Machine learning device and machining time prediction device
AssigneeFanuc Corporation
InventorHisateru Ishiwari
Life of this patent
Estimating the term…
Abstract
A machine learning device acquires from a numerical controller information relating to machining when the machining is performed, and further acquires an actual delay time due to servo control and due to machine movement which are caused in the machining when the machining is performed. Then, the device performs supervised learning using the acquired machining-related information as input data, and using the acquired actual delay time due to servo control and due to machine movement as supervised data, and constructs a learning model, thereby predicting the machine delay time caused in a machine with high precision.
Claims4
A machine learning device for learning a delay time due to servo control and due to machine movement, which are caused when performing machining on a workpiece based on a machining program in a numerical controller for controlling a machine having at least one axis based on the machining program, the machine learning device comprising:
- a processor configured to
- when the machining is performed, acquire information relating to the machining from the numerical controller at learning timings at which a command of a plurality of commands of the machining program is executed and at which a subsequent command of the plurality of commands is executed upon completion of execution of the command of the plurality of commands,
- acquire an actual delay time due to servo control and due to machine movement, which are caused in the machining when the machining is performed,
- perform supervised learning using the acquired information relating to the machining as input data, and using the acquired actual delay time due to servo control and due to machine movement as supervised data, wherein the processor is configured to omit, among the learning timings, a learning timing not causing the actual delay time from the supervised learning,
- construct a learning model based on the supervised learning, and
- store the learning model.
A machine learning device for predicting a delay time due to servo control and due to machine movement, which are caused when performing machining on a workpiece based on a machining program in a numerical controller controlling a machine having at least one axis based on the machining program, the machine learning device comprising:
- a processor configured to
- store a learning model constructed by
- acquiring information relating to the machining at learning timings at which a command of a plurality of commands is executed and at which a subsequent command of the plurality of commands is executed upon completion of execution of the command of the plurality of commands,
- acquiring an actual delay time due to servo control and due to machine movement, which are caused in the machining, and
- performing supervised learning using the acquired information relating to the machining as input data, and using the acquired actual delay time due to servo control and due to machine movement as supervised data, while omitting, among the learning timings, a learning timing not causing the actual delay time from the supervised learning,
- acquire information relating to machining of the workpiece based on the machining program calculated from machining simulation based on the machining program, and
- predict, based on the learning model, a delay time due to servo control and due to machine movement from the acquired information relating to the machining.
A system, comprising:
- a machine learning device for predicting a delay time due to servo control and due to machine movement, which are caused when performing machining on a workpiece based on a machining program in a numerical controller controlling a machine having at least one axis based on the machining program, the machine learning device comprising a processor configured to
- store a learning model constructed by
- acquiring information relating to the machining at learning timings at which a command of a plurality of commands is executed and at which a subsequent command of the plurality of commands is executed upon completion of execution of the command of the plurality of commands,
- acquiring an actual delay time due to servo control and due to machine movement, which are caused in the machining, and
- performing supervised learning using the acquired information relating to the machining as input data, and using the acquired actual delay time due to servo control and due to machine movement as supervised data, while omitting, among the learning timings, a learning timing not causing the actual delay time from the supervised learning,
- acquire information relating to machining of the workpiece based on the machining program calculated from machining simulation based on the machining program, and
- predict, based on the learning model, a delay time due to servo control and due to machine movement from the acquired information relating to the machining;
- and
- a machining time prediction device, comprising a further processor configured to
- predict a reference machining time not including the delay time due to servo control and due to machine movement based on the machining program,
- calculate a correction time for correcting the reference machining time based on the delay time due to servo control and due to machine movement predicted from the machining program by the machine learning device, and
- calculate a predicted machining time obtained by correcting the reference machining time by the correction time.
Dependencies are inferred from the claim text.