The problem of cogging torque is due to a magnetic behavior, intrinsic to synchronous machines and due to the presence of permanent magnets themselves. Cogging torque is a significant problem when the servo drive is used for applications where high precision in terms of position control is required. In this paper we present a method of cogging torque reduction by means of a control technique based on mathematical modeling of the cogging phenomenon itself in order to exploit this knowledge directly in the controller design. The mathematical model is inserted in the dynamic model of the synchronous machine in order to exploit the feedback linearization, providing an expression of the control law in which the contribution of the deterministic knowledge of the phenomenon is directly present. The cogging phenomenon physically depends on the angular position of the rotor, as well as the deterministic model we use to define the control vector. This makes it interesting and innovative to determine whether the control algorithm can be inserted within a sensor-less architecture, where rotor position and angular velocity measurements are not available. For this purpose, we present the use of an extended Kalman filter (EKF) in the continuous-time domain, discussing the advantages of an observer design based on a dynamic motor model in three-phase and direct-square axes. Results are presented through very accurate simulation for a trajectory-tracking problem, completing with variational analysis in terms of variation of initial conditions between EKF and motor dynamics, and in terms of parametric variation to verify the robustness of the proposed algorithm. Moreover, a computational analysis based on Simulink Profiler is proposed, which provides some indication for possible implementation on an embedded platform.
This is an open access article distributed under the Creative Commons Attribution License
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited