Solovév equilibrium

For most choices of $ P (\Psi)$ and $ g (\Psi)$, the GS equation has to be solved numerically. For the particular choice of $ P$ and $ g$ profiles,

$\displaystyle \frac{d P}{d \Psi} = - \frac{c_1}{\mu_0},$ (69)

$\displaystyle g \frac{d g}{d \Psi} = - c_2 R_0^2,$ (70)

analytical solution to the GS equation can be found, which is given by[9]

$\displaystyle \Psi = \frac{1}{2} (c_2 R_0^2 + c_0 R^2) Z^2 + \frac{1}{8} (c_1 - c_0) (R^2 - R_0^2)^2,$ (71)

where $ c_0$, $ c_1$, $ c_2$, and $ R_0$ are arbitrary constants. [Proof: By direct substitution, we can verify $ \Psi $ of this form is indeed a solution to the GS equation (53).] A useful choice for tokamak application is to set $ c_0 = B_0 / (R_0^2 \kappa_0 q_0)$, $ c_1 = B_0 (\kappa_0^2 + 1) / (R_0^2
\kappa_0 q_0)$, and $ c_2 = 0$. Then Eq. (71) is written

$\displaystyle \Psi = \frac{B_0}{2 R_0^2 \kappa_0 q_0} \left[ R^2 Z^2 + \frac{\kappa_0^2}{4} (R^2 - R_0^2)^2 \right],$ (72)

which can be solved analytically to give the explicit form of the contour of $ \Psi $ on $ (R, Z)$ plane:

$\displaystyle Z = \pm \frac{1}{R} \sqrt{\frac{2 R_0^2 \kappa_0 q_0}{B_0} \Psi - \frac{\kappa_0^2}{4} (R^2 - R_0^2)^2},$ (73)

which indicates the magnetic surfaces are up-down symmetrical. Using Eq. (69), i.e.,

$\displaystyle \frac{d P}{d \Psi} = - \frac{c_1}{\mu_0} = - \frac{B_0 (\kappa_0^2 + 1)}{\mu_0 R_0^2 \kappa_0 q_0},$ (74)

the pressure is written

$\displaystyle P = P_0 - \frac{B_0 (\kappa_0^2 + 1)}{\mu_0 R_0^2 \kappa_0 q_0} \Psi,$ (75)

where $ P_0$ is a constant of integration. Note Eq. (72) indicates that that $ \Psi = 0$ at the magnetic axis ( $ R = R_0, Z = 0$). Therefore, Eq. (75) indicates that $ P_0$ is the pressure at the magnetic axis. The toroidal field function $ g$ is a constant in this case, which implies there is no poloidal current in this equilibrium. (For the Solovev equilibrium (72), I found numerically that the value of the safety factor at the magnetic axis ( $ R = R_0, Z = 0$) is equal to $ q_0 g / (R_0 B_0)$. This result should be able to be proved analytically. I will do this later. In calculating the safety factor, we also need the expression of $ \vert \nabla \Psi \vert$, which is given analytically by
$\displaystyle \vert \nabla \Psi \vert$ $\displaystyle =$ $\displaystyle \sqrt{\left( \frac{\partial \Psi}{\partial R}
\right)^2 + \left( \frac{\partial \Psi}{\partial Z} \right)^2}$  
  $\displaystyle =$ $\displaystyle \frac{B_0}{2 R_0^2 \kappa_0 q_0} \sqrt{[2 R Z^2 + \kappa_0^2 (R^2 -
R_0^2) R]^2 + (2 R^2 Z)^2} .$ (76)

)

Define $ \Psi_0 = B_0 R_0^2$, and $ \overline{\Psi} = \Psi / \Psi_0$, then Eq. (72) is written as

$\displaystyle \overline{\Psi}$ $\displaystyle =$ $\displaystyle \frac{1}{2 \kappa_0 q_0} \left[ \overline{R}^2
\overline{Z}^2 + \frac{\kappa_0^2}{4} (\overline{R}^2 - 1)^2 \right],$ (77)

where $ \overline{R} = R / R_0$, $ \overline{Z} = Z / R_0$. From Eq. (77), we obtain

$\displaystyle \overline{Z} = \pm \frac{1}{\overline{R}} \sqrt{2 \kappa_0 q_0 \overline{\Psi} - \frac{\kappa_0^2}{4} (\overline{R}^2 - 1)^2} .$ (78)

Given the value of $ \kappa_0$, $ q_0$, for each value of $ \overline {\Psi }$, we can plot a magnetic surface on $ (\overline{R}, \overline{Z})$ plane. An example of the nested magnetic surfaces is shown in Fig. 6.

Figure 6: Flux surfaces of Solovév equilibrium for $ \kappa _0 = 1.5$ and $ q_0 = 1.5$, with $ \overline {\Psi }$ varying from zero (center) to 0.123 (edge). The value of $ \overline {\Psi }$ on the edge is determined by the requirement that the minimum of $ \overline {R}$ is equal to zero. (To prevent ``divided by zero'' that appears in Eq. (78) when $ R = 0$, the value of $ \overline {\Psi }$ on the edge is shifted to $ 0.123 - \varepsilon $ when plotting the above figure, where $ \varepsilon $ is a small number, $ \varepsilon = 10^{- 3}$ in this case.)
\includegraphics{/home/yj/project_new/solovev_equilibrium/fig1/plt.eps}

The minor radius of a magnetic surface of the Solovev equilibrium can be calculated by using Eq. (73), which gives

$\displaystyle R_{\ensuremath{\operatorname{in}}} = \sqrt{R_0^2 - \sqrt{A \Psi}},$ (79)

$\displaystyle R_{\ensuremath{\operatorname{out}}} = \sqrt{R_0^2 + \sqrt{A \Psi}},$ (80)

and thus

$\displaystyle a = \frac{R_{\ensuremath{\operatorname{out}}} - R_{\ensuremath{\o...
...}}{2} = \frac{\sqrt{R_0^2 + \sqrt{A \Psi}} - \sqrt{R_0^2 - \sqrt{A \Psi}}}{2} .$ (81)

where $ A = 8 R_0^2 q_0 / (B_0 \kappa_0)$. In my code of constructing Solovev magnetic surface, the value of $ a$ is specified by users, and then Eq. (81) is solved numerically to obtain the value of $ \Psi $ of the flux surface. Note that the case $ \Psi = 0$ corresponds to $ R_{\ensuremath{\operatorname{in}}} =
R_{\ensuremath{\operatorname{out}}} = R_0$, i.e., the magnetic axis, while the case $ \Psi = R_0^2
B_0 \kappa_0 / (8 q_0)$ corresponds to $ R_{\ensuremath{\operatorname{in}}} = 0$. Therefore, the reasonable value of $ \Psi $ of a magnetic surface should be in the range $ 0
\leqslant \Psi < R_0^2 B_0 \kappa_0 / (8 q_0)$. This range is used as the interval bracketing a root in the bisection root finder.

Using Eq. (81), the inverse aspect ratio of a magnetic surface labeled by $ \Psi $ can be approximated as[9]

$\displaystyle \varepsilon \approx \sqrt{\frac{2 q_0 \Psi}{\kappa_0 R_0^2 B_0}} .$ (82)

Therefore, the value of $ \Psi $ of a magnetic surface with the inverse aspect ratio $ \varepsilon $ is approximately given by

$\displaystyle \Psi = \frac{\varepsilon^2 \kappa_0 R_0^2 B_0}{2 q_0} .$ (83)

yj 2018-03-09