On Extended Kalman Filters with augmented state vectors for the stator flux estimation in SPMSMs

Abstract

Abstract-The demand for highly dynamic electrical drives, characterized by high quality torque control, in a wide variety of applications has grown tremendously during the past decades. Direct torque control (DTC) for permanent magnet synchronous motors (PMSM) can provide this accurate and fast torque control. When applying DTC the change of the stator flux linkage vector is controlled, based on torque and flux errors. As such the estimation of the stator flux linkage is essential. In the literature several possible solutions for the estimation of the stator flux linkage are proposed. In order to overcome problems associated with the integration of the back-emf, the use of state observers has been advocated in the literature. Several types of state observers have been conceived and implemented for PMSMs, especially the Extended Kalman Filter (EKF) has received much attention. In most reported applications however the EKF is only used to estimate the speed and rotor position of the PMSM in order to realize field oriented current control in a rotor reference frame. Far fewer publications mention the use of an EKF to estimate the stator flux linkage vector in order to apply DTC. Still the performance of the EKF in the estimation of the stator flux linkage vector has not yet been thoroughly investigated. In this paper the performance of the EKF for stator flux linkage is studied and simulated. The possibilities to improve the estimation by augmenting the state vector and the consequences of these alterations are explored. Important practical aspects for FPGA implementation are discussed.

Sensorless Direct Torque Control of Surface-Mounted Permanent Magnet Synchronous Motors with Nonlinear Kalman Filtering

“…The extended Kalman filter (EKF) is a popular approach for sensorless control scheme. However, the performance of EKF deteriorates or may even become unstable under measurement failure conditions [21][22][23][24][25][26][27].…”

On the stator flux linkage estimation of an PMSM with Extended Kalman Filters

“…In [10] and [11] the effects of incorrect values for the stator resistance R s , stator inductance L s , and permanent magnet flux Ψ f on the performance of EKFC and EKFF in a DTC drive have been investigated. The influence of an incorrect value of the stator resistance R s is found to be very small and only important during transients, even for large deviations of R s .…”

“…As discussed in [10], [11] the EKFs fail to estimate the stator flux vector correctly if motor parameters used in h(x) and o(x) deviate from the true values. A possible solution to overcome this problem is investigated in [11].…”

“…A possible solution to overcome this problem is investigated in [11]. The most important motor parameters in the EKF are estimated as well.…”

“…The EKF is often discussed for the sensorless control of PMSMs, however focused on the sensorless position estimation for current control in the rotor flux reference frame. Few publications discuss the EKF for the estimation of the stator flux linkage vector [3], [4], [10], [11].…”