The objective of this paper is to design a high performance torque controller for a fifth order model of a three phase switched reluctance motor (SRM) which includes both electrical and mechanical dynamics. The same model was considered in where only position control was achieved. Industrial requirements (speed and torque control) were not necessarily fulfilled though. By controlling the instantaneous torque of SRM's, both speed and position control can be achieved and the fastest possible response as well as a reduction of torque ripple obtained.
The proposed nonlinear control scheme is based on the recently developed adaptive nonlinear techniques which combine the geometric concepts of input-output linearization with the Lyapunov design of adaptive linear control. Once the SRM model is linearized, a multivariable decoupling linear controller for torque and unselected currents stabilization is designed. The parameters uncertainty is compensated by the adaptive scheme.
The problem of accurately measuring the SRM torque is also addressed and a method which deduces the torque through flux observation is used. Simulation results are given to demonstrate the effectiveness of the control method.