Synchronous Reluctance Motor Drive with Advanced Control: Comparative Study of PI, Predictive Model-Based Control, Model-Free Predictive Control and INDI Control

This thesis compares control strategies for synchronous reluctance motors (SynRMs), a machine technology of growing industrial relevance. Being free of rare-earth materials, SynRMs avoid the supply-chain constraints and the environmental and economic burden associated with rare-earth extraction, which is confined to a limited number of sites worldwide. These advantages, however, come at the cost of a more demanding control problem: the absence of permanent-magnet excitation, together with pronounced magnetic saturation and cross-coupling between the axes, makes the machine model strongly nonlinear and parameter-dependent over the operating range. Against this background, the thesis examines the proportional-integral (PI) regulator, still the reference solution in industrial drives, alongside two strategies of current research interest: predictive control and incremental nonlinear dynamic inversion (INDI). The latter, originally developed for fixed-wing flight control, constitutes the main contribution of this work. Its incremental formulation relies on measured or estimated derivatives of the state rather than on a complete model of the plant, which limits the sensitivity to parameter mismatch that typically affects model-based schemes applied to SynRM drives. All strategies are formulated in the dq reference frame and first assessed in simulation under comparable operating conditions. The experimental implementation of the PI and INDI controllers is then described in detail, from the discrete-time formulation to the tuning procedure adopted on the test bench, providing a direct comparison between the state of the art and the proposed approach. The measured responses confirm the advantages of INDI in terms of tuning effort and dynamic performance, consistently with the preliminary simulation study.