Study on the Neural Network SMVS Control in Cam Grinding

Xiu-mei CHEN, Bao-ying PENG, Qi-guang LI, Huai-jian XIA

Abstract


The contour of cam is created by X-C linkage motions. It is important that the X axis with grinding wheel frame accurately tracks the motion of the C axis. The coordination motion of the two axes is of great importance to the accuracy of the cam profile. The sliding mode variable structure (SMVS) control method can be applied to the servo axis position tracking control in the cam contour grinding. The SMVS control method has better performance than the PID control method both in control precision and disturbance resistance. But the chattering under SMVS control has a bad influence on the servo tracking performance. To solve this problem, the SMVS control based on neural network is applied to the motion of each servo axis in cam grinding. The displacement tracking of each servo axis is simulated and analyzed with Matlab/Simulink software. The error curve under neural network SMVS control is compared with the curve of the SMVC control. The simulation results show that the neural network SMVS control can inhibit the chattering effectively, and provide a new idea for improving the precision in cam grinding.

Keywords


X-C linkage motion, Neural network, Sliding mode variable structure (SMVS), Servo system, Tracking performance


DOI
10.12783/dtetr/amme2017/19505

Full Text:

PDF

Refbacks

  • There are currently no refbacks.