Self-learning control applied to vibration control of a rotating spindle by piezopusher bearings

Abstract

Abstract

Vibration suppression of rotating machinery is an important engineering problem. For several years, researchers have demonstrated that this vibration could be greatly reduced for machines by active control using active actuators. In this paper, active control that uses innovative active piezoceramic pusher bearings and self-learning control for the rotor system with varying initial state errors is presented and applied. To deal with the problems of initial state errors at each iteration, the D-type iterative learning control is combined with feedback compensation. The rule library of self-learning control according to all criteria, for example operating speed, is set up by hybrid control. The algorithms are coded in Visual Basic. Their performance is examined experimentally on a laboratory rotor rig. The results clearly demonstrate the more effective vibration suppression of unknown synchronous, subsynchronous and other transient and stochastic disturbances that can be achieved using innovative active piezoceramic pusher bearings with self-learning control.