Analytical Design of Self-Sensing Control for PMSM Using Quasi-Direct Calculation

Abstract

The analytical description and parameterization of a self-sensing control (SSC) for an electrical machine is an important step toward easier commissioning of these systems. In this article, the advantages of high bandwidth position estimation via numerical optimization and the filtering characteristics of a phaselocked loop are combined in the quasi-direct (QD) calculation. The QD calculation uses two parameters for estimation. With the help of the maximum possible acceleration of the drive train, an interdependency between these two parameters is derived. The remaining degree of freedom is used to tune the dynamics of the estimation. Using the transfer function of the estimator, which is derived analytically, the parameters of the speed control are selected, and a specified phase-margin is implemented. With the help of the analytical parameterization, no empirical or numerical tuning needs to be done, which is unique for SSC. All results are experimentally validated.