Experiment-based thermal error modeling method for dual ball screw feed system of precision machine tool

Abstract

In order to investigate the thermal behavior of a dual ball screw feed drive system of a precision boring machine tool, experimental study on and theoretical modeling of thermally induced error along with heat generation characteristics are conducted in this paper. Experiments are carried out on the machine tool to measure and collect the thermodynamic information with its feed drive system operating under different working conditions. Based on the real-time data of the thermal expansion of ball screw in the axial direction, relationships between the thermal error and axial elongation are established to predict the thermal error distribution. The results show that the thermal error varies with different working position through the ball screw length linearly and with working time nonlinearly. In addition, another simplified way to model thermal error is presented to overcome the difficulties existing in the ball screw feed drive system of which the axial elongation is hard to collect. Fuzzy clustering and linear regression methods are employed to carry out the theoretical modeling of thermal error and optimization to sift out the critical heat sources. With the temperature data of these critical heat generation points, the thermally induced error of the ball screw feed drive system can be predicted easily alternatively. Experiments under a different condition are preformed to verify both prediction and modeling methods. It turns out that the proposed prediction methods are effective and practical to be used in the machining process as well.