Back to Search Start Over

Thermal error modelling and compensation of CNC lathe feed system based on positioning error measurement and decoupling.

Authors :
Shi, Hu
Zhang, Boyang
Mei, Xuesong
Wang, Haitao
Zhao, Fei
Geng, Tao
Source :
Measurement (02632241). May2024, Vol. 231, pN.PAG-N.PAG. 1p.
Publication Year :
2024

Abstract

[Display omitted] • Feed system thermal error model based on positioning error decoupling is proposed. • Thermally induced error is affected by motion and non-motion heating. • Feed system already has thermal errors when it is not working. • Thermal error model based on MLR has physical meaning and good robustness. • MLR has better accuracy than data-driven models over long time prediction. Thermal error modeling and compensation considering of the measured positioning error decoupling for the lathe with a slant bed is proposed. Due to the complex coupling factors of thermal error, the current compensation methods exhibit limited applicability and generalization. The linear fitting method is used to decouple geometric error and thermal error from the positioning error. The motion heating and non-motion heating experiments of X-direction feed system were carried out to investigate the principle of thermal error generation and the impact factors. The multiple linear regression model of the thermal error was established based on the analysis of the thermal error measurement results. To verify the effectiveness of the proposed model, the thermal error at two working conditions was predicted, and then the error compensation was carried out. The results show that the positioning error is reduced by more than 84% and 92% with the present method compared with that without compensation. This research provides effective method for the generation principle and compensation of thermally induced errors of lathe with a slant bed. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
02632241
Volume :
231
Database :
Academic Search Index
Journal :
Measurement (02632241)
Publication Type :
Academic Journal
Accession number :
176630737
Full Text :
https://doi.org/10.1016/j.measurement.2024.114633