Research On Parameter Identification Of Surface Mounted Permanent Magnet Synchronous Motor

Permanent magnet synchronous motor (PMSM) servo system has a lot ofapplications in many areas of modern society. However, with the expanding anddeepening applications, scholars have higher requirements on the servo system.It must be have better tracking accuracy and more favorable robustness. For thisreason, many experts and scholars have conducted a lot of research onalgorithms and control strategies. However, mostof current methods requireprecise motor parameters as a precondition, including electrical parameters andmechanical parameters. Therefore, parameter identification has becomemeaningful and important work.For a surface mounted AC servo motor, a Matlab simulation platform hasbeen made based on the mathematical model of PMSM andvector controlscheme. Not only the simulation of double-loop control system has been done,but also the analysis of dynamic performance.Determine the following identifying program. The speed of motor and thestator resistance are known, the stator inductance and the flux should beidentified. The model reference adaptive algorithm and Popov ultra stabilitytheoryis the main theoretical basis. And according to the formula of adjustablemodel, a detailed analysis offactors which affect the dynamic performance ofidentification has been made. Based on the analysis, the dynamic performanceof identification process (including convergence speed and accuracy) has beenoptimized. Simulation results show the effectiveness of the algorithms.The changes of inertia will result a large impact on the performance ofsystem. Use FFRLS method and Landau algorithm to identify inertia. Thesimulation results show that both methods can accurately identify theinertia.The convergence time and convergence precision meet the requirements.Load disturbance torque, another important factor affecting the dynamicperformance of the system, has been studued in this paper. By design ing areduced-order disturbance observer, disturbance torque has been identifiedaccurately. Simulation results show the effectiveness of the reduced-orderobserver. In addition, a sudden increase or decrease of loads will cause thechange of instantaneous speed, By studying the principle of compensation ofinstantaneous speed, the load disturbance torque which served as feedforward isadded to the output of the speed regulator. By this compensation measures,anti-disturbance capacity of the system has been improved.