High accuracy motion control of linear motor drive wire-EDM machines

• Published in: International journal of advanced manufacturing technology Vol. 40; no. 9-10; pp. 918 - 928
• Main Authors: Yan, Mu-Tian, Cheng, Tun-Hua
• Format: Journal Article
• Language: English
• Published: London Springer-Verlag 01.02.2009; Springer Nature B.V
• Subjects: Accuracy; CAE) and Design; Cogging; Computer-Aided Engineering (CAD; Contouring; Control theory; Controllers; Dynamic models; Electric motors; Engineering; Feedforward control; Industrial and Production Engineering; Machine shops; Machining; Mechanical Engineering; Media Management; Micromachining; Motion control; Neural networks; Original Article; Parameter robustness; Permanent magnets; Proportional integral derivative; Robust control; System identification; Wire; Linear motor drive; Neural network-based feedforward controller; Motion control; Wire-EDM
• ISSN: 0268-3768; 1433-3015
• DOI: 10.1007/s00170-008-1398-4

This paper aims to develop and investigate a permanent magnet linear-motor-driven table system for a wire-EDM machine. A dynamic model and system identification of `the linear motor system have been derived and analyzed. The linear motor drive system has nonlinear and time-varying behaviors because of the effect of irregular friction of the sliding surface and cogging force. Therefore, a conventional digital controller may not suffice to provide a high contouring accuracy as well as adequate disturbance rejection and parameter variation robustness. An indirect adaptive controller (IAC), combined with a neural network-based feedforward controller (NNBFC) is proposed to improve the contouring performance of the linear motor system. Experimental results not only indicate that the proposed control scheme can achieve a high contouring accuracy of ± 0.3 μm, they also demonstrate that the developed linear motor drive wire-EDM machine can meet the requirement for micro machining. The maximum contour error of circular trajectory at a feed-rate of 0.1 mm/min was significantly reduced from 8 μm to 0.5 μm in comparison with proportional-integral-derivative (PID) controller.