Answers and Discussions for

ECE Ph.D. Qualifying Exam ES-1 August 2008



Problem 2

To move from stator phase variables to the weird reference frame, pre-multiply by $ \mathbf{K}_s^w = \begin{bmatrix} 1 & -t^2 \\ +t^2 & 1 \end{bmatrix} $. To do the opposite and move from the weird reference frame to stator phase variables, pre-multiply by $ \mathbf{K}_w^s = \left(\mathbf{K}_s^w\right)^{-1} = \frac{1}{1 + t^4} \begin{bmatrix} 1 & +t^2 \\ -t^2 & 1 \end{bmatrix} $. The inverse matrix is found with the explicit formula the the inverse of a 2x2 matrix and its determinant as well.

The voltage equations in stator phase variables are put in vector form.

$ \begin{equation} \vec{v}_{abs} = \mathbf{r}_s \vec{i}_{abs} + \mathit{p}\vec{\lambda}_{abs} = r_s \mathbf{I}_2 \vec{i}_{abs} + \mathit{p}\vec{\lambda}_{abs} \end{equation} $

The transformation proceeds next.

$ \begin{align} \vec{v}_{qds}^{w} &= \mathbf{K}_s^w \vec{v}_{abs} \\ \vec{v}_{qds}^{w} &= \mathbf{K}_s^w r_s \mathbf{I}_2 \left(\mathbf{K}_w^s \vec{i}_{qds}^{w}\right) + \mathbf{K}_s^w \mathit{p} \left(\mathbf{K}_w^s \vec{\lambda}_{qds}^{w}\right) \\ \vec{v}_{qds}^{w} &= r_s \cancelto{\mathbf{I}_2}{\mathbf{K}_s^w \mathbf{I}_2 \mathbf{K}_w^s} \vec{i}_{qds}^{w} + \mathbf{K}_s^w \left(\mathit{p}\mathbf{K}_w^s\right) \vec{\lambda}_{qds}^{w} + \cancelto{\mathbf{I}_2}{\mathbf{K}_s^w \mathbf{K}_w^s} \mathit{p}\vec{\lambda}_{qds}^{w} \end{align} $

The Product Rule is applied for $ \mathit{p} \left(\mathbf{K}_w^s \vec{\lambda}_{qds}^{w}\right) $ to produce two terms, the latter of which also simplifies due to $ \mathbf{K}_s^w $ and $ \mathbf{K}_w^s $ being matrix inverses. A detour is needed to simplify the middle term.

$ \begin{align} \mathbf{K}_s^w \left(\mathit{p}\mathbf{K}_w^s\right) &= \begin{bmatrix} 1 & -t^2 \\ +t^2 & 1 \end{bmatrix} \mathit{p} \left(\frac{1}{1 + t^4} \begin{bmatrix} 1 & +t^2 \\ -t^2 & 1 \end{bmatrix}\right) \\ \mathbf{K}_s^w \left(\mathit{p}\mathbf{K}_w^s\right) &= \begin{bmatrix} 1 & -t^2 \\ +t^2 & 1 \end{bmatrix} \frac{-4t^3}{\left(1 + t^4\right)^2} \begin{bmatrix} 1 & +t^2 \\ -t^2 & 1 \end{bmatrix} + \begin{bmatrix} 1 & -t^2 \\ +t^2 & 1 \end{bmatrix} \frac{1}{1 + t^4} \begin{bmatrix} 0 & +2t \\ -2t & 0 \end{bmatrix} \\ \mathbf{K}_s^w \left(\mathit{p}\mathbf{K}_w^s\right) &= \frac{-4t^3}{\left(1 + t^4\right)^2} \begin{bmatrix} 1 + t^4 & 0 \\ 0 & 1 + t^4 \end{bmatrix} + \frac{1}{1 + t^4} \begin{bmatrix} +2t^3 & +2t \\ -2t & +2t^3 \end{bmatrix} \\ \mathbf{K}_s^w \left(\mathit{p}\mathbf{K}_w^s\right) &= \frac{1}{1 + t^4} \begin{bmatrix} -4t^3 & 0 \\ 0 & -4t^3 \end{bmatrix} + \frac{1}{1 + t^4} \begin{bmatrix} +2t^3 & +2t \\ -2t & +2t^3 \end{bmatrix} \\ \mathbf{K}_s^w \left(\mathit{p}\mathbf{K}_w^s\right) &= \frac{1}{1 + t^4} \begin{bmatrix} -2t^3 & +2t \\ -2t & -2t^3 \end{bmatrix} \end{align} $

Recall by the Quotient Rule or the Chain Rule + Power Rule that $ \frac{d}{dt} \frac{a}{\sum_{k=0}^K b_k t^k} = \frac{-a\sum_{k=1}^K k b_k t^{k - 1}}{\left(\sum_{k=0}^K b_k t^k\right)^2} $ for $ a, b_k, K \in \mathbb{R} $. The vector voltage equation is finished.

$ \begin{equation} \vec{v}_{qds}^{w} = r_s \vec{i}_{qds}^{w} + \frac{1}{1 + t^4} \begin{bmatrix} -2t^3 & +2t \\ -2t & -2t^3 \end{bmatrix} \vec{\lambda}_{qds}^{w} + \mathit{p}\vec{\lambda}_{qds}^{w} \end{equation} $

The weird reference frame voltage equations can be expressed separately.

$ \begin{equation} \boxed{v_{qs}^w = r_s i_{qs}^w - \frac{2t^3}{1 + t^4} \lambda_{qs}^w + \frac{2t}{1 + t^4} \lambda_{ds}^w + \mathit{p}\lambda_{qs}^w} \end{equation} $

$ \begin{equation} \boxed{v_{ds}^w = r_s i_{ds}^w - \frac{2t}{1 + t^4} \lambda_{qs}^w - \frac{2t^3}{1 + t^4} \lambda_{ds}^w + \mathit{p}\lambda_{ds}^w} \end{equation} $


Discussion



Back to ES-1, August 2008

Alumni Liaison

Sees the importance of signal filtering in medical imaging

Dhruv Lamba, BSEE2010