Research On Sensorless Control Strategy Of Synchronous Reluctance Motor In Full Speed Range

Synchronous reluctance motor(SynRM)is considered to be a good candidate for replacing the traditional low-efficiency asynchronous motor in the field of industrial applications.The realization of low-cost and high-performance sensorless drive system of SynRM has great significance in theoretical research and practical application.In this thesis,high frequency rotating voltage injection method for zero-low speed operation and flux observer method for medium-high speed operation in sensorless drive system of SynRM are studied in depth.Since the inductance of SynRM changes greatly under different load conditions,highfrequency voltage signal with fixed amplitude will increase the amplitude of high-frequency(HF)current response under heavy load and reduce the amplitude under light load or no-load.Higher or lower HF current response will cause adverse impacts on the system.To solve this problem,this thesis proposes an HF injection voltage amplitude adjustment strategy based on on-line identification of incremental inductance,which can significantly reduce the HF current response amplitude of SynRM under heavy load,so as to reduce unnecessary HF loss,noise and torque ripple.Firstly,the amplitude of positive and negative sequence components of HF current response are extracted by discrete Fourier transform(DFT)algorithm,then the incremental inductance of the SynRM can be obtained.Finally,the HF voltage amplitude reference is calculated according to the identified inductance information.In addition,the influence of cross saturation effect on the position estimation accuracy of traditional HF rotating voltage injection method is studied,a new position extraction signal is deduced to suppress the position estimation error caused by cross saturation effect.The traditional flux observer method has poor robustness,to solve this problem,an apparent inductance identification method apply to the sensorless operation of SynRM is proposed,which can effectively compensate the rotor position estimation error caused by the mismatch of flux observer parameters.Compared with the traditional methods,the proposed method requires no additional hardware,and the identification results will not be affected by the position estimation error.Firstly,the inductance characteristics of SynRM are analyzed,then,the specific formula for identifying q-axis apparent inductance from q-axis incremantal inductance is deduced.After the q-axis apparent inductance is obtained,the fundamental frequency voltage equation of SynRM is simplified.Finally,the estimated value of d-axis apparent inductance is solved by Newton-Raphson iterative method.In addition,the weighted switching algorithm is used to realize the smooth transition of HF rotating voltage injection method and flux observer method.Finally,the correctness and effectiveness of the proposed algorithms are verified by MATLAB simulation and a set of 1.5KW SynRM experimental platform.