DE112020007089T5 - Numerical control device and machine learning device - Google Patents

Abstract

A numerical control device (1A) comprises a control unit (21) which controls a machine tool (2A), a state observation unit (11A) which observes state variables indicating a motor load current value (41) of a motor which is a cutting tool used by the machine tool (2A). a cutting coordinate value (40) which is a cutting position of a workpiece for the cutting tool, a tool type (42) which is a type of the cutting tool, a workpiece type (43) which is a type of the workpiece, and a workpiece temperature (44) which is a temperature of the workpiece, a data acquisition unit (12) which acquires a wear amount measurement result (45) which is a result of measuring a wear amount of the cutting tool, and a learning unit (13) which reads the wear amount of the cutting tool Learns based on a data set, which is based on a combination of the state variables and the wear amount measurement result (45).

Description

Area

The present disclosure relates to a numerical control device and a machine learning device (machine learning device) that estimate a wear amount of a cutting tool used by a machine tool.

background

A machine tool, which machines a workpiece using a cutting tool, is controlled by a numerical control device. When this machine tool performs machining for a long time, a cutting edge of the cutting tool is worn, and therefore a size error of a finished work increases. In order to reduce this size error, an operator of the machine tool repeatedly measures a wear amount of the cutting tool.

In order to save the labor of such wear amount measurement by the operator, various efforts have been made. A numerical control apparatus described in Patent Literature 1 estimates an actual wear amount of a cutting tool from data indicating a correlation between a rate of change of a motor load current and a wear amount of the cutting tool when the cutting tool touches a workpiece.

citation list

patent literature

Patent Literature 1: Japanese Patent Application Publication No. H10-20911 ( JP H10-20 911 A )

short description

Technical problem

However, in the technique described in the above Patent Literature 1, the wear amount of the cutting tool is estimated without considering an attachment deviation of the workpiece, a shape deviation of the workpiece, and the like. That is, with the technique described in the above Patent Literature 1, it is not possible to distinguish between an increase in motor load current associated with wear of the cutting tool and an increase in motor load current caused by a change in a workpiece cutting area associated with the attachment deviation of the work, the shape deviation of the work, or the like, and it is not possible to estimate an accurate amount of wear.

The present disclosure has been made with the above circumstances in mind, and an object of the present disclosure is to provide a numerical control apparatus that can accurately estimate a wear amount of a cutting tool.

the solution of the problem

In order to solve the above problems and achieve the object, the present disclosure provides a numerical control apparatus including: a control unit that controls a machine tool based on a machining program; a state observation unit that observes state variables including a load current value of a motor that drives a cutting tool used by the machine tool, a cutting position of a workpiece for the cutting tool, a tool type that is a type of the cutting tool, a workpiece type that is a type of the workpiece, and a workpiece temperature, which is a temperature of the workpiece; a data obtaining unit that obtains a wear amount measurement result which is a result of measuring a wear amount of the cutting tool; and a learning unit that creates a learning model used for estimating the wear amount of the cutting tool from the state variables based on a data set generated based on a combination of the state variables and the wear amount measurement result.

Advantageous Effects of the Invention

A numerical control device according to the present disclosure has an advantageous effect of being able to accurately estimate a wear amount of a cutting tool.

character list

• 1 12 is a diagram showing a configuration of a control system including a numerical control device according to a first embodiment.
• 2 14 is a flowchart showing a processing procedure of machine learning performed by a machine learning apparatus according to the first embodiment.
• 3 14 is a flowchart showing an estimation processing procedure of an estimated wear amount, which is performed by the machine learning apparatus according to the first embodiment.
• 4 14 is a diagram showing a configuration of a neural network used by the machine learning device according to the first embodiment.
• 5 14 is a diagram showing a hardware configuration example by which the machine learning device according to the first embodiment is implemented.
• 6 14 is a diagram showing a configuration of a control system including a numerical control device according to a second embodiment.
• 7 14 is a diagram for explaining cutting start coordinates for a workpiece without attachment errors, which are detected by the numerical control apparatus according to the second embodiment.
• 8th 14 is a diagram for explaining cutting start coordinates for a workpiece with an attachment error, which are detected by the numerical control apparatus according to the second embodiment.

Description of embodiment

Hereinafter, a numerical control device and a machine learning device according to embodiments of the present disclosure will be described in detail with reference to the drawings.

First embodiment.

1 12 is a diagram showing a configuration of a control system including a numerical control device according to a first embodiment. A control system 100A includes a numerical control device (NC device) 1A and a machine tool 2A.

The numerical control device 1A is a computer configured to control the machine tool 2A. The machine tool 2A is an apparatus configured to machine a workpiece, which is an object to be machined, using a cutting tool. The machine tool 2A includes a drive unit 31, a man-machine interface (HMI) screen 32, a temperature sensor 33, and a wear-amount gauge 34.

The drive unit 31 drives one or more motors. The motors driven by the drive unit 31 include a servo motor and a main spindle motor. In a case where the servo motor drives the cutting tool, the drive unit 31 transmits a load current value of the servo motor as a motor load current value 41 to the numerical controller 1A. In a case where the main spindle motor drives the cutting tool, the drive unit 31 transmits a load current value of the main spindle motor as a motor load current value 41 to the numerical control device 1A.

The MMS screen 32 is a screen set to display information entered by an operator. The MMS screen 32 is connected to an input device (not shown) configured to receive information entered by the operator and to display information transmitted by the input device. Examples of the input device include a mouse, a keyboard, or the like.

The information entered by the operator into the input device includes a tool type 42 and a workpiece type 43. The tool type 42 and the workpiece type 43 displayed on the HMI screen 32 are transmitted to the numerical control device 1A. The tool type 42 is information indicating a type of the cutting tool, and the workpiece type 43 is information indicating a type of the workpiece. The workpiece type 43 includes information on a material of the workpiece, information on a shape of the workpiece, information on a size of the workpiece, and the like. Note that the tool type 42 and the workpiece type 43 can be set by any method. In the following description, a case will be described in which the tool type 42 and the workpiece type 43 input to the input device are transmitted to the HMI screen 32, and the tool type 42 and workpiece type 43 are transmitted from the HMI screen 32 to the numerical controller 1A.

The temperature sensor 33 is an example of a temperature detection device capable of measuring a temperature of a workpiece. The temperature sensor 33 transmits the measured temperature to the numerical controller 1A as a workpiece temperature 44.

The wear amount measuring device 34 is a device configured to measure a wear amount of the cutting tool. In a case where the amount of wear- Measuring device 34 is a vernier caliper or the like used for manually measuring a wear amount of the cutting tool, a wear amount measurement result 45, which is a measurement result of the wear amount, is inputted into the numerical control device 1A by the operator. In a case where the wear amount gauge 34 is a device that automatically measures the wear amount of the cutting tool, the wear amount gauge 34 transmits the wear amount measurement result 45 to the numerical control device 1A.

In a case where the wear-amount gauge 34 is a manual gauge, the wear-amount gauge 34 is arranged outside of the machine tool 2A, whereas in a case where the wear-amount gauge 34 is an automatic gauge, the wear-amount gauge is the wear-amount gauge 34 arranged within the machine tool 2A. In the following description, a case where the wear amount measuring device 34 is an automatic measuring device and the wear amount measurement result 45 is transmitted from the wear amount measuring device 34 to the numerical control device 1A will be described.

The numerical control device 1A includes a machine learning device 10, a control unit 21, an estimated wear amount considering unit 22. The control unit 21 controls the machine tool 2A using a machining program 20. In addition, the control unit 21 calculates a machining program 20 based on the machining program 20 when executing the machining program Cutting coordinate value 40, which indicates a cutting position (tool coordinates) on the workpiece by the cutting tool. The control unit 21 transmits the cutting coordinate value 40 to the machine learning device 10.

The machine learning device 10 is a computer configured to learn a wear amount of the cutting tool based on information acquired when the workpiece is machined by the cutting tool. The machine learning device 10 has a function of learning the wear amount of the cutting tool and a function of estimating the wear amount of the cutting tool using a learning result. The machine learning device 10 outputs an estimated wear amount 75 , which is an estimation result, to the estimated wear amount considering unit 22 .

The machine learning apparatus 10 includes a state observation unit 11A, a data acquisition unit 12 and a learning unit 13. The state observation unit 11A acquires the motor load current value 41, the tool type 42, the workpiece type 43 and the workpiece temperature 44 from the machine tool 2A.

Specifically, the state observation unit 11A obtains the motor load current value 41 from the drive unit 31, and obtains the tool type 42 and the workpiece type 43 from the HMI screen 32. Further, the state observation unit 11A obtains the workpiece temperature 44 from the temperature sensor 33. Further, the state observation unit 11A obtains the cutting coordinate value 40 from of the control unit 21. The state observation unit 11A transmits the related cutting coordinate value 40, motor load current value 41, tool type 42, workpiece type 43 and workpiece temperature 44 to the learning unit 13.

The data acquisition unit 12 acquires the wear-amount measurement result 45 from the wear-amount measurement device 34. The data acquisition unit 12 transmits the acquired wear-amount measurement result 45 to the learning unit 13.

The learning unit 13 learns the estimated wear amount 75 based on a data set based on a combination of the cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43 and the workpiece temperature 44 output from the state observation unit 11A, and the Amount of wear measurement result 45 output from the data acquisition unit 12 is generated. Here, the data set is data in which a state variable and determination data are associated with each other. In the first embodiment, the cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43, and the workpiece temperature 44 are state variables, and the wear amount measurement result 45 is determination data.

In a learning phase, the learning unit 13 learns the estimated wear amount 75 by updating a learning model such as a neural network. The learning unit 13 adapts the learning model such that when the state variable is input to the learning model, the wear amount measurement result 45 is output from the learning model. The learning unit 13 has this learning model stored in itself.

When the state variable is received from the state observation unit 11A in an application phase (estimation phase), there is the learning unit 13 enters this state variable into the learning model. In this case, the estimated amount of wear 75 associated with the state variable is output from the learning model. The learning unit 13 transmits the estimated wear amount 75 to the estimated wear amount considering unit 22.

The estimated wear amount considering unit 22 calculates a correction amount 76 based on the estimated wear amount 75. The correction amount 76 is a position correction amount of the cutting tool, which is set to correct a workpiece machining position during machining performed by the cutting tool. The correction amount 76 is used to eliminate work machining error caused by wear of the cutting tool.

The estimated wear amount consideration unit 22 , which is a consideration unit, is configured to consider the correction amount 76 at a position of the cutting tool by transmitting the correction amount 76 to the control unit 21 . The control unit 21 controls the machine tool 2A while correcting the position of the cutting tool using the correction amount 76.

Next, a processing procedure of machine learning performed by the machine learning device 10 and an estimation processing procedure of the estimated wear amount 75 performed by the machine learning device 10 will be described. 2 14 is a flowchart showing a processing procedure of machine learning performed by a machine learning apparatus according to the first embodiment.

The tool type 42 and the workpiece type 43, which are used for machining a workpiece, are set using the HMI screen 32 in advance. The state observation unit 11A acquires the tool type 42 and the workpiece type 43 from the HMI screen 32 (step S10).

When execution of the machining program 20 starts, the control unit 21 calculates the cutting coordinate value 40 based on the machining program 20. The state observation unit 11A acquires the cutting coordinate value 40 from the control unit 21 (step S20). In addition, the state observation unit 11A acquires the motor load current value 41 from the drive unit 31 (step S30), and acquires the workpiece temperature 44 from the temperature sensor 33 (step S40). The state observation unit 11A transmits the acquired data to the learning unit 13 (step S50). The data transmitted to the learning unit 13 by the state observation unit 11A includes the cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43, and the workpiece temperature 44.

In machine learning, the data acquisition unit 12 acquires the wear amount measurement result 45 from the wear amount meter 34 (step S60). The data acquisition unit 12 transmits the acquired data to the learning unit 13 (step S70). That is, the data acquisition unit 12 transmits the wear amount measurement result 45 to the learning unit 13 . It should be noted that the processing from step S10 to step S70 can be executed in any order.

The learning unit 13 learns the estimated wear amount 75 based on the data acquired by the state observation unit 11A and the data acquisition unit 12 (step S80). This means that the learning unit 13 creates a learning model based on the data set in which the state variable and the determination data are associated with one another. The learning unit 13 has the learning model stored in itself.

3 14 is a flowchart showing an estimation processing procedure of an estimated wear amount, which is performed by the machine learning apparatus according to the first embodiment. The description of those of in 3 Processes shown that correspond to the in 2 processes shown are the same is omitted.

In the processing performed by the machine learning device 10 in estimating the estimated wear amount 75, steps S10 to S50 are the same as those in machine learning. In step S50, after the state observing unit 11A transmits the acquired data to the learning unit 13, the learning unit 13 estimates the estimated wear amount 75 based on the learning model and the data acquired by the state observing unit 11A (step S90). That is, the learning unit 13 estimates the estimated wear amount 75 based on the state variable and the learning model. In this way, the learning unit 13 estimates the estimated wear amount 75 from the state variable obtained from the state observation unit 11A based on the learning result (learning model) created using the data set in which the State variable and the determination data are associated with each other.

Note that when estimating the estimated wear amount 75, the learning unit 13 may update the learning model. In other words, the learning unit 13 can perform learning of the estimated wear amount 75 while estimating the estimated wear amount 75 .

The learning unit 13 transmits the estimated wear amount 75 obtained by the estimation to the estimated wear amount considering unit 22. The estimated wear amount considering unit 22 calculates the correction amount 76 used to correct the position of the cutting tool based on the estimated wear amount 75 obtained The estimated wear amount considering unit 22 transmits the calculated correction amount 76 to the control unit 21. The control unit 21 controls the machine tool 2A using the correction amount 76 and the machining program 20.

A relation between each piece of data acquired by the state observation unit 11A and the wear amount measurement result 45 acquired by the data acquisition unit 12 will now be described. Because resistance between the workpiece and the cutting tool increases as the cutting tool wears and its blade loses sharpness, the motor load current value 41 increases. In addition, the motor load current value 41 may change when a cutting area generated by the cutting tool changes due to an attachment error of the work, a variation in the shape of the work, an expansion of the work, or the like.

The machine learning apparatus 10 uses the cutting coordinate value 40 to thereby determine whether the change in the motor load current value 41 is caused by the wear of the cutting tool or the change in the cutting area.

When the motor load current value 41 changes with the cutting coordinate value 40 being a normal coordinate set in advance, the machine learning device 10 determines that the motor load current value 41 has changed due to wear of the cutting tool. In this case, the machine learning device 10 adapts a learning model so that the estimated wear amount 75 approximates the wear amount measurement result 45 . That is, as the cutting coordinate value 40 approaches the normal coordinate, the machine learning device 10 adjusts the learning model so that the estimated wear amount 75 approaches the wear amount measurement result 45 .

On the other hand, when the motor load current value 41 changes and the cutting coordinate value 40 is not a normal coordinate set in advance, the machine learning apparatus 10 determines that the motor load current value 41 has changed due to the change in the cutting area caused by an attachment error of the workpiece or the like is caused. In this case, the machine learning device 10 adjusts the learning model so that the estimated wear amount 75 does not approximate the wear amount measurement result 45 . That is, as the cutting coordinate value 40 moves further away from the normal coordinate, the machine learning device 10 adjusts the learning model so that the estimated wear amount 75 becomes further away from the wear amount measurement result 45 .

In addition, when the materials or the shapes of the cutting tool and the workpiece change, the amount of wear of the cutting tool for the same machining may change. Therefore, the machine learning device 10 adapts the learning model based on the tool type 42 and the workpiece type 43 .

In addition, because the workpiece itself deforms when the workpiece temperature 44 changes, the cutting area changes when the workpiece temperature 44 changes. For example, because the workpiece expands as the workpiece temperature 44 increases, the cutting area increases. In this case, the motor load current value 41 also increases in association with an increase in the workpiece temperature 44 . This means that the motor load current value 41 changes accordingly when the workpiece temperature 44 changes. Therefore, the machine learning device 10 adapts the learning model based on the workpiece temperature 44 .

In this way, the machine learning device 10 is used to learn the estimated amount of wear 75 associated with the cutting coordinate value 40 , the motor load current value 41 , the tool type 42 , the workpiece type 43 and the workpiece temperature 44 . Note that the machine learning device 10 may be, for example, a device that is connected to the numerical control device 1A via a network and is provided separately from the numerical control device 1A. Further, the machine learning device 10 may be built in the numerical control device 1A. Furthermore, the machine learning device 10 can exist on a cloud server.

The learning unit 13 learns the estimated wear amount 75 by so-called supervised learning according to a neural network model, for example. In this example, the supervised learning means a model that gives a learning device a large number of data sets from specified inputs and results (labels) to thereby learn characteristics of these data sets and estimate a result from the input.

The neural network consists of an input layer consisting of several neurons, an intermediate layer (hidden layer) consisting of several neurons, and an output layer consisting of several neurons. The number of intermediate layers may be one, or may be two or more.

4 14 is a diagram showing a configuration of a neural network used by the machine learning device according to the first embodiment. For example, in a case of a three-layer neural network as shown in 4 As shown, when two or more inputs are input to the input layers X1 to X3, these values are multiplied by weights w11 to w16, and the multiplication results are input to intermediate layers Y1 and Y2, and further the results are multiplied by weights w21 to w26 and the multiplication results are output from output layers Z1 to Z3. These output results vary according to values of the weights w11 to w16 and the weights w21 to w26.

The neural network according to the first embodiment learns the estimated wear amount 75 by so-called supervised learning based on the data set calculated based on the combination of the cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43, the workpiece temperature 44 and the wear amount Measurement result 45 is generated.

That is, the neural network learns the estimated wear amount 75 by adjusting the weights w11 to w16 and w21 to w26 such that the result obtained by inputting the cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43 and of the workpiece temperature 44 and output from the output layers Z1 to Z3 approximates the wear amount measurement result 45 . The learning unit 13 has the neural network stored in it, the weights w11 to w16 and w21 to w26 of which have been adjusted.

Furthermore, the neural network can learn the estimated amount of wear 75 by so-called unsupervised learning. The unsupervised learning is a method of learning how input data is distributed by giving the machine learning device only a large amount of input data and learning a device that performs compression, classification or shaping or the like on the input data without providing corresponding teaching data (output data). performs. The unsupervised learning can perform processing of clustering features included in these data sets into same features and the other processing like these. Using this result, unsupervised learning can realize prediction of an output by establishing a criterion and assigning outputs such that the criterion is optimized. Further, there is a learning tool called semi-supervised learning as an intermediate problem that lies between unsupervised learning and supervised learning, and this is categorized as a case where there are only some pairs of input and output data and data the data pairs are different are data from inputs only.

Further, the learning unit 13 can learn the estimated wear amount 75 based on a data set generated for two or more numerical control devices 1A. It should be noted that the learning unit 13 can obtain the data sets from two or more machine tools 2A used in the same site, or can learn the estimated wear amount 75 using the data sets collected from two or more machine tools 2A that work independently from each other at different locations. In addition, it is possible to meanwhile add a numerical control device that collects the data set to the objects, or, on the contrary, remove the numerical control device from the objects. Further, the machine learning device that has learned the estimated wear amount 75 for one numerical control device can be attached to the other numerical control device, and an estimated wear amount 75 on the other numerical control device can be relearned and updated accordingly.

Further, as a learning algorithm used in the learning unit 13, deep learning (DL) for learning extraction from a feature set itself can be used, and the learning unit 13 can perform machine learning according to another publicly known method such as genetic programming, funk national logic programming or support vector machine.

A hardware configuration of the machine learning device 10 will now be described. 5 14 is a diagram showing a hardware configuration example for implementing the machine learning device according to the first embodiment. The machine learning device 10 can be implemented by an input device 103, a processor 101, a memory 102, a display device 105 and an output device 104.

Examples of the processor 101 include a central processing unit (CPU), a central processing device, a processing device, an arithmetic device, a microprocessor, a microcomputer, a digital signal processor (DSP), and a large scale integrated circuit system (LSI). Memory 102 is, for example, random access memory (RAM) and/or read only memory (ROM).

The machine learning device 10 is implemented by the processor 101 reading and executing a tutorial stored in the memory 102, the tutorial being configured to be executed by a computer to perform operations of the machine learning device 10. It can also be said that the learning program, which is a program for executing the operations of the machine learning device 10, causes a computer to execute a procedure or method for the machine learning device 10.

The learning program executed by the machine learning device 10 has a modular configuration including the state observation unit 11A, the data obtaining unit 12 and the learning unit 13, these units being loaded into a main storage device and created on the main storage device.

The input device 103 receives the motor load current value 41, the tool type 42, the workpiece type 43 and the workpiece temperature 44, which are state variables, from the machine tool 2A and inputs them to the processor 101. Further, the input device 103 receives the wear amount measurement result 45 which is the determination data from the machine tool 2A and inputs the wear amount measurement result 45 into the processor 101 . Further, the input device 103 receives the cutting coordinate value 40 which is a state variable from the control unit 21 and inputs the cutting coordinate value 40 into the processor 101 .

The memory 102 is used as temporary storage when the processor 101 executes various kinds of processing. The memory 102 stores the cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43, the workpiece temperature 44, the wear amount measurement result 45, the estimated wear amount 75, etc. The output device 104 outputs the estimated wear amount 75 and the like to the estimated Amount of wear consideration unit 22 off.

The display device 105 displays the cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43, the workpiece temperature 44, the wear amount measurement result 45, the wear amount estimated 75, and the like. An example of the display device 105 is a liquid crystal monitor.

The tutorial may be stored on a computer-readable storage medium in a file in an installable format or in an executable format and provided as a computer program product. Furthermore, the learning program can be provided to the machine learning device 10 via a network such as the Internet. It should be noted that some of the functions of the machine learning device 10 may be implemented by dedicated hardware, such as dedicated circuitry, and the others may be implemented by software or firmware. In addition, the numerical control device 1</b>A can also be implemented by a hardware configuration similar to the machine learning device 10 .

The machine tool 2A performs machining in a state where a workpiece is fixed in the machine tool 2A. However, there may be a case where deformation of the work or the like occurs due to the size of the work itself, a variation in the shape of the work, or a change in temperature of the work, in addition to the attachment error in attaching the work. Due to these factors, the cutting area of the workpiece changes even if the same machining is performed. Therefore, the amount of wear of the cutting tool also changes. Further, because a resistance between the workpiece and the cutting tool increases in association with wear of the cutting tool, the motor load current value 41 necessary for machining increases. In addition, in a case where the attachment error of the work, the variation in work shape, or the like is large, there is a concern that a work accuracy of a finished product is adversely affected.

In the first embodiment, the estimated wear amount 75 of the cutting tool can be estimated with high accuracy because the numerical controller 1A estimates the estimated wear amount 75 based on the motor load current value 41, the workpiece temperature 44 and the cutting coordinate value 40 to reduce the influence of the attachment error of the workpiece, the Variation of the workpiece shape or the like to be considered. Therefore, the numerical control device 1A can calculate the correction amount 76 with high accuracy based on the estimated wear amount 75 estimated with high accuracy.

Because the calculated correction amount 76 is automatically taken into account in the control unit 21, the numerical control apparatus 1A can be operated continuously for a long period of time without an operator's manual action, resulting in an improvement in productivity.

In this way, in the first embodiment, the numerical control apparatus 1A includes the state observation unit 11A, the data acquisition unit 12, and the learning unit 13. Then, the state observation unit 11 observes the state variables showing the motor load current value 41, the cutting coordinate value 40, showing a cutting position on the workpiece for the cutting tool for making a cut include the tool type 42, the workpiece type 43 and the workpiece temperature 44. In addition, the data acquisition unit 12 acquires the wear amount measurement result 45 which is the result of measuring the wear amount of the cutting tool. In addition, the learning unit 13 creates the learning model used for estimating the wear amount of the cutting tool from the state variables based on the data set generated based on the combination of the state variables and the wear amount measurement result 45 . Because the numerical controller 1A estimates the wear amount considering an attachment deviation of the workpiece, a shape deviation of the workpiece, or the like using the cutting coordinate value 40, the motor load current value 41, and the workpiece temperature 44, the wear amount of the cutting tool can be accurately estimated in this way.

Second embodiment.

Next, a second embodiment will be referred to 6 until 8th described. In the second embodiment, a numerical control apparatus detects a deviation of a workpiece, such as a deviation in an attachment position of the workpiece or a deviation in a shape of the workpiece.

6 12 is a diagram showing a configuration of a control system including the numerical control device according to the second embodiment. Components under the components in 6 , which has the same functions as the numerical control device 1A of Fig 1 achieve the first embodiment shown are denoted by the same reference numerals, redundant description will be omitted.

A control system 100B includes a numerical control device 1B and a machine tool 2B. The numerical control device 1B is a computer configured to control the machine tool 2B. The numerical control device 1</b>B has a function of detecting a deviation of the attachment position of the work or the like and a function of calculating an attachment error amount of the work. The machine tool 2B is an apparatus configured to machine the workpiece using a cutting tool, similar to the machine tool 2A. The machine tool 2A includes the drive unit 31 and the MMS screen 32.

Information entered by an operator into the input device connected to the HMI screen 32 includes the tool type 42, the workpiece type 43 and an error threshold 46. The error threshold 46 is a threshold used to determine whether the attachment error amount of the work is within an allowable range or not. That is, the error threshold 46 is a threshold used to determine whether or not to issue a warning regarding the mounting error amount of the workpiece. The attachment error amount of the workpiece is calculated based on the cutting start coordinates corresponding to the cutting start position of the workpiece for the cutting tool to perform cutting. A difference between cutting start coordinates in the normal situation and actual cutting start coordinates is the mounting error amount of the workpiece.

The numerical control device 1B includes the control unit 21, a state observation unit 11B, an deviation determination device 50 and an error amount calculation device 60. The deviation determination device 50, the error amount calculation device 60 and the numerical control device 1B can be implemented by a hardware configuration that Machine learning device 10 is similar. The state observation unit 11B of the numerical controller 1B acquires the motor load current value 41, the tool type 42, the workpiece type 43 and the error threshold value 46 from the machine tool 2B.

Specifically, the state observation unit 11B obtains the motor load current value 41 from the drive unit 31, and obtains the tool type 42, the workpiece type 43, and the error threshold value 46 from the HMI screen 32. Further, the state observation unit 11B obtains the cutting coordinate value 40 from the control unit 21. The state observation unit 11B transmits the related cutting coordinate value 40, the motor load current value 41, the tool type 42, the workpiece type 43 and the error threshold value 46 to the deviation determination device 50.

The deviation determination device 50 is a device configured to determine whether or not the attachment error amount of the workpiece, a workpiece shape, or the like exceeds the error threshold value 46 set in advance. The error-amount calculation device 60 is a device configured to calculate the attachment error amount of the workpiece.

The deviation determination device 50 includes an attachment deviation determination unit 52 and a warning display unit 51. In addition, the error amount calculation device 60 includes an error amount calculation unit 62 and a calculation result display unit 61.

The attachment deviation determination unit 52 receives the motor load current value 41, the cutting coordinate value 40, the tool type 42, the workpiece type 43, and the error threshold value 46, which are output from the state observation unit 11B. The attachment deviation determination unit 52 detects the cutting start coordinates corresponding to the cutting start position based on the motor load current value 41 and the cutting coordinate value 40.

The cutting start coordinates will now be described. 7 14 is a diagram for explaining cutting start coordinates on a workpiece without attachment errors, which are detected by the numerical control apparatus according to the second embodiment. 8th 14 is a diagram for explaining cutting start coordinates on a workpiece with an attachment error, which are detected by the numerical control apparatus according to the second embodiment.

In 7 and 8th a case will be described in which a Z-axis direction is a vertical direction and an XY plane which is parallel to an upper surface of a machining table 85 on which a workpiece 80 is mounted is a horizontal plane. That is, two axes in a plane parallel to the upper surface of the machining table 85, which are orthogonal to each other, are set as an X-axis and a Y-axis. In addition, an axis orthogonal to the X-axis and the Y-axis is set as a Z-axis.

In 7 a case is shown in which the workpiece 80 is mounted straight with no inclination with respect to the machining table 85. FIG. In 8th a case is shown in which the workpiece 80 is mounted in a state in which the workpiece 80 is inclined with respect to the machining table 85. FIG.

For example, a cutting tool 71 machines the workpiece 80 in the Z-axis direction. In this case, the cutting tool 71 is moved from an upper surface of the workpiece 80 and contacts the workpiece 80 .

In a case where there is no mounting error of the workpiece 80 on the machining table 85, the cutting tool 71 starts to machine the workpiece 80 after coming into contact with the workpiece 80 at a point of desired cutting start coordinates (X1, Z1). . In this case, since the cutting tool 71 has come into contact with the workpiece 80 at a point of the cutting start coordinates (X1, Z1), the motor load current value 41 increases rapidly at the cutting start coordinates (X1, Z1). Then, by performing the machining on the workpiece 80 in the Z-axis direction, the cutting tool 71 machines a machining portion 81 of the workpiece 80, the portion 81 extending in the Z-axis direction.

In contrast, in a case where there is an attachment error of the workpiece 80 on the machining table 85, the cutting tool 71 starts to machine the workpiece 80 after colliding with the workpiece 80 at a point of cutting start coordinates (X2, Z2) that different from the desired cutting start coordinates (X1, Z1) has come into contact. Because the cutting tool 71 has come into contact with the workpiece 80 at a point of the cutting start coordinates (X2, Z2), the motor load current value becomes 41 in In this case, it increases rapidly at the cutting start coordinates (X2, Z2). Then, by performing the machining on the workpiece 80 in the Z-axis direction, the cutting tool 71 machines a machining region 82 of the workpiece 80, the region 82 being in a lower portion. The machining area 82 is an area different from the machining area 81 .

As seen above, in a case where there is an attachment error of the workpiece 80, the cutting start coordinates change compared to a case where there is no attachment error. Likewise, in a case where the shape of the workpiece 80 varies, the cutting start coordinates change compared to a case where the shape of the workpiece 80 does not vary. In addition, even in a case where the workpiece 80 swells or contracts in size, the cutting start coordinates change compared to a case where the workpiece 80 does not swell or contract.

In a case where a coordinate difference, which is a difference between the changed cutting start coordinates (X2, Z2) and the cutting start coordinates (X1, Z1) in the normal situation, exceeds the error threshold value 46 set in advance, the attachment deviation determination unit 52 determines, that the workpiece 80 has a deviation. The deviation of the workpiece 80 includes an attachment deviation of the workpiece 80 on the machining table 85, a shape deviation of the workpiece 80, or the like.

The cutting start coordinates in the normal situation differ for each combination of the tool type 42 and the workpiece type 43. Therefore, the attachment deviation determination unit 52 determines whether the workpiece 80 has a deviation or not using the cutting start coordinates in the normal situation which of the combination from the tool type 42 and the workpiece type 43 are assigned.

When a deviation is detected, the attachment deviation determination unit 52 transmits deviation information indicating a deviation determination to the warning display unit 51. In addition, the attachment deviation determination unit 52 transmits an initial position error, which is a coordinate difference between the changed cutting start coordinates and the cutting start coordinates in the normal situation , the tool type 42 and the workpiece type 43 to the error quantity calculation unit 62.

When the deviation information is received from the attachment deviation determination unit 52, the warning display unit 51 displays a warning indicating that the workpiece 80 has deviation and warns an operator that there is a concern about the machining accuracy of the workpiece 80 in a finished product. Note that the warning display unit 51 can issue a warning by any method, not limited to displaying the warning. For example, the warning display unit 51 may output a warning sound.

When the initial position error is received from the attachment deviation determination unit 52, the error amount calculation unit 62 calculates a workpiece attachment error amount based on the initial position error, the tool type 42 and the workpiece type 43, assuming that the shape of the workpiece 80 is normal. The error amount calculation unit 62 transmits the workpiece attachment error amount, which is a calculation result, to the calculation result display unit 61. When the workpiece attachment error amount is received by the error amount calculation unit 62, the calculation result display unit 61 displays the received workpiece attachment error amount and warns an operator that concerns with respect to the machining accuracy of the workpiece 80 in a finished product. In this way, since the numerical controller 1</b>B warns the operator in a case where there is an abnormality such as a workpiece attachment error, it is possible to prevent a defective product of the workpiece 80 from being sold as a finished product.

Note that the numerical control devices 1A and 1B can be combined with each other. That is, the numerical control device 1</b>A may include the deviation determination device 50 , or the numerical control device 1</b>A may include the deviation determination device 50 and the error amount calculation device 60 . In addition, the numerical control device 1</b>B may include the machine learning device 10 , or the numerical control device 1</b>B may include the machine learning device 10 and the estimated wear-amount considering unit 22 .

As described above, in the second embodiment, the numerical control device 1B detects the deviation of the workpiece, such as for example, the deviation of the workpiece attachment position or the deviation of the workpiece shape, using the cutting coordinate value 40, the motor load current value 41 and the error threshold value 46. Therefore, in a case where deviation of the workpiece occurs, the numerical control device 1B can issue a warning or the like.

The configurations shown in the above embodiments are only examples, and each may be combined with other publicly known techniques, or the embodiments may be combined with each other. In addition, each of these configurations may be partially omitted and/or modified without departing from the scope of the present disclosure.

Reference List

1A, 1B

numerical control device;

2A, 2B

machine tool;

10

machine learning device;

11A, 11B

condition observation unit;

12

data reference unit;

13

learning unit;

20

machine processing program;

21

control unit;

22

estimated wear amount consideration unit;

31

drive unit;

32

MMS screen;

33

temperature sensor;

34

Wear Amount Gauge;

40

cutting coordinate value;

41

motor load current value;

42

tool type;

43

workpiece type;

44

workpiece temperature;

45

wear quantity measurement result;

46

error threshold;

50

deviation determination device;

51

warning display unit;

52

attachment deviation determination unit;

60

Defect Quantity Calculator;

61

calculation result display unit;

62

Defect Quantity Calculation Unit;

71

cutting tool;

75

Estimated amount of wear;

76

correction quantity;

80

Workpiece;

81, 82

machine processing area;

85

machining table;

100A, 100B

control system;

101

Processor;

102

Storage;

103

input device;

104

output device;

105

display device;

X1 to X3

input layer;

Y1, Y2

interlayer;

Z1 to Z3

output layer.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• JP H1020911A [0004]

Claims (6)

Numerical control device, comprising: a control unit that controls a machine tool based on a machining program; a state observation unit which observes state variables, which a load current value of a motor that drives a cutting tool used by the machine tool, a cutting position of a workpiece for the cutting tool, a tool type, which is a type of cutting tool, a workpiece type, which is a type of the workpiece, and a workpiece temperature, which is a temperature of the workpiece; a data obtaining unit that obtains a wear amount measurement result which is a result of measuring a wear amount of the cutting tool; and a learning unit that creates a learning model used for estimating the wear amount of the cutting tool from the state variables based on a data set generated based on a combination of the state variables and the wear amount measurement result. Numerical control device according to claim 1 , wherein the learning unit estimates the wear amount of the cutting tool from the state variables using the learning model. Numerical control device according to claim 2 , further comprising: a consideration unit that calculates a position correction amount used for correcting a position of the cutting tool based on the wear amount estimated by the learning unit and transmits the position correction amount to the control unit, thereby considering the position correction amount in the position of the cutting tool. Numerical control device according to one of Claims 1 until 3 , further comprising: a deviation determination device that calculates a cutting start coordinate, which is a coordinate at which the cutting tool starts to machine the workpiece, based on the cutting position and the load current value, and outputs a warning when a coordinate difference between a coordinate at which the The cutting tool starts to machine the workpiece in a state where the workpiece is normal and the cutting start coordinate exceeds a threshold value. Numerical control device according to claim 4 , further comprising: an error amount calculator that estimates an attachment error amount of the workpiece on the assumption that a shape of the workpiece is normal based on the coordinate difference and outputs the attachment error amount. Machine learning device comprising: a state observation unit which observes state variables, which a load current value of a motor that drives a cutting tool used by a machine tool, a cutting position of a workpiece for the cutting tool, a tool type, which is a type of cutting tool, a workpiece type, which is a type of the workpiece, and a workpiece temperature, which is a temperature of the workpiece; a data obtaining unit that obtains a wear amount measurement result which is a result of measuring a wear amount of the cutting tool; and a learning unit that creates a learning model used for estimating the wear amount of the cutting tool from the state variables based on a data set generated based on a combination of the state variables and the wear amount measurement result.

Images