Modelling and compensation of thermal deformation for machine tool based on the real-time data of the CNC system

Abstract

As one of the main issue affecting the precision and stability of the machine tool, thermal deformation is determined by the environmental temperature as well as the heat generated in the working process of machine tool. Currently, research work on the modelling of thermal deformation mostly rely on the data retrieved from temperature sensors embedded or amount on the machine tool, where the thermal deformation is modelled as function of temperature field in the machine tool. Thermal deformation thus modelled is normally impractical in the thermal compensation process because of the high cost and low stability it may incur when a set of temperature sensors are utilized. In this paper, a novel approach for modelling and predicting the thermal deformation of ball screw is proposed. Instead of relying on the data from temperature sensors, the real-time data of the CNC system, e.g., current and velocity of driving motor, and displacement of axis, is utilized directly to derive the thermal deformation model. With such data, a model is firstly built with considering the heat generated from friction motion as well as the heat dissipation with the machining environment; the parameters of the proposed model were then identified by a multiple linear regression method. Based on the proposed model, the thermal deformation of ball screw can be precisely estimated in both the warming up and cooling down state of the machine tool, as validated by experiments on a ball screw of a small size machine tool.