Abstract
A nonlinear mathematical model for the dynamics of permanent magnet synchronous machines with interior magnets is discussed. The model of the current dynamics captures saturation and dependency on the rotor angle. Based on the model, a flatness-based field-oriented closed-loop controller and a feed-forward compensation of torque ripples are derived. Effectiveness and robustness of the proposed algorithms are demonstrated by simulation results.
Highlights
As modern three-phase current-fed machines are increasingly used for industrial applications, efficient and accurate current control of such machines is of great economic interest. This is due to the fact that the current controller influences the operational behaviour of the machine to a very large extent, namely the energy efficiency and the occurrence of undesirable side-effects, e.g., torque ripples
Due to increasingly involved designs of the rotor [3,4] and the stator [5], and due to higher power density [6], effects such as magnetic saturation of the iron material, cross-coupling [7], and the dependency of the system dynamics on the rotor position become more important for high-precision control design [8]
In order to approximate the current dynamics of Equation (6) as well, surrogate models for the partial derivatives ψdd, ψdq, ψqd, and ψqq of the flux linkages are required in addition to Equation (13)
Summary
For PMSMs, the consideration of the rotor position in modelling and control design is useful for compensating ripples on the torque and the current, see e.g., [8]. The model is derived in rotor-fixed coordinates, and a feed-forward control scheme based on look-up tables is proposed to compensate ripples on the current and the torque. The proposed machine model is exploited in order to derive a flatness-based field-oriented control (FOC) scheme (Section 3.1) as well as a feed-forward compensation of torque ripples (Section 3.2). The proposed scheme for current control and compensation of torque ripples explicitly decouples the dq-currents
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have