Control of a switched reluctance motor

Switched Reluctance Motors (SRMs) are getting increasing attention in variablespeed applications. The SRMs dynamic characteristics vary as conditionchange. The most important drawback of SRMs is ripples in the torque at lowspeed. The torque in the SRM is a function of flux-linkage current and rotorposition. Therefore, modeling of the flux-linkage characteristics of the SRM isvery important to minimize tork ripples. So far, several methods of modeling ofthe flux-linkage characteristics of the SRMs have been proposed. Due tononlinear flux-linkage characteristics of the SRM, this cannot be adequatelydescribed by these models. In this thesis, simulations results show that anArtificial Neural Network (ANN) based solution minimizes torque ripple in aSwitched Reluctance Motor (SRM).Science Code : 704Key Words :Switched reluctance motor, Artificial neural network,Nonlinear modelling.Page number :95Adviser :Prof. Dr. Çetin ELMAS