ a target pulse number, which is the number of times that a second unit irradiation operation should be performed to irradiate the processing light EL a unit number of times to the irradiation target position C ｎ and a target pulse number, which is the number of times that the n-th unit irradiation operation of irradiating the processing light EL a unit number of times should be performed.

[0153] In the above description, the data generating server 2 generates control data indicating a target number of pulses. However, the data generating server 2 may generate control data other than the control data indicating the target number of pulses. For example, the data generating server 2 may generate control data specifying arbitrary machining conditions (machining recipes) for the machining device 1. In this case, the data generating server 2 may generate model information LM indicating the relationship between the arbitrary machining conditions for machining the workpiece W and the post-machining shape of the workpiece W based on test machining condition information indicating the arbitrary machining conditions and the test workpiece shape, and calculate target values ​​of the arbitrary machining conditions (target machining conditions, machined workpiece machining conditions) based on the generated model information and machining target shape information. In other words, the data generating server 2 may generate control data indicating target values ​​of the arbitrary machining conditions (target machining conditions). The machining conditions may include conditions (irradiation conditions) of the machining light EL irradiated on the workpiece W. The irradiation conditions may include at least one of the following: a condition regarding the intensity of the processing light EL, a condition regarding the energy (e.g., pulse energy) of the processing light EL, a condition regarding the number of irradiations of the processing light EL, a condition regarding the irradiation position of the processing light EL, and a condition regarding the burst mode of the processing light EL. Note that the burst mode may mean an operation mode in which each pulse constituting the processing light EL can be divided into a desired number of pulses, as described in, for example, JP-A-2016-524864. In this case, the condition regarding the burst mode may include a condition regarding the number of divisions of the pulsed light. In addition, the irradiation target position C １ From C Ｎ The target pulse number, which is the number of times the processing light EL should be irradiated to each of the target positions, may be considered equivalent to the conditions regarding the number of irradiations and irradiation position of the processing light EL. In this case, the control data indicating the target pulse number may be considered to be control data specifying the processing conditions. The processing conditions may include movement conditions for at least one of the processing head 112 and the stage 132. The movement conditions may include, for example, at least one of a condition regarding the movement speed, a condition regarding the movement amount, a condition regarding the movement direction, and a condition regarding the movement timing.

[0154] In the above description, the processing device 1 processes the workpiece W by irradiating the workpiece W with a processing beam including a single processing light EL. However, the processing device 1 may process the workpiece W by irradiating the workpiece W with processing beams including a plurality of processing lights EL. In this case, the data generating server 2 may also generate control data by performing the data generating operation shown in FIG. 12. In this case, however, the model information LM may include the irradiation target position C for processing the workpiece W. １ From C Ｎ The model information LM may indicate a relationship between the number of times that a processing beam including a plurality of processing lights EL is irradiated onto each of the irradiation target positions C and the shape of the workpiece W after processing. The prediction information PI included in the model information LM may indicate a predicted value of the processing amount of the test workpiece W_test when a processing beam including a plurality of processing lights EL is irradiated onto the irradiation target position C a unit number of times.

[0155] When the processing device 1 processes the workpiece W by irradiating the workpiece W with a processing beam including a plurality of processing lights EL, the processing device 1 may process the workpiece W by forming interference fringes on the surface of the workpiece W, which are obtained by interfering with the plurality of processing lights EL. Note that an example of a processing device that processes the workpiece W by forming interference fringes on the surface of the workpiece W is described in U.S. Patent Application Publication No. 2022 / 0258289. Even in this case, the data generation server 2 may generate control data by performing the data generation operation shown in FIG. 12. However, in this case, the model information LM may not include the irradiation target position C for processing the workpiece W. １ From C Ｎ may indicate the relationship between the number of times interference fringes are formed and the shape of the workpiece W after machining. The prediction information PI included in the model information LM may indicate a predicted value of the machining amount of the test workpiece W_test when interference fringes are formed a unit number of times at the irradiation target position C.

[0156] In the above description, the processing apparatus 1 processes the workpiece W by irradiating the workpiece W with the processing light EL. However, the processing apparatus 1 may process the workpiece W by irradiating the workpiece W with an arbitrary energy beam. In this case, the processing apparatus 1 may be provided with a beam source capable of irradiating the arbitrary energy beam in addition to or instead of the processing light source 111. Examples of the arbitrary energy beam include at least one of a charged particle beam and an electromagnetic wave. Examples of the charged particle beam include at least one of an electron beam and an ion beam.

[0157] In the above description, the measurement unit 12 is provided separately from the processing unit 11. However, the measurement unit 12 may be integrated with the processing unit 11, and the workpiece W may be measured via a focusing optical system (fθ lens 1123) that focuses the processing light EL of the processing unit 11. Such a processing and measurement device is disclosed, for example, in International Publication No. 2021 / 024480.

[0158] In the above description, the information regarding the processing light EL contained in the reference information may include information regarding the light intensity distribution and fluence distribution of the processing light EL in the direction of propagation of the processing light EL (in other words, the irradiation direction).

[0159] In the above-described embodiment, instead of or in addition to calculating the optical penetration length from the processing results of the test workpiece Wt, the data generating server 2 may calculate the optical penetration length from the processing results of the workpiece W. In this case, the data generating server 2 may process and measure the portion of the workpiece W that is scheduled to be removed to obtain the optical penetration length, or may measure the workpiece W that has been processed to remove the optical penetration length to obtain information about the initial shape of the workpiece W.

[0160] In the above description, the data generation server 2 determines the optimal processing conditions using response surface methodology, but this is not limited to response surface methodology, and other methods such as experimental design, machine learning, Bayesian estimation, etc. can also be used.

[0161] In the above description, the stage unit 13 may include a plurality of stages 132 .

[0162] (5) Supplementary Notes The following supplementary notes are further disclosed regarding the above-described embodiments: [Supplementary Note 1] A data generation method for generating control data for controlling a processing device capable of removing and processing a workpiece to be processed by irradiating a surface of the workpiece with a pulse energy beam, the data generation method including: inputting first information on a target shape whose height varies in at least one direction, one or more pieces of second information on processing conditions for processing the target shape, and one or more pieces of third information on a height profile in the at least one direction of an object processed using the processing conditions; and outputting one or more processing conditions and processed shape prediction results corresponding to each of the one or more processing conditions, the one or more processing conditions being obtained by inputting the first information into a machine learning model trained using the one or more pieces of second information and the one or more pieces of third information.

[0163] The requirements of the above-described embodiments may be combined as appropriate. Some of the requirements of the above-described embodiments may not be used. The requirements of the above-described embodiments may be replaced with requirements of other embodiments as appropriate. Furthermore, to the extent permitted by law, the disclosures of all publications and U.S. patents relating to the devices, etc. cited in the above-described embodiments are incorporated herein by reference.

[0164] Furthermore, the present invention can be modified as appropriate within the scope that does not contradict the gist or idea of ​​the invention that can be read from the claims and the entire specification, and data generation methods, cloud systems, data generation devices, computer programs, recording media, and processing devices that involve such modifications are also included in the technical idea of ​​the present invention.

[0165] SYS Machining system 1 Machining device 11 Machining unit 12 Measuring unit 13 Stage unit 14 Control unit 2 Data generation server 21 Arithmetic unit 211 Model generation section 212 Control data generation section 3 Client terminal device EL Machining light W Workpiece W_test Test workpiece W_target Workpiece to be machined C Irradiation target position LM Model information

Claims

1. A data generation method for generating control data for controlling a machining device capable of removing a workpiece to be machined by irradiating a surface of the workpiece with a pulse energy beam, the data generation method including: acquiring test machining condition information including a number of pulses at which the pulse energy beam should be irradiated to each of a plurality of irradiation positions of the test workpiece in order to machine the test workpiece into a target shape; measuring a test workpiece shape, which is the shape of the test workpiece after machining based on the test machining condition information; and calculating a target number of pulses at which the pulse energy beam should be irradiated to each of a plurality of irradiation positions of the workpiece to be machined, based on the test machining condition information, the test workpiece shape, and predicted information on the shape of a portion to be machined by irradiating the pulse energy beam with a unit number of pulses.

2. The data generation method according to claim 1, wherein the prediction information includes at least one of information on the width and depth of the portion to be processed by the pulse energy beam of the unit pulse number.

3. A data generation method according to claim 1 or 2, wherein the prediction information indicates a predicted value of the amount of machining of the test work when the pulse energy beam is irradiated by a unit number of pulses.

4. The data generation method of claim 3, wherein the prediction information includes a predicted value of the amount of machining at a first test position and a predicted value of the amount of machining at a second test position of the test work when the pulse energy beam is irradiated to a first test position and a second test position different from the first test position by a unit pulse number.

5. A data generation method as described in claim 4, wherein the target number of pulses is calculated using the difference between the target shape and a predicted result of the shape of the workpiece to be machined based on the prediction information.

6. A data generation method according to any one of claims 1 to 5, comprising generating the prediction information so as to reduce a difference between the test work shape and a prediction result of the test work shape based on the prediction information.

7. A data generation method according to any one of claims 1 to 6, wherein calculating the target number of pulses includes: generating prediction information based on the test processing condition information and the test work shape; and calculating the target number of pulses based on the prediction information.

8. A data generation method according to any one of claims 1 to 7, wherein generating the target pulse number further includes generating target pulse number information indicating a target pulse number for irradiating the pulse energy beam at each of the multiple irradiation positions based on the target shape in order to change the shape of the workpiece to the target shape by removal processing.

9. A data generation method as described in any one of claims 1 to 8, further comprising predicting a predicted shape of the test work based on the test processing condition information, the prediction information including model parameters, and the model parameters are generated so as to reduce a difference between the test work shape and the predicted shape.

10. The data generation method described in claim 9, further comprising updating the model parameters so that a difference between the predicted shape and the target shape becomes smaller, and predicting the predicted shape comprises predicting the shape of the test work when removal processing is performed by irradiating the test work with the pulse energy beam based on the prediction information.

11. A data generation method as described in claim 9 or 10, wherein generating the target number of pulses includes generating the target number of pulses based on the target shape and the prediction information by assuming that the relationship between the shape of the test work and the number of test pulses indicates the relationship between the target shape and the target number of pulses.

12. A data generation method according to any one of claims 1 to 8, wherein generating the target pulse number includes generating the target pulse number information so that a predicted shape of the workpiece to be machined predicted from the prediction information when assuming that removal processing is performed by irradiating the pulse energy beam to each of the multiple irradiation positions for the number of pulses indicated by the target pulse number approaches the target shape.

13. A data generation method according to any one of claims 9 to 12, wherein the prediction information includes information that can be used to predict a predicted value of the machining amount of the test work when the pulse energy beam is irradiated to a first irradiation position among the multiple irradiation positions by a unit pulse number, the predicted value of the machining amount at the first irradiation position, and a predicted value of the machining amount at a position different from the first irradiation position.

14. The data generation method according to claim 13, wherein the prediction information includes information that can be used to predict a predicted amount of machining of the test work when the pulse energy beam is irradiated to a second irradiation position, which is different from the first irradiation position among the multiple irradiation positions, by a unit pulse number, the predicted amount of machining at the second irradiation position and a predicted amount of machining at a position different from the second irradiation position.

15. The data generating method according to any one of claims 9 to 14, wherein generating the target pulse number includes generating at least one piece of target pulse number information.

16. The data generating method according to claim 15, wherein generating the at least one piece of target pulse number information includes: generating a plurality of pieces of target pulse number information based on the prediction information and the target shape; and selecting, from the plurality of pieces of target pulse number information, at least one piece of target pulse number information that satisfies a predetermined selection criterion as the at least one piece of target pulse number information.

17. The data generation method described in claim 16, wherein the selection criteria include a first criterion related to the difference between the target shape and a predicted value of the shape of the workpiece to be machined predicted from the prediction information when assuming that removal processing is performed by irradiating the pulse energy beam to each of the multiple irradiation positions for the number of pulses indicated by the target pulse number information.

18. A data generation method as described in claim 16 or 17, wherein the processing device forms a riblet structure on the surface of the workpiece to be processed that can reduce friction of the surface of the workpiece against a fluid by removing the workpiece, and the selection criterion includes a second criterion regarding the characteristics of the riblet structure formed on the surface of the workpiece to be processed when it is assumed that removal is performed by irradiating the pulse energy beam at each of the multiple irradiation positions for the number of pulses indicated by the target pulse number information.

19. A data generation method according to any one of claims 1 to 18, wherein the test work is a first test work, the test machining condition information is first test machining condition information, and the test work shape is a first test work shape, the data generation method includes: acquiring second test machining condition information indicating a number of pulses to be irradiated with the pulse energy beam at each of a plurality of irradiation positions of a second test work different from the first test work; irradiating the second test work with the pulse energy beam based on the second test machining condition information; and measuring a second test work shape related to the shape of the second test work machined into the second test work by the pulse energy beam, and the prediction information is generated based on the second test machining condition information and the second test work shape.

20. A data generation method for generating control data for controlling a processing device capable of removing a workpiece to be processed by irradiating a surface of the workpiece with a pulse energy beam, the data generation method comprising: inputting test processing condition information indicating the number of pulses to be irradiated with the pulse energy beam at each of a plurality of irradiation positions of the test workpiece in order to process the test workpiece into a target shape; measuring a test workpiece shape related to the shape of the test workpiece processed with the pulse energy beam based on the test processing condition information; performing machine learning to generate model information including prediction information of a shape to be processed by the pulse energy beam of a unit number of pulses based on the test processing condition information and the test workpiece shape; and calculating a target number of pulses to be irradiated with the pulse energy beam at the plurality of irradiation positions of the workpiece to be processed based on the model information generated by the machine learning and the target shape.

21. The data generating method according to claim 20, wherein the model information includes information on at least one of the width and depth of a portion processed by the pulse energy beam of the unit pulse number.

22. The data generating method according to claim 20 or 21, wherein the model information indicates a predicted value of the machining amount of the test workpiece when the pulse energy beam is irradiated by a unit number of pulses.

23. The data generation method according to claim 22, wherein the model information includes a predicted value of the amount of machining at a first test position and a predicted value of the amount of machining at the second test position of the test work when the pulse energy beam is irradiated to a first test position and a second test position different from the first test position by a unit pulse number.

24. The data generation method according to claim 23, wherein the target number of pulses is calculated using the difference between the target shape and a predicted result of the shape of the workpiece to be machined based on the model information.

25. A data generation method according to any one of claims 20 to 24, comprising generating the model information so as to reduce a difference between the test work shape and a prediction result of the shape of the test work based on the model information.

26. A data generation method according to any one of claims 20 to 25, wherein calculating the target number of pulses includes: generating model information based on the test processing condition information and the test work shape; and calculating the target number of pulses based on the model information.

27. A data generation method as claimed in any one of claims 20 to 26, wherein generating the target pulse number further includes generating target pulse number information indicating a target pulse number for irradiating the pulse energy beam at each of the multiple irradiation positions based on the target shape in order to change the shape of the workpiece to the target shape by removal processing.

28. A data generation method as described in any one of claims 20 to 27, further comprising predicting a predicted shape of the test work based on the test processing condition information, the model information including model parameters, and the model parameters are generated so as to reduce a difference between the test work shape and the predicted shape.

29. The data generation method described in claim 28, further comprising updating the model parameters so that a difference between the predicted shape and the target shape becomes smaller, and predicting the predicted shape comprises predicting the shape of the test work when removal processing is performed by irradiating the test work with the pulse energy beam based on the model information.

30. A data generation method as described in claim 28 or 29, wherein generating the target number of pulses includes generating the target number of pulses based on the target shape and the prediction information by assuming that the relationship between the shape of the test work and the number of test pulses indicates the relationship between the target shape and the target number of pulses.

31. A data generation method according to any one of claims 20 to 27, wherein generating the target pulse number includes generating the target pulse number information so that a predicted shape of the workpiece to be machined predicted from the model information approaches the target shape when it is assumed that removal processing is performed by irradiating the pulse energy beam to each of the multiple irradiation positions for the number of pulses indicated by the target pulse number.

32. A data generation method according to any one of claims 28 to 31, wherein the model information includes information usable for predicting a predicted machining amount of the test workpiece when the pulse energy beam is irradiated to a first irradiation position among the multiple irradiation positions by a unit pulse number, the predicted machining amount at the first irradiation position and a predicted machining amount at a position different from the first irradiation position.

33. The data generation method according to claim 32, wherein the model information includes information that can be used to predict a predicted machining amount of the test work when the pulse energy beam is irradiated to a second irradiation position, which is different from the first irradiation position among the multiple irradiation positions, by a unit pulse number, including a predicted machining amount at the second irradiation position and a predicted machining amount at a position different from the second irradiation position.

34. The data generating method according to any one of claims 28 to 33, wherein generating the target pulse number includes generating at least one piece of target pulse number information.

35. The data generation method according to claim 34, wherein generating the at least one piece of target pulse number information includes: generating a plurality of pieces of target pulse number information based on the model information and the target shape; and selecting, from the plurality of pieces of target pulse number information, at least one piece of target pulse number information that satisfies a predetermined selection criterion as the at least one piece of target pulse number information.

36. A data generation method as described in claim 35, wherein the selection criteria include a first criterion relating to the difference between the target shape and a predicted value of the shape of the workpiece to be machined predicted from the prediction information when it is assumed that removal processing is performed by irradiating the pulse energy beam to each of the multiple irradiation positions for the number of pulses indicated by the target pulse number information.

37. A data generation method as described in claim 35 or 36, wherein the processing device forms a riblet structure on the surface of the workpiece to be processed that can reduce friction of the surface of the workpiece against a fluid by removing the workpiece, and the selection criterion includes a second criterion regarding the characteristics of the riblet structure formed on the surface of the workpiece to be processed when it is assumed that removal is performed by irradiating the pulse energy beam at each of the multiple irradiation positions for the number of pulses indicated by the target pulse number information.

38. A data generation method according to any one of claims 20 to 37, wherein the test work is a first test work, the test machining condition information is first test machining condition information, and the test work shape is a first test work shape, and the data generation method includes: acquiring second test machining condition information indicating a number of pulses to be irradiated with the pulse energy beam at each of a plurality of irradiation positions of a second test work different from the first test work; irradiating the second test work with the pulse energy beam based on the second test machining condition information; and measuring a second test work shape related to the shape of the second test work machined into the second test work by the pulse energy beam, and the model information is generated from the second test machining condition information and the second test work shape.

39. A data generation method for generating control data for controlling a processing device capable of removing a workpiece to be processed by irradiating a surface of the workpiece with a pulse energy beam, the data generation method including: acquiring target pulse number information indicating a target number of pulses with which the pulse energy beam should be irradiated to each of a plurality of irradiation positions on the workpiece to be processed, and target shape information relating to a target shape of the workpiece after removal processing; and generating the target pulse number information as the control data based on the model information and the target shape information.

40. A data generation method for generating control data for controlling a machining device capable of removing a surface of a workpiece to be machined by irradiating the surface of the workpiece with a pulse energy beam, the data generation method comprising: acquiring model data usable for predicting a predicted machining amount of the workpiece to be machined at a first irradiation position among a plurality of irradiation positions on the workpiece to be machined when the pulse energy beam is irradiated at the first irradiation position by a unit pulse number, and a predicted machining amount at a position different from the first irradiation position; and generating, as the control data, target pulse number data indicating a target number of pulses with which the pulse energy beam should be irradiated at each of the plurality of irradiation positions on the workpiece to be machined, based on the model data and target shape data relating to a target shape of the workpiece to be machined after removal processing.

41. A data generation method for generating control data for controlling a machining device capable of removing and machining a workpiece to be machined by irradiating a surface of the workpiece with a pulse energy beam, the data generation method including: acquiring test machining condition information for machining a test workpiece to a target shape; measuring a test workpiece shape, which is the shape of the test workpiece after machining based on the test machining condition information; and calculating machining conditions for the workpiece to be machined based on the test machining condition information, the test workpiece shape, and predicted information on the shape to be machined by the pulse energy beam.

42. A data generation method for generating control data for controlling a machining device capable of removing a surface of a workpiece to be machined by irradiating a pulse energy beam onto the surface of the workpiece, the data generation method comprising: generating first test machining condition information indicating the number of pulses with which the pulse energy beam should be irradiated to each of a plurality of irradiation positions of a first test workpiece; acquiring first test shape information on a shape of the first test workpiece which has been removed by irradiating the first test workpiece with the pulse energy beam based on the first test machining condition information; generating second test machining condition information indicating the number of pulses with which the pulse energy beam should be irradiated to each of a plurality of irradiation positions of a second test workpiece for machining a second test workpiece into a target shape; acquiring second test shape information on the shape of the second test workpiece which has been removed by irradiating the second test workpiece with the pulse energy beam based on the second test machining condition information; and calculating a target number of pulses with which the pulse energy beam should be irradiated to each of the irradiation positions of the workpiece to be machined, based on the first test shape information, the second test shape information, the second test machining conditions, and the target shape.

43. A computer program causing a computer to carry out the data generation method according to any one of claims 1 to 42.

44. A recording medium on which the computer program according to claim 43 is recorded.

45. A processing device that performs removal processing on the workpiece to be processed by using the control data generated by the data generation method according to any one of claims 1 to 42.

46. ​​The processing device according to claim 45, further comprising a control device which generates said control data by carrying out said data generating method.

Images