Title :
Modeling hysteresis using hybrid method of improved describing function and neural networks
Author :
Zhao, Tong ; Zhai, Yawei
Author_Institution :
Sch. of Autom. & Electr. Eng., Qingdao Univ. of Sci. & Technol., Qingdao, China
Abstract :
A. new method of establishing hysteresis model using describing function of backlash and neural networks is proposed in this paper. With the describing function of backlash N(A), a mathematical expression is written, which combined with sinusoidal function working as input function can well establish major hysteresis loop. Finally, we established the modeling of neural network based on the new method. The established modeling can well approximate a given real data. The simulation result proves the validity and feasibility of the scheme.
Keywords :
hysteresis; neurocontrollers; nonlinear control systems; backlash describing function; hybrid method; hysteresis loop; hysteresis model; mathematical expression; neural networks; sinusoidal function; Artificial neural networks; Biological neural networks; Data models; Hysteresis; Magnetic hysteresis; Mathematical model; Backlash; Describing Function; Hysteresis Loop; Neural Network;
Conference_Titel :
Electrical and Control Engineering (ICECE), 2011 International Conference on
Conference_Location :
Yichang
Print_ISBN :
978-1-4244-8162-0
DOI :
10.1109/ICECENG.2011.6057903