Search

This work investigates different control strategies for the induction motor (IM) drives in the electrical car application filed. Three control strategies, namely, a sequential direct model predictive control (SDMPC), a field oriented control (FOC) and a nonlinear hysteresis based controller (bang-bang control), were investigated and compared in this work. Following conclusions are drawn: the newly proposed SDMPC and the hysteresis controller will outperform the classical FOC control strategy in terms of the control dynamics; while during steady state, FOC achieves the best performance via using a modulator; In comparison with the conventional hysteresis controller, the SDMPC provides good flexibility via cost functions, and less tuning efforts. Thanks to its cascaded structure, no weighting factors are required in comparison with the classical MPC. At the end simulations are conducted, which confirm the above conclusion.

This paper presents a simple and effective cascaded predictive control (CPC) strategy for three-level NPC (3L-NPC) power converter fed induction machine (IM) drives. Scalar weighting factors, which are generally used in classic direct model predictive control (CDMPC), are difficult to select. Studies over weighting factor elimination is therefore of considerable significance. The proposed method in this work uses cascaded cost functions, which represent control objectives, without requiring any penalty coefficients, thereby reducing the tedious weighting factor tuning process. Meanwhile, the proposed strategy maintains the merits of CDMPC with good control dynamics and shows robustness for its control coefficients. Lastly, the validity and effectiveness of the proposed CPC are proved by simulation data.