This paper uses active flux concept to review fundamental frequency sensorless algorithms for both induction and permanent magnet motors in one framework. Fundamentally, sensorless torque estimation can be directly solved using voltage model (VM) estimator, or indirectly solved using current model (CM) estimator. The latter turns the torque estimation problem into a speed estimation problem. The stator flux in VM and the d-axis angle in CM are deemed as the two sets of original states for sensorless drive. Through change of states, the direct torque estimation can be realized via observer designs; whereas the speed dependency of the unknown state (e.g., active flux and emf) gives rise to a class of speed estimation methods, known as model reference adaptive system (MRAS). The idea of a general speed observer is proposed to summarize various separate speed estimation methods needed for direct torque estimation. It is suggested to adopt inherently sensorless designs such that two-way coupling between torque estimation and speed estimation is avoided. For induction motors, it turns out the unmodelled voltage in the active flux dynamics reveals current flowing in rotor bars and can be further modelled, for which the solutions to regeneration instability problem are discussed, and change of states is recommended to attain global stability. Finally discussed are the results of slow reversal test, where local weak observability of ac motors can be potentially preserved.

This paper designs parameter adaptation algorithms for online simultaneous identification of a two-parameter sigmoid inverter model for compensating inverter nonlinearity to reduce the voltage error in flux estimation for a position sensorless motor drive. The inverter model has two parameters, a2 and a3, where a2 is “plateau voltage”, and a3 is a shape parameter that mainly accounts for the stray capacitor effect. Parameter a3 is identified by the (6k ± 1)-th order harmonics in measured current. Parameter a2 is identified by the amplitude mismatch of the estimated active flux. It is found that the classic linear flux estimator, i.e., the hybrid of voltage model and current model, cannot be used for a2 identification. This paper proposes to use a saturation function based nonlinear flux estimator to build an effective indicator for a2 error. The coupled identifiability of the two parameters is revealed and analyzed, which was not seen in literature. The concept of the low current region where the two-way coupling between a2 and a3 occurs is established. In theory, it is suggested to stop the inverter identification in the low current region. However, the experimental results in which dc bus voltage variation and load change are imposed, have shown the effectiveness of the proposed online inverter identification and compensation method, even in low current region.