A DSOGI-FLL Based Resolver-to-Digital Conversion Method With Error Calibration Capability Used In PMSM

Resolver is widely used for angle and speed detection in electric vehicle drive application. Resolver generates angular position related voltage signals from sine and cosine windings, induced by high-frequency excitation signals. Due to the imperfect resolver installation, post-processing circuits and manufacturing tolerance, the inevitable amplitude imbalance error, offset error and non-orthogonal error might occur, which influences the angle accuracy from resolver to digital conversion (RDC) results. In order to improve accuracy and calibrate the resolver error, adaptive filter is proposed to remove DC offset errors, and a double second order generalized integrator frequency-locked loop (DSOGI-FLL) based RDC method is proposed to solve the imbalance amplitude error and achieve good frequency tracking. At last, third-order rational fraction polynomial approximation (TRFPA) are used to reconstruct the angle signals. Since the system has error suppression capabilities, the hardware cost of RDC can be further reduced. Both simulation and experimental results demonstrate the good dynamic tracking capability and error suppression capability. The proposed solution can be easily integrated an 'add-on' system to any resolvers based permanent-magnet synchronous motor (PMSM) system.