Improved Observer-Free Deadbeat Control of SPMSM for Dynamic Performance Optimization

To alleviate motor parameter dependence of deadbeat predictive current control (DBPCC), model-free method based on ultra-local model has attracted much attention. Unlike complex strategies utilizing observer or data-driven methods which rely on process resources in model-free control to estimate disturbance and identify parameters, an improved model-free and observer-free (MF-OF) deadbeat control is proposed to simplify the algorithm and improve dynamic performance. Traditional observer-free method adopts two-period information to extract inductance and does not consider EMF item, leading to slower dynamic response. To solve this problem, the proposed MF-OF method includes ultra-local model fully utilizing acknowledged EMF item, to relieve disturbance estimation stress. Meanwhile, the inductance extract algorithm in the proposed observer-free method only needs the last-period information as converging d-axis disturbance to zero, which leads to high-precision inductance estimation. Moreover, different from traditional MF-OF whose characteristic function pole influenced by rotor speed, proposed MF-OF eliminate this item when inductance is precise, so as enhancing dynamic response. Finally, on a 1 kW surface-mounted permanent magnet synchronous motor (SPMSM) experimental testbed, comparative studies prove the proposed MF-OF method has better current response compared with conventional MF-OF and improved model-free extended state observer (MF-ESO) method.