Sensorless Vector Control Of PMSM Based On Improved Sliding Model Observer

Permanent magnet synchronous motor(PMSM)is widely used in electric vehicles,military,civil and other driving fields as well as the new energy wind power generation.It is well known that accurate position and speed information of the rotor are necessary for the high-performance vector control of PMSM.Position sensorless control was proposed to reduce the series of problems caused by the use of mechanical position sensors and there are many different types of position identification method.In this paper,the position identification method of PMSM based on sliding mode observer(SMO)is researched.An improved SMO based method is proposed in this paper to improve position estimation accuracy and ultimately improve the vector control performance.The work done is as follows:First of all,the background and research status?development trend of PMSM control system are analyzed,Several commonly used sensorless control methods for PMSM and applicable conditions are introduced.Mathematical model and motion equation of PMSM in different coordinate axes are derived by using coordinate transformation theory,than several types of vector control methods for PMSM are introduced,advantages and disadvantages of each method are also analyzed,among of which,the vector control strategy of_di(28)0 is analyzed in detail.Secondly,the position sensorless control method of PMSM based on SMO is deeply studied.The sliding mode variable structure theory and the mathematical realization process are introduced in detail,the SMO model of PMSM is derived and the rotor position identification principle is analyzed.Considering the harmonic distortion occurs in back electromotive force of PMSM under actual operating conditions,harmonic analysis of the back electromotive force is carried out,aiming at the 5th and 7th harmonics existed in back electromotive force,an improved sliding mode observer structure with adaptive notch filter is proposed to eliminate the specified harmonic and obtain back electromotive force with minimized harmonic content,which can improve the position identification accuracy;Then,a normalized quadrature phase-locked loop is used to obtain the position and rotational speed information.The Simulink simulation model is built to verify the proposed improved SMO algorithm,simulation results confirm the feasibility of the method.Finally,based on theoretical analysis and simulation research,the experimental platform of PMSM drive control system based on TMS320F2808 is built.The static and dynamic performance test experiments are carried out for the proposed method,experimental results demonstrate the feasibility of the proposed improved SMO algorithm.