-
公开(公告)号:US20210036640A1
公开(公告)日:2021-02-04
申请号:US17065589
申请日:2020-10-08
发明人: Tomoya FUJITA , Toshiaki KIMATA , Tatsuya ITO , Yukihiro IUCHI , Goh SATO
摘要: A machine learning apparatus for learning a correction parameter used in correction of a command value that controls a motor in a motor drive system including a plurality of kinds of correction functions includes: a state observation unit that observes, as a state variable, each of a feature calculated on the basis of drive data and the kind of any of the correction functions of the motor drive system and the correction parameter; and a learning unit that learns the correction parameter for each of the correction functions according to a training data set created on the basis of the state variable.
-
公开(公告)号:US20200012253A1
公开(公告)日:2020-01-09
申请号:US16340385
申请日:2018-07-05
发明人: Toshiaki KIMATA , Tomoya FUJITA , Toshihiro AZUMA , Tatsuya ITO
IPC分类号: G05B19/402 , G05B13/02 , G05B13/04
摘要: A numerical control device includes: a tool-side displacement measurement unit; a workpiece-side displacement measurement unit; a drive signal measurement unit; a relative displacement calculation unit between the tool and the workpiece; a relative displacement prediction unit calculating a relative displacement predicted value from the drive signal, from a prediction model representing a relationship between the drive signal and the relative displacement; a model parameter operation unit generating prediction model parameters constituting the prediction model, from the drive signal, the relative displacement, and the predicted value; and a command value correction unit outputting a post-correction position command obtained by correcting a position command to the drive unit using the prediction model parameters. The model parameter operation unit changes the prediction model parameters to reduce a difference between the relative displacement and the predicted value.
-
3.
公开(公告)号:US20190386595A1
公开(公告)日:2019-12-19
申请号:US16336472
申请日:2018-06-14
发明人: Tomoya FUJITA , Toshiaki KIMATA , Tatsuya ITO , Yukihiro IUCHI , Goh SATO
摘要: A machine learning apparatus for learning a correction parameter used in correction of a command value that controls a motor in a motor drive system including a plurality of kinds of correction functions includes: a state observation unit that observes, as a state variable, each of a feature calculated on the basis of drive data and the kind of any of the correction functions of the motor drive system and the correction parameter; and a learning unit that learns the correction parameter for each of the correction functions according to a training data set created on the basis of the state variable.
-
公开(公告)号:US20160349732A1
公开(公告)日:2016-12-01
申请号:US15101218
申请日:2013-12-13
发明人: Toshiaki KIMATA , Kotaro NAGAOKA , Naoki NAKAMURA
IPC分类号: G05B19/404 , G05B19/402
CPC分类号: G05B19/404 , G05B19/402 , G05B19/416 , G05B2219/37388 , G05B2219/37621 , G05B2219/41056
摘要: To dynamically change the acceleration without trial executions of a machining program, a numerical control device, which controls a motor on the basis of a machining program that specifies a path for a drive target of the motor, includes a changing unit that changes the acceleration of the motor under the control of the motor on the basis of the inertia ratio, the acceleration-deceleration factor that can be externally input, or the current value of the motor.
摘要翻译: 为了在没有加工程序的试执行的情况下动态地改变加速度,基于指定电动机的驱动目标的路径的加工程序来控制电动机的数控装置包括改变单元,其改变加速度的加速度 电机根据惯性比的电机控制,加速减速系数可以从外部输入,也可以是电机的当前值。
-
-
-