Intelligent Machine Tool – A Thermal Diagnostic System for a CNC Pretensioned Ball Screw

The paper presents a compensation system of thermal deformation for conventional feed axes applied in CNC machine tools allowing for an effective reduction in the impact of heat generated during its operation on the positioning accuracy of the axis. The result has been achieved by equipping feed screws with thermistor temperature sensors. Wiring sensors was led out through an axial bore in the screw and through a rotating electrical connector to an acquisition device coupled with the control system of the CNC machine. An algorithm based on neural networks was implemented in the machine control system, which allows for the online calculation and compensation of heat deformation of feed screws. The algorithm takes into account a variation of thermal deformation values as a function of the table position and the current distribution of the temperature field of the screw and machine. The paper presents a user-friendly method for implementing algorithms containing neural networks in the machine control system. The proposed compensation method has been verified by measuring the linear accuracy of the feed axis positioning. The obtained results confirm the effectiveness of the proposed method in reducing the impact of thermal deformation errors on the positioning accuracy of the axis in CNC machine tools.