Title: Full compensation method of thermal error of NC machine tool based on sequence depth learning

Authors: MingXun Zhu; Zhigang Meng

Addresses: School of Economics and Management, Changsha Normal University, Hunan Changsha 410003, China ' School of Computer Engineering and Applied Mathematics, Changsha University, Hunan Changsha 410003, China

Abstract: In this paper, a full compensation method of thermal error of NC machine tools based on sequence depth learning is proposed. Firstly, the main structure of NC machine tool is analysed and the principle of thermal error compensation is determined. Then, multiple temperature test points are set at the machine body and main shaft to collect thermal error data. Secondly, the correlation degree of thermal error difference sequence is determined according to the running space state. Finally, the thermal error full compensation model is constructed with the help of RNN network algorithm in sequence depth learning, and the model is modified with the help of Lagrange function to realise full compensation. The results show that the total compensation error of thermal error of NC machine tool is less than 2%.

Keywords: sequence depth learning; compensation effect; numerical control machine; thermal error; full compensation.

DOI: 10.1504/IJMTM.2023.131301

International Journal of Manufacturing Technology and Management, 2023 Vol.37 No.2, pp.138 - 150

Received: 18 Sep 2021
Received in revised form: 27 Oct 2021
Accepted: 12 Apr 2022
Published online: 06 Jun 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article