Back to results
US 12,311,431 B2Granted
Control device, control system, and control method for wire straightener
AssigneeOMRON Corporation
InventorTakashi Fujii
Life of this patent
Estimating the term…
Abstract
Accuracy of feedforward control for a wire straightener is improved. A feedback operation amount to a straightener is determined by a feedback control means based on an error between a target curvature and a straightening curvature. A feedforward compensation means determines a feedforward compensation value from a measurement value of the straightener using a prediction model. A learning means performs machine learning on the prediction model using teaching data. The learning means adds at least one combination including the measurement value and a manipulated variable when an absolute value of an error is smaller than a reference value to the teaching data.
Claims10
A control device of a straightener, the control device comprising:
- a processor configured to determine a feedforward compensation value corresponding to a feedback operation amount of the straightener such that a straightening curvature of a wire straightened by the straightener approaches a target curvature,
- wherein the straightener includes:
- a winding axis about which the wire is wound;
- a plurality of rollers that are alternately disposed so as to sandwich and press the wire that is drawn from the winding axis;
- and
- a motor configured to draw the wire from the winding axis through the plurality of rollers,
- wherein the straightening curvature is a curvature of a first part of the wire that has passed through the plurality of rollers,
- wherein the feedback operation amount is determined by the processor based on an error between the target curvature and the straightening curvature,
- wherein a sum of the feedback operation amount and the feedforward compensation value is output by the processor to the straightener as an operation amount, and
- wherein the processor of the control device is configured to:
- determine the feedforward compensation value from a measurement value obtained from the straightener, using a predictive model;
- and
- perform a machine learning on the predictive model using teaching data by:
- acquiring at least one first combination including the target curvature, the operation amount corresponding to the target curvature, the measurement value, and the straightening curvature corresponding to both the measurement value and the operation amount;
- and
- adding at least one second combination in the at least one first combination to the teaching data, the at least one second combination including the measurement value and the operation amount when an absolute value of the error is smaller than a first reference value.
13 more elements
A control system that outputs a sum of a feedback operation amount to a straightener and a feedforward compensation value as an operation amount to the straightener such that a straightening curvature of a wire straightened by the straightener approaches a target curvature,
- wherein the straightener includes:
- a winding axis around which the wire is wound;
- and
- a plurality of rollers that are alternately disposed so as to sandwich and press the wire that is drawn from the winding axis;
- and
- a motor configured to draw the wire from the winding axis through the plurality of rollers,
- wherein the straightening curvature is a curvature of a part of the wire that has passed through the plurality of rollers,
- wherein the control system comprises:
- a first processor configured to determine the feedback operation amount based on an error between the target curvature and the straightening curvature;
- a second processor configured to determine the feedforward compensation value from a measurement value obtained from the straightener, using a predictive model;
- and
- a third processor configured to perform a machine learning on the predictive model using teaching data by:
- acquiring at least one first combination including the target curvature, the operation amount corresponding to the target curvature, the measurement value, and the straightening curvature corresponding to both the measurement value and the operation amount;
- and
- adding at least one second combination in the at least one first combination to the teaching data, the at least one second combination including the measurement value and the operation amount when an absolute value of the error is smaller than a first reference value.
12 more elements
A method for controlling a straightener, the method comprising:
- determining a feedforward compensation value corresponding to a feedback operation amount of the straightener such that a straightening curvature of a wire straightened by the straightener approaches a target curvature,
- wherein the straightener includes:
- a winding axis around which the wire is wound;
- a plurality of rollers that are alternately disposed so as to sandwich and press the wire drawn from the winding axis;
- and
- a motor configured to draw the wire from the winding axis through the plurality of rollers,
- wherein the straightening curvature is a curvature of a part of the wire that has passed through the plurality of rollers,
- wherein the feedback operation amount is determined based on an error between the target curvature and the straightening curvature,
- wherein a sum of the feedback operation amount and the feedforward compensation value is output to the straightener as an operation amount, and
- wherein the method comprises:
- determining the feedforward compensation value from a measurement value obtained from the straightener, using a predictive model;
- and
- performing a machine learning on the predictive model using teaching data,
- wherein performing the machine learning includes:
- acquiring at least one first combination including the target curvature, the operation amount corresponding to the target curvature, the measurement value, and the straightening curvature corresponding to both the measurement value and the operation amount;
- and
- adding at least one second combination in the at least one first combination to the teaching data, the at least one second combination including the measurement value and the operation amount when an absolute value of the error is smaller than a first reference value.
14 more elements
Dependencies are inferred from the claim text.