Abstract: A control device for a die cushion mechanism, for carrying out force control with high accuracy and at a high speed. The control device for controlling the force generated by the die cushion mechanism utilizes a press working cycle which is repeatedly carried out. A correcting part of the control device corrects each deviation in one working cycle based on a time-series of deviation data in the just before press working cycle. By repeating this operation, the deviation between the detected value and the commanded value may be converged toward zero. Therefore, the deviation may become smaller than that of the conventional feedback control, whereby it is possible to respond to a change in the commanded value in a short time.

Abstract: A controller including a learning control unit for determining learning data based on a positional deviation between a target-position command commanding superimposed-type motion including repetitive motion and a positional fed-back variable obtained from an output portion of an electric motor; and an operation control section for controlling the electric motor based on a corrected positional deviation. The learning control unit includes a first learning section for periodically determining, based on the positional deviation, and storing, first learning data according to a first learning period; a learning-data correcting section for correcting the first learning data to eliminate an influence of a local change included in the target-position command or the positional fed-back variable and periodically arising according to a period different from the first learning period; and a positional-deviation correcting section for correcting the positional deviation by using corrected learning data.

Abstract: A motor control apparatus 1 that controls a rotation speed of a motor driven according to a speed command includes an abnormal-rotation detector 21 which determines that the motor is in an abnormal-rotation state, such as a runaway, when a speed deviation Ver obtained by subtracting a detection speed TSA of a motor rotation from a speed command VCMD is equal to or larger than a threshold value, when a sign of a product of acceleration ?TSA obtained by subtracting a last detection speed from a detection speed this time that is detected in a predetermined cycle and a torque command Tc obtained by proportionally integrating the speed deviation is negative, and when an integration value VerSUM of the speed deviation Ver exceeds a predetermined value ALMlevel. With this arrangement, an abnormal rotation of the motor can be instantly detected even during acceleration or deceleration of the motor, thereby increasing the performance of protecting the motor.

Abstract: A workpiece is rotated by a master motor and a tool is linearly moved by a slave motor to cut a thread in the workpiece. Position feedback of the master motor is multiplied by a coefficient K and the result used as the position command of the slave motor. Provision is made of an angle synchronization learning control unit for storing one pattern cycle's worth of the correction data of the threading and adding the same to the position deviation. This control unit stores one pattern cycle's worth of the correction data corresponding to the position feedback of the master motor. The position is converted to the correction data corresponding to the time at that time based on the stored correction data to find the correction data and this is added to the position deviation.

Abstract: A controller for controlling a position of a driven element with respect to a machine effecting end so as to be in accord with a command. An acceleration sensor is mounted to a member of the machine effecting end to which a tool is attached. Acceleration detected by the sensor is subjected to second-order integration by a torsion estimator to obtain displacement ?? of the machine effecting end from the original position. Position feedback P1 of a driven element is subtracted from position command Pc to obtain first position deviation ?1. The displacement ?? is added to the first position deviation ?1 to obtain second position deviation ?2. The second position deviation ?2 is subjected to learning control of a learning controller to obtain a correction value, which is added to the first position deviation ?1 to obtain velocity command Vc.

Abstract: Disclosed is a controller for restraining vibration of a driven element driven by a servomotor. The driven element driven by the servomotor is provided with acceleration detecting means. A correction value is obtained by multiplying a detected acceleration value detected by the acceleration detecting means by a coefficient. Correction is made by subtracting the correction value from a velocity command, and velocity feedback control is executed in a velocity control processing section to obtain a current command. Further, the servomotor is driven by current control processing, whereupon the driven element is moved. If the detected acceleration value is increased by vibration of the driven element, the velocity command is corrected to restrain the vibration, so that the vibration of the driven element can be restrained.

Abstract: A position-of-magnetic-pole detecting device includes first and second units for inferring a position of a magnetic pole. The first unit applies a current command while changing an excitation phase, infers a domain, where the position of a magnetic pole lies and which has the width thereof determined with a range of electrical angles, from the direction of rotation in which a motor rotates, and the excitation phase, and reduces the domain so as to infer the position of a magnetic pole. The second unit infers the position of a magnetic pole from a fed-back current returned responsively to a voltage command applied with the excitation phase sequentially changed. Thereafter, a voltage high enough to cause magnetic saturation is applied in order to determine the orientation of a magnetic pole, whereby the position of a magnetic pole is inferred. Although the first unit is selected, if the motor does not move due to friction or the like, the second unit is used to infer the position of a magnetic pole.

Abstract: There is provided a servo controller for synchronously controlling a master side drive source to drive one drive axis and a slave side drive source to drive the other drive axis. The servo controller includes a correction data calculation means for calculating correction data to correct a positional deviation of a slave side drive source according to a synchronization error which is a difference between a positional deviation of a master side drive source and a positional deviation of the slave side drive source, in which the correction data is added to the positional deviation of the slave side drive source.

Abstract: A high-frequency voltage whose amplitude is small is applied as a d-phase command voltage, and an excitation phase is changed to a predetermined degree at predetermined intervals, whereby a motor is driven. As the d-phase command voltage has the small amplitude and high frequency, the rotor of the motor does not rotate. A d-phase feed back current is detected, and the product of a derivative of the d-phase feed back current by the high-frequency voltage command is calculated. A high-frequency component is removed from the product. An excitation phase (direction of a magnetic flux) associated with a deviation of 0 or ? from the position of a magnetic flux and also associated with a peak value assumed by the product having the high-frequency component removed therefrom is detected. A plurality of thus detected excitation phases is averaged in order to determine the direction of a magnetic flux. The excitation phase associated with one of two peak values assumed by the product is adopted on a fixed basis.

Abstract: A controller for controlling a position of a driven element with respect to a machine effecting end so as to be in accord with a command. An acceleration sensor is mounted to a member of the machine effecting end to which a tool is attached. Acceleration detected by the sensor is subjected to second-order integration by a torsion estimator to obtain displacement ?? of the machine effecting end from the original position. Position feedback P1 of a driven element is subtracted from position command Pc to obtain first position deviation ?1. The displacement ?? is added to the first position deviation ?1 to obtain second position deviation ?2. The second position deviation ?2 is subjected to learning control of a learning controller to obtain a correction value, which is added to the first position deviation ?1 to obtain velocity command Vc.

Abstract: Disclosed is a controller for restraining vibration of a driven element driven by a servomotor. The driven element driven by the servomotor is provided with acceleration detecting means. A correction value is obtained by multiplying a detected acceleration value detected by the acceleration detecting means by a coefficient. Correction is made by subtracting the correction value from a velocity command, and velocity feedback control is executed in a velocity control processing section to obtain a current command. Further, the servomotor is driven by current control processing, whereupon the driven element is moved. If the detected acceleration value is increased by vibration of the driven element, the velocity command is corrected to restrain the vibration, so that the vibration of the driven element can be restrained.

Abstract: A control device for a die cushion mechanism, for carrying out force control with high accuracy and at a high speed. The control device for controlling the force generated by the die cushion mechanism utilizes a press working cycle which is repeatedly carried out. A correcting part of the control device corrects each deviation in one working cycle based on a time-series of deviation data in the just before press working cycle. By repeating this operation, the deviation between the detected value and the commanded value may be converged toward zero. Therefore, the deviation may become smaller than that of the conventional feedback control, whereby it is possible to respond to a change in the commanded value in a short time.

Abstract: A workpiece is rotated by a master motor and a tool is linearly moved by a slave motor to cut a thread in the workpiece. Position feedback of the master motor is multiplied by a coefficient K and the result used as the position command of the slave motor. Provision is made of an angle synchronization learning control unit for storing one pattern cycle's worth of the correction data of the threading and adding the same to the position deviation. This control unit stores one pattern cycle's worth of the correction data corresponding to the position feedback of the master motor. The position is converted to the correction data corresponding to the time at that time based on the stored correction data to find the correction data and this is added to the position deviation.

Abstract: A switch is turned on by a learning control start command from a master control unit, and the positional deviation at respective cycles is read in. Correction data read out from a learning memory is added to the positional deviation, and the result is filtered by band limiting filter and then stored in the learning memory as correction data. The correction data read out from the memory is compensated for phase delay, fall in gain, and the like, by a dynamics compensating element, and is added to the positional deviation and input to a positional control section. When the command pattern for the same shape is completed, and a learning control end command is output, whereupon the switch is turned off and learning control terminates.

Abstract: A high-frequency voltage whose amplitude is small is applied as a d-phase command voltage, and an excitation phase is changed to a predetermined degree at predetermined intervals, whereby a motor is driven. As the d-phase command voltage has the small amplitude and high frequency, the rotor of the motor does not rotate. A d-phase feed back current is detected, and the product of a derivative of the d-phase feed back current by the high-frequency voltage command is calculated. A high-frequency component is removed from the product. An excitation phase (direction of a magnetic flux) associated with a deviation of 0 or ? from the position of a magnetic flux and also associated with a peak value assumed by the product having the high-frequency component removed therefrom is detected. A plurality of thus detected excitation phases is averaged in order to determine the direction of a magnetic flux. The excitation phase associated with one of two peak values assumed by the product is adopted on a fixed basis.

Abstract: A position-of-magnetic-pole detecting device includes first and second units for inferring a position of a magnetic pole. The first unit applies a current command while changing an excitation phase, infers a domain, where the position of a magnetic pole lies and which has the width thereof determined with a range of electrical angles, from the direction of rotation in which a motor rotates, and the excitation phase, and reduces the domain so as to infer the position of a magnetic pole. The second unit infers the position of a magnetic pole from a fed-back current returned responsively to a voltage command applied with the excitation phase sequentially changed. Thereafter, a voltage high enough to cause magnetic saturation is applied in order to determine the orientation of a magnetic pole, whereby the position of a magnetic pole is inferred. Although the first unit is selected, if the motor does not move due to friction or the like, the second unit is used to infer the position of a magnetic pole.

Abstract: Learning control is performed when carrying out processing by repeating instructions in a pattern cycle. Time/position converting means determines a positional deviation for a prescribed position with respect to a reference position, from the positional deviation determined by sampling, and the reference position output in synchronization with the drive of the servo motor. Corresponding correction data stored in the memory means is added to the positional deviation, and then the result is subjected to filtering processing to update the correction data corresponding to the position. Position/time converting means then determines correction data for the current sampling time, on the basis of the correction data corresponding to the position as stored in the memory means, and the detected reference position. This correction data is processed to compensate for dynamic properties, thereby deriving a correctional quantity, which is added to the positional deviation.

Abstract: A gear machining controller capable of performing a gear machining of high precision. A gear machining tool and a workpiece are driven by a first motor and a second motor, respectively, to be synchronized with each other, to perform gear machining on the workpiece such as gear cutting and gear finishing. A position amending section is provided at least one of position control loops of servo systems for controlling position/velocity of the first and second motors. The position amendment section obtains amendment amounts based on position deviations and amends a position command of a present processing period by the amendment amount obtained at the processing period preceding by one cycle period of variation of a load torque. An error of synchronization of the first and second motors is reduced so that the gear machining of high precision is performed.

Abstract: A switch is turned on by a learning control start command from a master control unit, and the positional deviation at respective cycles is read in. Correction data read out from a learning memory is added to the positional deviation, and the result is filtered by band limiting filter and then stored in the learning memory as correction data. The correction data read out from the memory is compensated for phase delay, fall in gain, and the like, by a dynamics compensating element, and is added to the positional deviation and input to a positional control section. When the command pattern for the same shape is completed, and a learning control end command is output, whereupon the switch is turned off and learning control terminates.

Abstract: Learning control is performed when carrying out processing by repeating instructions in a pattern cycle. Time/position converting means determines a positional deviation for a prescribed position with respect to a reference position, from the positional deviation determined by sampling, and the reference position output in synchronization with the drive of the servo motor. Corresponding correction data stored in the memory means is added to the positional deviation, and then the result is subjected to filtering processing to update the correction data corresponding to the position. Position/time converting means then determines correction data for the current sampling time, on the basis of the correction data corresponding to the position as stored in the memory means, and the detected reference position. This correction data is processed to compensate for dynamic properties, thereby deriving a correctional quantity, which is added to the positional deviation.