CN114746203A

CN114746203A - Control device, electric discharge machine, and machine learning device

Info

Publication number: CN114746203A
Application number: CN201980102614.1A
Authority: CN
Inventors: 森田一成; 黑川聪昭
Original assignee: Mitsubishi Electric Corp
Current assignee: Mitsubishi Electric Corp
Priority date: 2019-12-03
Filing date: 2019-12-03
Publication date: 2022-07-12
Anticipated expiration: 2039-12-03
Also published as: WO2021111530A1; JPWO2021111530A1; JP6739690B1; CN114746203B

Abstract

The control device has: a processing condition output unit (27) that outputs processing conditions used in the machining of the wobbled-machined embossed discharge to a control unit that controls a control target; an actual performance calculation unit (23) that calculates, for each machining area, a machining actual performance of the electrical discharge machining of the shape relief of the specific section based on a control result that is a result of controlling, in the specific section, each machining area obtained by dividing the area subjected to the weaving machining using the machining condition; an evaluation calculation unit (22) that calculates an evaluation point representing the evaluation of the machining performance of the specific section, based on the machining performance for each machining area; and an evaluation output unit (26) that outputs the evaluation points and displays the evaluation points.

Description

Control device, electric discharge machine, and machine learning device

Technical Field

The present invention relates to a control device, an electric discharge machine, and a machine learning device applied to shape-carving electric discharge machining.

Background

An electric discharge machine is a device that performs electric discharge machining on a workpiece according to machining conditions, i.e., machining voltage. Whether or not a machining result obtained by electric discharge machining is acceptable can be determined based on machining performance such as time required for actual machining.

The electric discharge machine described in patent document 1 detects physical quantities indicating machining results, and determines whether or not machining results are acceptable based on a comparison result between the detected physical quantities and reference values set for the respective physical quantities.

Patent document 1: international publication No. 2000/32342

Disclosure of Invention

Since the electric discharge machine of patent document 1 is a wire electric discharge machine, it is possible to determine whether or not the machining result is acceptable when the machining in a specific direction is controlled in the 2-dimensional plane. However, the electric discharge machine of patent document 1 cannot be applied to the engraving electric discharge machining for performing the machining in the 3-dimensional direction. The wobble processing is performed in the case of the engraving electrical discharge machining. Since the machining result changes in a complex manner for each machining area in the 2-dimensional plane if the weaving machining is performed, the electric discharge machine of patent document 1 cannot determine whether the machining result is acceptable or not in consideration of the machining result for each machining area in the 2-dimensional plane that changes in a complex manner.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a control device capable of determining whether or not a machining result is acceptable in consideration of the machining result for each machining area in a 2-dimensional plane for the relief discharge machining.

In order to solve the above problems and achieve the object, a control device according to the present invention includes: a machining condition output unit that outputs a machining condition used in a shape-engraved electric discharge machining that performs wobbling to a control unit that controls a control target; and an actual result calculation unit that calculates a machining actual result of the electrical discharge machining of the specific section for each machining area based on a control result that is a result of controlling the machining conditions in the specific section for each machining area obtained by dividing the area subjected to the weaving machining. Further, a control device of the present invention includes: an evaluation calculation unit that calculates an evaluation point indicating an evaluation of the machining performance of the specific section based on the machining performance for each machining area; and an output unit that outputs the evaluation points and displays the evaluation points.

ADVANTAGEOUS EFFECTS OF INVENTION

The control device according to the present invention has an effect that, in the case of the die sinking electric discharge machining, whether or not the machining result is acceptable can be determined in consideration of the machining result for each machining area in the 2-dimensional plane.

Drawings

Fig. 1 is a diagram showing a configuration of an electric discharge machine according to embodiment 1.

Fig. 2 is a diagram showing a configuration of a control device according to embodiment 1.

Fig. 3 is a diagram for explaining a machining quadrant in electric discharge machining by the electric discharge machine according to embodiment 1.

Fig. 4 is a diagram for explaining weaving performed by the electric discharge machine according to embodiment 1.

Fig. 5 is a diagram for explaining a machining result in the weaving machining explained in fig. 4.

Fig. 6 is a flowchart showing a procedure of evaluation processing of a machining result by the electric discharge machine according to embodiment 1.

Fig. 7 is a diagram for explaining a relationship between the ratio of the machining time calculated by the control device according to embodiment 1 and the evaluation point.

Fig. 8 is a diagram for explaining the evaluation points for each machining quadrant calculated by the control device according to embodiment 1.

Fig. 9 is a diagram showing a correspondence relationship between the machining elapsed time and the machining depth calculated by the control device according to embodiment 1.

Fig. 10 is a diagram showing machining conditions corrected based on the evaluation points for each machining depth by the control device according to embodiment 1.

Fig. 11 is a diagram for explaining the relationship between the machining time and the evaluation point calculated by the control device according to embodiment 1.

Fig. 12 is a diagram for explaining the evaluation points for each machining quadrant calculated based on the target time by the control device according to embodiment 1.

Fig. 13 is a diagram showing the machining conditions corrected based on the evaluation point for each machining time by the control device according to embodiment 1.

Fig. 14 is a diagram showing a configuration of a control device according to embodiment 2.

Fig. 15 is a flowchart showing a procedure of evaluation processing of a machining result by the electric discharge machine according to embodiment 2.

Fig. 16 is a diagram for explaining a machining speed corresponding to a machining condition corrected by the electric discharge machine according to embodiment 2.

Fig. 17 is a diagram for explaining a machining speed for each machining quadrant corresponding to the machining condition corrected by the electric discharge machine according to embodiment 2.

Fig. 18 is a diagram showing a configuration of a control device according to embodiment 3.

Fig. 19 is a diagram showing another configuration example of the control device according to embodiment 3.

Fig. 20 is a diagram showing an example of a hardware configuration of a processing result evaluation unit included in the control devices according to embodiments 1 to 3.

Detailed Description

Next, a control device, an electric discharge machine, and a machine learning device according to embodiments of the present invention will be described in detail with reference to the drawings. The present invention is not limited to these embodiments.

Embodiment 1.

Fig. 1 is a diagram showing a configuration of an electric discharge machine according to embodiment 1. The electric discharge machine 1 is a device for performing a shape-engraved electric discharge machining. The electric discharge machine 1 applies a high-frequency pulse voltage between the tool electrode E as a machining electrode and the workpiece 17, thereby generating electric discharge between the tool electrode E and the workpiece 17. The electric discharge machine 1 sequentially removes the workpiece 17 by the generated electric discharge in a minute amount to machine the workpiece 17 into a shape corresponding to the shape of the tool electrode E. In the following description, the electric discharge machining of the workpiece 17 is sometimes referred to as machining.

The electric discharge machine 1 includes a base 19, a control device 2, a drive unit 12, a display unit 13, and a table 18. The control device 2 includes a machine control unit 14, a power supply control unit 15, and a machining result evaluation unit 16A. The electric discharge machine 1 machines a workpiece 17 placed on a table 18 on a base 19 using a tool electrode E attached to a drive unit 12. The tool electrode E is located at a position facing the workpiece 17.

The power supply control unit 15 and the machine control unit 14 are control units that control a control target. The power source control unit 15 controls electric power, and the machine control unit 14 controls the drive unit 12 and the like.

The power supply control unit 15 controls the power supplied between the tool electrode E and the workpiece 17. The power supply control unit 15 controls the electric discharge between the tool electrode E and the workpiece 17 based on the machining conditions transmitted from the machining result evaluation unit 16A.

The machine control unit 14 controls the position of the drive unit 12 and the like based on the machining conditions transmitted from the machining result evaluation unit 16A. The machine control unit 14 controls the drive unit 12 to control the distance between the tool electrode E and the workpiece 17 so that electric discharge occurs between the tool electrode E and the workpiece 17.

The machining result evaluation unit 16A outputs machining conditions to the machine control unit 14 and the power supply control unit 15 based on the machining position, the machining depth, and the like input by the operator of the electric discharge machining. An example of the machining condition output from the machining result evaluation unit 16A to the machine control unit 14 is the coordinates of the tool electrode E. The electric discharge machine 1 machines a position corresponding to the coordinates of the tool electrode E. An example of the machining condition output from the machining result evaluation unit 16A to the power supply control unit 15 is a voltage value of a high-frequency pulse voltage used for electric discharge machining.

The machining result evaluation unit 16A evaluates the machining result based on the control result transmitted from at least one of the machine control unit 14 and the power supply control unit 15. The control result transmitted from the machine control unit 14 is a machining condition or a control value actually used by the machine control unit 14 at the time of machining. The control result transmitted from the power supply control unit 15 is a machining condition or a control value actually used by the power supply control unit 15 at the time of machining. The control result for each section is transmitted from at least one of the machine control unit 14 and the power supply control unit 15 to the machining result evaluation unit 16A. The interval here is defined by a machining depth or a machining time. That is, the machining result evaluation unit 16A obtains a control result for each machining depth or a control result for each machining time (time required for machining). The machining result evaluation unit 16A calculates a machining result based on the control result transmitted from at least one of the machine control unit 14 and the power supply control unit 15. The machining result evaluation unit 16A calculates a machining result such as a machining time when a control result for each machining depth (a specific range of machining depths) is acquired as a control result of the specific section. Further, the machining result evaluation unit 16A calculates a machining result of the machining depth when a control result for each machining time (a specific range of machining time) is acquired as a control result of the specific section.

The machining result evaluation unit 16A evaluates the machining result for each machining area of the workpiece 17. That is, the machining result evaluation unit 16A calculates an evaluation point indicating evaluation of the machining result for each machining area of the workpiece 17. The evaluation point becomes higher when the machining is performed close to the target. The evaluation point is, for example, the higher the machining time is closer to the target. The evaluation point is, for example, such that the machining depth becomes higher as the machining depth approaches the target machining depth. In addition, the evaluation point is such that the fluctuation of the difference from the set machining time becomes higher as it becomes smaller between the machining areas. Next, a case where the machining time set for the workpiece 17 is the same in each machining area or a case where the machining depth set for the workpiece 17 is the same in each machining area will be described. The machining area is described later. As described above, the machining performance may be a machining depth or a machining time.

The machining result evaluation unit 16A corrects the machining condition based on the evaluation point if the evaluation point indicating the evaluation of the machining result is calculated. When the evaluation point is low, the machining result evaluation unit 16A corrects the machining conditions for each machining area so that the machining time as the actual machining result becomes the same between the machining areas.

The driving unit 12 moves in the X direction, the Y direction, and the Z direction in accordance with a command from the machine control unit 14. In the present embodiment, a case where the Z direction is a vertical direction and the XY plane is a horizontal plane will be described.

The display unit 13 displays various information transmitted from the machining result evaluation unit 16A. An example of the display unit 13 is a liquid crystal monitor. The display unit 13 displays, for example, the evaluation points calculated by the machining result evaluation unit 16A, the machining conditions corrected by the machining result evaluation unit 16A, and the like. The display unit 13 may display information transmitted from the machine control unit 14 and the power supply control unit 15.

Fig. 2 is a diagram showing a configuration of a control device according to embodiment 1. Fig. 2 also shows the display unit 13 and the machining specifications in addition to the machining result evaluation unit 16A, the power supply control unit 15, and the machine control unit 14, which are components of the control device 2. The machining result evaluation unit 16A includes an input unit 21, a machining condition storage unit 25, a machining condition output unit 27, an evaluation calculation unit 22, an actual result calculation unit 23, a machining condition correction unit 24A, and an evaluation output unit 26.

The input unit 21 receives a machining specification used when the workpiece 17 is subjected to electric discharge machining, and sends the machining specification to the machining condition output unit 27. The machining specification is a physical quantity of the workpiece 17 at the time of electric discharge machining. Examples of the machining specifications are a machining position, a machining depth, a finish surface roughness, and a reduction amount. Further, machining specifications other than these may be input to the input section 21. Since the number of the tool electrodes E, the specification of the workpiece 17, information for correcting the position of the tool electrodes E, a swing method, a lift method, and the like are different depending on the processing content, various processing specifications are input to the input unit 21 for each processing content.

The machining position is a position in the XY plane of the tool electrode E. The machining position corresponds to the position of the upper surface of the workpiece 17 and is indicated by the center position of the tool electrode E with respect to the center position of the workpiece 17. The machining depth is a depth of machining to the workpiece 17. The machining depth is represented by the distance from the top surface of the workpiece 17 before machining.

The finish surface roughness is the surface roughness of the finish surface after the workpiece 17 is machined. The reduction amount is a machining magnification to the workpiece 17 with respect to the tool electrode E. In the die-sinking electrical discharge machining, the workpiece 17 is machined in a shape obtained by reversing the shape of the tool electrode E, but the machining magnification at this time is the machining magnification.

The machining condition storage unit 25 stores machining conditions associated with machining specifications. The machining conditions are conditions of machining used in electric discharge machining. The machining conditions include a position condition for specifying the position of the tool electrode E and a power condition for specifying the power to be generated by the high-frequency pulse voltage.

The power conditions include, for example, a current peak value, a current on state, a current off time, a voltage value, a height to lift the tool electrode E from the workpiece 17, a lifting speed of the tool electrode E, a time (a stop time) to bring the tool electrode E close to the workpiece 17 and discharge, an average value of the voltage value, and the like. The downtime is a time interval from the lifting action of the tool electrode E to the next lifting action.

The position condition includes an X coordinate, a Y coordinate, and a Z coordinate of the tool electrode E. The X coordinate and the Y coordinate correspond to a machining position specified by a machining specification, and the Z coordinate corresponds to a machining depth specified by the machining specification.

The machining condition output unit 27 reads out the machining condition corresponding to the machining specification from the machining condition storage unit 25. The machining condition output unit 27 outputs the position condition among the machining conditions to the machine control unit 14, and outputs the power condition among the machining conditions to the power supply control unit 15.

The actual result calculation unit 23 calculates a machining actual result for each machining area based on the control result transmitted from at least one of the machine control unit 14 and the power supply control unit 15. The control result is, for example, position information, power information, and the like, which are the results of the machining condition control. The actual result calculation unit 23 specifies the machining area that has been machined based on the position information transmitted from the machine control unit 14, and calculates the machining actual result corresponding to the machining area that has been machined based on the control result transmitted from at least one of the machine control unit 14 and the power supply control unit 15. The actual result calculation unit 23 calculates a region in the machined XY plane and a region in the machined Z direction based on the position information, and specifies a machined region based on the region in the machined XY plane and the region in the machined Z direction. The control result transmitted from the machine control unit 14 is position information indicating the position of the tool electrode E, and the control result transmitted from the power supply control unit 15 is electric power information indicating the electric power supplied to the tool electrode E. The performance calculation unit 23 calculates transition information (such as a machining speed) of the position information based on the control result transmitted from the machine control unit 14, and uses the calculated transition information as the control result of the machine control unit 14. The performance calculation unit 23 may specify the machining area in which machining is completed based on the position condition transmitted from the machine control unit 14. In this case, the position condition transmitted from the machine control unit 14 is a position condition used when the machine control unit 14 controls the position of the tool electrode E. An example of the machining performance calculated by the performance calculation unit 23 is a machining time for each machining area of the workpiece 17. The actual result calculation unit 23 transmits the machining actual results for each machining area to the evaluation calculation unit 22.

In the shape-carving electric discharge machining, machining in the Z direction is performed while performing wobbling machining in the XY plane. In this case, the larger the reduction amount is, the larger the movement amounts of the tool electrode E in the X direction and the Y direction during the wobbling machining become. The tool electrode E during the wobbling process moves in the positive X direction, the positive Y direction, the negative X direction, or the negative Y direction from the center position in the XY plane of the workpiece 17, that is, the center position at the initial position in the XY plane of the tool electrode E. In the case of the die-sinking electric discharge machining, for example, the machining is performed in the depth direction and the wobbling machining is performed in the 2-dimensional plane direction perpendicular to the depth direction. In the case of the die-sinking electric discharge machining, the machining in the depth direction and the wobbling machining in any direction can be performed. In the electric discharge machining, for example, the wobbling machining may be performed in a direction along the surface of the hemisphere or the like.

The entire machining region in the 2-dimensional plane in which the weaving machining is performed is divided into a plurality of machining regions in the 2-dimensional plane. An example of the divided machining area is a machining quadrant. When the center position of the tool electrode E at the initial position in the XY plane is (X, Y) — 0, the entire machining region in the 2-dimensional plane is divided into 4 parts by the X axis and the Y axis. The 4 divided regions are the 1 st to 4 th processing quadrants.

In the case of wire electric discharge machining, the wire electrode travels along a set machining path, but in the case of contour-carving electric discharge machining including weaving machining, the travel pattern of the tool electrode E changes in accordance with the stability of the machining state corresponding to the machining conditions. That is, in the case of the die-cut electrical discharge machining, when the same machining condition is set for each machining quadrant, the machining performance such as the machining time may be different for each machining quadrant due to the influence of machining chips and the like. Therefore, the machining quadrant may include a machining quadrant in which the machining time is shortened or a machining quadrant in which the machining time is lengthened.

The evaluation calculation unit 22 calculates an evaluation point of the machining result for each machining quadrant based on the machining time for each machining quadrant. The evaluation calculation unit 22 gives higher evaluation points as the difference between the machining time and the machining time of the other machining quadrant is smaller, for example. The evaluation calculation unit 22 assigns higher evaluation points as the machining time becomes closer to the target value. The evaluation calculation unit 22 determines whether or not the machining state is acceptable for each machining quadrant based on the evaluation point for each machining quadrant obtained by the calculation. The evaluation calculation unit 22 determines that the machining quadrant having the evaluation point higher than the threshold value is acceptable, and determines that the machining quadrant having the evaluation point equal to or lower than the threshold value is unacceptable.

The evaluation calculation unit 22 transmits the evaluation point for each machining quadrant and the machining result for each machining quadrant obtained by the calculation to the evaluation output unit 26 and the machining condition correction unit 24A. The evaluation calculation unit 22 also transmits the determination result of the pass/fail determination to the evaluation output unit 26. The actual machining results for each machining quadrant may be transmitted from the actual results calculating unit 23 to the evaluation output unit 26 and the machining condition correcting unit 24A.

The machining condition correction unit 24A corrects the power condition transmitted from the power supply control unit 15 or the position condition transmitted from the machine control unit 14 based on the evaluation point. The power condition transmitted from the power supply control unit 15 is a power condition used when the power supply control unit 15 controls power. The position condition transmitted from the machine control unit 14 is a position condition used when the machine control unit 14 controls the position of the tool electrode E. That is, the machining condition correction unit 24A corrects the machining condition used for the electric discharge machining. In the following description, a case where the machining condition correction unit 24A corrects the power condition among the machining conditions will be described.

The machining condition correction unit 24A corrects the machining condition for each machining quadrant based on the evaluation point for each machining quadrant. The machining condition correction unit 24A corrects the machining conditions so that the evaluation point of each machining quadrant is increased. For example, when there is a difference between the machining time in a specific machining quadrant and the machining time in another machining quadrant, the machining condition correction unit 24A corrects the machining condition so that the difference in the machining time is small. For example, when the number of machining quadrants is 4, the machining condition correction unit 24A corrects the machining conditions of the respective machining quadrants such that the machining times of the 4 machining quadrants become equal.

When the machining quadrant has a machining time longer than the other machining quadrants, the machining condition correction unit 24A reduces the machining time by correcting the machining conditions, for example, so that the height at which the tool electrode E is lifted from the workpiece 17 is reduced.

The machining condition correction unit 24A transmits the corrected machining condition to the machining condition storage unit 25, and the machining condition storage unit 25 stores the machining condition. The machining condition correction unit 24A sends the corrected machining condition to the evaluation output unit 26. The evaluation output unit 26 transmits at least 1 of the machining performance, the evaluation point, the determination result of the non-defective determination, and the corrected machining condition to the display unit 13, and causes the display unit 13 to display the result.

The machining condition correction unit 24A may correct the machining condition for each machining quadrant based on the machining result without using the evaluation point. In this case, the evaluation calculation unit 22 becomes unnecessary.

Fig. 3 is a diagram for explaining a machining quadrant in electric discharge machining by the electric discharge machine according to embodiment 1. In fig. 3, the initial position of the tool electrode E and the machining quadrant included in the overall machining area a0 on the XY plane are shown.

The machining area a0 is composed of a machining quadrant a1 as the 1 st machining quadrant, a machining quadrant a2 as the 2 nd machining quadrant, a machining quadrant A3 as the 3 rd machining quadrant, and a machining quadrant a4 as the 4 th machining quadrant. When the center position of the machining area a0 is (X, Y) ((0, 0)), the machining quadrant a1 is an area where X ≧ 0 and Y ≧ 0, and the machining quadrant a2 is an area where X < 0 and Y ≧ 0. The machining quadrant A3 is the region where X < 0 and Y < 0, and the machining quadrant A4 is the region where X.gtoreq.0 and Y < 0.

The initial position of the tool electrode E is a position where the center position C in the XY plane of the tool electrode E is the center of the machining area a 0. The tool electrode E moves in the machining area a0 so as not to protrude from the machining area a0, and performs the swing machining.

In fig. 3, the description has been given of the case where the machining quadrants are the machining quadrants a1 to a4 in the XY direction in which the coordinate plane is divided by orthogonal coordinate axes, but the machining quadrants of the coordinate plane composed of XYZ axes or other drive axes may be used as the machining quadrants. The entire processing area a0 may be subdivided into an arbitrary number of divisions. That is, the entire machining area a0 may be divided into 2 or 3, or may be divided into 5 or more.

Fig. 4 is a diagram for explaining weaving performed by the electric discharge machine according to embodiment 1. Here, an example of the movement path of the tool electrode E in the machining quadrants a1 to a4 during the oscillating machining will be described.

At the time of starting the swing machining, the center position C of the tool electrode E is at the center position (0, 0) of the machining area a0 (ST 1). If the swing machining is started, the tool electrode E is moved from the center position of the machining area a0 in the positive X direction. Thereby, the positive X-direction end of the tool electrode E reaches the positive X-direction end of the machining area a0 (ST 2).

Then, the tool electrode E is moved in the positive Y direction. Thereby, the end of the tool electrode E in the positive Y direction reaches the end of the machining area a0 in the positive Y direction (ST 3).

Then, the tool electrode E is moved in the negative X direction. Thereby, the end of the tool electrode E in the negative X direction reaches the end of the machining area a0 in the negative X direction (ST 4).

Then, the tool electrode E is moved in the negative Y direction. Thereby, the end of the tool electrode E in the negative Y direction reaches the end of the machining area a0 in the negative Y direction (ST 5).

Then, the tool electrode E is moved in the positive X direction. Thereby, the positive X-direction end of the tool electrode E reaches the positive X-direction end of the machining area a0 (ST 6).

Then, the tool electrode E is moved in the positive Y direction. Thereby, the center position C of the tool electrode E reaches the position where the Y coordinate becomes 0 (ST 7). The position of the tool electrode E in ST7 is the same as that in ST 2.

Then, the tool electrode E repeats the processing of ST3 to ST7 described above. The tool electrode E performs machining in the depth direction while repeating the processing from ST3 to ST7, and thus machines the workpiece 17 while moving in a spiral shape.

If the tool electrode E finishes machining to a specific machining depth, the tool electrode E is moved in the negative X direction from the position ST 7. Thereby, the center position C of the tool electrode E reaches the center position of the machining area a0 (ST 8).

The path of the oscillating process is not limited to the path shown in fig. 4. The path of the oscillating machining may also be circular. The path of the oscillating machining may be a reciprocating path from the center of the machining area a0 to the outer peripheral portion of the machining area a 0. In this case, the tool electrodes E reciprocate radially from the center of the machining region a0 toward the outer peripheral portion of the machining region a 0.

Fig. 5 is a diagram for explaining a machining result in the weaving machining explained in fig. 4. In the present embodiment, the boundary line between the machining quadrants a1 and a4 is defined as being included in the machining quadrant a1, and the boundary line between the machining quadrants a2 and a1 is defined as being included in the machining quadrant a 1. Similarly, the boundary line between the machining quadrants A3 and a2 is included in the machining quadrant a2, and the boundary line between the machining quadrants a4 and A3 is included in the machining quadrant a 4.

At the time of starting the swing machining, the center position C of the tool electrode E is at the center position of the machining area a0 (ST 11). If the swing machining is started, the tool electrode E is moved from the center position of the machining area a0 in the positive X direction. The positive X-direction end of the tool electrode E reaches the positive X-direction ends of the machining quadrants a1 and a4 (ST 12).

Then, the tool electrode E moves in the positive Y direction, and the end of the tool electrode E in the positive Y direction reaches the end of the machining quadrant a1 in the positive Y direction (ST 13). Then, the tool electrode E moves in the negative X direction. Thereby, the center position C of the tool electrode E reaches the boundary line between the machining quadrant a1 and the machining quadrant a2 (ST 14).

After the swing machining is started, the machining quadrant a1 is machined until the center position C of the tool electrode E reaches the boundary line between the machining quadrant a1 and the machining quadrant a 2. That is, the period in which the center position C of the tool electrode E is located within the machining quadrant a1 becomes the machining time in the machining quadrant a 1.

In the wobbling process of the 1 ST cycle until the machining quadrants a1 to a4, the machining time from ST11 to ST12 is included in the machining time of the machining quadrant a1, but in the wobbling process of the 2 nd cycle and thereafter, the machining time from ST11 to ST12 is not included in the machining time of the machining quadrant a 1. After the tool electrode E finishes machining to a predetermined machining depth, the machining time from ST7 to ST8 is included in the machining time of the machining quadrant a 1.

The performance calculation unit 23 calculates the position of the tool electrode E in the machining area a0 based on the control results such as the position information transmitted from the machine control unit 14, and calculates the machining time of each of the machining quadrants a1 to a4 based on the calculated position. That is, the performance calculation unit 23 calculates the accumulated machining time for each machining condition in each of the machining quadrants a1 to a4 based on the time during which the center position C of the tool electrode E stays in the machining quadrants a1 to a 4.

Further, the processing time on the boundary line of the processing quadrants a1, a4 may be set as the processing time of the processing quadrant a 4. Similarly, the machining time on the boundary line between the machining quadrants a2 and a1 may be set as the machining time of the machining quadrant a1, the machining time on the boundary line between the machining quadrants A3 and a2 may be set as the machining time of the machining quadrant a2, and the machining time on the boundary line between the machining quadrants a4 and A3 may be set as the machining time of the machining quadrant A3.

The machining time on the boundary line between the machining quadrants a1 and a4 may be sequentially allocated to the machining quadrants a1 and a4 at a specific ratio. Similarly, the machining time on the boundary line between the machining quadrants a2 and a1 may be sequentially allocated to the machining quadrants a2 and a1 at a specific ratio, the machining time on the boundary line between the machining quadrants A3 and a2 may be sequentially allocated to the machining quadrants A3 and a2 at a specific ratio, and the machining time on the boundary line between the machining quadrants a4 and A3 may be sequentially allocated to the machining quadrants a4 and A3 at a specific ratio.

Fig. 6 is a flowchart showing a procedure of evaluation processing of a machining result by the electric discharge machine according to embodiment 1. Before the machining is started, a machining specification necessary for machining the workpiece 17 is input to the input unit 21 by an operator.

The machining condition output unit 27 selects a machining condition corresponding to the machining specification in the machining condition storage unit 25. That is, the machining condition output unit 27 selects the machining condition most suitable for machining using the machining specification, and reads the selected machining condition from the machining condition storage unit 25. The machining condition output unit 27 outputs the position condition among the machining conditions to the machine control unit 14, and outputs the power condition among the machining conditions to the power supply control unit 15.

The electric discharge machine 1 starts machining using the position condition and the power condition (step S11). The machining conditions used for machining include a plurality of machining conditions in accordance with a combination of productivity and finished surface roughness. That is, during machining, various machining conditions are used in order of machining order. Here, a case where N (N is a natural number) machining conditions are used will be described. That is, the electric discharge machine 1 performs machining using the machining conditions of N ═ 1, 2, ·, and N in this order. In other words, the first processing condition is a processing condition where N is 1, and the last processing condition is N.

The machining result evaluation unit 16A sets the machining conditions in which n is 1 in the machine control unit 14 and the power supply control unit 15 (step S12). The electric discharge machine 1 sequentially performs machining in the machining quadrants a1 to a4 using machining conditions where n is 1. The processing of the processing quadrant a1 is the processing from ST11 to ST14 described in fig. 5. The series of processing until quadrants a1 to a4 is the processing from ST2 to ST7 described in fig. 4.

The actual results calculation unit 23 determines whether or not the machining in the machining quadrant a1 is completed (step S13). If the machining in the machining quadrant a1 is not completed (No in step S13), the electric discharge machine 1 continues the machining in the machining quadrant a 1. When the machining of the machining quadrant a1 is completed (Yes at step S13), the actual results calculation unit 23 calculates the machining actual results of the machining quadrant a1 (step S14). After the electric discharge machine 1 performs machining in the machining quadrant a1, the electric discharge machine performs machining in the machining quadrant a 2.

The actual results calculation unit 23 determines whether or not the machining in the machining quadrant a2 is completed (step S15). If the machining in the machining quadrant a2 is not completed (No in step S15), the electric discharge machine 1 continues the machining in the machining quadrant a 2. When the machining of the machining quadrant a2 is completed (Yes at step S15), the actual result calculation unit 23 calculates the machining actual result of the machining quadrant a2 (step S16). After the electric discharge machine 1 performs machining in the machining quadrant a2, the electric discharge machine performs machining in the machining quadrant A3.

The actual results calculation unit 23 determines whether or not the machining in the machining quadrant a3 is completed (step S17). If the machining in the machining quadrant A3 is not completed (No in step S17), the electric discharge machine 1 continues the machining in the machining quadrant A3. When the machining of the machining quadrant A3 is completed (Yes at step S17), the actual result calculation unit 23 calculates the machining actual result of the machining quadrant A3 (step S18). After the electric discharge machine 1 performs machining in the machining quadrant A3, the electric discharge machine performs machining in the machining quadrant a 4.

The actual results calculation unit 23 determines whether or not the machining in the machining quadrant a4 is completed (step S19). If the machining in the machining quadrant a4 is not completed (No in step S19), the electric discharge machine 1 continues the machining in the machining quadrant a 4. When the machining of the machining quadrant a4 is completed (Yes at step S19), the actual result calculation unit 23 calculates the machining actual result of the machining quadrant a4 (step S20).

As described above, the control device 2 selects the first machining condition, that is, the machining condition where n is 1, continues the machining until the machining in the machining quadrant a1 is completed, and calculates the machining result at the machining completion stage. Then, the controller 2 calculates the machining performance for the machining quadrants a2 to a4 in the same manner as for the machining quadrant a 1. That is, the actual result calculation unit 23 calculates the machining time as the machining actual result for each of the machining quadrants a1 to a 4.

Then, the performance calculation unit 23 calculates the machining time ratio for each of the machining quadrants a1 to a4 (step S21). The ratio of the machining time is a ratio of the machining time in each of the machining quadrants a1 to a4 to the machining time in the entire machining quadrants a1 to a 4.

The evaluation calculation unit 22 calculates the evaluation point based on the ratio of the machining time in each of the machining quadrants a1 to a4 (step S22). The evaluation calculation unit 22 calculates evaluation points for each of the machining quadrants a1 to a4, for example. The actual result calculation unit 23 determines whether or not the machining of the workpiece 17 has reached the set machining depth (step S23). Specifically, the actual result calculation unit 23 acquires information of an actual machining depth from an encoder (not shown) of a motor included in the electric discharge machine 1, and the like, and determines whether or not the machining to the workpiece 17 has reached the machining depth set as the machining condition based on the acquired information. The actual result calculation unit 23 here determines whether or not the machining depth set under the machining condition where n is 1 is reached.

If the set machining depth is not reached (No at step S23), the electric discharge machine 1 repeats the processing from step S13 to step S23 until the set machining depth is reached.

When the machining depth reaches the set machining depth (Yes at step S23), the machining condition correction unit 24A determines that the machining using the machining condition is completed, and corrects the machining condition reaching the set machining depth. The machining condition correction unit 24A corrects the machining condition where n is 1. The processing condition correction unit 24A corrects the processing condition based on the evaluation point (step S24). The machining condition correction unit 24A corrects the machining conditions for each of the machining quadrants a1 to a4, for example, based on the evaluation points for each of the machining quadrants a1 to a 4. When the machining condition for calculating the machining result is an appropriate machining condition, the machining condition correction unit 24A may not correct the machining condition.

The machining condition correcting unit 24A transmits the corrected machining condition to the machining condition storage unit 25, and the machining condition storage unit 25 stores the corrected machining condition. Thus, the machining condition storage unit 25 stores the machining condition before the correction of n-1 and the machining condition of n-1 corrected by the machining condition correction unit 24A (step S25).

The display unit 13 displays the machining condition in which n is 1 corrected by the machining condition correction unit 24A (step S26). The display unit 13 displays the machining results calculated by the results calculating unit 23 (step S27). The display unit 13 displays the machining result in the case where the machining is performed under the machining condition before the correction of n being 1. The display unit 13 displays the machining time as the machining result. The display unit 13 may display a machining time ratio as a machining result. The display unit 13 displays the evaluation point calculated by the evaluation calculation unit 22 (step S28).

The control device 2 determines whether or not the machining condition is corrected for the machining condition where N is N (step S29). That is, the control device 2 determines whether or not to calculate the machining performance for all the machining conditions corresponding to the machining specifications.

If the machining condition is not corrected for the machining condition where N is N (No at step S29), the control device 2 sets N to N +1 (step S30) and executes the processing of steps S13 to S29. The control device 2 herein executes the processing of steps S13 to S29 for the processing condition where n is 2, and then executes the processing of steps S13 to S29 for the processing condition where n is 3. The control device 2 ends the control of the wobbling process if the processes of steps S13 to S29 are executed for the process condition of N ═ N (step S29, Yes). The display unit 13 may display the machining conditions, the machining results, and the evaluation points in an arbitrary order.

As described above, the controller 2 can evaluate the machining performance for each of the machining quadrants a1 to a4 even for complicated machining involving weaving machining, and can correct the machining conditions for each of the machining quadrants a1 to a4 based on the evaluation results. Further, since the machining results can be evaluated for each of the machining quadrants a1 to a4, the machining conditions can be easily corrected to appropriate machining conditions for not only machining of a simple shape but also machining of a complex shape. Further, by setting the machining time as an index for determining whether machining is acceptable or not, the machining efficiency can be directly determined.

Further, since the control device 2 evaluates the machining for each of the machining quadrants a1 to a4, the machining conditions, the machining results, and the evaluation points can be displayed on the display unit 13 for each of the machining quadrants a1 to a 4. In addition, when the influence of machining chips, the influence of electrode rigidity, the influence of changes in the machining area, and the like during the die sinking electric discharge machining are large, the control device 2 can appropriately correct the machining conditions in a short time because the machining results are evaluated for each of the machining quadrants a1 to a 4.

Note that, if the machining of the machining quadrant Ap (any of which p is 1 to 4) is completed, the actual result calculation unit 23 may calculate the machining actual result of the machining quadrant Ap at any timing. The actual results calculation unit 23 may calculate the actual results of the machining in the machining quadrants a1 to a4 in a lump after the machining in the machining quadrants a1 to a4 corresponding to 1 week is completed, for example. The actual results calculation unit 23 may calculate the machining actual results of the machining quadrants a1 to a4 after the set machining depth is reached.

In addition, the evaluation calculation unit 22 may calculate the evaluation point for the nth machining condition at an arbitrary timing, if the machining result is calculated for the nth machining condition (any one of N is 1 to N). The evaluation calculation unit 22 may calculate the machining results for all the machining conditions, and then calculate the evaluation points for each machining condition.

In addition, the machining condition correction unit 24A may correct the nth machining condition at an arbitrary timing if the evaluation point is calculated for the nth machining condition. The machining condition correction unit 24A may calculate evaluation points for all machining conditions, and then correct each machining condition. The machining result evaluation unit 16A may calculate the evaluation point based on the ratio of the machining depth for each of the machining quadrants a1 to a 4.

Fig. 7 is a diagram for explaining a relationship between the ratio of the machining time calculated by the control device according to embodiment 1 and the evaluation point. Here, a case where the same machining conditions are used in the machining quadrants a1 to a4 will be described. For example, when the machining condition Q1 is used, the machining condition Q1 is used for all of the machining quadrants a1 to a 4. That is, the evaluation points shown in fig. 7 are evaluation points calculated based on the machining results (the machining time ratios in fig. 7) in the case where the machining quadrants a1 to a4 are machined under the same machining conditions. Fig. 7 shows the correspondence between the machining conditions, the evaluation points, and the proportion of the machining time for each of the machining quadrants a1 to a 4. The machining time ratio is a ratio of the machining time of each machining quadrant in a certain machining condition. That is, the ratio of the machining time is a ratio of the machining time of each machining quadrant to the total machining time. In the following description, the ratio of the machining time in each machining quadrant is sometimes referred to as a machining time ratio.

The evaluation calculation unit 22 calculates a machining time ratio (%) of the machining quadrants a1 to a4, for example, assuming that the entire machining time under the machining condition Q1 is 100%. In the example shown in fig. 7, under the machining condition Q1, the machining time ratio in the machining quadrant a1 is 60%, and the machining time ratio in the machining quadrant a2 is 20%. In the machining condition Q1, the machining time ratio in the machining quadrant A3 was 10%, and the machining time ratio in the machining quadrant a4 was 10%.

The evaluation calculation unit 22 calculates a relative difference between the machining time ratios based on the machining time ratios of the machining quadrants, and calculates an evaluation point based on the difference. The evaluation calculation unit 22 gives a high evaluation point to a machining condition with small fluctuation between machining time ratios, for example. When the machining time ratios are equal to each other under ideal conditions, the evaluation calculation unit 22 assigns the highest evaluation point when all the machining time ratios in the machining quadrants a1 to a4 are 25%, and assigns smaller evaluation points when the difference from the ideal conditions becomes larger. The evaluation calculation unit 22 assigns an evaluation point of "6" to the machining condition Q1, for example.

For example, when the machining time ratio of 1 or a plurality of specific machining quadrants is extremely small relative to the machining time ratios of the other machining quadrants, the evaluation calculation unit 22 determines that the specific machining quadrant has a shape with a small machining amount. In this case, the machining condition correction unit 24A changes the swing direction, the axis movement speed, and the like of the machining condition used in the specific machining quadrant.

On the other hand, when the machining time ratio of 1 or a plurality of specific machining quadrants is extremely large relative to the machining time ratios of the other machining quadrants, the evaluation calculation unit 22 determines that the machining of the specific machining quadrant is unstable. In this case, the machining condition correction unit 24A changes the machining condition used in the specific machining quadrant so as to eliminate unstable machining. As described above, the control device 2 evaluates the machining performance for each of the machining quadrants a1 to a4 based on the machining time ratio, and thus can accurately determine whether or not the machining is acceptable for each of the machining quadrants a1 to a 4.

The evaluation calculation unit 22 of the present embodiment may calculate the evaluation points for each of the machining quadrants a1 to a 4. That is, the evaluation calculation unit 22 of the present embodiment may calculate the evaluation point for each machining condition, or may calculate the evaluation point for each machining quadrant. Fig. 8 is a diagram for explaining the evaluation points for each machining quadrant calculated by the control device according to embodiment 1. Fig. 8 shows the correspondence between the machining conditions and the evaluation points for each of the machining quadrants a1 to a 4.

Fig. 8 shows the evaluation points for each machining quadrant calculated by the evaluation calculation unit 22 based on the machining time ratio for each machining quadrant shown in fig. 7. The evaluation calculation unit 22 compares the machining time ratios of the machining quadrants a1 to a4 with respect to the machining condition Q1, for example, and calculates evaluation points of the machining quadrants a1 to a4 in the machining condition Q1 based on the comparison result. The machining condition correction unit 24A changes the machining conditions for each of the machining quadrants a1 to a4 so that the evaluation points of the machining quadrants a1 to a4 are increased.

The evaluation output unit 26 causes the display unit 13 to display the evaluation points calculated by the evaluation calculation unit 22 for each of the machining quadrants a1 to a 4. The display unit 13 displays the correspondence relationship shown in fig. 8, for example. The machining condition correction unit 24A causes the display unit 13 to display the machining conditions corrected for each of the machining quadrants a1 to a 4. The display unit 13 displays the evaluation points for each of the machining quadrants a1 to a4, thereby providing the operator with accurate evaluation points with high accuracy.

The actual result calculation unit 23 may calculate a correspondence relationship between the machining elapsed time and the machining depth. Fig. 9 is a diagram showing a correspondence relationship between the machining elapsed time and the machining depth calculated by the control device according to embodiment 1. The horizontal axis of the graph shown in fig. 9 represents the processing elapsed time for the entire processing quadrants a1 to a4 (processing region a0), and the vertical axis represents the processing depth.

The actual result calculation unit 23 calculates a graph such as that shown in fig. 9 indicating the correspondence between the machining elapsed time and the machining depth, and displays the graph on the display unit 13. The performance calculation unit 23 may calculate a graph indicating a correspondence relationship between the machining elapsed time and the machining depth for each of the machining quadrants a1 to a4, and may display the graph on the display unit 13.

The machining condition correction unit 24A may correct the machining condition based on the evaluation point for each machining depth. Fig. 10 is a diagram showing machining conditions corrected based on the evaluation points for each machining depth by the control device according to embodiment 1. Fig. 10 shows an example of the correspondence between the machining depth and the evaluation point in any of the machining quadrants a1 to a4 and the machining condition after correction.

The machining depths are sequentially separated by a specific dimension. In fig. 10, the case where the processing depths are sequentially separated by 2mm is shown. "0" in the machining depth means that the machining depth is 0mm to 2mm, and the evaluation point in this case is "7". Similarly, "-2" of the machining depth means that the machining depth is 2mm to 4mm, and the evaluation point in this case is "3".

The evaluation calculation unit 22 calculates evaluation points for each of the machining quadrants a1 to a4 based on the correspondence between the machining elapsed time and the machining depth for each of the machining quadrants a1 to a 4. The evaluation calculation unit 22 calculates a machining time corresponding to each machining depth in the machining quadrant a1 based on the correspondence between the machining elapsed time and the machining depth in the machining quadrant a 1.

Similarly, the evaluation calculation unit 22 calculates the machining time corresponding to the machining depth for the machining quadrants a2 to a4 based on the correspondence relationship between the machining elapsed time and the machining depth in the machining quadrants a2 to a 4. The evaluation calculation unit 22 calculates evaluation points of the machining quadrants a1 to a4 based on the machining elapsed times in the machining quadrants a1 to a 4.

"IP", "ON", "OFF", "Volt", "JUMP", "UP", "DN", and "SV" shown in FIG. 10 are processing conditions. "IP" is the peak current value, "ON" is the current ON, "OFF" is the current OFF time, and "Volt" is the voltage value.

Further, "JUMP" is the lifting speed of the tool electrode E, and "UP" is the height at which the tool electrode E is lifted from the workpiece 17. "DN" is the time for which the tool electrode E and the workpiece 17 are brought close to each other and discharged, and "SV" is the average value of the voltage values.

For example, the initial values of "IP", "ON", "OFF", "Volt", "JUMP", "UP", "DN", and "SV" are set to "10", "6", "100", "10", and "50", respectively. Further, the initial values of the processing conditions may be different for each processing depth.

The processing condition correction unit 24A corrects the processing condition based on the evaluation point for these initial values. The machining condition correction unit 24A does not correct the machining condition for the machining condition whose evaluation point is "10", for example. On the other hand, the machining condition correction unit 24A greatly corrects the machining condition for a machining condition as small as the evaluation point "3".

The machining result evaluation unit 16A may calculate the evaluation point for each of the machining quadrants a1 to a4 based on the result of comparison between the machining time calculated as the machining result and the target machining time, i.e., the target time.

Fig. 11 is a diagram for explaining the relationship between the machining time and the evaluation point calculated by the control device according to embodiment 1. Fig. 11 shows the correspondence between the machining conditions, the evaluation points, and the machining time for each of the machining quadrants a1 to a 4.

The evaluation calculation unit 22 calculates the machining time (minutes) of the machining quadrants a1 to a4 for each machining condition, for example. In the example shown in fig. 11, under the machining condition Q1, the machining time in the machining quadrant a1 is 60 minutes, and the machining time in the machining quadrant a2 is 20 minutes. In the processing condition Q1, the processing time in the processing quadrant A3 was 10 minutes, and the processing time in the processing quadrant a4 was 10 minutes.

The evaluation calculation unit 22 compares the machining time for each of the machining quadrants a1 to a4 with the target time for each of the machining quadrants a1 to a4, and calculates an evaluation point based on the comparison result in each of the machining quadrants a1 to a 4. The evaluation calculation unit 22 may calculate the evaluation point based on the difference between the actual machining time and the target time, or may calculate the evaluation point based on the ratio between the actual machining time and the target time. The evaluation calculation unit 22 gives a higher evaluation point to a machining quadrant having a smaller difference between the actual machining time and the target time as the machining actual result, for example.

The evaluation calculation unit 22 calculates the evaluation point in the machining condition Q1 based on, for example, the difference between the machining time in the machining quadrant a1 and the target time in the machining quadrant a1, the difference between the machining time in the machining quadrant a2 and the target time in the machining quadrant a2, the difference between the machining time in the machining quadrant A3 and the target time in the machining quadrant A3, and the difference between the machining time in the machining quadrant a4 and the target time in the machining quadrant a 4. In the example shown in fig. 11, the evaluation point in the processing condition Q1 is "6". In addition, the target time may be different for each of the machining quadrants a 1-a 4. The target time may be different for each processing condition.

When calculating the evaluation points based on the target time, the evaluation calculation unit 22 also calculates the evaluation points for each of the machining quadrants a1 to a 4. Fig. 12 is a diagram for explaining the evaluation points for each machining quadrant calculated based on the target time by the control device according to embodiment 1. Fig. 12 shows the correspondence among the machining conditions, the evaluation points for each machining quadrant, the target time for each of the machining quadrants a1 to a4, and the machining time for each of the machining quadrants a1 to a 4.

The evaluation calculation unit 22 compares, for example, 20 minutes, which is a target time in the machining quadrant a1, with 60 minutes, which is an actual machining time in the machining quadrant a1, for the machining condition Q1, and calculates an evaluation point of the machining quadrant a1 in the machining condition Q1 based on the comparison result. Similarly, the evaluation calculation unit 22 calculates evaluation points of the machining quadrants a2 to a4 under the machining condition Q1.

The machining condition correction unit 24A changes the machining conditions for each of the machining quadrants a1 to a4 so that the evaluation points of the machining quadrants a1 to a4 are increased. The machining condition correction unit 24A corrects the machining condition so that the machining time approaches the target time. That is, the machining condition correcting unit 24A corrects the machining condition so that the machining time becomes longer when the machining time is shorter than the target time, and corrects the machining condition so that the machining time becomes shorter when the machining time is longer than the target time. The machining condition correction unit 24A increases the current peak value, for example, when the machining time is longer than the target time, that is, when the machining speed is slow.

The evaluation output unit 26 causes the display unit 13 to display the evaluation points calculated by the evaluation calculation unit 22 for each of the machining quadrants a1 to a 4. The machining condition correction unit 24A causes the display unit 13 to display the machining conditions corrected for each of the machining quadrants a1 to a 4.

The machining condition correction unit 24A may correct the machining condition based on the evaluation point for each machining time. Fig. 13 is a diagram showing the machining conditions corrected based on the evaluation point for each machining time by the control device according to embodiment 1. Fig. 13 shows the correspondence between the machining time and the evaluation point in any of the machining quadrants a1 to a4 and the machining conditions after correction.

The processing times are separated in sequence by specific times. In fig. 13, the processing time is shown divided in 20 minutes in order. "0" in the processing time means that the processing elapsed time is 0 to 20 minutes, and the evaluation point in this case is "7". Similarly, "20" of the processing time means that the processing elapsed time is 20 to 40 minutes, and the evaluation point in this case is "3".

The evaluation calculation unit 22 calculates evaluation points for each of the machining quadrants a1 to a4 based on the correspondence between the machining elapsed time and the machining depth for each of the machining quadrants a1 to a 4. The evaluation calculation unit 22 calculates a machining depth corresponding to the machining time based on the correspondence between the machining elapsed time and the machining depth in the machining quadrant a1 for the machining quadrant a 1.

Similarly, the evaluation calculation unit 22 calculates the machining depth corresponding to the machining time also for the machining quadrants a2 to a4 based on the correspondence between the machining elapsed time and the machining depth in the machining quadrants a2 to a 4. The evaluation calculation unit 22 calculates evaluation points of the machining quadrants a1 to a4 based on the machining depths in the machining quadrants a1 to a 4. In this case, the machining condition correction unit 24A corrects the machining conditions based on the evaluation points for each of the machining time points in the machining quadrants a1 to a 4. For example, the evaluation calculation unit 22 calculates evaluation points of the machining quadrants a1 to a4 based on the ratio of the machining depth for each machining quadrant. In this case, the evaluation calculation unit 22 lowers the evaluation point when the machining depth of the machining quadrant is extremely shallow or extremely deep with respect to the machining depth of the other machining quadrant. The evaluation calculation unit 22 may calculate the evaluation points of the machining quadrants a1 to a4 based on the difference between the target machining depth for each of the machining quadrants a1 to a4 and the actual machining depth for each of the machining quadrants a1 to a 4.

As described above, in embodiment 1, the machining result evaluation unit 16A calculates the machining results for each of the machining quadrants a1 to a4 based on the machining conditions of the machining quadrants a1 to a4, and calculates the evaluation points based on the machining results of the machining quadrants a1 to a 4. Therefore, for the engraving electrical discharge machining, it is possible to determine whether the machining result is acceptable or not for each of the machining quadrants a1 to a 4. In addition, whether or not the machining result is acceptable can be determined in consideration of the machining results for each of the machining quadrants a1 to a4 within the 2-dimensional plane that changes in a complex manner.

Embodiment 2.

Next, embodiment 2 of the present invention will be described with reference to fig. 14 to 17. In embodiment 2, the machining condition is corrected based on the result of comparison between the machining speed as the machining result and the target speed, which is the target machining speed.

Fig. 14 is a diagram showing a configuration of a control device according to embodiment 2. Among the components of fig. 14, those that realize the same functions as those of the control device 2 of embodiment 1 shown in fig. 2 are denoted by the same reference numerals, and redundant description thereof is omitted.

Fig. 14 illustrates the display unit 13 and the machining specifications in addition to the machining result evaluation unit 16B, the power supply control unit 15, and the machine control unit 14, which are components of the control device 2. The machining result evaluation unit 16B includes a target speed storage unit 28 and a speed difference calculation unit 35 in addition to the components of the machining result evaluation unit 16A.

The target speed storage unit 28 stores target speeds for machining in the machining quadrants a1 to a4 for each machining condition. The speed difference calculation unit 35 calculates a machining speed as a machining actual result for each of the machining quadrants a1 to a4 based on a control result such as position information transmitted from the machine control unit 14. The processing speed is a processing depth per unit time of processing.

The speed difference calculation unit 35 calculates the machining time for each of the machining quadrants a1 to a4 by the same processing as the actual result calculation unit 23. The speed difference calculation unit 35 acquires information on the actual machining depth from an encoder of a motor included in the electric discharge machine 1. The speed difference calculation unit 35 calculates a machining speed as a machining result based on the acquired information of the machining depth. The speed difference calculation unit 35 calculates a difference between the machining speed as the machining result and the target speed stored in the target speed storage unit 28 (hereinafter, referred to as a speed difference). The speed difference is a positive value when the machining speed as the machining result is greater than the target speed, and a negative value when the machining speed as the machining result is less than the target speed.

The actual result calculation unit 23 may calculate the machining speed as the actual result of machining. The actual result calculation unit 23 may calculate the machining time for each of the machining quadrants a1 to a4, and the speed difference calculation unit 35 may calculate the machining speed as the machining actual result.

The speed difference calculation unit 35 transmits the calculated speed difference to the evaluation output unit 26, and displays the speed difference on the display unit 13. The speed difference calculation unit 35 transmits the calculated speed difference to the evaluation calculation unit 22.

The evaluation calculation unit 22 calculates evaluation points for the respective machining quadrants a1 to a4 based on the speed differences for the respective machining quadrants a1 to a 4. The evaluation calculation unit 22 sets the evaluation point to be lower as the speed difference of the machining quadrant to be evaluated is larger.

The evaluation calculation unit 22 transmits the calculated evaluation point to the evaluation output unit 26, and displays the evaluation point on the display unit 13. The display unit 13 displays evaluation points for each of the machining quadrants a1 to a4 in a table format shown in fig. 12, for example. The display unit 13 displays the machining speed, the target speed, and the speed difference as the machining results for each of the machining quadrants a1 to a 4. Thus, the user can accurately determine whether machining is acceptable or not, even for machining contents in which machining actual results for each machining condition are unlikely to differ.

The evaluation calculation unit 22 sends the calculated evaluation point to the processing condition correction unit 24A. The machining condition correction unit 24A corrects the machining conditions for each of the machining quadrants a1 to a4 based on the evaluation point sent thereto from the evaluation calculation unit 22. The machining condition correction unit 24A corrects the machining conditions for each of the machining quadrants a1 to a4 so that the speed difference between the machining quadrants a1 to a4 approaches 0.

Fig. 15 is a flowchart showing a procedure of evaluation processing of a machining result by the electric discharge machine according to embodiment 2. Among the processes in fig. 15, the same processes as those described in fig. 6 are denoted by the same step numbers, and the description thereof will be omitted.

The processing up to steps S11 to S21 in the flowchart shown in fig. 15 is the same processing as the flowchart described in fig. 6. In step S21, the performance calculator 23 calculates the machining time ratio for each of the machining quadrants a1 to a4, and then calculates the speed difference for each of the machining quadrants a1 to a4 by the speed difference calculator 35 (step S31).

Then, the machining result evaluation unit 16B performs the same processing as the processing of steps S22 to S27 described with reference to fig. 6. In step S22 of fig. 15, the evaluation calculation unit 22 calculates an evaluation point based on the speed difference between the machining quadrants a1 to a 4. The evaluation calculation unit 22 calculates evaluation points for each of the machining quadrants a1 to a4, for example. In step S24 of fig. 15, the machining condition correction unit 24A corrects the machining condition based on the evaluation point calculated from the speed difference. The machining condition correction unit 24A corrects the machining conditions for each of the machining quadrants a1 to a4, for example, based on the evaluation points for each of the machining quadrants a1 to a 4.

In step S27, after the display unit 13 displays the machining result, the display unit 13 displays the speed difference for each of the machining quadrants a1 to a4 (step S32). The display unit 13 displays the evaluation point calculated by the evaluation calculation unit 22 (step S28).

The control device 2 determines whether or not the machining condition is corrected for the machining condition where N is N (step S29).

If the machining condition is not corrected for the machining condition of N-N (No at step S29), the control device 2 sets N-N +1 (step S30) and executes the processing of steps S13 to S29. If the processing of steps S13 to S29 is executed for the processing condition where N is N (step S29, Yes), the control device 2 ends the control of the weaving processing. The display unit 13 may display the machining conditions, the machining results, the speed difference, and the evaluation points in an arbitrary order.

The display unit 13 may graphically display the machining speed as the machining result. The display unit 13 may graphically display the machining speed before the machining condition is corrected and the machining speed after the machining condition is corrected. The display unit 13 may graphically display the speed difference as the actual result of the machining. The display unit 13 may graphically display the speed difference before the machining condition is corrected and the speed difference after the machining condition is corrected.

Fig. 16 is a diagram for explaining a machining speed corresponding to a machining condition corrected by the electric discharge machine according to embodiment 2. Fig. 17 is a diagram for explaining a machining speed for each machining quadrant corresponding to the machining condition corrected by the electric discharge machine according to embodiment 2. Here, a case will be described where the machining conditions are corrected so that the machining speed is constant and the speed difference is constant.

The pattern 51 shown in fig. 16 is the same pattern as that shown in fig. 9. The horizontal axis of the

graphs

51 and 52 shown in fig. 16 represents the machining elapsed time, and the vertical axis represents the machining depth. In addition, in the

graphs

61 and 62 shown in fig. 16, the horizontal axis represents the machining depth, and the vertical axis represents the velocity difference.

Fig. 51 shows machining speed information 71, which is a correspondence relationship between the machining elapsed time and the machining depth before the machining condition is corrected. Fig. 52 shows machining speed information 72, which is a correspondence relationship between the machining elapsed time and the machining depth after the machining conditions have been corrected.

The

processing speed information

71 and 72 is information of a processing depth to be processed per unit time. When the machining speed is constant and the speed difference is constant, the machining depth is proportional to the machining elapsed time. In the machining speed information 71, the machining depth is not proportional to the machining elapsed time.

The graph 61 shows velocity difference information 81, which is a correspondence relationship between the machining depth and the velocity difference before the machining condition is corrected. The graph 62 shows velocity difference information 82, which is a correspondence relationship between the machining depth and the velocity difference after the machining condition has been corrected.

The

velocity difference information

81 and 82 shows the velocity difference for each block of the machining depth of the machining area a 0. Each block of the machining depth (hereinafter referred to as a depth block) shows a range from a certain machining depth to a certain machining depth. In each depth block, the velocity difference of the processing area a0 is shown. For example, the

velocity difference information

81 and 82 shows the velocity difference for depth blocks of 2mm width, as shown by the velocity difference in depth blocks of 2mm to 4mm in depth of machining, and the velocity difference in depth blocks of 4mm to 6mm in depth of machining.

The "0" on the vertical axis in the graph 61 is a predicted value of the velocity difference when the machining is performed under the machining condition before the correction. That is, when machining is performed under the machining condition before correction, the predicted value of the speed difference is "0". In other words, when machining is performed under the machining conditions before correction, the machining speed is the same as the target speed. In actual machining, as shown in graph 61, the speed difference may become positive at the time of starting machining and at the time of finishing machining, and the speed difference may become negative during machining.

The "0" on the vertical axis in the graph 62 is a predicted value of the velocity difference when machining is performed under the machining condition after correction. That is, when machining is performed under the machining conditions after correction, the predicted value of the speed difference is "0". In actual machining, as shown in graph 62, the speed difference is mostly not 0.

The machining condition correction unit 24A corrects the machining condition so that the machining speed becomes slower for the machining depth at which the speed difference becomes positive in the speed difference information 81. On the other hand, the machining condition correction unit 24A corrects the machining condition so that the machining speed becomes faster for the machining depth at which the speed difference becomes negative in the speed difference information 81.

The machining condition correction unit 24A corrects the machining condition corresponding to the pattern 51, and thereby the machining speed information 71 becomes the machining speed information 72 and the speed difference information 81 becomes the speed difference information 82. As shown in the speed difference information 81, the difference between the speed difference before the machining condition is corrected and the predicted value of the speed difference is large. As shown in the speed difference information 82, the speed difference after the machining condition is corrected is smaller than the speed difference before the machining condition is corrected in difference from the predicted value of the speed difference.

The machining condition correction unit 24A repeats the correction of the machining condition so that the speed difference approaches "0". That is, the machining condition correction unit 24A corrects the machining condition so that the machining speed becomes slower for the machining depth at which the speed difference becomes positive in the speed difference information 82. On the other hand, the machining condition correction unit 24A corrects the machining condition so that the machining speed becomes higher for the machining depth at which the speed difference becomes negative in the speed difference information 82.

This makes the actual speed difference approach the predicted value of the speed difference. Since the control device 2 continuously corrects the machining conditions so that the predicted value of the speed difference becomes 0, the actual speed difference approaches 0 by repeating the correction of the machining conditions. That is, the actual machining speed is approaching the target speed. The display unit 13 displays the

graphics

51, 52, 61, and 62.

The pattern 53 shown in fig. 17 is the same pattern as that shown in fig. 16. The horizontal axis of the

graphs

53 and 54 shown in fig. 17 represents the machining elapsed time, and the vertical axis represents the machining depth. In addition, in the

graphs

63 and 64 shown in fig. 17, the horizontal axis represents the machining depth, and the vertical axis represents the velocity difference.

Fig. 53 shows machining speed information 73, which is a correspondence relationship between the machining elapsed time and the machining depth before the machining condition is corrected. The graph 54 shows the machining speed information 74, which is the correspondence between the machining elapsed time and the machining depth after the machining conditions have been corrected for each of the machining quadrants a1 to a 4.

The

machining speed information

73 and 74 is information of the machining depth to be machined per unit time. In the machining speed information 73, the machining depth is not proportional to the machining elapsed time.

Fig. 63 shows velocity difference information 83, which is a correspondence relationship between the machining depth and the velocity difference before the machining condition is corrected. The graph 64 shows velocity difference information 84, which is a correspondence relationship between the machining depth and the velocity difference after the machining conditions are corrected for each of the machining quadrants a1 to a 4.

The

velocity difference information

83 and 84 shows the velocity difference for each depth block of the machining quadrants a1 to a 4. Each depth block shows 4 speed differences, which are speed differences for each of the machining quadrants a1 to a 4.

The left histogram shown in each depth block is the velocity difference of the machining quadrant a1, and the 2 nd histogram from the left shown in each depth block is the velocity difference of the machining quadrant a 2. The 3 rd histogram from the left shown in each depth block is the velocity difference of the machining quadrant A3, and the right histogram shown in each depth block is the velocity difference of the machining quadrant a 4.

The "0" on the vertical axis in the graph 63 is a predicted value of the velocity difference when the machining is performed under the machining condition before the correction. That is, when machining is performed under the machining condition before correction, the predicted value of the speed difference is "0". In actual machining, as shown in graph 63, the speed difference may become positive or negative in a specific machining quadrant of a specific depth block.

The vertical axis "0" in the graph 64 is a predicted value of the velocity difference obtained by correcting the machining conditions for each of the machining quadrants a1 to a 4. That is, when machining is performed under the machining condition after correction, the predicted value of the speed difference is "0". In actual machining, as shown in graph 64, the speed difference is often not 0.

The machining condition correction unit 24A corrects the machining condition so that the machining speed is reduced for the machining quadrant of the depth block for which the speed difference is positive in the speed difference information 83. On the other hand, the machining condition correction unit 24A corrects the machining condition so that the machining speed becomes higher for the machining quadrant of the depth block for which the speed difference is negative in the speed difference information 83.

The machining condition correction unit 24A corrects the machining conditions corresponding to the pattern 53 for each of the machining quadrants a1 to a4, so that the machining speed information 73 becomes the machining speed information 74 and the speed difference information 83 becomes the speed difference information 84. As shown in the speed difference information 83, the difference between the speed difference before the machining condition is corrected and the predicted value of the speed difference is large. As shown in the velocity difference information 84, the difference between the velocity difference after the machining condition correction and the predicted value of the velocity difference before the machining condition correction is smaller for each of the machining quadrants a1 to a 4.

The machining condition correction unit 24A repeats the correction of the machining conditions so that the speed difference approaches "0" for each of the machining quadrants a1 to a 4. This makes the actual speed difference approach the predicted value of the speed difference. Since the control device 2 continuously corrects the machining conditions so that the predicted value of the speed difference becomes 0, the actual speed difference in each of the machining quadrants a1 to a4 is also continuously close to 0 by repeating the correction of the machining conditions for each of the machining quadrants a1 to a 4. The display unit 13 displays the

graphics

53, 54, 63, and 64.

As described above, in embodiment 2, since the machining result evaluation unit 16B corrects the machining condition based on the result of comparing the machining speed as the machining result with the target speed, the machining condition of the die-sinking electrical discharge machining can be corrected to the machining condition corresponding to the target speed. Further, since the machining result evaluation unit 16B corrects the machining conditions for each of the machining quadrants a1 to a4, it is possible to derive the machining conditions that can achieve the target machining for each of the machining quadrants a1 to a 4.

Embodiment 3.

Next, embodiment 3 of the present invention will be described with reference to fig. 18. In embodiment 3, the machining condition correction unit includes a machine learning device that learns the machining conditions that can achieve the target machining for each of the machining quadrants a1 to a4 based on the machining conditions used for the actual machining and the machining results obtained by the actual machining. In the present embodiment, a case where the machining result is the machining time will be described. In the following description, the difference between the machining time and the target time may be referred to as a time difference. The time difference is a positive value when the machining time is longer than the target time, and is a negative value when the machining time is less than or equal to the target time.

Fig. 18 is a diagram showing a configuration of a control device according to embodiment 3. Among the components in fig. 18, those that realize the same functions as those of the control device 2 in embodiment 1 shown in fig. 2 are given the same reference numerals, and redundant description thereof is omitted.

Fig. 18 also shows the display unit 13 and the machining specifications in addition to the machining result evaluation unit 16C, the power supply control unit 15, and the machine control unit 14, which are the components of the control device 2. The machining result evaluation unit 16C includes a machining condition correction unit 24B instead of the machining condition correction unit 24A, as compared with the machining result evaluation unit 16A.

The processing condition correction unit 24B includes a machine learning device 29 and an intention determination unit 33. The machine learning device 29 includes a learning unit 30 and a state observation unit 40. The state observation unit 40 acquires evaluation points for each of the machining quadrants a1 to a4 from the evaluation calculation unit 22. The evaluation points acquired by the state observing unit 40 are the evaluation points before the machining condition correction and the evaluation points after the machining condition correction. The state observation unit 40 acquires the position conditions for each of the machining quadrants a1 to a4 from the machine control unit 14, and acquires the power conditions for each of the machining quadrants a1 to a4 from the power supply control unit 15. That is, the state observation unit 40 acquires the machining conditions for each of the machining quadrants a1 to a4 from the machine control unit 14 and the power supply control unit 15. The machining conditions acquired by the state observing unit 40 are machining conditions before machining condition correction and machining conditions after machining condition correction. The state observation unit 40 observes the obtained evaluation points for the respective machining quadrants a1 to a4 and the obtained machining conditions for the respective machining quadrants a1 to a4, and transmits the observed results to the learning unit 30 as state variables.

The learning unit 30 learns the machining conditions to be used next for each of the machining quadrants a1 to a4, in accordance with the state variables. That is, the learning unit 30 learns the machining conditions in which the time difference is small, that is, the evaluation point is high, for each of the machining quadrants a1 to a 4. Specifically, the learning unit 30 learns the next action, which is the machining condition, in accordance with a data set created based on state variables including the machining condition and the evaluation point.

The learning unit 30 includes a function updating unit 31 and a reward calculating unit 32. The reward calculation unit 32 calculates a reward based on the state variable. That is, the return calculation unit 32 calculates the return of the machining condition based on the machining condition and the evaluation point. The return calculation unit 32 increases the return of the machining condition as the evaluation point is higher. That is, the reward calculation unit 32 compares the evaluation points before and after the correction of the machining condition, and gives a high reward to the machining condition having a high evaluation point.

The function update unit 31 stores a function for deciding an action, and updates the function for deciding an action based on the return. An example of the function for determining an action is an action cost function Q(s) described later_t，a_t). The function update unit 31 of the present embodiment performs the electric discharge machining on the action merit function Q(s) so as to determine the machining condition in which the evaluation point becomes high each time the electric discharge machine 1 performs the electric discharge machining_t，a_t) And (6) updating. The function update unit 31 transmits the calculated action to the intention determination unit 33.

That is, the electric discharge machine 1 performs electric discharge machining under a certain machining condition, and changes a specific item among the machining conditions by a specific value according to the size of the evaluation point. The state observation unit 40 may observe the machining condition and the time difference as the state variables. In this case, the learning unit 30 learns the action, which is the next processing condition, in accordance with a data set created based on the state variables including the processing conditions and the time difference.

When the evaluation point is lower than the reference value, the machine learning device 29 corrects the machining condition so that the evaluation point exceeds the reference value. For example, when the time difference is a negative value, the machine learning device 29 increases the numerical value of the item that affects the machining time in the machining condition by a specific value so that the machining time becomes longer. On the other hand, when the time difference is a positive value, the machine learning device 29 increases the numerical value of the item that affects the machining time in the machining condition by a specific value so that the machining time becomes short. The machine learning device 29 repeatedly learns these processes, and controls the next electric discharge machining using the result of the learning. As described above, the machine learning device 29 continuously learns the machining conditions so that the machining time approaches the target time.

The intention determining unit 33 uses the action merit function Q(s) updated by the function updating unit 31_t，a_t) I.e. the trained model calculates the action. Specifically, the intention determining unit 33 uses the action cost function Q(s) if the evaluation point and the machining condition observed by the state observing unit 40 are input_t，a_t) The optimum corrected machining condition is calculated by increasing the evaluation point to be higher than the current evaluation pointAnd (4) calculating. The intention determining unit 33 transmits the machining condition, which is the calculated action, to the machining condition storage unit 25 and the evaluation output unit 26. In this case, the machining condition storage unit 25 stores the machining conditions learned by the learning unit 30, and the evaluation output unit 26 displays the machining conditions learned by the learning unit 30 on the display unit 13. If the machining condition is out of the allowable range, the intention determining unit 33 does not send the machining condition to the machining condition storage unit 25, but sends it to the evaluation output unit 26. In this case, the evaluation output unit 26 also causes the display unit 13 to display the machining condition learned by the learning unit 30.

The user can also determine whether or not the machining conditions calculated by the function update unit 31 are acceptable. In this case, the intention determining unit 33 determines whether or not machining conditions are permitted in accordance with an instruction (a permissible instruction or an impermissible instruction) input to the machining result evaluating unit 16C by the user.

Here, the learning process of the action of the learning unit 30 will be described. The learning algorithm used by the learning unit 30 may be any learning algorithm. Here, a case where Reinforcement Learning (Reinforcement Learning) is applied to the Learning algorithm will be described. Reinforcement learning is an action to be taken by an agent, which is an action agent in a certain environment, by observing a current state represented by a state variable and deciding an action to be taken based on the observation result. The agent receives a response from the environment by selecting an action, and learns a countermeasure that is most highly reported by a series of actions. As typical reinforcement Learning methods, Q-Learning (Q-Learning) and TD-Learning (TD-Learning) are known. For example, in the case of Q learning, the action merit function Q(s)_t，a_t) Is represented by the following formula (1). That is, an example of the action value table is the action value function Q(s) of the formula (1)_t，a_t)。

[ formula 1 ]

In formula (1), s_tRepresenting the environment at time t, a_tIndicating the action at time t. By action a_tThe environment becomes s_t+1。r_t+1Represents the return by the change of its environment, γ represents the discount rate, and α represents the learning coefficient. When Q learning is applied, the next processing condition becomes action a_t。

The update represented by equation (1) is such that if the action value of the best action a at time t +1 is greater than the action value Q of action a performed at time t, the action value Q is increased, and conversely, the action value Q is decreased. In other words, the action value function Q(s) is applied so that the action value Q of the action a at the time t is close to the best action value Q at the time t +1_t，a_t) And (6) updating. Thus, the best action value in a certain environment is in turn propagated to the action values in its previous environment.

For example, the electric discharge machine 1 performs machining under the machining conditions before correction, and calculates an evaluation point (here, the 1 st evaluation point) in the machining conditions before correction by the control device 2. Then, the control device 2 corrects the machining condition based on the 1 st evaluation point, and the electric discharge machine 1 performs machining under the corrected machining condition, and calculates an evaluation point (here, the 2 nd evaluation point) in the machining condition corrected by the control device 2. In this case, if the 2 nd evaluation point is higher than the 1 st evaluation point, the reward calculation unit 32 increases the reward for the machining condition after the correction. At this time, the reward calculation unit 32 gives a reward of "1", for example, to the machining condition after the correction. On the other hand, if the 2 nd evaluation point is smaller than the 1 st evaluation point, the reward calculation unit 32 decreases the reward with respect to the machining condition after the correction. At this time, the reward calculation unit 32 gives a reward of "-1", for example, to the machining condition after the correction. For example, when the evaluation point is the highest point, the return calculation unit 32 sets the return as the maximum return. The reward calculation unit 32 sends the calculated reward to the function update unit 31.

The function update unit 31 updates the function for determining the action in accordance with the reward calculated by the reward calculation unit 32. For example, in the case of Q learning, a line represented by the formula (1)Dynamic value function Q(s)_t，a_t) Is a function for calculating an action, and is updated by the function update unit 31.

The machining performance may be a difference between the actual machining time ratio and the target machining time ratio for each of the machining quadrants a1 to a4, or may be a difference between the actual machining depth and the target machining depth for each of the machining quadrants a1 to a 4. The machining performance may be a difference between the actual machining speed and the target machining speed for each of the machining quadrants a1 to a 4.

The machining condition correction unit 24B described in embodiment 3 may be applied to the machining result evaluation unit 16B described in embodiment 2. Fig. 19 is a diagram showing another configuration example of the control device according to embodiment 3. Of the components shown in fig. 19, those that realize the same functions as those of the control device 2 described in

embodiments

1 and 2 are given the same reference numerals, and redundant description thereof is omitted.

Fig. 19 also shows the display unit 13 and the machining specifications in addition to the machining result evaluation unit 16D, the power supply control unit 15, and the machine control unit 14, which are components of the control device 2. The machining result evaluation unit 16D includes a machining condition correction unit 24B instead of the machining condition correction unit 24A, as compared with the machining result evaluation unit 16B. The processing condition correction unit 24B of the processing result evaluation unit 16D also includes an intention determination unit 33 and a machine learning device 29, similarly to the processing condition correction unit 24B of the processing result evaluation unit 16C.

The machine learning device 29 of the machining result evaluation unit 16D also learns the machining conditions that can achieve the target machining for each of the machining quadrants a1 to a4 based on the machining results obtained by the actual machining, as in the machine learning device 29 of the machining result evaluation unit 16C.

In the present embodiment, the case where the machine learning device 29 performs machine learning by reinforcement learning has been described, but the machine learning device 29 may perform machine learning by other known methods, for example, neural networks, genetic programming, functional logic programming, support vector machines, and the like.

As described above, in embodiment 3, the machine learning device 29 observes the machining conditions and the evaluation points as state variables, and learns the machining conditions having the higher evaluation points for each of the machining quadrants a1 to a4 in accordance with the data set created based on the state variables. Therefore, the machine learning device 29 can learn the machining conditions that can achieve the target machining for each of the machining quadrants a1 to a 4.

Here, the hardware configuration of the machining result evaluation units 16A to 16D will be described. Fig. 20 is a diagram showing an example of a hardware configuration of a processing result evaluation unit included in the control device according to the embodiments 1 to 3. Note that since the machining result evaluation units 16A to 16D have the same hardware configuration, the hardware configuration of the machining result evaluation unit 16A will be described here.

The processing result evaluation unit 16A can be realized by an input device 151, a processor 152, a memory 153, and an output device 154 shown in fig. 20. Examples of the processor 152 are a CPU (also referred to as a Central Processing Unit, a Processing Unit, an arithmetic Unit, a microprocessor, a microcomputer, a processor, a dsp (digital Signal processor)), or a system lsi (large Scale integration). Examples of the memory 153 are a ram (random Access memory), a rom (read Only memory).

The machining result evaluation unit 16A is realized by the processor 152 reading out and executing a machining result evaluation program stored in the memory 153 for executing the operation of the machining result evaluation unit 16A, which can be executed by a computer. The machining result evaluation program, which is a program for executing the operation of the machining result evaluation unit 16A, can be said to cause a computer to execute the procedure or method of the machining result evaluation unit 16A.

The machining result evaluation program executed by the machining result evaluation unit 16A has a module configuration including the evaluation calculation unit 22, the actual result calculation unit 23, and the machining condition correction unit 24A, and these are placed on the main storage device and are generated on the main storage device.

The memory 153 is used as a temporary memory when various processes are executed by the processor 152. The memory 153 stores, for example, a machining result evaluation program, machining conditions, and the like. The output device 154 outputs the machining result, the evaluation point, the machining condition, and the like to an external device such as the display unit 13.

The machining result evaluation program may be provided as a computer program product by being stored in a computer-readable storage medium in an installable form or an executable form of a file. The machining result evaluation program may be supplied to the machining result evaluation unit 16A via a network such as the internet.

The function of the machining result evaluation unit 16A may be partially implemented by dedicated hardware such as a dedicated circuit, and partially implemented by software or firmware. The control device 2 may have the same hardware configuration as the machining result evaluation unit 16A.

The configuration described in the above embodiment is an example of the content of the present invention, and may be combined with other known techniques, and a part of the configuration may be omitted or modified without departing from the scope of the present invention.

Description of the reference symbols

An electric discharge machine 1, a control device 2, a drive unit 12, a display unit 13, a machine control unit 14, a power supply control unit 15, a machining result evaluation unit 16A to 16D, a workpiece 17, a table 18, a base 19, an input unit 21, an evaluation calculation unit 22, an actual result calculation unit 23, a machining condition correction unit 24A, 24B, a machining condition storage unit 25, an evaluation output unit 26, a machining condition output unit 27, a target speed storage unit 28, a machine learning unit 29, a learning unit 30, a function update unit 31, a reporting calculation unit 32, an intention determination unit 33, a speed difference calculation unit 35, a state observation unit 40, an input unit 151, a processor 152, a 153 memory 153, an output unit 154, an a0 machining area, a1 to a4 machining quadrants, a C center position, and E tool electrodes.

Claims

1. A control device, characterized by comprising:

a machining condition output unit that outputs a machining condition used in a shape-engraved electric discharge machining that performs wobbling to a control unit that controls a control target;

an actual result calculation unit that calculates a machining actual result of the electrical discharge machining for each machining area in a specific section based on a control result that is a result of controlling each machining area obtained by dividing the area subjected to the weaving machining in the specific section using the machining condition;

an evaluation calculation unit that calculates an evaluation point indicating an evaluation of the machining result for the specific section based on the machining result for each of the machining regions; and

and an output unit that outputs and displays the evaluation point.

2. The control device according to claim 1,

the evaluation calculation unit calculates the evaluation point for each of the machining areas.

3. The control device according to claim 1 or 2,

further comprising a processing condition correcting unit for correcting the processing condition so that the evaluation point is raised,

the machining condition output unit outputs the corrected machining condition to the control unit.

4. The control device according to claim 3,

the specific section is defined by the machining depth of the engraving electrical discharge machining,

the actual performance calculation unit calculates a machining actual performance of the engraving electrical discharge machining for each machining depth of the engraving electrical discharge machining,

the machining condition correcting unit corrects the machining condition for each machining depth of the engraving electrical discharge machining.

5. The control device according to claim 4,

the machining performance is a machining time for each of the machining areas.

6. The control device according to claim 4,

the machining performance is a ratio of a machining time for each machining region to an entire machining time for each machining depth of a 2-dimensional plane perpendicular to a depth direction.

7. The control device according to claim 3,

the specific interval is defined by the machining time of the engraving electrical discharge machining,

the actual performance calculation unit calculates a machining actual performance of the engraving electrical discharge machining for each machining time of the engraving electrical discharge machining,

the machining condition correcting unit corrects the machining condition for each machining time of the engraving electrical discharge machining.

8. The control device according to claim 7,

the machining performance is a machining depth for each of the machining regions.

9. The control device according to any one of claims 3 to 8,

the machining condition correcting unit includes:

a machine learning device that machine-learns the machining condition such that the evaluation point becomes higher, based on the evaluation point and the machining condition; and

and an intention determining unit that, if the evaluation point and the machining condition are input, calculates a machining condition higher than the evaluation point using a learning result obtained by machine learning by the machine learning device.

10. The control device according to any one of claims 1 to 9,

further comprising a speed difference calculation unit for calculating a machining speed based on the machining depth detected during the engraving electrical discharge machining and the control result for each of the machining areas, and calculating a difference between the calculated machining speed and a target machining speed,

the output unit outputs and displays the difference.

11. An electric discharge machine is characterized by comprising:

a driving unit that drives a tool electrode that generates electric discharge with a workpiece;

a machine control unit that controls the driving unit to control a distance between the tool electrode and the workpiece during the wobble electric discharge machining;

a power supply control unit that controls power supplied between the tool electrode and the workpiece;

a machining condition output unit that outputs machining conditions used in the engraving electrical discharge machining to the machine control unit and the power supply control unit;

and an output unit that outputs and displays the evaluation point.

12. A machine learning device for learning processing conditions used in a shape-engraved electric discharge process for performing a wobbling process,

the machine learning device is characterized by comprising:

a state observation unit that observes, as state variables, evaluation points representing evaluations of machining results of the die-sinking electrical discharge machining and the machining conditions for each of the machining regions obtained by dividing the region in which the weaving machining is performed; and

and a learning unit that learns the machining condition in which the evaluation point increases for each of the machining areas in accordance with a data set created based on the state variable.

13. The machine learning apparatus of claim 12,

the learning unit includes:

a return calculation unit that calculates a return of the machining condition based on the machining condition and the evaluation point; and

and a function updating unit that updates a function for determining an action as a next processing condition based on the return.

14. The machine learning apparatus of claim 13,

the reward calculation unit compares a1 st evaluation point in a1 st processing condition with a2 nd evaluation point in a2 nd processing condition, and gives a high reward to a processing condition having a higher evaluation point out of the 1 st evaluation point and the 2 nd evaluation point,

the function updating unit updates the function so as to determine a machining condition in which the evaluation point increases.