Title :
Neuro sliding mode control of timing belt servo-system
Author :
Yildiz, Yildiray ; Sabanovic, Asif
Author_Institution :
Fac. of Eng. & Natural Sci., Sabanci Univ., Istanbul, Turkey
Abstract :
In this paper, the employment of neural networks with sliding mode control in the control of a linear drive with flexible transmission element is described. Linear drives with flexible transmission elements are cheaper and also more efficient than the ones with rigid transmissions like power screw systems. Hence, these devices play an important role in the industry. A neuro-sliding mode controller cascaded with a discrete sliding mode controller is used to control the system. Neuro-sliding mode controller is used in the outer loop and produces reference for the discrete sliding mode controller which serves as a force controller, in the inner loop. The control signal of the neuro-controller is obtained by minimizing an error function which is derived from Lyapunov stability analysis. The controller performance is tested with different loading conditions and different friction torques and the results are presented.
Keywords :
Lyapunov methods; control system synthesis; discrete systems; force control; linear motors; neurocontrollers; position control; servomechanisms; variable structure systems; Lyapunov stability analysis; control system synthesis; controller performance test; discrete sliding mode controller; error function minimization; flexible transmission elements; force controller; linear drives; neural networks; neuro sliding mode controller; position control; power screw systems; timing belt servo system; Belts; Control systems; Electrical equipment industry; Employment; Error correction; Fasteners; Force control; Neural networks; Sliding mode control; Timing;
Conference_Titel :
Advanced Motion Control, 2004. AMC '04. The 8th IEEE International Workshop on
Print_ISBN :
0-7803-8300-1
DOI :
10.1109/AMC.2004.1297659