Inertial Measurement Unit for Measuring and Processing of Axial Thermal Displacement Signal of a Machine Tool

The machining accuracy of a machine tool is affected by several factors, including temperature variations of the environmental space, thermal deformation in dynamic operations. However, with the development of In-dustry 4.0, the manufacturing industry has moved towards digital, intelligent and predictive technologies. In terms of eliminating thermal errors, the neural network method is utilized to obtain the thermal error compensation model for machine tools to improve the machining accuracy. However, when thermal and dynamic er-rors caused by the movement of a machine tool are abnormally large, it will lead to shutting down of the ma-chine tool for the elimination or handling of this problem. Shutdown detection has an impact on the production cycle and capacity of the factory. In this study, the inertial measurement unit (IMU) with accelerometers and gyroscopes was employed to measure the accuracy of a machine tool. When measuring the acceleration signals of a machine tool in the dynamic process, the acceleration signals were filtered and integrated by mathematical operations to obtain the velocity and displacement from IMU signals. The velocity and dis-placement data were combined through data fusion to eliminate information errors caused by multiple integration in mixed data. In the verification experiment, the machine tool was set with the error values of 15µm and 50µm to verify the signal measurement and processing accuracy of the IMU module. Under 10mm moving distance, the displacement of a machine tool could be detected by the errors of 20.58µm and 47.66µm, re-spectively. The errors in IMU measurement accuracy were 37.2% and 4.7%, respectively. The results from this study disclosed that this method produced highly reliable thermal displacement values in real-time and could be applied to development of functions such as instant fault identification and self-compensation control.