Title :
Polynomial neural network based modeling of Switched Reluctance Motors
Author :
Vejian, R. ; Gobbi, R. ; Sahoo, N.C.
Author_Institution :
Metronic Eng. Sdn Bhd, Shah Alam
Abstract :
Switched reluctance motor (SRM) has double salient structure which makes its magnetic characteristics; i.e. flux linkage and torque to be a nonlinear function of stator current and rotor position. For this reason, modeling and control of the SRM is by no means a trivial task. It was proven by many researchers in this area, that a simple mathematical model has never able to represent the complete overall magnetic characteristics. Moreover, there is no distinct guideline about what sort of mathematical model would be suitable. To overcome this modeling problem, a self-organizing polynomial neural network is projected in this paper. With this scheme incorporated, the model is let to evolve iteratively and progressively without any prior knowledge of the plant. Subsequently, MATLAB/SIMULINK is used to model the SRM drive system. Finally, experimental results for both static and dynamic conditions are presented.
Keywords :
electric machine analysis computing; neural nets; polynomials; reluctance motors; SRM; double salient structure; flux linkage; magnetic characteristics; polynomial neural network based modeling; rotor position; self-organizing polynomial neural network; stator current; switched reluctance motors; Couplings; Magnetic flux; Magnetic switching; Mathematical model; Neural networks; Polynomials; Reluctance machines; Reluctance motors; Stators; Torque; Modeling; Switched Reluctance Motor (SRM); flux linkage; polynomial neural networks (PNN); torque;
Conference_Titel :
Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century, 2008 IEEE
Conference_Location :
Pittsburgh, PA
Print_ISBN :
978-1-4244-1905-0
Electronic_ISBN :
1932-5517
DOI :
10.1109/PES.2008.4596075