Simulation an‎d Improvement of Model-Free Predictive Current Control for Permanent Magnet Machine

3/16/2026 12:00:00 AM

To address the model dependency issue in model-based predictive current control, a model-free predictive current control approach has been proposed. This method demonstrates excellent performance in the presence of parameter dependency an‎d mismatches between controller parameters an‎d system parameters. However, the model-free predictive control based on the ultra-local model still depends on its input gain coefficient. An improper selec‎tion of this input gain can introduce errors into the system input an‎d consequently affect overall system performance. To overcome this limitation, a parameter estimation method is employed in this study to estimate the input gain coefficient online an‎d in real time, taking into account variations in the operating point an‎d motor operating conditions. The coefficient is determined based on instantaneous current, voltage, an‎d speed measurements. This feature enables the proposed algorithm to completely eliminate dependency on the system model. Furthermore, the system equations an‎d modeling are first examined in detail, an‎d subsequently, the proposed algorithm is validated through simulation an‎d implementation on a real experimental setup.

تخمين پارامتر , موتور سنكرون مغناطيس دائم , كنترل پيشبين جريان , كنترل پيشبين جريان مستقل از مدل , مشاهده گر حالت سطح

Permanent Magnet Synchronous Motor , Predictive Current Control , Model-Free Predictive Current Control , Extended State Observer , Parameter Estimation

Amirhossein Fallah Khorshidi

Dr.Davood Arab Khaburi