Hysteresis modeling based on the neural network identification

A novel model based on the physical properties is proposed for hysteresis modeling. The model consists of two parts. One is the extend value and the other is the shorten value. The output of the model is determined by the two effects. Combining the chaos optimization learning rule, the model can be identified by the neural networks. The modeling experiment of the piezoelectric ceramic is realized by the model. The simulation results prove that the error of the modeling can be reduced, the precision can be improved, and the effect is better than conventional ones.