Diagnostic device, diagnostic system, diagnostic method, and program

Abstract

The learning device (10) includes a simulation data acquisition unit (102) that performs a machining simulation using a first machining program to acquire first simulation data, an actual machining data acquisition unit (103) that performs actual machining using the first machining program to acquire first actual machining data, and a model generation unit (104) that learns the relationship between the first simulation data and the first actual machining data and generates a learned model for predicting second actual machining data obtained by performing actual machining using the second machining program from second simulation data obtained by performing a machining simulation using the second machining program.

Description

[Technical field]

[0001] This disclosure , Diagnosis Cutting System and Diagnosis By law Regarding. [Background technology]

[0002] In the field of machine tools, cutting is performed by controlling the positional relationship between a tool attached to the machine tool and a workpiece (piece to be machined) based on a machining program, and cutting the workpiece into a desired shape with the tool. In cutting, if the tool is not in the specified shape (for example, if the tool is worn out), it may not be possible to machine the workpiece into the desired shape. In recent years, various attempts have been made to diagnose the machining state of machine tools in order to achieve the desired machining.

[0003] As a technology related to the above circumstances, Patent Document 1 discloses a machining state diagnosis device that learns the relationship between estimated machining state information obtained by virtually machining a workpiece based on control information and actual machining state information obtained by actually machining based on the control information. This machining state diagnosis device can diagnose whether the machining state is abnormal by comparing machining state information obtained by machining based on predetermined control information with estimated machining state information estimated from the corresponding control information and actual machining state information derived from the learned relationship. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-026598 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the machining state diagnosis device described in Patent Document 1 learns the relationship between estimated machining state information including multiple machining processes such as straight line machining, groove machining, hole machining, and corner R part machining, and actual machining state information, and there is a large error between the machining state information obtained by machining based on predetermined control information and the actual machining state information derived from the learned relationship and the estimated machining state information estimated from the corresponding control information.In addition, since the machining state information with a large error is compared with the actual machining state information to diagnose whether the machining state is abnormal or not, the diagnosis result of the machining state is unreliable.

[0006] In view of the above circumstances, an object of the present disclosure is to provide a diagnosis device that accurately estimates machining state information and reliably diagnoses whether or not the machining state is abnormal. [Means for solving the problem]

[0007] In order to achieve the above object, the diagnosis of the present disclosure system teeth, A prediction device and a diagnosis device are provided, The prediction device comprises: A storage means for storing a trained model generated by the learning device; a simulation data acquisition means for performing a machining simulation based on the second machining program and acquiring second simulation data; A prediction means for predicting second actual processing data based on the trained model and the second simulation data; Equipped with The diagnostic device comprises: The above The second machining program was used for actual machining. The above An actual processing data acquisition means for acquiring second actual processing data; The prediction device predicts A predicted data acquisition means for acquiring predicted data of the second actual processing data; Diagnosing whether or not there is an abnormality in the machining based on the feature amount of the second actual machining data and the feature amount of the predicted data of the second actual machining data. In addition to determining whether or not there is an abnormality in the machining, an approximation curve relating to the feature amount of the second actual machining data and the feature amount of the predicted data of the second actual machining data is obtained, and based on the obtained approximation curve, a diagnosis is made as to whether or not there is an abnormality in the tool. A diagnostic tool; Equipped with 、 The trained model is The aboveA trained model generated by a learning device, the trained model being obtained by learning a relationship between first simulation data and first actual processing data, for predicting the second actual processing data from the second simulation data; the first simulation data is obtained by a machining simulation using a first machining program by the learning device, The first actual machining data is obtained by performing actual machining using the learning device in accordance with the first machining program. Effect of the Invention

[0008] According to the present disclosure, actual machining data can be predicted with high accuracy, and it can be diagnosed with high reliability as to whether or not the machining is abnormal. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing an overall configuration of a processing system according to a first embodiment of the present disclosure. [Diagram 2] FIG. 1 is a diagram for explaining a functional configuration of a learning device according to a first embodiment of the present disclosure. [Diagram 3] FIG. 1 is a diagram illustrating a functional configuration of a prediction device according to a first embodiment of the present disclosure. [Figure 4] FIG. 1 is a diagram illustrating a functional configuration of a diagnostic device according to a first embodiment of the present disclosure. [Diagram 5] FIG. 1 is a diagram illustrating a functional configuration of a terminal device according to a first embodiment of the present disclosure. [Figure 6] FIG. 1 is a diagram illustrating an example of the hardware configuration of a learning device, a prediction device, a diagnosis device, and a terminal device according to a first embodiment of the present disclosure. [Figure 7] 1 is a flowchart illustrating an example of a model generation operation performed by a learning device according to a first embodiment of the present disclosure. [Figure 8] 1 is a flowchart illustrating an example of an operation of predicting actual processing data by a prediction device according to a first embodiment of the present disclosure. [Figure 9] 11 is a flowchart illustrating an example of an operation of diagnosing an abnormality in a tool by a diagnosing device according to a second embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment in which a learning device, a prediction device, a diagnosis device, and a diagnosis system according to the present disclosure are applied to a processing system will be described with reference to the drawings. In each drawing, the same or equivalent parts are denoted by the same reference numerals.

[0011] (Embodiment 1) A processing system 1 according to a first embodiment will be described with reference to FIG.

[0012] The machining system 1 includes a learning device 10, a prediction device 20, a diagnosis device 30, a sensor 31, a machine tool 40, a tool 41, a CNC (Computer Numerical Controller) 42, and a terminal device 50. The machining system 1 is a machining system that is introduced, for example, into a production site in a factory.

[0013] As described later, the machining system 1 executes each of the following phases: a machining simulation phase, an actual machining phase, a learning phase, a prediction phase, and a diagnosis phase. Note that the term "actual machining" is used to emphasize the difference from "machining simulation" which means virtual machining.

[0014] Next, the functions of each device will be outlined, and the details will be described later.

[0015] The learning device 10 generates a learned model required for predicting actual machining data in the prediction device 20 based on simulation data obtained by performing a machining simulation using a machining program and actual machining data obtained by performing actual machining using the machining program. The simulation data is data such as axis positions, tool axial cutting depths, tool radial cutting depths, and cutting volumes obtained in a time series manner during the machining simulation. The actual machining data is data such as axis positions, torque axis speeds, and the like obtained in a time series manner during actual machining. The learning device 10 is realized by, for example, a personal computer, a server, or the like. The learning device 10 is an example of a learning device according to the present disclosure.

[0016] The prediction device 20 predicts actual processing data from simulation data based on the learned model generated by the learning device 10. The prediction device 20 is realized by, for example, a personal computer, a server, etc. The prediction device 20 is an example of a prediction device according to the present disclosure.

[0017] The diagnostic device 30 compares actual machining data obtained from the sensor 31 during actual machining with actual machining data predicted by the prediction device 20 (prediction data), and diagnoses whether or not there is an abnormality in the machining. The sensor 31 is a variety of sensors installed on the machine tool 40 to detect the state of the machine tool 40. The sensor 31 obtains time-series data such as axis position and torque during actual machining. The diagnostic device 30 is realized by, for example, a PLC (Programmable Logic Controller), a personal computer, etc. The diagnostic device 30 is an example of a diagnostic device according to the present disclosure.

[0018] Machine tool 40 operates under the control of CNC 42, which controls the positional relationship between tool 41 attached to a spindle of machine tool 40 and a workpiece (piece to be machined) fixed to a table or a turning spindle of machine tool 40. Then, the tool 41 or the workpiece rotates as the spindle or turning spindle rotates, and cutting is performed in which tool 41 comes into contact with the workpiece and cuts off a part of the workpiece.

[0019] The terminal device 50 is, for example, a personal computer for FA (Factory Automation). The user of the terminal device 50 is, for example, an operator in a factory. The terminal device 50 is equipped with CAM (Computer Aided Manufacturing) and generates a machining program for realizing a desired machining operation by a user. The generated machining program is used to obtain simulation data by machining simulation or to obtain actual machining data by actual machining.

[0020] In the above description, various data are exchanged between multiple devices as appropriate. The data may be exchanged by communication between the devices or through removable media. In the following description, for simplicity, it is assumed that the devices are connected to each other by communication lines as shown in FIG. 1 as necessary.

[0021] Next, each of the above-mentioned machining simulation phase, actual machining phase, learning phase, prediction phase and diagnosis phase will be briefly described.

[0022] (1) Machining simulation phase In the machining simulation phase, a machining program is input to the terminal device 50, and the terminal device 50 performs a machining simulation using the input machining program to obtain simulation data. In the machining simulation, the positional relationship between the tool and the workpiece in the virtual space is controlled according to control information obtained by executing the machining program, and the shape of the workpiece is evaluated by removing the area through which the tool passes from the workpiece. The simulation data is data such as the axis position, the tool axial cutting depth, the tool radial cutting depth, and the cutting volume obtained in a time series manner during the machining simulation. The simulation data is used to generate a trained model in the learning phase, which will be described later.

[0023] The machining process performed by the machining program includes a plurality of machining processes such as linear machining, groove machining, and hole machining, but the machining program often does not include a description for identifying each machining process. In this machining simulation phase, each machining process is identified from the acquired simulation data, and the simulation data is divided into each machining process. Specifically, the time-series change in the simulation data is captured, and the simulation data is divided into each machining process. For example, it can be identified that the machining process has transitioned to groove machining at the point where the cutting depth in the tool diameter direction changes to a value close to the tool diameter. For example, it can be identified that the machining process has transitioned to hole machining at the point where the tool is changed to a drill. Furthermore, based on the information on the machining process identified in this machining simulation phase, the actual machining data is divided into each machining process in the actual machining phase described later.

[0024] In addition, the machining program executed during the machining simulation in this machining simulation phase is different from the machining program executed when predicting actual machining data in the prediction phase described later. For example, this means that in this machining simulation phase, the machining program that is the target of the machining simulation is machining program X, and in the prediction phase, the machining program that is the target of the prediction of actual machining data is machining program Y. Hereinafter, for the sake of distinction, the machining program in this machining simulation phase will be referred to as the "first machining program" and the simulation data as the "first simulation data".

[0025] (2) Actual machining phase In the actual machining phase, the first machining program is input to the machine tool 40, and the machine tool 40 performs actual machining. The diagnosis device 30 acquires data obtained from the sensor 31 during the actual machining as actual machining data. The actual machining data is data on axis positions, torque, etc. obtained in time series during the actual machining. The actual machining data is used to generate a trained model in the learning phase described later.

[0026] In addition, the actual machining data is divided into each machining process based on the information of the machining process identified in the above (1) machining simulation phase. Specifically, the time-series phases of the simulation data and the actual machining data may be aligned, and the actual machining data may be divided into each machining process in the same manner as the simulation data. Hereinafter, for the sake of distinction, the actual machining data obtained in this actual machining phase is referred to as the "first actual machining data".

[0027] In this actual machining phase, it is preferable that the tool 41 of the machine tool 40 is not deteriorated. The first actual machining data obtained in this actual machining phase is used to generate a trained model in a learning phase described later, and if the tool 41 is deteriorated, the relationship between the first simulation data and the first actual machining data is learned using the first actual machining data cut by the deteriorated tool, so that it is not possible to reliably diagnose whether the machining is abnormal or not.

[0028] (3) Learning Phase In the learning phase, the first simulation data for each machining process acquired in the above (1) machining simulation phase and the first actual machining data for each machining process acquired in the above (2) actual machining phase are input to the learning device 10, and the learning device 10 generates a trained model that associates the input first simulation data with the first actual machining data. The generated trained model is used to predict the actual machining data from the simulation data in the prediction phase described later. The "simulation data" here refers to simulation data that was not used to generate the trained model. By using this trained model, the actual machining data can be predicted from the simulation data without performing actual machining.

[0029] In order to improve the accuracy of prediction in the prediction phase described below, that is, to generate a trained model for making highly accurate predictions, training may be performed using a plurality of first simulation data and first actual processing data.

[0030] (4) Prediction Phase In the prediction phase, similarly to the (1) machining simulation phase, a machining simulation is performed on the terminal device 50 using a machining program different from the first machining program to obtain simulation data. Hereinafter, for the sake of distinction, the machining program in this prediction phase will be referred to as the "second machining program" and the simulation data will be referred to as the "second simulation data."

[0031] Next, the second simulation data for each machining process acquired by the terminal device 50 and the trained model generated in the above (3) learning phase are input to the prediction device 20, which predicts the actual machining data. The predicted actual machining data is used to diagnose whether or not the machining is abnormal in the diagnosis phase described later. Hereinafter, for the sake of distinction, the predicted actual machining data in this prediction phase will be referred to as "second actual machining data (predicted data)". Also, the actual machining data obtained by performing actual machining according to the second machining program in the diagnosis phase described later will be referred to as "second actual machining data".

[0032] (5) Diagnostic phase In the diagnosis phase, similarly to the (2) actual machining phase, the diagnosis device 30 performs actual machining according to the second machining program to obtain second actual machining data. Next, the diagnosis device 30 obtains the second actual machining data (prediction data) predicted in the above (4) prediction phase. The diagnosis device 30 compares the second actual machining data with the second actual machining data (prediction data) to diagnose whether or not there is an abnormality in the machining.

[0033] According to the above-mentioned phases (1) to (5), the processing system 1 according to the first embodiment can predict the second actual processing data (prediction data) for each processing step from the second simulation data for each processing step without performing actual processing. Also, according to the processing system 1 according to the first embodiment, it is possible to diagnose whether or not there is an abnormality in processing by comparing the second actual processing data with the second actual processing data (prediction data).

[0034] Next, the functional configuration of the learning device 10 will be described with reference to Fig. 2. The learning device 10 includes a communication unit 101, a simulation data acquisition unit 102, an actual processing data acquisition unit 103, a model generation unit 104, and a storage unit 105.

[0035] The communication unit 101 communicates with an external device and transmits and receives various data as necessary. The communication unit 101 is realized by, for example, a network interface.

[0036] The simulation data acquisition unit 102 acquires first simulation data for each machining process generated by the terminal device 50 in the (1) machining simulation phase. The simulation data acquisition unit 102 acquires the first simulation data generated by the terminal device 50, for example, by communication via the communication unit 101. The simulation data acquisition unit 102 is an example of a simulation data acquisition means according to the present disclosure.

[0037] The actual processing data acquisition unit 103 acquires first actual processing data for each processing step acquired by the diagnosis device 30 in the (2) actual processing phase. The actual processing data acquisition unit 103 acquires the first actual processing data acquired by the diagnosis device 30, for example, by communication via the communication unit 101. The actual processing data acquisition unit 103 is an example of an actual processing data acquisition means according to the present disclosure.

[0038] The model generation unit 104 generates a trained model for predicting the second actual processing data from the second simulation data based on the first simulation data for each processing step acquired by the simulation data acquisition unit 102 and the first actual processing data for each processing step acquired by the actual processing data acquisition unit 103. The model generation unit 104 stores the generated trained model in the storage unit 105. The model generation unit 104 generates the model by a machine learning method such as multiple regression analysis, support vector machine, random forest, gradient boosting tree (decision tree), etc. The model generation unit 104 is an example of a model generation means according to the present disclosure.

[0039] As described above, the model generation unit 104 generates a trained model based on the first simulation data for each machining process and the first actual machining data for each machining process. Therefore, the trained model generated is optimized for each machining process. Therefore, as described later, by predicting the second actual machining data (prediction data) for each machining process, the actual machining data can be predicted with high accuracy.

[0040] The storage unit 105 stores the trained model generated by the model generation unit 104. The storage unit 105 is an example of a storage means according to the present disclosure.

[0041] Next, the functional configuration of the prediction device 20 will be described with reference to Fig. 3. The prediction device 20 includes a communication unit 201, a model acquisition unit 202, a simulation data acquisition unit 203, a prediction unit 204, and a storage unit 205.

[0042] The communication unit 201 communicates with an external device and transmits and receives various data as necessary. The communication unit 201 is realized by, for example, a network interface.

[0043] The model acquisition unit 202 acquires the trained model generated by the learning device 10 in the (3) learning phase, and stores it in the storage unit 205. The model acquisition unit 202 acquires the trained model stored in the storage unit 105 of the learning device 10, for example, via the communication unit 201, and stores it in the storage unit 205.

[0044] The simulation data acquisition unit 203 acquires second simulation data for each processing step generated by the terminal device 50 in the (4) prediction phase. The simulation data acquisition unit 203 acquires the second simulation data generated by the terminal device 50, for example, by communication via the communication unit 201.

[0045] The prediction unit 204 refers to the trained model stored in the storage unit 205, inputs the second simulation data for each machining process acquired by the simulation data acquisition unit 203 to the trained model, and predicts the second actual machining data (prediction data) for each machining process. The prediction unit 204 is an example of a prediction means according to the present disclosure.

[0046] The storage unit 205 stores the trained model acquired by the model acquisition unit 202.

[0047] Next, the functional configuration of the diagnostic device 30 will be described with reference to Fig. 4. The diagnostic device 30 includes a communication unit 301, an actual processing data acquisition unit 302, a predicted data acquisition unit 303, a diagnostic unit 304, and a storage unit 305.

[0048] The communication unit 301 communicates with an external device and transmits and receives various data as necessary. The communication unit 301 is realized by, for example, a network interface.

[0049] The actual machining data acquisition unit 302 acquires data obtained from the sensor 31 via the communication unit 301 as second actual machining data for each machining process. Specifically, the actual machining data acquisition unit 302 acquires data obtained from the sensor 31 during actual machining by the machine tool 40 as second actual machining data for each machining process. The actual machining data acquisition unit 302 stores the acquired second actual machining data in the storage unit 305.

[0050] The predicted data acquisition unit 303 acquires the second actual processing data (predicted data) predicted by the prediction device 20. The predicted data acquisition unit 303 stores the second actual processing data (predicted data) in the storage unit 305. The acquired predicted data acquisition unit 303 is an example of a predicted data acquisition means according to the present disclosure.

[0051] The diagnosis unit 304 compares the second actual processing data acquired by the actual processing data acquisition unit 302 with the second actual processing data (predicted data) acquired by the predicted data acquisition unit 303 to diagnose whether or not there is an abnormality in processing. The diagnosis unit 304 is an example of a diagnosis means according to the present disclosure.

[0052] The storage unit 305 stores the second actual processing data for each processing step acquired by the actual processing data acquisition unit 302 and the second actual processing data (predicted data) acquired by the predicted data acquisition unit 303.

[0053] Next, the functional configuration of the terminal device 50 will be described with reference to Fig. 5. The terminal device 50 includes a communication unit 501, a simulation execution unit 502, and a storage unit 503.

[0054] The communication unit 501 communicates with an external device and transmits and receives various data as necessary. The communication unit 501 is realized by, for example, a network interface.

[0055] The simulation execution unit 502 receives a machining program, performs a machining simulation based on the input machining program, and acquires simulation data. The simulation execution unit 502 stores the acquired simulation data in the storage unit 503.

[0056] The storage unit 503 stores the simulation data for each processing step acquired by the simulation execution unit 502 .

[0057] Next, an example of the hardware configuration of the learning device 10, prediction device 20, diagnosis device 30, and terminal device 50 (hereinafter referred to as "learning device 10, etc.") will be described with reference to Fig. 6. The learning device 10, etc. shown in Fig. 6 is realized by a computer such as a personal computer, a microcontroller, or a PLC.

[0058] The learning device 10 etc. includes a processor 1001, a memory 1002, an interface 1003, and a secondary storage device 1004, which are connected to each other via a bus 1000.

[0059] The processor 1001 is, for example, a CPU (Central Processing Unit). The processor 1001 loads an operating program stored in a secondary storage device 1004 into a memory 1002 and executes the program, thereby implementing each function of the learning device 10 and the like.

[0060] The memory 1002 is a main storage device formed of, for example, a RAM (Random Access Memory). The memory 1002 stores an operating program that the processor 1001 reads from the secondary storage device 1004. The memory 1002 also functions as a working memory when the processor 1001 executes the operating program.

[0061] The interface 1003 is an I / O (Input / Output) interface such as a serial port, a USB (Universal Serial Bus) port, a network interface, etc. The interface 1003 realizes the functions of the communication unit 101, the communication unit 201, the communication unit 301, and the communication unit 501.

[0062] The secondary storage device 1004 is, for example, a flash memory, a hard disk drive (HDD), or a solid state drive (SSD). The secondary storage device 1004 stores an operation program executed by the processor 1001. The secondary storage device 1004 realizes the functions of the storage unit 105, the storage unit 205, the storage unit 305, and the storage unit 503.

[0063] Next, an example of the operation of model generation by the learning device 10 will be described with reference to Fig. 7. The operation shown in Fig. 7 is executed, for example, when a user operates the learning device 10 and instructs generation of a learning model. It is also assumed that (1) the machining simulation phase and (2) the actual machining phase have already been executed before starting the operation shown in Fig. 7.

[0064] The simulation data acquisition unit 102 of the learning device 10 acquires first simulation data for each machining process (step S101).

[0065] Next, the actual processing data acquisition unit 103 of the learning device 10 acquires the first actual processing data for each processing step (step S102).

[0066] The model generation unit 104 of the learning device 10 generates a trained model based on the first simulation data acquired in step S101 and the first actual processing data acquired in step S102, and stores the trained model in the storage unit 105 (step S103). Then, the learning device 10 ends the model generation operation.

[0067] Next, an example of the operation of predicting actual processing data by the prediction device 20 will be described with reference to Fig. 8. The operation shown in Fig. 8 is executed, for example, when a user operates the prediction device 20 and instructs prediction of actual processing data. In addition, it is assumed that (3) the learning phase has already been executed and a learned model has been generated before starting the operation shown in Fig. 8.

[0068] The model acquisition unit 202 of the prediction device 20 acquires a trained model and stores it in the storage unit 205 (step S201). However, for example, if the latest trained model is already stored in the storage unit 205, this operation does not necessarily have to be performed.

[0069] The simulation data acquisition unit 203 of the prediction device 20 acquires second simulation data for each processing step (step S202).

[0070] The prediction unit 204 of the prediction device 20 refers to the learned model stored in the storage unit 205 in step S201, and predicts the second actual processing data (prediction data) for each processing step from the second simulation data for each processing step acquired in step S202 (step S203). Then, the prediction device 20 ends the operation of predicting the actual processing data.

[0071] The above describes the machining system 1 according to the embodiment 1. According to the machining system 1 according to the embodiment 1, by referring to a trained model generated based on the first simulation data for each machining process obtained by performing a machining simulation using the first machining program and the first actual machining data for each machining process obtained by performing actual machining using the first machining program, it is possible to predict the second actual machining data (prediction data) for each machining process from the second simulation data for each machining process obtained by performing a machining simulation using the second machining program without performing actual machining.

[0072] Furthermore, according to the processing system 1 of embodiment 1, the second actual processing data (predicted data) is predicted for each processing step using a learned model generated based on the first simulation data for each processing step and the first actual processing data for each processing step, so that the diagnostic device 30 can compare the second actual processing data for each processing step with the second actual processing data (predicted data) to diagnose whether or not there is an abnormality in the processing.

[0073] (Embodiment 2) In the second embodiment, as described below, the actual machining data (predicted data) predicted in the first embodiment is utilized for abnormality diagnosis of a tool.

[0074] When a machine tool performs actual machining many times, the tools of the machine tool wear out. Therefore, a technology is required to diagnose whether there is an abnormality in the tools based on the actual machining state. For example, it is possible to diagnose the tools based on the sensor values ​​detected by the sensors of the machine tool during the execution of actual machining.

[0075] When the products manufactured by the machine tool are mass-produced products, the machining conditions are the same for each actual machining. Therefore, for example, by generating a trained model that correlates the sensor value with the presence or absence of an abnormality, it is possible to diagnose an abnormality in the tool based on the trained model and the sensor value.

[0076] However, when the product manufactured by the machine tool is a one-off item rather than a mass-produced item, the machining shape and machining process are different for each actual machining. Therefore, a trained model used for abnormality diagnosis of a tool used in manufacturing a one-off item cannot be generated from multiple identical machining data in the same way as for the above-mentioned mass-produced items.

[0077] In the second embodiment, tool abnormality diagnosis is performed based on actual machining data obtained during actual machining and actual machining data (prediction data) predicted from simulation data obtained by performing a machining simulation using a machining program.

[0078] The overall configuration of the machining system 1 according to the second embodiment is similar to that of the first embodiment shown in Fig. 1. However, as described below, the functions of the diagnostic device 30 are partially different. In the second embodiment, (1) the machining simulation phase, (2) the actual machining phase, and (3) the learning phase are assumed to be executed when the tool 41 is not deteriorated. The machining system 1 according to the second embodiment is an example of a diagnostic system according to the present disclosure.

[0079] In the second embodiment, the diagnostic device 30 diagnoses whether or not there is an abnormality in the tool 41 of the machine tool 40, based on the second actual machining data obtained by the sensor 31 and the second actual machining data (prediction data) predicted by the prediction device 20. The diagnostic device 30 according to the second embodiment is an example of a diagnostic device according to the present disclosure.

[0080] Among the functional configurations of a diagnostic device 30 according to the second embodiment, differences from the first embodiment will be described.

[0081] The diagnosing unit 304 diagnoses whether or not there is an abnormality in the tool 41 based on the feature amount of the second actual machining data acquired by the actual machining data acquiring unit 302 and the feature amount of the second actual machining data (predicted data) acquired by the predicted data acquiring unit 303. The diagnosing unit 304 is an example of a diagnosing means according to the present disclosure.

[0082] As the above feature quantity, for example, the average value, variance, maximum / minimum value, intermediate value, skewness, convexity, etc. of the torque during actual machining can be used. As a simple example, the diagnosing unit 304 may diagnose that the tool 41 is abnormal when the difference between the average value of the torque in the predicted data and the average value of the torque in the actual machining data is equal to or greater than a threshold value. For example, if the characteristics of the tool 41 are known and the torque increases when the tool 41 deteriorates, it is considered that the tool 41 is abnormal when the average value of the torque in the actual machining data is significantly higher than the average value of the torque in the predicted data. Note that the deviation of the feature quantity between the predicted data and the actual machining data may not only be higher but also lower depending on the machining process.

[0083] More complicated examples are also possible, as described below. For example, based on the feature amount of the second actual machining data that changes in time series and the feature amount of the second actual machining data (predicted data), an approximation curve relating to the distribution when the horizontal axis is elapsed time and the vertical axis is the difference between the feature amount of the second actual machining data and the feature amount of the second actual machining data (predicted data) is obtained, and an abnormality judgment of the tool 41 is performed based on the approximation curve. Note that the approximation curve here includes a "straight line" such as a regression line.

[0084] For example, a regression line is obtained as an approximation curve, and when the slope of the regression line is equal to or greater than a threshold, the tool 41 is diagnosed as having an abnormality. This is because the slope of the regression line increases as the difference between the feature amount of the second actual machining data and the feature amount of the second actual machining data (predicted data) increases. In other words, the greater the slope of the regression line, the greater the deviation between the second actual machining data and the second actual machining data (predicted data), which means that the tool 41 is degraded.

[0085] Alternatively, for example, an approximation curve may be obtained by curve fitting, and when an inflection point appears on the obtained approximation curve, the tool 41 may be diagnosed as having an abnormality. An inflection point is a point where the rate of change of the difference in the feature quantity reverses from positive to negative or from negative to positive, and therefore can be said to be a point where the tendency of the difference in the feature quantity changes significantly. It is considered that such a large change does not occur when the tool 41 is not deteriorated, so when an inflection point appears, it is considered that the tool 41 is deteriorated.

[0086] When diagnosing whether or not there is an abnormality in the tool 41, the diagnosis unit 304 notifies the diagnosis result. The diagnosis unit 304 outputs the diagnosis result as an alarm sound or a message to, for example, a display / alarm unit (not shown) connected to the diagnosis device 30. The diagnosis result is notified by displaying a message on the screen of the CNC 42 of the machine tool 40, outputting an alarm sound, or stopping new machining using the tool 41.

[0087] Next, an example of the operation of diagnosing a tool abnormality by the diagnosing device 30 will be described with reference to Fig. 9. The operation shown in Fig. 9 is executed when actual machining is performed after the above-mentioned phases (1) to (4) have been executed.

[0088] The actual processing data acquisition unit 302 of the diagnosis device 30 acquires second actual processing data from the sensor 31 (step S301).

[0089] The predicted data acquisition unit 303 of the diagnosis device 30 acquires the second actual processing data (predicted data) predicted by the prediction device 20 (step S302).

[0090] The diagnosis unit 304 of the diagnosis device 30 diagnoses whether or not there is an abnormality in the tool 41 based on the features of the second actual machining data acquired in step S301 and the features of the second actual machining data (predicted data) acquired in step S302 (step S303).

[0091] The diagnosing unit 304 notifies the result of the diagnosis in step S303 (step S304), and the diagnosing device 30 then ends the operation of diagnosing abnormalities in the tool.

[0092] The machining system 1 according to the second embodiment has been described above. According to the machining system 1 according to the second embodiment, whether or not there is an abnormality in the tool 41 is diagnosed based on the feature amount of the second actual machining data and the feature amount of the second actual machining data (prediction data) predicted from the second simulation data. Since the second actual machining data and the second actual machining data (prediction data) are data classified for each machining process, even if the workpiece machined by the machine tool 40 is not a mass-produced product, whether or not there is an abnormality in the tool 41 can be appropriately diagnosed.

[0093] (Modification) In the second embodiment, the diagnostic device 30 diagnoses the abnormality of the tool 41, but a device other than the diagnostic device 30 may diagnose the abnormality. For example, the diagnostic device 30 may output the second actual machining data to a personal computer, and the personal computer may acquire the second actual machining data and the second actual machining data (prediction data) to diagnose the abnormality of the tool 41. In this case, the personal computer is provided with functions corresponding to the actual machining data acquisition unit 302, the prediction data acquisition unit 303, and the diagnostic unit 304 shown in FIG. 4. The personal computer is an example of a diagnostic device according to the present disclosure.

[0094] In each embodiment, the learning device 10, the prediction device 20, the diagnosis device 30, and the terminal device 50 are each separate devices. However, the functions of two or more of these devices may be integrated into one device. For example, a server, a personal computer, or the CNC 42 of the device 40 may include the functions of each device, and various necessary data may be exchanged via communication via a network, a CPU bus, or the like.

[0095] 6, the learning device 10 etc. includes a secondary storage device 1004. However, this is not limiting, and the secondary storage device 1004 may be provided outside the learning device 10 etc., and the learning device 10 etc. and the secondary storage device 1004 may be connected via an interface 1003. In this configuration, removable media such as a USB flash drive or a memory card may also be used as the secondary storage device 1004.

[0096] 6, the learning device 10, etc. may be configured by a dedicated circuit using an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), etc. In the hardware configuration shown in Fig. 6, a part of the functions of the learning device 10, etc. may be realized by a dedicated circuit connected to the interface 103, for example.

[0097] The programs used in the learning device 10 and the like can be distributed by storing them on computer-readable recording media such as CD-ROMs (Compact Disc Read Only), DVDs (Digital Versatile Discs), USB flash drives, memory cards, HDDs, etc. Then, by installing such programs on a specific or general-purpose computer, the computer can function as the learning device 10 and the like.

[0098] Furthermore, the above-mentioned program may be stored in a storage device owned by another server on the Internet, and the above-mentioned program may be downloaded from that server.

[0099] Various embodiments and modifications of the present disclosure are possible without departing from the broad spirit and scope of the present disclosure. The above-described embodiments are for explaining the present disclosure and do not limit the scope of the present disclosure. In other words, the scope of the present disclosure is indicated by the claims, not the embodiments. Various modifications made within the scope of the claims and within the scope of the disclosure equivalent thereto are considered to be within the scope of the present disclosure. [Explanation of symbols]

[0100] 1 Machining system, 10 Learning device, 20 Prediction device, 30 Diagnosis device, 31 Sensor, 40 Machine tool, 41 Tool, 42 CNC, 50 Terminal device, 101 Communication unit, 102 Simulation data acquisition unit, 103 Actual machining data acquisition unit, 104 Model generation unit, 105 Memory unit, 201 Communication unit, 202 Model acquisition unit, 203 Simulation data acquisition unit, 204 Prediction unit, 205 Memory unit, 301 Communication unit, 302 Actual machining data acquisition unit, 303 Prediction data acquisition unit, 304 Diagnosis unit, 305 Memory unit, 501 Communication unit, 502 Simulation execution unit, 503 Memory unit, 1000 Bus, 1001 Processor, 1002 Memory, 1003 Interface, 1004 Secondary storage device.

Claims

1. an actual machining data acquiring means for acquiring second actual machining data by performing actual machining according to the second machining program; A predicted data acquisition means for acquiring predicted data of the second actual processing data; a diagnosis means for diagnosing whether or not there is an abnormality in machining based on a feature amount of the second actual machining data and a feature amount of the predicted data of the second actual machining data; Equipped with the diagnosing means obtains an approximation curve relating to a feature amount of the second actual machining data and a feature amount of the predicted data of the second actual machining data, and diagnoses whether or not there is an abnormality in a tool based on the obtained approximation curve; The predicted data of the second actual machining data is predicted by a prediction device based on a trained model and second simulation data, the second simulation data is obtained by a machining simulation by the prediction device using the second machining program, The trained model is a trained model generated by a learning device, and is obtained by learning a relationship between first simulation data and first actual processing data, and is a trained model for predicting the second actual processing data from the second simulation data; the first simulation data is obtained by a machining simulation using a first machining program by the learning device, The first actual machining data is obtained by performing actual machining according to the first machining program by the learning device. Diagnostic equipment.

2. A diagnostic device according to claim 1, the prediction device, and Equipped with The prediction device comprises: A storage means for storing the trained model generated by the learning device; a simulation data acquisition means for performing a machining simulation based on the second machining program and acquiring the second simulation data; A prediction means for predicting the second actual processing data based on the trained model and the second simulation data; Equipped with Diagnostic system.

3. performing actual machining according to a second machining program to obtain second actual machining data; Diagnosing whether or not there is an abnormality in the machining based on the feature amount of the second actual machining data and the feature amount of the predicted data of the second actual machining data; In the diagnosis, an approximation curve relating to a feature amount of the second actual machining data and a feature amount of predicted data of the second actual machining data is obtained, and whether or not there is an abnormality in the tool is diagnosed based on the obtained approximation curve; The predicted data of the second actual machining data is predicted based on a learned model and second simulation data, the second simulation data is obtained by a machining simulation based on the second machining program, The trained model is a trained model for predicting the second actual processing data from the second simulation data, the trained model being obtained by learning a relationship between the first simulation data and the first actual processing data; the first simulation data is obtained by a machining simulation based on a first machining program, The first actual machining data is obtained by performing actual machining according to the first machining program. Diagnostic methods.

4. Computer, an actual machining data acquiring means for acquiring second actual machining data by performing actual machining according to the second machining program; A predicted data acquisition means for acquiring predicted data of the second actual processing data; a diagnosis means for diagnosing whether or not there is an abnormality in machining based on a feature amount of the second actual machining data and a feature amount of the predicted data of the second actual machining data; Function as a the diagnosing means obtains an approximation curve relating to a feature amount of the second actual machining data and a feature amount of the predicted data of the second actual machining data, and diagnoses whether or not there is an abnormality in a tool based on the obtained approximation curve; The predicted data of the second actual machining data is predicted based on a learned model and second simulation data, the second simulation data is obtained by a machining simulation based on the second machining program, The trained model is a trained model for predicting the second actual processing data from the second simulation data, the trained model being obtained by learning a relationship between the first simulation data and the first actual processing data; the first simulation data is obtained by a machining simulation based on a first machining program, The first actual machining data is obtained by performing actual machining according to the first machining program. program.

Images