| Permanent magnet synchronous motors,as a kind of high-performance motor device,have been widely used in rail transportation,variable frequency air conditioning,ship propulsion and other fields.In the permanent magnet synchronous motor control system,the accurate identification of the electromagnetic parameters of the motor plays an important role in improving the motor’s operating performance and early fault diagnosis.In actual working conditions,the electromagnetic parameter of the motor is an amount that changes from moment to moment,and the method of parameter identification is needed to achieve the purpose of real-time parameter acquisition.In this paper,the surface-mounted permanent magnet synchronous motor(Surface Mounted PMSM,SMPMSM)is used as the object,and the online parameter identification of the motor is studied.First,the coordinate transformation was used to derive the mathematical model of the permanent magnet synchronous motor in the αβ and dq coordinate systems.Based on this,the vector control principle with id=0 was introduced,and the characteristics of SMPMSM were combined in the MATLAB simulation environment.The simulation of the vector control system is performed to prepare a basic platform for parameter identification.Secondly,two online identification algorithms are introduced:the recursive least square method with forgetting factor and the identification principle of model reference adaptive algorithm are introduced.Combining the two algorithms with the SMPMSM model,the identification equations for each parameter are derived.Among them,in order to solve the problem of under-ranking of the model reference adaptive identification algorithm,this paper uses the given resistance value to identify the inductor and the permanent magnet flux linkage.When the identification value of the two reaches a stable value,the two-step method MRAS is used to identify the resistance value.Realize online identification of three motor parmeters.Finally,the two algorithms are compared and analyzed through simulation,and it is verified that the two-step method MRAS has a better identification effect.Next,the effect of the inverter dead zone effect on the parameter identification when the actual system is running is analyzed.It is difficult to judge the current zero point during zone compensation.This paper uses a harmonic voltage injection method to reduce the effects of current distortion and torque ripple caused by the dead zone effect,improve the accuracy of the dq axis voltage,and compare it with the The combination of identification algorithms and simulations verify the effectiveness of the dead-band compensation method to improve identification accuracy.Finally,an experimental platform for SMPMSM parameter identification based on TMS320F28335 is set up,and it is designed from two parts:hardware circuit and software program.Through system debugging and experiments,the experimental waveforms of related identification parameters such as speed,current,drive waveform of the inverter module,stator resistance,quadrature axis inductance,and permanent magnet flux linkage were obtained.Among them,the parameter identification error was within 5%,reaching The expected target further verifies that the designed vector control system is stable and reliable,and the parameter identification algorithm used is reasonable and feasible. |
