Comparison between various approaches without weighting factors with conventional MPTC for linear induction motors

Abstract

Model predictive control (MPC) is widely recognized as an effective control strategy for linear induction motor (LIM) drives due to its direct switching capability, multivariable regulation, and online optimization. However, conventional MPC (MPC-Conv.) requires carefully tuned weighting factors to balance thrust and primary flux, resulting in increased design complexity and significant tuning effort. Although a normalized average ranking MPC (MPC-NAR) has been proposed to eliminate weighting factors, it fails to guarantee stable operation across the full speed range. This paper proposes two MPC strategies for LIM drives that completely avoid weighting factors by employing a single cost function combined with hard constraints. The first strategy enforces a hard flux boundary (MPC-Flux Constr.), while the second applies a hard thrust boundary (MPC-Thrust Constr.). These formulations significantly reduce cost-function complexity while maintaining robust control performance. Comprehensive simulation results obtained from a 3-kW arc induction motor demonstrate that the flux-based MPC with a hard thrust constraint consistently outperforms existing approaches, achieving faster speed regulation, lower thrust ripple, stable full-speed operation, and reduced computational burden under diverse operating conditions.