Neural adaptive control of dynamic sandwich systems with hysteresis

An adaptive control strategy is presented for dynamic sandwich systems with hysteresis. The so-called sandwich system with hysteresis is the system that can be represented by a cascade of a dynamic smooth nonlinear (DSNL1), a hysteresis nonlinear (HNL) and another dynamic smooth nonlinear (DSNL2) subsystem. In this control strategy, a neural network based inverse model is constructed to compensate for the effect of the first dynamic block (i.e. DSNL1) of the sandwich system. Thus, the sandwich system can be transformed into a dynamic nonlinear subsystem preceded by hysteresis. Then a novel hysteretic operator is proposed to transform the multi-valued mapping of hysteresis into a one-to-one mapping. Base on the proposed hysteretic operator, a neural adaptive controller is developed for the modified system. One of the advantages of the controller is that it does not need to construct the inverse model of hysteresis to cancel the hysteretic effect