Neural network based modeling of traveling wave ultrasonic motor using genetic algorithm

Ultrasonic motors (USM) have heavy nonlinear and load dependent characteristics such as dead-zone and saturation reverse effects, which vary with driving conditions. These features make serious problem in modeling of this kind of motors. In this paper, Modeling of Traveling Wave type Ultrasonic Motor (TWUSM) by means of genetic algorithm & neural network-based Hammerstein model is presented, in which the nonlinear static part is approximated by a genetic algorithm (GA) based radial basis function neural network (RBFNN) and the linear dynamic part is modeled by a transfer function model. GA is also adopted to optimize the hidden centers, the radial basis function widths and the weights of the RBFNN. The simulation results verify significantly improved matching between measured and simulated data.