The purpose of the present invention is to provide a control device that enables learning control according to the angle-synchronization method to be applied even when oscillation operation is performed in a processing machine that does not have a main shaft. According to a configuration having a phase-data-creating unit and a learning control unit that performs learning control on the basis of the phase data, learning control according to the angle synchronization method can be applied even in processing machines that do not have a main shaft. The phase-data-creating unit creates servo-control-period-specific phase data from the servo control periods of a servo control device, and data relating to periods of non-repeatable normal movement commands and repeatable oscillation commands as communicated from a higher control device.

The purpose of the present invention concerns a control command for a machine tool, and is to make it possible to adopt a high-frequency repetitive oscillation command, without restrictions on communication capacity, when acquiring a control command created by superimposing a command with repeatability on a normal move command such as a move command having no repeatability. The purpose can be achieved by: acquiring, from an higher-order control device, only information about the form of a command, and parameter information on numerical information relating to an amplitude, a period, and other feature amount; creating a repetitive movement command that is a command with repeatability only from the parameter information; and superimposing the repetitive movement command on a normal move command such as a move command having no repeatability.

The present invention is a servo control device that controls a servomotor based on a command position. The servo control device includes a correction unit that corrects the command position using a first neural network that performs processing based on parameters representing a network structure, and a servo amplifier that controls the servomotor based on a corrected command position outputted from the correction unit. The correction unit corrects the command position based on the corrected command position and the actual position of the servomotor.

An evaluation value E.sub.k on a trajectory error e.sub.k between an actual trajectory y.sub.k and a target trajectory x is calculated. In a case where the calculated evaluation value E.sub.k is better than a best evaluation value E.sub.best, the best evaluation value E.sub.best is updated by the evaluation value E.sub.k and is stored. A commanded trajectory u.sub.k in this situation is employed as a best commanded trajectory u.sub.best and stored. In a case where the calculated evaluation value E.sub.k is worse than the best evaluation value E.sub.best, a compensator that calculates a correction of the trajectory u.sub.k+1 is changed to another compensator and the correction of the trajectory u.sub.k+1 is calculated. A commanded trajectory in the next-time operation u.sub.k+1 is calculated from the correction of the trajectory u.sub.k+1 and the best commanded trajectory u.sub.best.

An evaluation value E.sub.k on a trajectory error e.sub.k between an actual trajectory y.sub.k and a target trajectory x is calculated. In a case where the calculated evaluation value E.sub.k is better than a best evaluation value E.sub.best, the best evaluation value E.sub.best is updated by the evaluation value E.sub.k and is stored. A commanded trajectory u.sub.k in this situation is employed as a best commanded trajectory u.sub.best and stored. In a case where the calculated evaluation value E.sub.k is worse than the best evaluation value E.sub.best, a compensator that calculates a correction of the trajectory u.sub.k+1 is changed to another compensator and the correction of the trajectory u.sub.k+1 is calculated. A commanded trajectory in the next-time operation u.sub.k+1 is calculated from the correction of the trajectory u.sub.k+1 and the best commanded trajectory u.sub.best.

The purpose of the present invention concerns a control command for a machine tool, and is to make it possible to adopt a high-frequency repetitive oscillation command, without restrictions on communication capacity, when acquiring a control command created by superimposing a command with repeatability on a normal move command such as a move command having no repeatability. The purpose can be achieved by: acquiring, from an higher-order control device, only information about the form of a command, and parameter information on numerical information relating to an amplitude, a period, and other feature amount; creating a repetitive movement command that is a command with repeatability only from the parameter information; and superimposing the repetitive movement command on a normal move command such as a move command having no repeatability.