an investigation of dual stator winding induction machines

een discussed in [10.31-10.32] using Routh-Hurwitz criterion while a design strategy of
both observer gains and speed estimator parameters has been proposed and discussed. It
has been shown that the instability region can be reduced by properly designing the
observer gains and speed estimator parameters. In [10.39], the Butterworth method was
used to select the observer gains and the D-decomposition method [10.66, 10.67] defines
the boundary of stability area and instability area. However, the effectivities of the
observer gains and speed estimator parameters within the whole speed range have not
been ensured and checked by author.
Since the effect of stator resistance is so significant at low speed while temperature
affects the stator resistance value so much, the machine parameter adaptive scheme
becomes one of the ways to improve the MRAS adaptive algorithm at low speed range.
The MRAS scheme has also been used to estimate the stator resistance. This estimation
method was firstly presented in [10.21] and then adopted in [10.34, 10.36, 10.37]. The
design methods for observer gains and estimator parameters can only ensure the stability
of two estimators individually, however the overall estimation algorithm may be
unstable. To solve this problem, in [10.33, 10.36], the speed and stator resistance
estimators are combined to be a multi-input, multi-output system (MIMO) so that the
observer gains and the parameters of estimators can be designed properly to ensure the
stability of the whole system. A comparison study of three different MRAS schemes has
been presented in [10.35].
The speed identification problem at low speed range has received considerable
attention and some new approaches can be found in the recent publications. In [10.40], a
different speed adaptive scheme has been proposed, in which the current difference, its
38