The Krasnosel'skii-Pokrovkii (KP) is used to build the rate-independent hysteresis model for the Giant Magnetostrictive Actuator (GMA). Firstly, an off-line method of identifying discretization density function for KP hysteresis model is proposed. Secondly, in order to combine the hysteresis model with adaptive control theory, neural network is adopted to realize adaptive on-line identification of the density function of the KP hysteresis model. In the BP algorithm of the neural network, the discretization KP operator is used as operational function of the neural network and the density functi...
