Diagnostic device, processing system, diagnostic method and program

Abstract

To enhance determination accuracy for determining a state of a processing tool.SOLUTION: A diagnostic device includes a communication control part for receiving a physical amount that undergoes time change due to processing operation to a workpiece of a processing tool provided in a processing device, a characteristic extraction part for extracting characteristic information related to detection information from a result of frequency analysis to the detection information shown by the physical amount, and a determination part for determining a state of the processing tool on the basis of the characteristic information, and the characteristic extraction part specifies an extraction section for extracting the characteristic information according to at least either of a tool kind representing the kind of the processing tool or a processing kind representing a kind of processing to the workpiece from a processing section representing the processing operation, and extracts the characteristic information in the specified extraction section.SELECTED DRAWING: Figure 11

Description

[Technical Field]

[0001] The present invention relates to a diagnostic device, a processing system, a diagnostic method, and a program. [Background technology]

[0002] 2. Description of the Related Art Conventionally, techniques have been disclosed for detecting the state of a machining tool provided in a machining device during machining work using the machining tool. Summary of the Invention [Problem to be solved by the invention]

[0003] For example, a technique is known in which detection information of physical quantities emitted by a tool attached to a rotating shaft constituting a processing device while performing a processing operation on a workpiece is frequency-analyzed, and characteristic information relating to the detection information is extracted from the results of the frequency analysis (see Patent Document 1).

[0004] However, conventional technology does not take into consideration the identification of extraction sections for frequency analysis according to the type of machining tool, the type of machining performed on the workpiece, etc., and therefore ends up extracting characteristic information from sections that are not necessary when determining the state of the machining tool, resulting in the problem that it is not possible to improve the accuracy of determining the state of the machining tool.

[0005] The present invention has been made in view of the above, and has an object to improve the accuracy of determining the state of a machining tool. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, the present invention comprises a communication control unit that receives, from a processing device, physical quantities that change over time in accordance with the processing operation of a processing tool provided on a workpiece, a feature extraction unit that extracts feature information related to the detection information from the results of frequency analysis of the detection information indicated by the physical quantities, and a judgment unit that judges the state of the processing tool based on the feature information, wherein the feature extraction unit identifies an extraction section from a processing section representing the processing operation for extracting the feature information according to at least one of a tool type representing the type of the processing tool and a processing type representing the type of processing on the workpiece, and extracts the feature information in the identified extraction section. [Effects of the Invention]

[0007] According to the present invention, it is possible to improve the accuracy of determining the state of a machining tool. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a machining system to which a diagnostic device according to a first embodiment is applied. [Figure 2] FIG. 2 is a block diagram illustrating an example of a hardware configuration of the processing device. [Figure 3] FIG. 3 is a block diagram illustrating an example of a hardware configuration of the diagnostic device. [Figure 4] FIG. 4 is a block diagram illustrating an example of a functional configuration of the diagnostic device. [Figure 5] FIG. 5 is a diagram illustrating an example of the correspondence between the context information stored in the diagnostic device and the learning model. [Figure 6] FIG. 6 is a diagram showing an example of an average spectrum obtained by frequency analysis by the frequency analysis unit of the diagnostic device when there is no continuous cutting chipping. [Figure 7] FIG. 7 is a diagram showing an example of an average spectrum obtained by frequency analysis by the frequency analysis unit of the diagnostic device when continuous cutting chipping occurs. [Figure 8] FIG. 8 is a diagram showing an example of an average spectrum obtained by frequency analysis by the frequency analysis unit of the diagnostic device in the case where there is no chipping due to intermittent cutting. [Figure 9] FIG. 9 is a diagram showing an example of an average spectrum obtained by frequency analysis by the frequency analysis unit of the diagnostic device when chipping occurs during intermittent cutting. [Figure 10] FIG. 10 is a flowchart showing an example of the flow of diagnostic processing in the diagnostic device. [Figure 11] Figure 11 shows an example of determining the state of a machining tool of a machining device, where Figure 11(a) shows the change over time in the tapping process, and Figure 11(b) shows the ladder signal and time waveform related to the tapping process of Figure 11(a). [Figure 12] FIG. 12 is a diagram showing a ladder signal, a time waveform, and a frequency waveform related to the tapping process of FIG. 11(a). [Figure 13] FIG. 13 is a flowchart showing an example of the flow of a model generation process performed by the diagnostic device. [Figure 14] FIG. 14 is a flowchart showing an example of the flow of the process of extracting feature information according to the BPF by the diagnostic device. [Figure 15] FIG. 15 is a diagram for explaining an example of a method for selecting a BPF by the diagnostic device. [Figure 16] FIG. 16 is an enlarged view of the BPF vicinity of the average spectrum calculated by the diagnostic device. [Figure 17] FIG. 17 is a diagram showing an example of machining scores for all sections in a conventional tapping process. [Figure 18] FIG. 18 is a diagram showing an example of a machining score in an actual machining section according to the tapping process of this embodiment. [Figure 19] FIG. 19 is a diagram illustrating another example of the correspondence between the context information and the learning model stored in the diagnostic device according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a diagnostic device, a processing system, a diagnostic method, and a program will be described in detail with reference to the accompanying drawings.

[0010] (First embodiment) Fig. 1 is a block diagram showing a configuration example of a processing system to which a diagnostic device according to the first embodiment is applied. As shown in Fig. 1, the processing system according to the present embodiment includes a processing device 200 and a diagnostic device 100.

[0011] The processing device 200 and the diagnostic device 100 may be connected in any manner. For example, the processing device 200 and the diagnostic device 100 may be connected by a dedicated connection line, a wired network such as a wired LAN (Local Area Network), a wireless network, or the like.

[0012] The machining device 200 includes a machine control unit 201, a tool changer 202, a display unit 203, a memory unit 204, a communication control unit 205, a numerical control unit 206, a tool information input unit 207, an alarm unit 208, an input / output unit 209, a machine tool 220, etc.

[0013] 1 and equipped with a drive unit. Z-axis stage 226 is equipped with a rotation spindle 221, which is an example of a rotation axis that constitutes processing apparatus 200. A tool holder 222 that holds a processing tool 223 is attached to rotation spindle 221. Machine tool 220 is equipped with an XY-axis stage 225, which is below rotation spindle 221 and is equipped with a drive unit and is movable in two axial directions in a plane perpendicular to Z-axis stage 226. XY-axis stage 225 holds workpiece 224, which is an object to be processed by processing tool 223.

[0014] The numerical control unit 206 executes machining by the machining device 200 using computer numerical control (CNC). For example, the numerical control unit 206 reads a machining program from the input / output unit 209, and generates and outputs numerical control data for controlling the rotation of the main spindle and the position of each axis stage. The machining program also describes the storage number of the tool changer 202, and the numerical control unit 206 performs tool change according to that description.

[0015] The numerical control unit 206 outputs the context information to the communication control unit 205. The context information is information that defines the operation of the machining tool 223 of the machining device 200, and a plurality of pieces of context information are defined for each type of operation of the machining tool 223. In this embodiment, the context information includes, for example, tool information that identifies the machining tool 223 of the machining device 200, rotation information of the rotating spindle 221 (for example, the spindle rotation speed that is the rotation speed of the rotating spindle 221), movement information of the Z-axis stage 226 and the XY-axis stage 225 (movement speed, information during movement), etc.

[0016] The tool information includes at least information such as the type of machining tool 223, such as a drill, tap, reamer, or end mill, and the number of cutting edges of the machining tool 223. This tool information is input by the operator from the tool information input unit 207 in accordance with the information displayed on the display unit 203. Alternatively, the tool information can be obtained by reading a list file of the tool information from the input / output unit 209, or by inputting the information from an external computer (not shown) via the communication control unit 205. Moreover, the tool information may be stored in the storage unit 204 and made referable from the machining program.

[0017] The tool information may also include a processing type that indicates the type of processing performed on the workpiece 224. The processing type may be drilling, threading, or the like.

[0018] The numerical control unit 206 transmits, for example, context information defining the current operation of the machining tool 223 to the diagnostic device 100 via the communication control unit 205. When machining the workpiece 224 according to the machining program, the numerical control unit 206 controls the type of the machining tool 223, the positions of the Z-axis stage 226 and the XY-axis stage 225, the rotation speed of the rotary spindle 221, etc., depending on the machining process. The numerical control unit 206 transmits, from the context information, context information corresponding to a predetermined operation to the diagnostic device 100 via the communication control unit 205. Here, the predetermined operation is a preset operation of the machining tool 223. In this embodiment, every time the numerical control unit 206 changes the type of operation of the machining tool 223, it sequentially transmits context information corresponding to the changed type of operation to the diagnostic device 100 via the communication control unit 205.

[0019] The communication control unit 205 controls communication with an external device such as the diagnostic device 100. For example, the communication control unit 205 transmits context information corresponding to the current operation of the machining tool 223 to the diagnostic device 100.

[0020] The physical quantity information detection unit 227 has a sensor that detects, as an analog signal, a time-varying physical quantity emitted by the machining tool 223 while the machining tool 223 is performing a machining operation on the workpiece 224. The physical quantity information detection unit 227 also has a function of appropriately amplifying the analog signal detected by the sensor, cutting off a desired frequency range, and then converting the analog signal into a digital signal. The physical quantity information detection unit 227 also functions as an example of a transmission unit that transmits the digital signal to the diagnostic device 100 as detection information. The type of sensor included in the physical quantity information detection unit 227 and the physical quantity detected may be any type. For example, the sensor included in the physical quantity information detection unit 227 may be a microphone, an acceleration sensor, an AE (acoustic emission) sensor, or the like, and each sensor outputs acoustic data, acceleration data, or data indicating AE waves as detection information. The diagnostic device 100 may have any number of physical quantity information detection units 227. For example, the diagnostic device 100 may include multiple sensors that detect different physical quantities.

[0021] In FIG. 1 , the physical quantity information detection unit 227 includes a sensor attached to the side of the structure that holds the rotating spindle 221 and a sensor attached to the side of the XY-axis stage 225 or via a vise. The vise is a device similar to a vice that is attached to the processing device 200 and clamps the workpiece 224. The sensor is preferably installed near the spindle and the processing tool 223, where cutting vibrations can be minimized. The sensor included in the physical quantity information detection unit 227 includes an acceleration sensor. When processing by the processing device 200 begins, the physical quantity information detection unit 227 detects the acceleration of vibrations generated by the rotation of the rotating spindle 221. When the processing tool 223 and the workpiece 224 come into contact with each other and actual cutting begins, a cutting force is generated, which acts as an excitation force, vibrating the processing tool 223 and the workpiece 224, and the vibrations propagate between them. The physical quantity information detection unit 227 transmits the acceleration of the vibration and the like as detection information to the diagnostic device 100. As a result, the communication control unit 101 of the diagnostic device 100 receives from the processing device 200 the physical quantity that changes over time in accordance with the processing operation of the processing tool 223 provided in the processing device 200 on the workpiece 224.

[0022] For example, if the cutting edge of the machining tool 223 breaks (breaks) or chips (chips) during machining, the cutting force that was uniform for each cutting edge during normal machining becomes uneven, and the generated vibration changes.

[0023] Damage to the machining tool 223 can be broadly divided into wear type and breakage type. Breakage type basically occurs due to brittle fracture or plastic deformation of the machining tool 223. Chipping is one type of breakage, and is a small chip that occurs on the cutting edge of the machining tool 223. In a tool with multiple blades, even if one blade is damaged by chipping, the remaining blades may be able to compensate for the damage. However, even if there is no effect on the machining quality, the load on the remaining blades increases from the perspective of the lifespan of the machining tool 223, so there is a demand for immediate damage detection of the machining tool 223 when chipping occurs.

[0024] As described above, the diagnostic device 100 receives, via the communication control unit 101, detection information indicated by a time-varying physical quantity (such as vibration acceleration) associated with the machining operation of the machining tool 223 provided on the machining apparatus 200 on the workpiece 224. In addition, the communication control unit 101 controls communication with the machining apparatus 200 to receive context information from the machining apparatus 200. The determination unit 102 determines the state of the machining tool 223 (e.g., whether the machining tool 223 is normal or not) based on feature information associated with the detection information. Specifically, the determination unit 102 determines whether the state of the machining tool 223 of the machining apparatus 200 is normal or not by referring to the context information and the detection information. Furthermore, when the diagnostic device 100 diagnoses that the state of the machining tool 223 of the machining apparatus 200 is abnormal, it transmits alert information to the machining apparatus 200 via the communication control unit 101. When the machining apparatus 200 receives the alert information via the communication control unit 205, the machining apparatus 200 displays the alert information on the display unit 203 and activates the alarm unit 208. The alarm unit 208 is a patrol lamp, a buzzer, a speaker, etc. Furthermore, the machine control unit 201 can interrupt the operation of the processing device 200 according to the processing program and stop the processing of the processing device 200.

[0025] Furthermore, the diagnostic device 100, as will be described in detail later, changes the analysis section for analyzing the state of the machining tool 223 associated with machining depending on at least one of the tool type indicating the type of machining tool 223 provided on the machining device 200 and the machining type indicating the type of machining performed on the workpiece 224.

[0026] Fig. 2 is a block diagram showing an example of a hardware configuration of a processing apparatus. As shown in Fig. 2, the processing apparatus 200 according to this embodiment includes a CPU (Central Processing Unit) 251, a ROM (Read Only Memory) 252, a RAM (Random Access Memory) 253, a communication I / F (Interface) 254, a drive control circuit 255, a motor 256, an input / output I / F 257, an input device 258, and a display 259, which are connected via a bus 260.

[0027] The CPU 251 controls the entire processing apparatus 200. The CPU 251 controls the entire operation of the processing apparatus 200 and realizes various functions of the processing apparatus 200 by executing programs stored in the ROM 252 or the like using the RAM 253 as a work area (working region), for example.

[0028] The communication I / F 254 is an interface for communicating with external devices such as the diagnostic device 100. The drive control circuit 255 is a circuit that controls the driving of the motor 256. The rotating spindle 221, the Z-axis stage 226, and the XY-axis stage 225 each include a drive unit such as the motor 256. The sensor 270 is attached to the processing device 200 and converts a physical quantity that changes in response to the operation of the processing device 200 into an electrical signal. The signal conversion circuit 271 amplifies the electrical signal output from the sensor 270 to a desired level, removes noise components contained in the electrical signal, and converts it into a digital signal. The signal conversion circuit 271 then outputs the digital signal to the diagnostic device 100 as detection information. In other words, the sensor 270 and the signal conversion circuit 271 correspond to, for example, the physical quantity information detection unit 227 shown in FIG. 1.

[0029] The numerical control unit 206 and the communication control unit 205 shown in FIG. 1 may be realized by the CPU 251 executing a program stored in the ROM 252, i.e., by software, or may be realized by hardware such as an IC (Integrated Circuit), or may be realized by a combination of software and hardware.

[0030] Fig. 3 is a block diagram showing an example of the hardware configuration of the diagnostic device 100 according to the present embodiment. As shown in Fig. 3, the diagnostic device 100 has a configuration in which a CPU 151, a ROM 152, a RAM 153, a communication I / F 154, an auxiliary storage device 155, and an input / output I / F 157 are connected via a bus 160.

[0031] The CPU 151 controls the entire diagnostic device 100. For example, the CPU 151 controls the operation of the entire diagnostic device 100 and realizes the diagnostic function of the processing device 200 by executing a program stored in the ROM 152 or the like using the RAM 153 as a work area.

[0032] The communication I / F 154 is an interface for communicating with external devices such as the machining device 200. The auxiliary storage device 155 stores various information such as setting information of the diagnostic device 100, context information received from the machining device 200, and detection information output from the physical quantity information detection unit 227. The auxiliary storage device 155 also stores various calculation results used to determine whether the state of the machining tool 223 of the machining device 200 is normal or not. The auxiliary storage device 155 is made up of non-volatile storage means such as an HDD (Hard Disk Drive), an EEPROM (Electrically Erasable Programmable Read-Only Memory), or an SSD (Solid State Drive).

[0033] The input / output I / F 157 sequentially displays the detection information input from the physical quantity information detection unit 227 on the display 159, and displays the determination result by the determination unit 102. The input / output I / F 157 also accepts settings required for diagnosing the processing device 200 that are input by the user while looking at the display 159 via an input device 158 such as a keyboard or a mouse.

[0034] 4 is a block diagram showing an example of the functional configuration of the diagnostic device 100 according to this embodiment. In addition to the communication control unit 101 and the determination unit 102, the diagnostic device 100 includes a storage unit 103, a generation unit 104, a display control unit 105, a display unit 106, an input unit 107, a reception unit 120, and a feature extraction unit 110.

[0035] The memory unit 103 stores various information required for the diagnostic function of the diagnostic device 100. The memory unit 103 is realized, for example, by the RAM 153 and the auxiliary storage device 155 shown in FIG. 3 . For example, the memory unit 103 stores one or more models (hereinafter referred to as learning models) used to determine whether the state of the machining tool 223 of the machining device 200 is abnormal. Here, the learning model is generated by learning, for example, using detection information output from the physical quantity information detection unit 227 when the state of the machining tool 223 of the machining device 200 is normal. The learning method and format of the learning model may be any method and format. For example, the learning model and the learning method for the learning model may be a learning model such as a GMM (Gaussian Mixture Model) or an HMM (Hidden Markov Model), and a model learning method corresponding to the learning model.

[0036] Furthermore, the storage unit 103 may set rules for the normal state and abnormal state of the machining tool 223 of the processing device 200 and store them as a learning model. For example, the rule stored in the storage unit 103 as the learning model may be that the first 10 machining operations after a new machining tool 223 is attached and machining is started constitute a learning period for determining diagnostic rules. The rules stored in the storage unit 103 as the learning model may be determined in advance separately from actual machining, and the determined rules may be stored in the storage unit 103 as the learning model.

[0037] In this embodiment, a learning model is generated for each piece of context information and stored in the storage unit 103. The storage unit 103 stores, for example, context information and a learning model corresponding to the context information in association with each other.

[0038] FIG. 5 is a diagram illustrating an example of the correspondence between the context information stored in the diagnostic device and the learning model.

[0039] As shown in Figure 5, for each machining process number, which is a number that identifies the machining process (machining type), the tool type of the machining tool 223, the tool number of the machining tool 223, the rotation speed [rpm] of the machining tool 223, the number of blades (Z) of the machining tool 223, the harmonic order of the machining tool 223, whether the actual machining section of the machining process can be extracted, and the identification model number of the learning model are associated and stored.

[0040] 5, machining process 4 and machining process 5 use the same end mill as machining tool 223 and have the same rotation speed. On the other hand, machining process 1, machining process 2, and machining process 3 use different machining tools 223, and the machining tools 223 rotate at different rotation speeds. In this embodiment, if the rotation speeds and tool types of the machining tools 223 are different, the diagnostic device 100 generates learning models (distinguished by identification model numbers) for each of them and stores them in the memory unit 103.

[0041] Furthermore, machining process 4 and machining process 5 are consecutive machining processes of the same part using the same end mill as the machining tool 223. However, machining process 4 and machining process 5 have different analysis conditions and therefore different vibration intensities. Therefore, even when machining is performed by rotating the same machining tool 223 at the same rotation speed, the diagnostic device 100 generates a separate learning model for each machining process and determines whether the state of the machining tool 223 of the machining device 200 is normal.

[0042] Regarding whether or not the actual machining section shown in FIG. 5 can be extracted, whether or not only the actual machining section in which machining is actually performed can be extracted is set for at least one of the tool type and machining type. That is, whether or not the actual machining section can be extracted can be set not only for each machining tool 223 but also for each machining process number. Also, whether or not the actual machining section can be extracted can be set not only for each machining tool 223 but also for each combination of machining tool 223 and machining process number. Alternatively, whether or not the actual machining section can be extracted can be set uniquely for each tool type. For example, in the case of a tap, since it is common for cutting to be performed at a constant speed on the way thereto and back, in the case of a tap, actual machining section extraction is automatically turned ON (extraction of the actual machining section is possible).

[0043] The harmonic order can be set for each machining tool 223. A harmonic is a frequency component that is an integer multiple of the fundamental frequency. The harmonic order represents an integer multiple of the fundamental frequency.

[0044] 4, the communication control unit 101 includes a receiving unit 101a and a transmitting unit 101b. The receiving unit 101a receives various types of information transmitted from the processing device 200 or an external device. For example, the receiving unit 101a receives context information corresponding to the current operation of the processing tool 223 and detection information output from the physical quantity information detecting unit 227. The transmitting unit 101b transmits various types of information to the processing device 200.

[0045] The feature extraction unit 110 extracts feature information related to the detection information from the results of frequency analysis of the detection information indicated by physical quantities that change over time as a result of a machining operation. The feature extraction unit 110 also changes the extraction section for extracting feature information from the machining section indicating the machining operation in accordance with at least one of a tool type indicating the type of machining tool 223 and a machining type indicating the type of machining performed on the workpiece, and extracts the feature information in the changed extraction section. The feature extraction unit 110 also extracts feature information related to the detection information from the results of frequency analysis of the detection information indicated by physical quantities. The feature extraction unit 110 also generates a learning model and extracts feature information (feature quantities) from the detection information to be used in the determination by the determination unit 102. The feature extraction unit 110 performs frequency analysis of the detection information during one machining operation for one or more frames and extracts the change over time of the frames as feature information (feature quantities). Here, the feature information may be any information that indicates the characteristics of the detection information. For example, when the detection information is acoustic data collected by a microphone, the feature extraction unit 110 extracts feature quantities such as energy, frequency spectrum, and MFCC (Mel-Frequency Cepstrum Coefficients) from the detection information. In this embodiment, the feature extraction unit 110 includes a BPF (Band Pass Filter) setting unit 111, a frequency shift estimation unit 114, a frequency analysis unit 115, a waveform-in-process extraction unit 116, and a section classification unit 119. Furthermore, the BPF setting unit 111 includes a bandwidth setting unit 112, a range setting unit 117, a band selection unit 113, and a natural frequency exclusion unit 118.

[0046] The generation unit 104 generates a learning model for determining the normal state of the machining tool 223 of the machining device 200 by learning using feature information extracted from detection information of the normal state of the machining tool 223 of the machining device 200. However, if the learning model is generated by an external device, the diagnostic device 100 does not need to include the generation unit 104. Specifically, the learning model may be generated by an external device, and the learning model generated by the external device may be received by the receiving unit 101a and stored in the storage unit 103. When context information for which a learning model is not defined and detection information corresponding to the context information are input, the generation unit 104 may generate a learning model corresponding to the context information using feature information extracted from the detection information.

[0047] The determination unit 102 determines the state of the machining tool 223 of the processing device 200 (whether the machining tool 223 is normal or not) based on the feature information extracted from the detection information. In the present embodiment, the determination unit 102 determines the state of the machining tool 223 of the processing device 200 using the feature information and a learning model corresponding to the context information. For example, the determination unit 102 requests the feature extraction unit 110 to extract the feature information from the detection information. The determination unit 102 calculates a likelihood indicating the likelihood that the feature information extracted from the detection information is normal, using the corresponding learning model. The determination unit 102 compares the likelihood with a predetermined threshold. If the likelihood is equal to or greater than the threshold, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is normal. If the likelihood is less than the threshold, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is abnormal.

[0048] The method for determining the state of the machining tool 223 of the processing device 200 is not limited to this, and any method may be used as long as it can determine the state of the machining tool 223 of the processing device 200 using feature information and a model. For example, instead of directly comparing the likelihood with a threshold, the determination unit 102 may compare a value indicating a variation in the likelihood with a threshold to determine whether the state of the machining tool 223 of the processing device 200 is normal. Alternatively, the determination unit 102 may calculate a score, which is a positive number greater than or equal to zero, by taking the logarithm of the likelihood and inverting the sign. The score is close to zero if the state of the machining tool 223 of the processing device 200 is normal, and increases as the degree of abnormality of the state of the machining tool 223 of the processing device 200 increases. Therefore, if the score is less than or equal to a predetermined threshold, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is normal. If the score is greater than or equal to the threshold, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is abnormal. That is, the determination unit 102 determines the state of the machining tool 223 of the machining device 200 by comparing at least one of the likelihood or a value calculated using the likelihood with a threshold value.

[0049] Each unit shown in FIG. 4 (communication control unit 101, judgment unit 102, reception unit 120, feature extraction unit 110, generation unit 104) may be realized by CPU 151 shown in FIG. 3 executing a program, i.e., by software, or by hardware such as an IC (Integrated Circuit), or by a combination of software and hardware.

[0050] The diagnostic device 100 according to this embodiment is characterized by a feature extraction unit 110 and a reception unit 120. In this embodiment, the feature extraction unit 110 includes a BPF (Band Pass Filter) setting unit 111, a frequency shift estimation unit 114, a frequency analysis unit 115, an in-machining waveform extraction unit 116, and a section classification unit 119. Furthermore, the BPF setting unit 111 includes a bandwidth setting unit 112, a range setting unit 117, a band selection unit 113, and a natural frequency exclusion unit 118. In this embodiment, the diagnostic device 100 requires, among the context information, context information such as a spindle rotation speed and tool information during or before and after the operation of the machining tool 200. Therefore, the reception unit 120 includes a spindle rotation speed reception unit 122, a tool information reception unit 121, and a machining process reception unit 123.

[0051] Next, the operation of the diagnostic device 100 according to this embodiment will be described in detail with reference to FIG.

[0052] In this embodiment, the machining device 200 has a physical quantity information detection unit 227 installed near the rotation main shaft 221, and an acceleration sensor is used as the sensor 270 of the physical quantity information detection unit 227. The physical quantity information detection unit 227 amplifies an analog signal detected by the sensor 270 using a preamplifier of the sensor 270, samples the analog signal at predetermined time intervals, and converts the sampled analog signal into a digital signal using an analog-to-digital (A / D) converter (signal conversion circuit 271). The diagnosis device 100 receives the digital signal output from the physical quantity information detection unit 227 as detection information using the receiving unit 101a. The digital signal output from the physical quantity information detection unit 227 is converted into acceleration units using a calibration value of the sensor 270 as necessary. However, these processes will be omitted here, and the description will be given in a state that does not depend on the sensitivity of the sensor 270 or the specifications of the A / D converter (signal conversion circuit 271). Therefore, the receiving unit 101a receives, as detection information, a waveform in the time domain of an observation value proportional to the acceleration detected by the sensor 270 of the physical quantity information detecting unit 227.

[0053] The reception unit 120 of the diagnostic device 100 requests the machining device 200 to transmit the respective context information from the spindle rotation speed reception unit 122, the tool information reception unit 121, and the machining process reception unit 123, and transmits and receives the context information via the respective communication control units 205 and 101. Here, the context information includes rotation information, machining process information, tool information, etc.

[0054] The rotation information may be either the spindle rotation speed set from the machining program read by the machining device 200 or the spindle rotation speed measured by a tachometer in the machining device 200. The rotation information is, for example, 7500 rpm set from the machining program. The machining process information includes a number identifying the machining process described in the machining program and information regarding the start and end of operation of the rotating spindle 221 and the stage (here, the XY-axis stage 225 and the Z-axis stage 226). The machining process information is, for example, information regarding the start and end of rotation of the XY-axis stage 225. The tool information includes the tool type of the machining tool 223, the diameter of the machining tool 223, the number of blades of the machining tool 223, and the machining type indicating the type of machining performed on the workpiece 224. However, the tool information is not limited to the context information from the machining device 200, and may be context information input from the input unit 158 ​​to the diagnostic device 100, context information stored in the auxiliary storage device 155, or context information received by the receiving unit 101a from an external device other than the machining device 200. The tool information is, for example, the number of blades of the machining tool 223 (for example, four).

[0055] The waveform-in-process extraction unit 116 of the feature extraction unit 110 extracts waveform data during processing, which is acceleration waveform data for time intervals corresponding to each of the three processing steps, namely, the rotational movement of the XY-axis stage 225, the start of the rotational movement, and the end of the rotational movement, from the detection information (for example, acceleration waveform data, which is acceleration waveform data) input from the physical quantity information detection unit 227, for each cutting depth of the processing tool 223.

[0056] The frequency analysis unit 115 is an example of a frequency analysis unit that performs frequency analysis on the detection information. The frequency analysis unit 115 performs a Fourier transform on a predetermined number of consecutive samples of the extracted waveform-in-process data using, for example, an FFT (Fast Fourier Transform) algorithm. The data string of the waveform-in-process data to be Fourier transformed may use the extracted waveform-in-process data for the entire data string, or may be a string in which a portion of the data string is replaced with zeros.

[0057] However, the waveform-in-processing extraction unit 116 determines the frequency resolution that can be analyzed by Fourier transform based on the time interval for sampling the detection information (acceleration waveform data) before A / D conversion by the signal conversion circuit 271 and the data length of the detection information (the number of data in the data string). Here, an example will be described on the assumption that the combination of the time interval and the data length is set to result in a frequency resolution of approximately 5.8 Hz.

[0058] 6 and 7 are diagrams showing examples of average spectra obtained by frequency analysis by the frequency analysis unit of the diagnostic device when continuous cutting chipping is present and absent. The average spectra shown in FIGS. 6 and 7 are obtained by performing a Fourier transform from the beginning of the waveform data during tapping to obtain amplitude power and then averaging this power. The average spectrum shown in FIG. 6 is the average spectrum when there is no chipping of the machining tool 223, and the average spectrum shown in FIG. 7 is the average spectrum when there is chipping of the machining tool 223. Only average spectra with frequencies of 150.0 Hz or less are shown in FIGS. 6 and 7.

[0059] When the rotation speed of the rotating spindle 221 is 600 rpm, this 600 rpm is converted to a frequency of 10.0 Hz. This frequency is called the fundamental rotation frequency and corresponds to the peak indicated by a square (□) in Figures 6 and 7. In this case, a cutting force is generated in the processing device 200 at each fundamental rotation frequency, causing vibrations in the processing device 200 and the workpiece 224, which then propagate. Therefore, acceleration waveform data corresponding to this is input to the diagnostic device 100 as detection information. Therefore, the average spectrum calculated by the frequency analysis unit 115 ideally has a spectral structure with sharp peaks in the TPF and its multiple harmonic components.

[0060] Figures 6 and 7 show the average spectra for taps with and without chipping, and are an example of continuous cutting. The peaks marked with a ▽ in the average spectrum indicate frequencies above integer harmonics of the rotational frequency. One cause of the other visible peaks is the runout of the cutting edge due to tool eccentricity. Comparing the cases with and without chipping, it can be seen that the level of the ▽ (second harmonic) at 20 Hz increases from the case without chipping in Figure 6 to the case with chipping in Figure 7.

[0061] In the average spectra shown in Figures 6 and 7, it is observed that the power is large at peaks other than the rotational frequency and the harmonic components of that rotational frequency. The peak marked with a square in the average spectra shown in Figures 6 and 7 is 10.0 Hz, which corresponds to the rotational speed of the rotating spindle 221, and this frequency is called the fundamental rotational frequency. Furthermore, when damage or chipping of the cutting edge occurs, the cutting force becomes uneven, just like when the cutting edge runs out, and the generated vibration also becomes uneven, causing changes not only in the fundamental rotational frequency but also in the harmonic components.

[0062] While Figures 6 and 7 illustrate continuous cutting with a tap, the following discussion will also cover harmonics in the case of side milling with an end mill as an example of intermittent cutting. Intermittent cutting refers to a situation in which multiple cutting edges independently contact the workpiece. Figures 8 and 9 show examples of average spectra for the presence and absence of chipping, determined by frequency analysis using the frequency analysis unit of a diagnostic device for cases with and without chipping in intermittent cutting. The average spectrum shown in Figure 8 is an average spectrum obtained by Fourier transforming a data string from side milling with a two-flute end mill to obtain amplitude power, and then averaging this power. Therefore, Figures 8 and 9 only show average spectra for frequencies below 1500.0 Hz.

[0063] When the rotation speed of the rotating spindle 221 is 10,000 rpm, this 10,000 rpm is converted to a frequency of 166.6 Hz. Furthermore, because the end mill, which is an example of the machining tool 223, has two cutting edges, the machining device 200 repeats intermittent cutting at 166.6 Hz × 2 = 333.3 Hz, generating a cutting force that generates and propagates vibrations in the machining device 200 and the workpiece 224. Therefore, acceleration waveform data corresponding to this is input to the diagnostic device 100 as detection information. The frequency of the acceleration waveform data is called the tool passing frequency (TPF). Therefore, the average spectrum calculated by the frequency analysis unit 115 ideally has a spectral structure with sharp peaks at the TPF and its multiple harmonic components.

[0064] In Figures 8 and 9, the peaks marked with a ▽ in the average spectrum represent the TPF, the second harmonic of the TPF, and the third harmonic of the TPF, demonstrating significant power. However, in reality, it is impossible to completely eliminate the runout of the cutting tool 223, so other peaks are observed. The peak marked with a square (□) in the average spectrum is at 166.6 Hz, which corresponds to the rotational speed of the rotating spindle 221. This frequency is called the fundamental rotation frequency. The peaks marked with a circle in the average spectrum shown in Figures 8 and 9 are components that are modulated by the fundamental rotation frequency from the TPF and its harmonic components, and are called sidebands. When the waveform of the TPF component is viewed over time, the amplitude of the waveform becomes uneven due to runout of the cutting edge, and these sidebands appear as a characteristic of the spectrum. Furthermore, when damage or chipping of the cutting edge occurs, the cutting force becomes uneven, similar to runout of the cutting edge, and the generated vibration also becomes uneven, resulting in a similar increase in sidebands.

[0065] 10 is a flowchart showing an example of the flow of diagnostic processing in the diagnostic device. The numerical control unit 206 of the machining device 200 sequentially transmits context information indicating the current operation of the machining tool 223 to the diagnostic device 100. The receiving unit 101a receives the context information transmitted from the machining device 200 (step S101). Next, the feature extraction unit 110 reads a learning model corresponding to the received context information from the storage unit 103 (step S102). The timing of step S102 may be immediately before step S106, as long as it is before step S106 in which feature information is extracted according to the BPF described below.

[0066] The physical quantity information detection unit 227 of the processing device 200 sequentially outputs the detection information during processing by the processing device 200.

[0067] The in-process waveform extraction unit 116 of the feature extraction unit 110 extracts in-process waveform data from the detection information (sensor data) received from the processing device 200 and context information relating to the processing operation (step S103).

[0068] Next, the section classification unit 119 of the feature extraction unit 110 divides (classifies) the analysis section in which the state of the machining tool 223 associated with machining is analyzed into an actual machining section and a non-machining section based on at least one of the tool type indicating the type of machining tool 223 provided on the machining device 200 and the machining type indicating the type of machining performed on the workpiece 224, thereby identifying the actual machining section as the extraction section from which feature information is extracted.

[0069] Specifically, the section classification unit 119 of the feature extraction unit 110 classifies the in-machining section (extraction section) from the extracted in-machining waveform data (step S104). For example, when a tap is used as the machining tool 223, the first half of the machining section (in-machining section) corresponds to cutting (or plastic deformation) processing to create a thread, and the second half of the machining section (punching section) corresponds to a process of removing the tap by rotating the machined portion in the reverse direction. In other words, when a tap is used as the machining tool 223, the second half of the machining section is a non-machining section that does not directly contribute to machining. Furthermore, the section classification unit 119 classifies the machining section into an actual machining section and a non-machining section based on information related to the machining operation when the machining device 200 executes the above-mentioned computer numerical control.

[0070] Next, the frequency analysis unit 115 performs frequency analysis of the waveform data being processed using an FFT algorithm or the like (step S105). This frequency analysis is performed by shifting a preset number of data samples and the start position of the data string from the waveform data being processed. The result of the frequency analysis of the waveform data being processed is three-dimensional data in which multiple spectra are arranged in time series. The feature extraction unit 110 extracts feature information, which is the fundamental or harmonic frequency of the rotation frequency, from the multiple spectra or an average spectrum obtained by averaging multiple spectra over an arbitrary time range (step S106). In this embodiment, since a BPF is recorded in the learning model read in step S102, the feature extraction unit 110 uses the BPF to extract feature information (fundamental or harmonic frequency of the rotation frequency) from the average spectrum.

[0071] The determination unit 102 determines the state of the machining tool 223 of the machining device 200 using the feature information extracted by the feature extraction unit 110 and a learning model corresponding to the received context information (step S107). This allows the state of the machining tool 223 of the machining device 200 to be determined using the machining tool 223, the feature information extracted according to the type of machining, and the learning model, making it possible to detect and monitor with high accuracy the occurrence of abnormalities in the state of the machining tool 223 of the machining device 200 for various machining operations performed in the machining center. The determination unit 102 outputs the determination result to the display unit 106 via the display control unit 105 (step S108). Alternatively, the determination unit 102 transmits alert information to the machining device 200 or an external device via the transmission unit 101b (step S108).

[0072] Here, a method for determining the state of the machining tool 223 of the machining device 200 will be described. Chipping detection in the machining device 200 equipped with a numerical control unit 206 that is a CNC will be described below. CNC is a method of numerically controlling the movement amount, movement speed, etc. of a tool in machining using a computer.

[0073] Fig. 11 is a diagram showing an example of determining the state of the machining tool 223 of the machining device 200. Fig. 11 shows a case of machining a fastening hole using a tap as an example of machining by the machining tool 223.

[0074] CNC uses NC programs, but there are also canned cycles that can command several frequently used machining cycles in one block. For tapping, the canned cycle commands used are G84 (tapping: rotating the spindle forward to cut in, then reversing to release at the bottom of the hole) and G74 (reverse tapping).

[0075] FIG. 11(a) is a diagram showing the change over time in the tapping process, and FIG. 11(b) is a diagram showing the ladder signal and time waveforms involved in the tapping process of FIG. 11(a).

[0076] As shown in FIG. 11( a), in the case of tapping, the first approximately half of the machining section corresponds to the actual machining portion, and the second approximately half corresponds to the punching section, where no actual machining is performed. Therefore, in this embodiment, the feature extraction unit 110 extracts feature information from the actual machining section as an extraction section from among the actual machining section in which the machining tool 223 performs machining operations on the workpiece 224 and the non-machining section in which the machining tool 223 does not perform machining operations on the workpiece 224. Specifically, the section classification unit 119 of the feature extraction unit 110 classifies the machining section into an actual machining section and a non-machining section based on at least one of context information that defines the operation of the machining tool 223 and a ladder signal that switches the signal level and outputs depending on the machining section. More specifically, the section classification unit 119 extracts (classifies) the actual machining section, which is approximately the first half of the machining section, from the machining section and uses it for score calculation. For example, as shown in Figure 11(b), when using a ladder signal that switches between "High" and "Low" signal levels depending on the processing section, approximately half of the first half of the ladder signal, which is the incoming processing section, corresponds to the actual processing section.

[0077] The determination unit 102 performs evaluation in units of machining in the actual machining section extracted by the in-machining waveform extraction unit 116, and diagnoses the state of the machining tool 223. More specifically, the determination unit 102 is able to detect chipping by evaluating changes from the learning model when no chipping occurs. Below, a method for extracting the actual machining section from the machining section will be described.

[0078] The method of extracting an actual machining section from a machining section in this embodiment is shown for the case where context information and ladder signals can be received from the numerical control unit 206 of the machining device 200.

[0079] As shown in Fig. 11(b), the ladder signal is output as a binary signal, for example, a "High" level at the start of machining and a "Low" level at the end of machining for each machining operation of each tool. Because the machining section itself can be determined from such a binary signal, the machining waveform extraction unit 116 can extract the actual machining section. In other words, the machining waveform extraction unit 116 extracts the actual machining section using the result of the determination unit 102 determining the actual machining section and the non-machining section from the signal level.

[0080] For example, when the processing waveform extraction unit 116 is executed in a fixed cycle, it automatically cuts out the first half of the processing section (= the section where the ladder signal is at a "High" level) and extracts the actual processing section, provided that the extraction of only the actual processing section within the processing section is set to ON (see Figure 5).

[0081] More specifically, the actual machining section is acquired by first extracting the machining section, storing it in a temporary storage device such as the RAM 153, and then extracting it up to a position corresponding to the first half of the time. The first half of the time can be acquired by acquiring the sample position for the first half from the position (or memory address) halfway through the memory usage of the machining section, or from the number of time samples in the section where the ladder signal is "High."

[0082] On the other hand, the processing waveform extraction unit 116 is set to OFF to extract only the actual processing section in which processing is actually being performed (see Figure 5), and when executed in a fixed cycle, it extracts the processing section (= ladder signal is at the "High" level) as the actual processing section.

[0083] Even if a fixed cycle is not used, if the entry section (actual machining section) and the exit section (non-machining section) of the machining tool 223 have the same speed, the above-mentioned method can be applied regardless of the type of machining tool 223.

[0084] In the above, the actual machining interval is extracted based on the time waveform associated with the tapping process, but this is not limiting. FIG. 12 illustrates the ladder signal, time waveform, and frequency waveform associated with the tapping process of FIG. 11(a). For example, the machining waveform extraction unit 116 detects a change in the portion of the ladder signal where it transitions from a low level to a high level, and extracts a change near the center of the ladder signal as a waveform change based on a value change in a specific band in the time waveform or frequency domain associated with the tapping process. The machining waveform extraction unit 116 also extracts the actual machining interval, or detects the position where the ladder signal transitions from a high level to a low level, and extracts a waveform change in the time waveform or frequency domain associated with the tapping process as a value change in a specific band in the frequency domain. The machining waveform extraction unit 116 may also extract the actual machining interval by calculating a position approximately half the length of the corresponding interval.

[0085] Furthermore, in FIG. 11, a case where a tap is used to make a fastening hole is shown as an example of processing using the processing tool 223, but it is not limited to this and needless to say, it can also be applied to processing a through hole.

[0086] An example of processing other than tapping is reaming, in which a reamer is passed through a pilot hole that is slightly smaller than the required dimensions to obtain a perfectly round hole with a smooth surface.

[0087] Next, an example of model generation processing by the diagnostic device 100 according to this embodiment will be described with reference to FIG. 13. FIG. 13 is a flowchart showing an example of the flow of model generation processing by the diagnostic device. In this embodiment, the generation unit 104 executes the model generation processing in advance, for example, before the diagnostic processing of the processing device 200. Alternatively, as described above, the generation unit 104 may execute the model generation processing when context information for which a learning model has not been defined is input. Furthermore, when the learning model is generated externally as described above, the model generation processing does not need to be executed in the diagnostic device 100.

[0088] The receiving unit 101a receives the context information transmitted from the processing device 200 (step S201). The receiving unit 101a also receives the detection information (sensor data) transmitted from the processing device 200 (step S202).

[0089] The context information and detection information received in this manner are used to generate a learning model. In this embodiment, the generation unit 104 generates a learning model for each piece of context information, so the detection information needs to be associated with the corresponding context information. For this reason, for example, the receiving unit 101a associates the received detection information with the context information received at the same timing and temporarily stores the association information in the storage unit 103 or the like. Then, the generation unit 104 confirms that the detection information stored in the storage unit 103 is normal information, and generates a learning model using only the normal detection information. That is, the generation unit 104 generates a learning model using detection information labeled as normal.

[0090] The confirmation of whether the detection information is normal (labeling) may be performed at any timing after the detection information is stored in the storage unit 103, etc., or may be performed in real time while the processing device 200 is operating. Alternatively, the generation unit 104 may generate a learning model assuming that the detection information is normal without labeling the detection information. If the detection information assumed to be normal is actually abnormal, the generated learning model will not correctly perform the process of determining whether the state of the processing tool 223 of the processing device 200 is normal. Therefore, it is possible to determine whether the learning model was generated using abnormal detection information based on, for example, the frequency with which the state of the processing tool 223 of the processing device 200 is determined to be abnormal, and measures can be taken, such as deleting the erroneously generated learning model. Alternatively, a learning model generated using abnormal detection information may be used as a learning model for determining abnormalities.

[0091] The waveform-in-process extraction unit 116 of the feature extraction unit 110 extracts waveform-in-process data based on the received detection information and context information during the processing operation (step S203).

[0092] Next, the waveform-in-process extraction unit 116 of the feature extraction unit 110 extracts the in-process section from the extracted waveform-in-process data (step S204).

[0093] Next, the frequency analysis unit 115 performs frequency analysis on the extracted waveform data being processed using an FFT algorithm or the like (step S205). The frequency analysis unit 115 performs frequency analysis on the waveform data being processed while shifting a preset number of data samples and the start position of the data string. The obtained result of the frequency analysis becomes three-dimensional data in which multiple spectra are arranged in time series. The feature extraction unit 110 extracts feature information from the multiple spectra or a spectrum obtained by averaging multiple spectra over an arbitrary time range according to the BPF (step S206). This method will be described later.

[0094] The generation unit 104 generates a learning model corresponding to the same context information by using feature information extracted from the detection information associated with the same context information (step S207). The generation unit 104 stores the generated learning model in the storage unit 103 (step S208).

[0095] Next, an example of the flow of the process of extracting feature information according to the BPF by the feature extraction unit 110 according to this embodiment will be described with reference to Fig. 14. Fig. 14 is a flowchart showing an example of the flow of the process of extracting feature information according to the BPF by the diagnostic device.

[0096] The BPF setting unit 111 receives tool information such as the number of cutting edges (number of teeth) Z and rotation information such as the spindle rotation speed r from the tool information receiving unit 121 and the spindle rotation speed receiving unit 122 of the receiving unit 120 (step S301). Then, the BPF setting unit 111 calculates the rotation fundamental frequency and the TPF using equations (1) and (2) (step S302). Here, the spindle rotation speed r is the rotation speed of the rotating spindle 221 set by the machining program. Fundamental rotation frequency [Hz] = r [rpm] / 60 (1) TPF = rotational fundamental frequency [Hz] × Z [sheets] (2) That is, the BPF setting unit 111 calculates the fundamental rotation frequency using the rotation information, and calculates the TPF using the fundamental rotation frequency and the number of teeth included in the tool information.

[0097] For example, in the case of continuous cutting, the BPF setting unit 111 calculates the TPF as follows. Continuous cutting is cutting in which the cutting edge continues to contact the material. Examples include drilling, tapping, and reaming. The fundamental rotation frequency [Hz] = S [rpm] / 60, and the number of blades = 1. TPF=S / 60[Hz]×1[sheet]=S / 60[Hz]

[0098] Furthermore, for example, in the case of intermittent cutting, the BPF setting unit 111 calculates the TPF as follows. Intermittent cutting is cutting in which the cutting edge repeatedly comes into contact with and out of contact with the material. An example is end mill side machining. The fundamental rotation frequency [Hz] = S [rpm] / 60, and the number of blades = Z. TPF=S / 60[Hz]×Z[sheets]=SZ / 60[Hz]

[0099] Next, the BPF setting unit 111 of the feature extraction unit 110 sets (calculates) a BPF center frequency (an example of a center frequency) using the rotation information, tool information, and frequency range. In this embodiment, the BPF setting unit 111 uses Equation (3) to set a BPF center frequency within a frequency range set by the range setting unit 117 of the BPF setting unit 111. Here, the range setting unit 117 sets a frequency range of interest. For example, the range setting unit 117 sets a lower limit frequency and an upper limit frequency, and calculates a natural number n such that the BPF center frequency of Equation (3) falls within this range. Alternatively, the natural number n may be any number that can identify the lower and upper limits of the frequency, such as the order of a harmonic of the rotation fundamental frequency [Hz] or the order of an overtone of the TPF. In other words, the BPF setting unit 111 sets the rotation fundamental frequency and an integer multiple of the rotation fundamental frequency as the BPF center frequencies. Alternatively, the BPF setting unit 111 may set the sidebands of the TPF and the frequencies of integer multiples of the TPF as the BPF center frequency.

[0100] Furthermore, the BPF setting unit 111 calculates the BPF from the BPF center frequency and the bandwidth [Hz] set by the bandwidth setting unit 112 (step S303). Here, the bandwidth setting unit 112 sets the bandwidth of the frequency band of interest within the frequency range set by the range setting unit 117. BPF center frequency = fundamental frequency [Hz] × n (3)

[0101] When the bandwidth setting unit 112 sets the bandwidth [Hz] to b, the BPF setting unit 111 calculates as many BPFs as the number of BPF center frequencies so as to satisfy equation (4). BPF center frequency -b / 2≦BPF(n)≦BPF center frequency+b / 2...(4)

[0102] Next, the band selection unit 113 (an example of a BPF selection unit) selects a BPF to be used for extracting feature information from among the plurality of BPFs (step S304). The band selection unit 113 selects the BPF by one of the following selection methods (a) to (e). (a) fullselect (b) Overtones of TPF (TPF, 2×TPF, …) and their sidebands (c) Only the harmonic sidebands of the TPF (d) Rotational fundamental frequency and (b) or (c) (e) Interactive selection

[0103] 15 is a diagram illustrating an example of a method for selecting a BPF by the diagnostic device. For example, when a BPF is selected by selection method (e), the display control unit 105 displays a BPF selection screen 300 on the display unit 106. The BPF selection screen 300 includes a context information display unit 310 that displays the fundamental rotation frequency and the number of blades, a range display unit 320 that displays the range set by the range setting unit 117, a bandwidth display unit 330 that displays the bandwidth set by the bandwidth setting unit 112, a band display unit 340 that displays the band selected by the band selection unit 113, a data display unit 350, etc.

[0104] Range display unit 320 displays the upper limit frequency of the range set by range setting unit 117. Specifically, range display unit 320 includes a frequency input text box 323 in which the upper limit frequency can be input, and an order input text box 324 in which the upper limit frequency is input in terms of the harmonic order of the TPF. A user of diagnostic device 100 can input the upper limit frequency exclusively into frequency input text box 323 and order input text box 324. Frequency radio button 321 and TPF order radio button 322 can exclusively set into which of frequency input text box 323 and order input text box 324 the upper limit frequency is input. In range display unit 320 shown in FIG. 15 , the TPF harmonic order is input into order input text box 324, and the upper limit frequency is input as a second-order TPF (2×TPF).

[0105] The bandwidth display unit 330 has a bandwidth input text box 331. In the bandwidth input text box 331 shown in FIG. 15, a bandwidth of 40.0 Hz is set. The data display unit 350 displays an average spectrum labeled as normal, which is an average spectrum obtained by averaging multiple spectra obtained by frequency analysis of the in-machining waveform data used for learning the model generation process shown in FIG. 13. The average spectrum displayed in the data display unit 350 may be an average spectrum obtained from test in-machining waveform data obtained by machining with simulated blade vibration and labeled as abnormal, or may be multiple average spectra labeled as both normal and abnormal (the average spectra shown in FIGS. 6 to 9).

[0106] 14. Corresponding to the BPF displayed in the BPF display section 351, the band display section 340 displays a TPF selection toggle button 341 for selecting the use of a TPF for extracting feature information, a sideband selection toggle button 342 for selecting the use of a sideband for extracting feature information, and an other harmonics toggle button 343. Of the TPF selection toggle button 341, the sideband selection toggle button 342, and the other harmonics toggle button 343, those used for extracting feature information in step S306, which will be described later, are turned on, and those not used for extracting feature information are turned off.

[0107] Next, the natural frequency excluding unit 118 excludes BPFs close to the natural frequencies of the machining device 200 and the machining tool 223 from the BPFs set by the BPF setting unit 111 (step S305). The machining tool 223, holder, rotating spindle 221, etc. have natural frequencies depending on their shapes, dimensions, and weights. The frequency components of these natural frequencies tend to be stronger than the power of other frequency components when the machining tool 223 of the machining device 200 is in an abnormal or normal state due to cutting tool damage or whirling. Therefore, if the frequency components of the natural frequency are included in the feature information, the accuracy of determining the state of the machining tool 223 of the machining device 200 decreases. Therefore, the natural frequencies are input in advance from the input unit 107 and stored in the memory unit 103. The natural frequency excluding unit 118 retrieves the natural frequencies from the memory unit 103. If a BPF including this natural frequency is found among the BPFs calculated using Equation (4), the BPF is excluded.

[0108] Fig. 16 is an enlarged view of the BPF vicinity of the average spectrum calculated by the diagnostic device. In Fig. 16, the solid line average spectrum is the average spectrum near the BPF (indicated by reference numeral 361) labeled as normal, and the dashed line average spectrum is the average spectrum near the BPF (indicated by reference numeral 361) labeled as abnormal.

[0109] 16, the peak indicated by reference numeral 362 is the natural frequency. When selecting a BPF using the above-described selection method (e), the user turns off the sideband selection toggle button 360 on the BPF selection screen 500, thereby excluding selection methods (b) and (c). Alternatively, when using the above-described selection methods (a) and (e), BPFs that include the natural frequency are automatically excluded from the BPFs.

[0110] The feature extraction unit 110 extracts only the power of the average spectrum obtained by the Fourier transform, whose center frequency is within the range of the BPF, as feature information of the average spectrum (step S306). That is, the feature extraction unit 110 extracts feature information using the BPF selected by the band selection unit 113. For example, the feature extraction unit 110 may set the bandwidth to zero, select the center frequency of the Fourier transform closest to the BPF center frequency in equation (3), and extract the power corresponding to the center frequency in the average spectrum as feature information. The feature extraction unit 110 converts the amplitude and power extracted as feature information from the average spectrum into optimal values, such as a linear scale or a logarithmic scale (dB), depending on the machining method and tool type.

[0111] In step S105 of Fig. 10, the frequency analysis unit 115 performs frequency analysis on the waveform data being processed using an FFT algorithm or the like, while shifting a preset number of data samples and the start position of the data string from among the waveform data being processed, similar to step S204 of Fig. 13. This results in a three-dimensional data group in which multiple spectra SPj(f) are arranged in chronological order. Here, j (= 1 to J) is the number of spectra, and corresponds to the number of spectra for which frequency analysis was performed while shifting the start position of the data string.

[0112] Next, a description will be given of a first method for determining the state of the machining tool 223 of the machining device 200. The first method is to compare the TPF or harmonics with a threshold value to determine whether an abnormality has occurred.

[0113] In the first determination method, the feature extraction unit 110 first calculates the average spectrum SP(f) of the multiple spectra SPj(f). Next, the feature extraction unit 110 extracts, from the average spectrum SP(f), the power or amplitude closest to the BPF center frequency as feature information. If the extracted feature information is the TPF and its harmonics, the determination unit 102 compares the TPF and its harmonics with preset thresholds for the TPF and its harmonics. If the feature information is less than the thresholds, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is abnormal. If the extracted feature information is sidebands and other harmonics of the fundamental rotation frequency, the determination unit 102 compares the sidebands and harmonics with preset thresholds for the sidebands and harmonics. If the feature information exceeds the thresholds, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is abnormal. Alternatively, if the extracted feature information is a sideband wave or other harmonic of the fundamental rotation frequency, the judgment unit 102 may calculate the rate at which the feature information exceeds a threshold, compare the rate with a preset threshold for the rate, and judge that the state of the machining tool 223 of the machining device 200 is abnormal if the rate exceeds the threshold.

[0114] Next, we will explain the second determination method for the state of the machining tool 223 of the machining device 200. The second determination method converts the TPF or harmonics into a multidimensional vector, detects outliers using one-class SVM, and determines whether an abnormality exists.

[0115] In the second determination method, the feature extraction unit 110 extracts feature information in the same manner as in the first determination method. Next, the determination unit 102 performs one-class SVM learning using the multidimensional feature information, and determines whether the state of the machining tool 223 of the machining device 200 is abnormal by outlier detection.

[0116] Next, we will explain a third determination method for the state of the machining tool 223 of the machining device 200. The third determination method performs abnormality determination using a GMM (Gaussian Mixture Model), an HMM (Hidden Markov Model), a machine learning model, AI identification, etc. in addition to the TPF.

[0117] In the third determination method, the determination unit 102 reads from the storage unit 103 a learning model generated by the model generation process shown in FIG. 13 (a learning model when the state of the machining tool 223 of the machining device 200 is normal). The learning model may be, for example, a probability density function P(X) such as a GMM (Gaussian Mixture Model). Here, X (= {x1, x2, ... xn}) is an n-dimensional feature extracted according to a BPF during training of the learning model. The BPF is stored in the storage unit 103 together with the learning model, and in step S106 of FIG. 13, the feature of each of the multiple spectra SPj(f) is extracted using this BPF.

[0118] 13, if the likelihood obtained by inputting the feature into the probability density function P(X) is equal to or greater than a preset threshold, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is normal, and if the likelihood is less than the threshold, the determination unit 102 determines that the state of the machining tool 223 of the processing device 200 is abnormal. Alternatively, as shown in the following formula (5), the determination unit 102 defines a value obtained by reversing the sign of the logarithmic likelihood as the abnormality score aj, and sets the abnormality score as an index value such that the abnormality score increases as the abnormality state of the processing device 200 becomes stronger, and obtains the abnormality score aj for the number j=1 to J of spectra. aj=-log(P(Xj)) (5)

[0119] The determination unit 102 selects, as the total score of the abnormality scores aj, for example, the maximum value of the abnormality scores aj, the average of the abnormality scores aj, or a value suitable for the tool and processing method, as shown in equation (6). A = (Σaj) / J (6)

[0120] Then, the judgment unit 102 compares the abnormality score aj with a preset threshold value, and if the abnormality score aj is equal to or greater than the threshold value, judges that the state of the machining tool 223 of the processing device 200 is abnormal, and if the abnormality score aj is less than the threshold value, judges that the state of the machining tool 223 of the processing device 200 is normal.

[0121] Here, a comparison example will be described between the machining score for the entire section (entry section + punching section) of a conventional tapping process and the machining score for the entire section (entry section only) of the tapping process of this embodiment.

[0122] FIG. 17 is a diagram showing an example of a machining score for the entire section (entry section + punching section) in a conventional tapping process, and FIG. 18 is a diagram showing an example of a machining score for the actual machining section (entry section only) in the tapping process of this embodiment.

[0123] As shown in Fig. 17, when the score is calculated for the entire section of the tapping process (entry section + punching section), the score value for "chipping occurs" in the tapping process is 0.2 to 0.6. On the other hand, as shown in Fig. 18, when the score is calculated for the actual machining section of the tapping process (entry section only), the score value for "chipping occurs" in the tapping process is 0.8 to 1.6. In other words, when the score is calculated for the actual machining section of the tapping process (entry section only), the score value increases by more than double compared to when the score is calculated for the entire section of the tapping process (entry section + punching section).

[0124] Therefore, by cutting out the waveform for the first half of the machining section as shown in Figure 18 and processing it, it is possible to improve the accuracy of determining the state of the machining tool and reduce the amount of calculation required to detect the state of the machining tool compared to Figure 17.

[0125] According to this embodiment, the diagnostic device changes the extraction section for extracting feature information associated with machining depending on at least one of the tool type, which indicates the type of tool installed in the machining device, and the machining type, which indicates the type of machining performed on the workpiece, and extracts feature information from the changed extraction section. This allows for detection of the state of the machining tool by excluding non-machining sections that do not directly contribute to machining for at least one of the tool type and machining type that perform machining for only a portion of the machining time, thereby improving the accuracy of determining the state of the machining tool. This further reduces the amount of calculation required for detecting the state of the machining tool. In other words, this embodiment saves on both manpower and time required for inspecting processed products.

[0126] Note that, although the feature extraction unit 110 of the diagnosis device 100 of this embodiment is configured to detect the actual machining section based on information from the numerical control unit 206, the present invention is not limited to this. For example, the section classification unit 119 of the feature extraction unit 110 may be configured to classify the machining section into an actual machining section and a non-machining section based on the current consumption of the machining device 200. Furthermore, for example, the section classification unit 119 of the feature extraction unit 110 may be configured to classify (classify) the machining section into an actual machining section and a non-machining section based on signals (vibration, power, acoustic energy signals, etc.) generated when the machining tool 223 performs a machining operation on the workpiece 224.

[0127] (Second embodiment) Next, a second embodiment will be described. The second embodiment will describe a case where the processing device 200 cannot output the ladder signal shown in Fig. 11(b) to the outside due to the specifications of the processing device, the old or new model of the processing device, etc. In other words, the second embodiment makes it possible to realize the same functions as the first embodiment even in a market (technical area) where there are policy or specification restrictions on the processing device 200.

[0128] As described above, in the second embodiment, as one of the specification or configuration constraints, the processing device 200 does not transmit a ladder signal but transmits only context information to the diagnostic device 100. In the following description of the second embodiment, the description of the same parts as in the first embodiment will be omitted, and only the parts that differ from the second embodiment will be described.

[0129] FIG. 19 is a diagram showing another example of the correspondence between the context information and the learning model stored in the diagnostic device according to the second embodiment.

[0130] The method of extracting an actual machining section (extraction section) from a machining section in this embodiment is shown for the case where only context information can be received from the numerical control unit 206 of the machining device 200. In other words, this corresponds to the case where there is no ladder signal shown in FIG. 11(b).

[0131] In this case, since no ladder signal is received from the numerical control unit 206 of the machining device 200, the machining section itself is unknown. Therefore, the machining waveform extraction unit 116 must first analyze the machining (cutting) signal of the machining section and estimate the section. Specifically, the machining waveform extraction unit 116 extracts a waveform representing the actual machining section and a waveform representing the non-machining section from the signal levels of the physical quantities.

[0132] The example shown in FIG. 19 differs from FIG. 5 in that it includes feed information indicating the feed distance of the machining tool 223 and machining depth information indicating the machining depth distance of the machining tool 223.

[0133] The processing waveform extraction unit 116 analyzes the processing signal (by time-frequency analysis such as spectrogram, or analysis from the time waveform) to estimate and extract the processing section.

[0134] Alternatively, the machining waveform extraction unit 116 detects the contact start period of the machining tool 223 and calculates the cutting time D / F [min] from the feed rate F [mm / min] and depth D [mm] from the contact start period. The machining waveform extraction unit 116 then determines the machining end time or estimates the extraction period by analyzing the machining (cutting) signal as a guide for the machining end time. The machining waveform extraction unit 116 may also detect the end point from the signal in the same way as the start point and estimate the extraction period. For example, in the frequency domain, the contact start period can be determined as the cutting start time (contact start period) when the S / 60 [Hz] component (S is the fundamental rotation frequency) or TPF component of the cutting signal, or a component including its harmonics, or a change in signal power is greater than or equal to a preset threshold. In the time domain (time waveform), the machining waveform extraction unit 116 may determine the cutting start period when a change in the signal level of a physical quantity is greater than or equal to a preset threshold.

[0135] The subsequent processing has been explained in the first embodiment, so the explanation will be omitted.

[0136] According to this embodiment, the diagnostic device changes the extraction section for extracting feature information associated with machining depending on at least one of the tool type, which indicates the type of tool installed in the machining device, and the machining type, which indicates the type of machining performed on the workpiece, and extracts feature information from the changed extraction section. This allows for detection of the state of the machining tool by excluding non-machining sections that do not directly contribute to machining for at least one of the tool type and machining type that perform machining for only a portion of the machining time, thereby improving the accuracy of determining the state of the machining tool. This further reduces the amount of calculation required for detecting the state of the machining tool. In other words, this embodiment saves on both manpower and time required for inspecting processed products.

[0137] The program executed by the diagnostic device 100 of this embodiment is provided by being pre-installed in the ROM 152 or the like. The program executed by the diagnostic device 100 of this embodiment may be provided as a computer program product by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, or a digital versatile disk (DVD).

[0138] Furthermore, the program executed by the diagnostic device 100 of this embodiment may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network. Also, the program executed by the diagnostic device 100 of this embodiment may be provided or distributed via a network such as the Internet.

[0139] The program executed by the diagnostic device 100 of this embodiment has a modular configuration including the above-mentioned units (communication control unit 101, judgment unit 102, generation unit 104, display control unit 105, feature extraction unit 110, reception unit 120, etc.), and in terms of actual hardware, the CPU 151 (an example of a processor) reads and executes the program from the ROM 152, thereby loading the above-mentioned units into a main storage device and generating the units on the main storage device. Also, the hardware of the diagnostic device 100 of this embodiment may be incorporated into the processing device 200, which is configured to execute the above-mentioned program, thereby forming a processing device with a diagnostic function.

[0140] Each function of each embodiment described above can be realized by one or more processing circuits. Here, the term "processing circuit" in this specification includes a processor programmed to perform each function by software, such as a processor implemented by an electronic circuit, an ASIC (Application Specific Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), a conventional circuit module, and other devices designed to perform each function described above.

[0141] The devices described in each embodiment represent only one of several computing environments for implementing the embodiments disclosed herein. In one embodiment, diagnostic device 100 includes multiple computing devices, such as a server cluster, configured to communicate with each other via any type of communication link, including a network, shared memory, etc., and to perform the processes disclosed herein.

[0142] It should be noted that diagnostic device 100 is not limited to a desktop PC as long as it is a device equipped with a communication function. Diagnostic device 100 may be, for example, an image forming device, a PJ (Projector), an IWB (Interactive White Board: a white board with an electronic blackboard function capable of mutual communication), an output device such as digital signage, a HUD (Head Up Display) device, industrial machinery, an imaging device, a sound collection device, medical equipment, a network home appliance, an automobile (Connected Car), a notebook PC (Personal Computer), a mobile phone, a smartphone, a tablet terminal, a game console, a PDA (Personal Digital Assistant), a digital camera, a wearable PC, or the like.

[0143] For example, aspects of the present invention are as follows. <1> a communication control unit that receives, from the processing device, a physical quantity that changes over time in accordance with a processing operation of a processing tool provided in the processing device on a workpiece; a feature extraction unit that extracts feature information related to the detection information from a result of frequency analysis of the detection information represented by the physical quantity; a determination unit that determines a state of the machining tool based on the characteristic information; Equipped with the feature extraction unit changes an extraction section for extracting the feature information from the machining section representing the machining operation in accordance with at least one of a tool type representing the type of the machining tool and a machining type representing the type of machining performed on the workpiece, and extracts the feature information in the changed extraction section. A diagnostic device characterized by: <2> the feature extraction unit extracts the feature information in the actual machining section from among an actual machining section in which the machining tool performs the machining operation on the workpiece and a non-machining section in which the machining tool does not perform the machining operation on the workpiece, which are included in the machining section; Characterized by <1> The diagnostic device described in <3> The machining operation is executed by a computer numerical control in the machining device, and the feature extraction unit further includes a section classification unit that classifies the machining section into the actual machining section and the non-machining section based on information related to the machining operation by the computer numerical control. Characterized by <2> The diagnostic device described in <4> the section classification unit classifies the machining section into the actual machining section and the non-machining section based on at least one of context information that defines the operation of the machining tool and a ladder signal that switches a signal level and outputs it depending on the machining section; Characterized by <3> The diagnostic device described in <5> The section classification unit classifies the processing section into the actual processing section and the non-processing section based on a ladder signal that switches the signal level between High and Low and outputs the ladder signal according to the processing section. Characterized by <4> The diagnostic device described in <6> The section classification unit classifies the machining section into the actual machining section and the non-machining section based on a current consumption of the machining device. Characterized by <3> The diagnostic device described in <7> the section classification unit classifies the machining section into the actual machining section and the non-machining section based on a signal generated during a machining operation of the machining tool on the workpiece; Characterized by <3> The diagnostic device described in <8> The feature extraction unit further includes a machining waveform extraction unit that extracts a waveform representing the actual machining section and a waveform representing the non-machining section from the signal level of the physical quantity. Characterized by <2> Or <7> 10. The diagnostic device according to claim 9, wherein the diagnostic device is a <9> The processing waveform extraction unit determines the signal level of the physical quantity from a time waveform. Characterized by <8> The diagnostic device described in <10> The processing waveform extraction unit extracts a value change of a specific band in a frequency domain of the signal level of the physical quantity as a waveform change. Characterized by <7> The diagnostic device described in <11> the feature extraction unit performs frequency analysis of the detection information during one processing operation for one or more frames and extracts a change over time of the frames as the feature information; Characterized by <1> Or <10> 10. The diagnostic device according to claim 9, wherein the diagnostic device is a <12> <1> Or <11> a diagnostic device according to any one of the preceding items; a machining device provided with a machining tool to be diagnosed by the diagnostic device; A processing system comprising: <13> A diagnostic method executed by a diagnostic device that diagnoses a state of a machining tool provided in a machining device, comprising: a communication control step of receiving, from the processing device, a physical quantity that changes over time in association with a processing operation on a workpiece; a feature extraction step of extracting feature information related to the detection information from a result of frequency analysis of the detection information represented by the physical quantity; a determining step of determining a state of the machining tool based on the characteristic information; Run The feature extraction step changes an extraction section for extracting the feature information from the machining section representing the machining operation according to at least one of a tool type representing the type of the machining tool and a machining type representing the type of machining performed on the workpiece, and extracts the feature information in the changed extraction section. A diagnostic method characterized by: <14> Computer, a communication control unit that receives, from the processing device, a physical quantity that changes over time in accordance with a processing operation of a processing tool provided in the processing device on a workpiece; a feature extraction unit that extracts feature information related to the detection information from a result of frequency analysis of the detection information represented by the physical quantity; a determination unit that determines a state of the machining tool based on the characteristic information; It functions as the feature extraction unit changes an extraction section for extracting the feature information from the machining section representing the machining operation according to at least one of a tool type representing the type of the machining tool and a machining type representing the type of machining performed on the workpiece, and extracts the feature information in the changed extraction section. program. [Explanation of symbols]

[0144] 100 Diagnostic equipment 102 Judgment section 110 Feature Extraction Unit 200 Processing equipment 206 Numerical Control Unit 223 Processing tools 224 Workpiece 227 Physical quantity information detection unit [Prior art documents] [Patent documents]

[0145] [Patent Document 1] Japanese Patent Publication No. 2022-072435

Claims

1. a communication control unit that receives, from the processing device, a physical quantity that changes over time in accordance with a processing operation of a processing tool provided in the processing device on a workpiece; a feature extraction unit that extracts feature information related to the detection information from a result of frequency analysis of the detection information represented by the physical quantity; a determination unit that determines a state of the machining tool based on the characteristic information; Equipped with the feature extraction unit specifies an extraction section for extracting the feature information from the machining section representing the machining operation according to at least one of a tool type representing the type of the machining tool and a machining type representing the type of machining performed on the workpiece, and extracts the feature information from the specified extraction section. A diagnostic device characterized by:

2. the feature extraction unit extracts the feature information in the actual machining section from among an actual machining section in which the machining tool performs the machining operation on the workpiece and a non-machining section in which the machining tool does not perform the machining operation on the workpiece, which are included in the machining section; The diagnostic device according to claim 1 .

3. the machining operation is performed by computer numerical control in the machining device; The feature extraction unit further includes a section classification unit that classifies the machining section into the actual machining section and the non-machining section based on information related to the machining operation by the computer numerical control.

3. The diagnostic device according to claim 2.

4. the section classification unit classifies the machining section into the actual machining section and the non-machining section based on at least one of context information that defines the operation of the machining tool and a ladder signal that switches a signal level and outputs it depending on the machining section; 4. The diagnostic device according to claim 3.

5. the section classification unit divides the machining section into the actual machining section and the non-machining section based on a ladder signal that switches between High and Low signal levels and outputs the ladder signal according to the machining section; 5. The diagnostic device according to claim 4.

6. The section classification unit classifies the machining section into the actual machining section and the non-machining section based on a current consumption of the machining device.

4. The diagnostic device according to claim 3.

7. the section classification unit classifies the machining section into the actual machining section and the non-machining section based on a signal generated during a machining operation of the machining tool on the workpiece; 4. The diagnostic device according to claim 3.

8. The feature extraction unit further includes a machining waveform extraction unit that extracts a waveform representing the actual machining section and a waveform representing the non-machining section from the signal level of the physical quantity.

3. The diagnostic device according to claim 2.

9. The processing waveform extraction unit determines the signal level of the physical quantity from a time waveform. The diagnostic device according to claim 8 .

10. The processing waveform extraction unit extracts a value change of a specific band in a frequency domain of the signal level of the physical quantity as a waveform change. The diagnostic device according to claim 8 .

11. the feature extraction unit performs frequency analysis of the detection information during one processing operation for one or more frames and extracts a change over time of the frames as the feature information; The diagnostic device according to claim 1 .

12. A diagnostic device according to any one of claims 1 to 11; a machining device provided with a machining tool to be diagnosed by the diagnostic device; A processing system comprising:

13. A diagnostic method executed by a diagnostic device that diagnoses a state of a machining tool provided in a machining device, comprising: a communication control step of receiving, from the processing device, a physical quantity that changes over time in association with a processing operation on a workpiece; a feature extraction step of extracting feature information related to the detection information from a result of frequency analysis of the detection information represented by the physical quantity; a determining step of determining a state of the machining tool based on the characteristic information; Run The feature extraction step specifies an extraction section for extracting the feature information from the machining section representing the machining operation according to at least one of a tool type representing the type of the machining tool and a machining type representing the type of machining performed on the workpiece, and extracts the feature information from the specified extraction section. A diagnostic method characterized by:

14. Computer, a communication control unit that receives, from the processing device, a physical quantity that changes over time in accordance with a processing operation of a processing tool provided in the processing device on a workpiece; a feature extraction unit that extracts feature information related to the detection information from a result of frequency analysis of the detection information represented by the physical quantity; a determination unit that determines a state of the machining tool based on the characteristic information; It functions as the feature extraction unit specifies an extraction section for extracting the feature information from the machining section representing the machining operation according to at least one of a tool type representing the type of the machining tool and a machining type representing the type of machining performed on the workpiece, and extracts the feature information from the specified extraction section. program.

Images