Adaptive Step-Size Tuning Combined with Ternary Search Optimization for FCS-MPC Sensorless Speed Estimation of PMSMs

To address the slow transient convergence, low steady-state estimation accuracy and heavy computational complexity in finite control set model predictive control (FCS-MPC)-based sensorless speed estimation for permanent magnet synchronous motors (PMSMs). This paper proposes an adaptive step-size tuning strategy combined with ternary search optimization. Firstly, a cost function based on the flux linkage error is constructed, and the prediction step size of the electrical angle is adaptively adjusted according to the normalized flux linkage error. Specifically, a larger step size is adopted under transient conditions to accelerate convergence, whereas a smaller step size is adopted under steady-state operation to improve estimation precision. Secondly, a ternary search strategy is introduced to optimize the electrical-angle search process, reducing the number of candidate evaluations and the computational complexity. Finally, the effectiveness of the proposed method is validated through MATLAB/Simulink simulations and experiments on a PMSM test bench. Comparative results demonstrate that, under various operating conditions, the proposed method achieves faster transient convergence, higher steady-state estimation accuracy, and lower computational complexity than the conventional fixed step-size FCS-MPC speed estimation method.