Control device of wire electric discharge machine and machine learning device

Patent metadata

Field	Value
Publication number	US-11267059-B2
Application number	US-201815928905-A
Country	US
Kind code	B2
Filing date	Mar 22, 2018
Priority date	Mar 30, 2017
Publication date	Mar 8, 2022
Grant date	Mar 8, 2022

Abstract

A control device of a wire electric discharge machine and a machine learning device are provided that can appropriately and readily determine a correction parameter. The control device, which optimizes the correction parameter for wire electrical discharge machining process, includes a machine learning device configured to learn the correction parameter for the wire electrical discharge machining process. The machine learning device includes a state observation unit configured to observe, as a state variable, condition data indicative of a condition for the wire electrical discharge machining process, a determination data acquisition unit configured to acquire determination data indicative of the correction parameter of the case where machining precision is favorable in the wire electrical discharge machining process, and a learning unit configured to learn the correction parameter in association with the condition for the wire electrical discharge machining process using the state variable and the determination data.

First claim

What is claimed is:

1. A controller for optimizing a proportional gain for performing a wire electrical discharge machining process in a wire electrical discharge machine, the controller comprising a processor configured to
learn the proportional gain for performing the wire electrical discharge machining process,
observe, as a state variable, condition data indicative of a condition for performing the wire electrical discharge machining process, the condition data including at least a machining shape defined by a machining program;
acquire determination data, the determination data being indicative of the proportional gain of a case where machining precision is favorable in the wire electrical discharge machining process; and
carry out supervised learning, using the state variable and the determination data, to associate the proportional gain with the condition for performing the wire electrical discharge machining process,
wherein the machining shape includes a section of a workpiece to be machined with a gap between the workpiece and a wire electrode of the wire electrical discharge machine, and
wherein an unevenness, which is undefined by the machining program, is on a surface of the section of the workpiece and the unevenness causes changes in the gap in the wire electrical discharge machining process defined by the machining program.

2. The controller according to claim 1 , wherein the processor is further configured to:
calculate an error between a correlation model that derives, from the state variable and the determination data, the proportional gain for performing the wire electrical discharge machining process and a correlation feature identified based on teacher data prepared in advance; and
update the correlation model such that the error is reduced.

3. The controller according to claim 1 , wherein the processor is further configured to carry out operation of the state variable and the determination data using a multilayer structure.

4. The controller according to claim 1 , wherein the processor is further configured to learn, using the state variables and pieces of the determination data obtained for each of a plurality of wire electric discharge machines, the proportional gain for performing the wire electrical discharge machining process on each of the wire electric discharge machines.

5. The controller according to claim 1 , wherein the machining shape is programmed in the machining program.

6. The controller according to claim 1 , wherein the machining shape includes at least one of concavity, convexity, an angle or a curvature, of a corner section of the workpiece to be machined.

7. The controller according to claim 1 , wherein the proportional gain is for controlling a feed rate of the wire electrode in the wire electrical discharge machining process.

8. The controller according to claim 1 , wherein a first value of the proportional gain at a corner portion of the machining shape defined by the machining program is different from a second value of the proportional gain at a linear portion of the machining shape defined by the machining program.

9. The controller according to claim 1 , wherein the processor is configured to carry out the supervised learning to learn a correction parameter sensitive to a change of the proportional gain.

10. The controller according to claim 9 , wherein the learned correction parameter is applicable to determine a first value of the proportional gain at a corner portion of the machining shape defined by the machining program, so that the first value is different from a second value of the proportional gain at a linear portion of the machining shape defined by the machining program.

Assignees

• Fanuc Corp

Inventors

• Nakashima Yasuo

Classifications

• B23H7/20

  for program control, e.g. adaptive · CPC title

• G05B2219/42018

  Pid learning controller, gains adapted as function of previous error · CPC title

• G05B2219/42152

  Learn, self, auto tuning, calibrating, environment adaptation, repetition · CPC title

• B23H7/02 Primary

  Wire-cutting · CPC title

• B23H1/024 Primary

  Detection of, and response to, abnormal gap conditions, e.g. short circuits (preventing short circuits or other abnormal discharges by altering machining parameters using adaptive control B23H7/16) · CPC title