Self-commissioning of electrical parameters for PMSM in sensorless drives

Abstract

PMSM applications in high-performance control, with or without sensors, involve well self-tuning of the controllers and observers to machine parameters with off/on-line identifications for "plug-in" controllers. This paper proposed self-commissioning methods to automatic identify the PMSM electrical parameters (stator resistance, dq-inductances and PM flux) in sensorless drives, sometimes the rotor is allowed to move. No added hardware or off-line manipulation is required. An observer-based equivalent stator resistance identification, including the inverter resistance and dead-time, is developed at standstill by using two dc current injection levels in phase a and the corresponding stator voltage references. A new dq-inductance identification strategy using HF voltage injection at standstill is proposed with two cascaded observers: i) observer of the currentproduced stator flux (without PM flux) -robust to dc offset and PM flux, and then ii) observer of the dq-inductancesbased on the error of the current-produced stator flux. The PM flux is estimated at constant speed with V/f control in two steps: the stator flux without integration, and the rotor position based on the active flux are estimated in αβ stator reference, and then the PM flux is computed in dq rotor reference. The identification methods are validated by simulation results with good results.

Rotor position estimation scheme with harmonic ripple attenuation for sensorless controlled permanent magnet synchronous motors

“…This kind of method mainly includes the disturbance observer, the sliding mode observer (SMO), the Kalman filtering method, model reference adaptive observer, and the non-linear observer etc. [9][10][11][12][13][14][15]. Compared with other strategies, the SMO is widely used because of its simple algorithm, high robustness, and excellent dynamic performance.…”