A numerical controller includes a machine learning device for performing machine learning of machining condition adjustment of a machine tool. The machine learning device calculates a reward based on acquired machining-state data on a workpiece, and determines an adjustment amount of machining condition based on a result of machine learning and machining-state data, and adjusts machining conditions based on the adjustment amount. Further, the machine learning of machining condition adjustment is performed based on the determined adjustment amount of machining condition, the machining-state data, and the reward.

A feedback control numerical machine tool and a method thereof are provided. The machine tool includes at least two spindles, an acoustic frequency detecting module, at least two spindle position detecting modules, and a control module. The spindles machine a workpiece. The acoustic frequency detecting module detects an acoustic frequency of the spindles when machining the workpiece. The spindle position detecting modules detects position information of the spindles when machining the workpiece. The control module acquires the acoustic frequency and the position information of the spindles, monitors whether any of the spindles chatters according to the acoustic frequency of the spindles, and performs chattering avoidance to the spindle that chatters according to the position information of the spindles. As such, the present disclosure performs chattering monitoring to a plurality of spindles and avoids the chattering immediately.

A management system including a network, plural manufacturing cells connected to the network, and a management device that is connected to the network and manages the plurality of manufacturing cells, in which the manufacturing cell includes: a machine tool; and a control device that controls the machine tool, analyzes a vibration state of a spindle positioning shaft of the machine tool, and sends an analysis result via the network; in which the management device includes: a communication unit that receives the analysis result sent by the control device; and a detection unit that compares the analysis results thus received, and compares vibration states of the machine tool of each of the manufacturing cells, so as to detect spindle failure of any of the machine tools; and in which the communication unit, in a case of the detection unit detecting the spindle failure, sends a failure signal via the network.

A numerical controller includes a machine learning device for performing machine learning of machining condition adjustment of a machine tool. The machine learning device calculates a reward based on acquired machining-state data on a workpiece, and determines an adjustment amount of machining condition based on a result of machine learning and machining-state data, and adjusts machining conditions based on the adjustment amount. Further, the machine learning of machining condition adjustment is performed based on the determined adjustment amount of machining condition, the machining-state data, and the reward.