1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
|
Description: Avoid extra alignment tab error in split environment
Author: Rafael Laboissière <rafael@debian.org>
Forwarded: https://sourceforge.net/p/xmds/mailman/message/37729681/
Applied-Upstream: https://sourceforge.net/p/xmds/code/3128/
Last-Update: 2022-11-02
--- xmds2-3.1.0+dfsg.orig/userdoc/src/reference_elements.rst
+++ xmds2-3.1.0+dfsg/userdoc/src/reference_elements.rst
@@ -667,8 +667,8 @@ which has the inverse transform:
This transform pair has the useful property that the Laplacian in cylindrical co-ordinates is diagonal in this basis:
.. math::
- \nabla^2 \left(f(r) e^{i m \theta}\right) &=& \left(\frac{\partial^2 f}{\partial r^2} +\frac{1}{r}\frac{\partial f}{\partial r} -\frac{m^2}{r^2} f \right) e^{i m \theta}\\
- &=& \left\{\mathcal{H}^{-1}_m \left[(-k^2) F_m(k)\right](r) \right\} e^{i m \theta}
+ \nabla^2 \left(f(r) e^{i m \theta}\right) &= \left(\frac{\partial^2 f}{\partial r^2} +\frac{1}{r}\frac{\partial f}{\partial r} -\frac{m^2}{r^2} f \right) e^{i m \theta}\\
+ &= \left\{\mathcal{H}^{-1}_m \left[(-k^2) F_m(k)\right](r) \right\} e^{i m \theta}
XMDS labels the variables in the transformed space with a prefix of 'k', just as for :ref:`Fourier transforms<dft_Transform>`. The order :math:`m` of the transform is defined by the ``order`` attribute in the ``<dimension>`` element, which must be assigned as a non-negative integer. If the order is not specified, it defaults to zero which corresponds to the solution being independent of the angular coordinate :math:`\theta`.
@@ -710,8 +710,8 @@ The "spherical-bessel" transform
When working in spherical coordinates, it is often useful to use the spherical Bessel functions :math:`j_l(x)=\sqrt{\frac{\pi}{2x}}J_{l+\frac{1}{2}}(x)` as a basis. These are eigenfunctions of the radial component of Laplace's equation in spherical coordinates:
.. math::
- \nabla^2 \left[j_l(k r)\; Y^m_l(\theta, \phi)\right] &=& \left[\frac{\partial^2 }{\partial r^2} +\frac{2}{r}\frac{\partial }{\partial r} -\frac{l(l+1)}{r^2}\right] j_l(k r) \; Y^m_l(\theta, \phi) \\
- &=& -k^2 j_l(k r)\; Y^m_l(\theta, \phi)
+ \nabla^2 \left[j_l(k r)\; Y^m_l(\theta, \phi)\right] &= \left[\frac{\partial^2 }{\partial r^2} +\frac{2}{r}\frac{\partial }{\partial r} -\frac{l(l+1)}{r^2}\right] j_l(k r) \; Y^m_l(\theta, \phi) \\
+ &= -k^2 j_l(k r)\; Y^m_l(\theta, \phi)
Just as the Bessel basis above, the transformed dimensions are prefixed with a 'k', and it is possible (and usually wise) to use the ``spectral_lattice`` attribute to specify a different lattice size in the transformed space. Also, the spacing of these lattices are again chosen in a non-uniform manner to Gaussian quadrature methods for spectrally accurate transforms. Finally, the ``order`` attribute can be used to specify the order :math:`l` of the spherical Bessel functions used.
|