Predictive controller considering electrical constraints: a case example for five-phase induction machines

Abstract

The modern control of power drives involves the consideration of electrical constraints in the regulator strategy, including voltage/current limits imposed by the power converter and the electrical machine, or magnetic saturation due to the iron core. This issue has been extensively analysed in conventional three-phase drives but rarely studied in multiphase ones, despite the current interest of the multiphase technology in high-power density, wide speed range or fault-tolerant applications. In this paper, a generalised controller using model-based predictive control techniques is introduced. The proposal is based on two cascaded predictive stages. First, a continuous stage generates the optimal stator current reference complying with the electrical limits of the drive to exploit its maximum performance characteristic. Then, a finite-control-set predictive controller regulates the stator current and generates the switching state in the power converter. A five-phase induction machine with concentrated windings is used as modern high-performance drive case example. This is a common multiphase drive that can be considered as a system with two frequency-domain control subspaces, where fundamental and third harmonic currents are orthogonal components involved in the torque production. Experimental results are provided to analyse the proposed controller, where optimal reference currents are generated and steady/transient states are studied.