Research On On-line Parameter Identification Of Permanent Magnet Synchronous Motor Considering The Variation Of Operating Condition

Permanent magnet synchronous motor(PMSM)has been widely used because of its good dynamic performance and high power density.Due to temperature change and magnetic saturation effect,the parameters of PMSM will inevitably change.Real-time access to accurate electromagnetic parameters of the motor,high-performance vector control and fault diagnosis are important help.In this paper,the on-line parameter identification of permanent magnet synchronous motor(PMSM)is studied.Firstly,derived the mathematical model of permanent magnet synchronous motor,and the parameter identification model is established based on the least square method.The parameter identification method based on dynamic model and continuous disturbance injection,and the parameter identification method based on steady-state model and current pulse injection were proposed to realize the online parameter identification.Studied the parameter identification of motor under different working conditions.The simulation analysis shows that the algorithm based on the steady-state model can only run under the stable condition because of errors when the working condition changes.The algorithm based on dynamic model can identify correctly when the working condition changes,but cannot converge when the working condition is stable.In order to cope with the change of working conditions,a parameter identification method based on the steady-state model and the weighted least square method was proposed.By adjusting the weight online,the algorithm could only run when the system was running stably.A parameter identification method based on dynamic model and weighted least square method is proposed,which only runs in the transient process of system.Finally,rules are designed to make the above steady-state and dynamic algorithms switch with the operating conditions,which can be used for parameter identification under any operating conditions.After derivation and simulation,it is concluded that the dead-zone effect of inverter will cause the given voltage to be not equal to the actual voltage,which will affect the accuracy of parameter identification.A dead-zone identification and compensation method based on direct axis voltage was proposed.The compensation voltage Δ V was adjusted in real time through direct axis characteristic harmonic analysis and closed-loop control to eliminate the dead-zone effect.Finally,an experimental platform is set up and the above theories are verified.Experiments show that the proposed dead-zone compensation method can effectively eliminate voltage and current harmonics,and the proposed parameter identification method can identify the resistance,inductance and flux flux more accurately.