Title :
Online Modeling for Switched Reluctance Motors Using B-Spline Neural Networks
Author :
Lin, Zhengyu ; Reay, Donald S. ; Williams, Barry W. ; He, Xiangning
Author_Institution :
Control Tech. PLC, Powys
Abstract :
A novel online-modeling scheme for the switched reluctance motor (SRM) using a B-spline neural network (BSNN) is proposed in this paper. A 2-D BSNN is designed to learn the nonlinear-flux-linkage characteristic of an SRM online and in real-time. Torque, incremental inductance, and back-emf estimates can be derived from the BSNN after training. The scheme does not require a priori knowledge of the SRM electromagnetic characteristics. Simulation and experimental results show that the scheme has a good estimation performance and robustness at low to medium motor speed.
Keywords :
neural nets; nonlinear control systems; reluctance motors; torque; B-spline neural networks; SRM online; back-emf estimates; incremental inductance; motor speed; nonlinear flux linkage; switched reluctance motors; torque; Couplings; Helium; Inductance; Neural networks; Reluctance machines; Reluctance motors; Robustness; Spline; Torque; Voltage; Modeling; neural networks; nonlinear estimation; reluctance motors;
Journal_Title :
Industrial Electronics, IEEE Transactions on
DOI :
10.1109/TIE.2007.904009
