Predictive torque control has gained prominence in motor drive applications due to its intuitive response and ability to simultaneously manage multiple control objectives. However, traditional finite control set implementations face significant practical barriers: excessive computational demands, sensitivity to motor parameter variations, and tedious tuning of weighting factors within cost functions. These challenges intensify when systems employ multi-level inverters or unconventional topologies like open winding configurations.
The proposed research addresses these limitations by introducing a streamlined FCSPTC method specifically designed for open winding permanent magnet synchronous motor (OWPMSM) drives powered by two dual two-level voltage source inverters. The key innovation lies in eliminating the need for flux weighting factors through a voltage-based cost function, reducing the control algorithm to evaluate only five prediction voltage vectors from a potential set of 19 possibilities.
This simplification delivers three primary benefits. First, the control structure becomes more computationally efficient with fewer operational steps and decision points—advantageous for real-time embedded systems. Second, robustness improves by depending solely on stator resistance identification, minimizing sensitivity to uncertain motor parameters. Third, switching frequency naturally decreases through intelligent selection of optimal switching states from redundant combinations, independent of frequency weighting factors.
Experimental validation demonstrates that the proposed technique maintains optimal performance across flux, torque, and current regulation while reducing computational load and switching losses. The method eliminates the lengthy parameter tuning process that traditionally slows controller deployment, making it more practical for industrial applications.
Open winding motor configurations offer advantages in fault tolerance and thermal distribution but present control challenges due to increased complexity. This weighting-factor-less approach represents meaningful progress toward simplifying advanced predictive control methods for specialized motor topologies, potentially accelerating adoption in variable-speed drive systems across industrial, automotive, and renewable energy sectors.
