Date of Award


Publication Type

Master Thesis

Degree Name



Electrical and Computer Engineering

First Advisor

Kar, Narayan (Electrical and Computer Engineering)


Engineering, Automotive.



Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.


Considering the extensive non-linearities in the switched reluctance motor (SRM) drive, variation in the DC bus voltage and specific requirements of the hybrid electric vehicles (HEVs) traction application, a feed-forward back propagation neural network (BPNN) based torque controller is proposed. By using proposed controller, the torque ripple has been effectively reduced at low speeds while the power efficiency has been optimized at high speeds range. The problem of multi-valuedness related with the neural network based direct inverse control has been targeted by designing a bank of two-hidden-layer neural network controllers. And the problem of torque oscillation due to the change of control mode and step change of firing angle has been solved by using dead-band filtering and nearly continuous changing of firing angle and phase currents. Computed results are presented to demonstrate the effectiveness of the proposed control scheme.