Open AccessThermal Deformation Test and Modeling of Main Spindle of Numerical Control Vertical Machining Center
Author(s) -
Qijun Xiao,
Zhizeng Luo,
Min Luo,
Shihsheng Liu,
Wei-Tai Hsu,
Chia-Wei Huang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1583/1/012005
Subject(s) - machining , machine tool , artificial neural network , deformation (meteorology) , thermal , numerical control , mechanical engineering , binary number , approximation error , computer science , engineering , materials science , artificial intelligence , algorithm , mathematics , physics , composite material , arithmetic , meteorology
Thermal error is the main factor affecting the manufacturing accuracy of the numerical control vertical machining center. The thermal deformation of the machine spindle is the main source of machine tool thermal error. In view of the deficiency of single factor modeling of thermal deformation of spindle of traditional machine tools, it is measured in this paper the temperature rise and Z-axis thermal deformation of a vertical machining center under several rotating speed conditions, the test data is analyzed and processed, and the relationship curves of thermal deformation of the spindle relative to time and temperature rise under several working conditions are drawn. Binary linear regression analysis model and artificial neural network model for thermal deformation of the spindle are presented. By comparing the residual error of these two modeling methods, the error for binary regression modeling is from −7.5μm to 10μm, however, the error for neural network modeling is from −5.5μm to 8.1μm, it is shown that neural network modeling is superior to binary regression modeling.