Recursive Wavelet Elman neural network for a synchronous reluctance motor

Author

Chao-Ting Chu ; Huann-Keng Chiang ; Tzu-Chieh Lin ; Chih-Ti Kung

Author_Institution

Grad. Sch. of Eng. Sci. & Technol., Nat. Yunlin Univ. of Sci. & Technol., Yunlin, Taiwan

fYear

2014

fDate

7-10 May 2014

Firstpage

1

Lastpage

4

Abstract

This paper proposed recursive Wavelet Elman neural network (RWENN) speed control for synchronous reluctance motor (SynRM). Wavelet neural network (WNN) activation function is replaced by wavelet functions which WNN combines wavelet transform with time domain, analytical capabilities and scale neural network. This paper proposed RWENN that has satisfactory control nonlinear problem in SynRM. We used the discrete Lyapunov theory to ensure network converges. Finally, the experimental results validated RWENN has satisfactory performance in SynRM.

Keywords

Lyapunov methods; angular velocity control; control nonlinearities; machine control; neurocontrollers; reluctance motors; time-domain analysis; wavelet neural nets; wavelet transforms; RWENN speed control; SynRM; WNN activation function; analytical capability; control nonlinear problem; discrete Lyapunov theory; network converges; recursive wavelet Elman neural network; scale neural network; synchronous reluctance motor; time domain; wavelet neural network activation function; Convergence; Equations; Mathematical model; Neural networks; Reluctance motors; Wavelet transforms; Elman neural network; Wavelet function; error back-propagation; recursive; synchronous reluctance motor;

fLanguage

English

Publisher

ieee

Conference_Titel

Next-Generation Electronics (ISNE), 2014 International Symposium on

Conference_Location

Kwei-Shan

Type

conf

DOI

10.1109/ISNE.2014.6839343

Filename

6839343