A nonlinear full model of switched reluctance motor with artificial neural network

This paper presents a novel nonlinear full model developed by using artificial neural networks (ANNs) for switched reluctance motors (SRMs). The proposed ANN based nonlinear model consists of two different models, namely forward and inverse model. The purpose of the forward model is to estimate the flux linkage and torque of the SRM as a function of stator current and rotor position. And, the purpose of the inverse model is to estimate stator current and flux linkage of the SRM as a function of torque and rotor position. Also conversions can be achieved between torque, stator current and flux linkage with these models. Computational load of the processor has been considered and minimized to use the developed model in real industrial applications. The experimental tests are realized to verify the accuracy and feasibility of the proposed model.