The Nonlinear Position Control Of USM Based On Its Hammerstein Model

Ultrasonic motor has obvious nonlinearity and time-varying characteristic, because of its special structure and operating mechanism. So that the control strategy of ultrasonic motor cannot be designed according to traditional electromagnetic motor. It spurs people continually to de research for proper control methods, continually to improve USM’s control effect.The usual control variables of USM are frequency, phase difference and the amplitude of voltage. Considering to the actual situation, the frequency of the driving voltage is used as the control variable and the position is the output in this paper to de research on the modelling and designing for USM’s control strategy.Considering to USM’s obvious nonlinearity, proper nonlinear model, directly used to establish the model of USM, can get a simple and effective model of motor. Intelligent Optimizations always have complex process of optimization, and generally need long time to do the optimization. Thus, the off-line optimization method is used to identify the model of USM. In order to satisfy the need of position control of USM, the Bacterial Foraging Optimization algorithm is used to identify the parameters and orders of the model. Then USM’s nonlinear Hammerstein model is established, in which the frequency is the input variable, the position of motor rotor is the output variable. The validity of the model has been verified by comparing the calculated data of the model established with the experimental data.On the basis of USM’s nonlinear Hammerstein model to research on the nonlinear position control strategy of USM. The inversion of the nonlinear part is done, as part of the controller, to compensate the nonlinear part of motor’s characteristic. Then design the direct generalized predictive control strategy for the remained linear part of motor to realize effective position control of USM. It can fully verify the efficiency of this control strategy by experimental results.