Editing
Appendix/Ramblings/NumericallyDeterminedEigenvectors
(section)
Jump to navigation
Jump to search
Warning:
You are not logged in. Your IP address will be publicly visible if you make any edits. If you
log in
or
create an account
, your edits will be attributed to your username, along with other benefits.
Anti-spam check. Do
not
fill this in!
==Finite-Difference Representation== ===General Approach=== Working with the Taylor series expansion, we can write, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~x(\xi)</math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~ x(a) + (\xi - a) x_a' + \tfrac{1}{2} (\xi-a)^2 x_a'' \, , </math> </td> </tr> </table> </div> and letting <math>~\xi_\pm = a \pm \Delta </math>, we have, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~x_+</math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~ x(a) + \Delta \cdot x_a' + \tfrac{1}{2} \Delta^2 x_a'' \, , </math> </td> </tr> </table> </div> and, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~x_-</math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~ x(a) - \Delta \cdot x_a' + \tfrac{1}{2} \Delta^2 x_a'' \, . </math> </td> </tr> </table> </div> Subtracting the second of these two expressions from the first gives, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~x_+ - x_-</math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~ 2 \Delta \cdot x_a' </math> </td> </tr> <tr> <td align="right"> <math>~\Rightarrow ~~~ x_a'</math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~ \frac{x_+ - x_-}{2 \Delta} \, ; </math> </td> </tr> </table> </div> while, adding the two expressions together gives, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~\frac{x_+ - 2x_a + x_-}{\Delta^2}</math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~ x_a'' \, . </math> </td> </tr> </table> </div> ===Integrating Outward Through the Core=== From the LAWE for the core, we have, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~a (g^2 - a^2) x_a''</math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ - ( 4g^2 - 6a^2 ) x_a' - a \mathfrak{F}_\mathrm{core} x_a \, . </math> </td> </tr> </table> </div> So, putting these last three expressions together gives an approximate relation between <math>~x_+</math> and the previous two values of the function, <math>~x_-</math> and <math>~x_a</math>, namely, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~a (g^2 - a^2) \biggl[ \frac{x_+ - 2x_a + x_-}{\Delta^2} \biggr]</math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~ - ( 4g^2 - 6a^2 ) \biggl[\frac{x_+ - x_-}{2 \Delta} \biggr] - a \mathfrak{F}_\mathrm{core} x_a </math> </td> </tr> <tr> <td align="right"> <math>~\Rightarrow~~~ a (g^2 - a^2) \biggl[ \frac{x_+ }{\Delta^2} \biggr] + ( 4g^2 - 6a^2 ) \biggl[\frac{x_+ }{2 \Delta} \biggr] </math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~a (g^2 - a^2) \biggl[ \frac{2x_a - x_-}{\Delta^2} \biggr] + ( 4g^2 - 6a^2 ) \biggl[\frac{x_-}{2 \Delta} \biggr] - a \mathfrak{F}_\mathrm{core} x_a </math> </td> </tr> <tr> <td align="right"> <math>~ \Rightarrow~~~x_+[2a (g^2 - a^2) + \Delta( 4g^2 - 6a^2 ) ] </math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~2a (g^2 - a^2) [ 2x_a - x_- ] + ( 4g^2 - 6a^2 ) [\Delta x_- ] - 2\Delta^2 a \mathfrak{F}_\mathrm{core} x_a </math> </td> </tr> <tr> <td align="right"> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~[4a (g^2 - a^2) - 2\Delta^2 a \mathfrak{F}_\mathrm{core} ]x_a + [ \Delta( 4g^2 - 6a^2 ) - 2a (g^2 - a^2)] x_- \, . </math> </td> </tr> </table> </div> Now, at the very center of the configuration, <math>~(a = 0)</math>, we expect the function, <math>~x(\xi)</math>, to be symmetric; that is, we expect <math>~x_- = x_+</math>. So for this case alone, we have, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~ x_+[2a (g^2 - a^2) + \Delta( 4g^2 - 6a^2 ) - \Delta( 4g^2 - 6a^2 ) + 2a (g^2 - a^2)] </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~[4a (g^2 - a^2) - 2\Delta^2 a \mathfrak{F}_\mathrm{core} ]x_a </math> </td> </tr> <tr> <td align="right"> <math>~\Rightarrow~~~ x_+[2(g^2 - \cancelto{0}{a^2}) + 2(g^2 - \cancelto{0}{a^2})] </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~[4 (g^2 - \cancelto{0}{a^2}) - 2\Delta^2 \mathfrak{F}_\mathrm{core} ]x_a </math> </td> </tr> <tr> <td align="right"> <math>~\Rightarrow~~~ x_+ </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~\biggl[ 1 - \frac{\Delta^2 \mathfrak{F}_\mathrm{core}}{2g^2} \biggr]x_a \, . </math> </td> </tr> </table> </div> For all other coordinate locations, <math>~a = \xi</math>, in the range, <math>~0 < \xi < 1</math>, we will use the general expression, namely, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~ \Rightarrow~~~x_+ </math> </td> <td align="center"> <math>~\approx</math> </td> <td align="left"> <math>~\frac{[4a (g^2 - a^2) - 2\Delta^2 a \mathfrak{F}_\mathrm{core} ]x_a + [ \Delta( 4g^2 - 6a^2 ) - 2a (g^2 - a^2)] x_- }{[2a (g^2 - a^2) + \Delta( 4g^2 - 6a^2 ) ] } \, . </math> </td> </tr> </table> </div> Keep in mind that, when we move across the interface at <math>~a = 1</math>, we want both the value of the function, <math>~x_q</math>, and its first derivative, <math>~x_q'</math>, to be the same as viewed from ''both'' the envelope and the core. In a numerical integration algorithm, it will be very straightforward to set the ''value'' of the eigenfunction at the interface. In order to properly handle the first derivative, I propose that we extend the core solution and evaluate the eigenfunction at one zone beyond the interface, and identify the values of the eigenfunction that straddles the interface as, <div align="center"> <math>~(x_-)_q</math> and <math>~(x_+)_q</math>. </div> Then define the slope of the eigenfunction at the interface by the expression, <div align="center" id="InterfaceSlope"> <table border="1" cellpadding="8"> <tr><td align="center">Slope at the Interface</td></tr> <tr><td align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~x_q'</math> </td> <td align="center"> <math>~\equiv</math> </td> <td align="left"> <math>~\frac{(x_+)_q - (x_-)_q}{2\Delta} \, .</math> </td> </tr> </table> </td></tr> </table> </div> ===Integrating Outward Through the Envelope=== From the LAWE for the envelope, we have, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~a^2( 1 - q^3 a^3 ) x_a'' </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ - ( 3 - 6q^3 a^3 ) a x_a' - [ q^3 \mathfrak{F}_\mathrm{env} a^3 -\alpha_e ]x_a \, . </math> </td> </tr> </table> </div> Inserting the same finite-difference expressions for the first and second derivatives, we therefore have, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~a^2( 1 - q^3 a^3 ) \biggl[ \frac{x_+ - 2x_a + x_-}{\Delta^2} \biggr] </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ - ( 3 - 6q^3 a^3 ) a \biggl[ \frac{x_+ - x_-}{2 \Delta} \biggr] - [ q^3 \mathfrak{F}_\mathrm{env} a^3 -\alpha_e ]x_a </math> </td> </tr> <tr> <td align="right"> <math>~\Rightarrow ~~~ a^2( 1 - q^3 a^3 ) \biggl[ \frac{x_+ }{\Delta^2} \biggr] + ( 3 - 6q^3 a^3 ) a \biggl[ \frac{x_+ }{2 \Delta} \biggr] </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ ( 3 - 6q^3 a^3 ) a \biggl[ \frac{x_-}{2 \Delta} \biggr] - a^2( 1 - q^3 a^3 ) \biggl[ \frac{x_- - 2x_a }{\Delta^2} \biggr] - [ q^3 \mathfrak{F}_\mathrm{env} a^3 -\alpha_e ]x_a </math> </td> </tr> <tr> <td align="right"> <math>~\Rightarrow ~~~ x_+ [2a^2( 1 - q^3 a^3 ) + \Delta ( 3 - 6q^3 a^3 ) a ] </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ [ \Delta ( 3 - 6q^3 a^3 ) a - 2a^2( 1 - q^3 a^3 ) ] x_- + [4a^2( 1 - q^3 a^3 ) - 2\Delta^2 ( q^3 \mathfrak{F}_\mathrm{env} a^3 -\alpha_e ) ] x_a </math> </td> </tr> </table> </div> Now, at the interface (only), we need to relate <math>~x_-</math> to <math>~x_+</math> in such a way that the slope gives the proper value at the interface. Specifically, we need to set, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~x_-</math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~x_+ - 2\Delta (x_q') \, ,</math> </td> </tr> </table> </div> where, <math>~x_q'</math> takes the [[#InterfaceSlope|value that was determined for the core]]. Hence, ''at'' the interface <math>~(a = 1)</math>, the first step into the envelope is special and demands that, <div align="center"> <table border="0" cellpadding="5" align="center"> <tr> <td align="right"> <math>~ x_+ [2a^2( 1 - q^3 a^3 ) + \Delta ( 3 - 6q^3 a^3 ) a ] </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ [ \Delta ( 3 - 6q^3 a^3 ) a - 2a^2( 1 - q^3 a^3 ) ] [x_+ - 2\Delta (x_q')] + [4a^2( 1 - q^3 a^3 ) - 2\Delta^2 ( q^3 \mathfrak{F}_\mathrm{env} a^3 -\alpha_e ) ] x_a </math> </td> </tr> <tr> <td align="right"> <math>~\Rightarrow ~~~ x_+ [4a^2( 1 - q^3 a^3 ) ] </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ [ \Delta ( 3 - 6q^3 a^3 ) a - 2a^2( 1 - q^3 a^3 ) ] [- 2\Delta (x_q')] + [4a^2( 1 - q^3 a^3 ) - 2\Delta^2 ( q^3 \mathfrak{F}_\mathrm{env} a^3 -\alpha_e ) ] x_a </math> </td> </tr> <tr> <td align="right"> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ -2\Delta [ \Delta ( 3 - 6q^3 a^3 ) a - 2a^2( 1 - q^3 a^3 ) ] x_q' + [4a^2( 1 - q^3 a^3 ) - 2\Delta^2 ( q^3 \mathfrak{F}_\mathrm{env} a^3 -\alpha_e ) ] x_a </math> </td> </tr> <tr> <td align="right"> and, setting, <math>~a = 1 ~~~~\Rightarrow ~~~ x_+ </math> </td> <td align="center"> <math>~=</math> </td> <td align="left"> <math>~ \frac{ 2\Delta [2( 1 - q^3 ) - \Delta ( 3 - 6q^3 ) ] x_q' + [4( 1 - q^3 ) - 2\Delta^2 ( q^3 \mathfrak{F}_\mathrm{env} -\alpha_e ) ] x_a }{ 4( 1 - q^3 ) } \, . </math> </td> </tr> </table> </div>
Summary:
Please note that all contributions to JETohlineWiki may be edited, altered, or removed by other contributors. If you do not want your writing to be edited mercilessly, then do not submit it here.
You are also promising us that you wrote this yourself, or copied it from a public domain or similar free resource (see
JETohlineWiki:Copyrights
for details).
Do not submit copyrighted work without permission!
Cancel
Editing help
(opens in new window)
Navigation menu
Personal tools
Not logged in
Talk
Contributions
Log in
Namespaces
Page
Discussion
English
Views
Read
Edit
View history
More
Search
Navigation
Main page
Tiled Menu
Table of Contents
Old (VisTrails) Cover
Appendices
Variables & Parameters
Key Equations
Special Functions
Permissions
Formats
References
lsuPhys
Ramblings
Uploaded Images
Originals
Recent changes
Random page
Help about MediaWiki
Tools
What links here
Related changes
Special pages
Page information