Recurrent fuzzy neural network controller for linear ultrasonic motor

The standing-wave linear ultrasonic motor (LUSM) has the advantages that are small size, and without noise and electromagnetic wave effect. The speed and position of the conventional LUSM can be manipulated by controlling the frequency, two phases difference and the voltage amplitude of the sinusoidal voltage waveforms. But the driver is affected easily by quality factor such that the two-phase voltages would be imbalance. Therefore, a good dynamic performance of the conventional LUSM is difficult to be obtained due to the unbalanced two-phase voltages. Because the standing-wave LUSM could be supplied a single-phase sinusoidal wave, there is an outstanding performance in practical application without the demerit of the unbalanced two-phase voltages. The driving circuit for the standing-wave LUSM combines the two control circuits to regulate the frequency and the amplitude of the driving voltage. Because the dynamic characteristics of the LUSM are nonlinear and the precise dynamic model is difficult to obtain, a recurrent fuzzy neural network (RFNN) controller is proposed to control the position of the moving table of the LUSM to achieve precision position control with robustness.