A Multi-parameter Estimation Method of PMSM based on Model Decoupling and State Tracking Considering Machine Nonlinearity

Permanent magnet synchronous motor(PMSM) has been widely used in the industrial field due to the high torque density and high power density. Since motor control relies on accurate electrical parameter information, efficient parameter estimation is of great significance for high-performance vector control. A novel multi-parameter tracking identification method based on decoupling strategy is proposed to estimate both stable and variable parameters, including resistance, permanent magnet flux and d-q axis inductance. The motor parameters are decoupled based on the relationship between power and torque to reduce the computational complexity and cross influence at first, then the stable parameters are estimated according to the steady-state measurement, and finally the inductance of d-q axis are tracked and estimated. This decoupling-based parameter tracking estimation method separates the current-affected inductance from the stable parameters, reduces the estimation error and improves the accuracy of the inductance model. The results show that the tracking estimation efficiency of the proposed method is significantly improved with experiments and comparisons on the PMSM driven system.