LAUSR

Abstract This paper presents a quantitative comparison of state of the art model predictive direct switching control (MPDSC) methods for electrical drives. In MPDSC the switching states of the inverter are directly computed via model predictive control (MPC). This eliminates the need for modulators and presents an attractive alternative to classical field oriented control (FOC) approaches. Three classes of MPDSC methods are compared to field oriented model predictive control (FOMPC). The investigated MPDSC approaches are: hysteresis-based MPDSC, finite control set MPC (FCS-MPC) and model predictive pulse pattern control (MP 3 C). The comparison is based on transient and stationary simulations of a permanent magnet synchronous machine (PSM) driven by a voltage-fed two-level inverter, representative for high-performance automotive applications.