Fritz John conditions: Difference between revisions

From formulasearchengine
Jump to navigation Jump to search
en>Bgwhite
m Do general fixes and cleanup. -, typos fixed: a optimal → an optimal using AWB (9429)
en>Leoberacai
No edit summary
Line 1: Line 1:
In mathematics, to approximate a derivative to an arbitrary order of accuracy, it is possible to use the [[finite difference]]. A finite difference can be '''central''',  '''forward''' or '''backward'''.
56 year-old Property Manager Chang Righter from Lorette, likes to spend time beatboxing, penomet and towards the disabled. In the recent couple of months has paid a try to spots such as Banc d'Arguin National Park.<br><br>Also visit my web-site: [http://Penometonline.net penius Enlarment]
 
==Central finite difference==
 
This table contains the coefficients of the central differences, for several order of accuracy:<ref name=fornberg>{{Citation | last1=Fornberg | first1=Bengt | title=Generation of Finite Difference Formulas on Arbitrarily Spaced Grids | doi=10.1090/S0025-5718-1988-0935077-0  | year=1988 | journal=[[Mathematics of Computation]] | issn=0025-5718 | volume=51 | issue=184 | pages=699–706}}.</ref>
<!-- replaces Image:Coeff_der_cent_eng.jpg -->
 
{| class="wikitable" style="text-align:center"
|-
! Derivative
! Accuracy
! &minus;4
! &minus;3
! &minus;2
! &minus;1
! 0
! 1
! 2
! 3
! 4
|-
| rowspan="4" | 1
|| 2 || &nbsp; || &nbsp; || &nbsp; || &minus;1/2 || 0|| 1/2|| &nbsp; || &nbsp; || &nbsp;
|-
|| 4 || &nbsp; || &nbsp; || 1/12 || &minus;2/3 || 0|| 2/3|| &minus;1/12 || &nbsp; || &nbsp;
|-
|| 6 || &nbsp; || &minus;1/60 || 3/20  || &minus;3/4 || 0 || 3/4 || &minus;3/20 || 1/60 || &nbsp;
|- style="border-bottom: 2px solid #aaa;"
|| 8 ||1/280 || &minus;4/105 || 1/5 || &minus;4/5 || 0 || 4/5 || &minus;1/5 || 4/105 || &minus;1/280
|-
| rowspan="4" | 2
|| 2 || &nbsp; || &nbsp; || &nbsp; || 1 || −2|| 1|| &nbsp; || &nbsp; || &nbsp;
|-
|| 4 || &nbsp; || &nbsp; || &minus;1/12 || 4/3 || &minus;5/2|| 4/3|| &minus;1/12 || &nbsp; || &nbsp;
|-
|| 6 || &nbsp; || 1/90 || &minus;3/20  || 3/2 || &minus;49/18 || 3/2 || &minus;3/20 || 1/90 || &nbsp;
|- style="border-bottom: 2px solid #aaa;"
|| 8 ||&minus;1/560 || 8/315 || &minus;1/5 || 8/5 || &minus;205/72 || 8/5 || &minus;1/5 || 8/315 || &minus;1/560
|-
| rowspan="3" | 3
|| 2 || &nbsp; || &nbsp; || &minus;1/2 || 1 || 0|| &minus;1|| 1/2 || &nbsp; || &nbsp;
|-
|| 4 || &nbsp; || 1/8 || &minus;1 || 13/8 || 0|| &minus;13/8|| 1 || &minus;1/8 || &nbsp;
|- style="border-bottom: 2px solid #aaa;"
|| 6 || &minus;7/240 || 3/10 || &minus;169/120  || 61/30 ||0 || &minus;61/30|| 169/120 || &minus;3/10 || 7/240
|-
| rowspan="3" | 4
|| 2 || &nbsp; || &nbsp; || 1 || &minus;4 || 6|| &minus;4|| 1 || &nbsp; || &nbsp;
|-
|| 4 || &nbsp; || &minus;1/6 || 2 || &minus;13/2 || 28/3|| &minus;13/2|| 2 || &minus;1/6 || &nbsp;
|-
|| 6 || 7/240 || &minus;2/5 || 169/60  || &minus;122/15 ||91/8 || &minus;122/15|| 169/60 || &minus;2/5 || 7/240
|}
 
For example, the third derivative with a second-order accuracy is
 
: <math>\displaystyle f'''(x_{0}) \approx \displaystyle \frac{-\frac{1}{2}f(x_{-2}) + f(x_{-1}) -f(x_{+1}) +\frac{1}{2}f(x_{+2})}{h^3_x} + O\left(h_x^2  \right)  </math>
 
where <math> h_x </math> represents a uniform grid spacing between each finite difference interval.
 
==Forward and backward finite difference==
 
This table contains the coefficients of the forward differences, for several order of accuracy:<ref name=fornberg/>
 
{| class="wikitable" style="text-align:center"
|-
! Derivative
! Accuracy
! 0
! 1
! 2
! 3
! 4
! 5
! 6
! 7
! 8
|-
| rowspan="6" | 1
|| 1 || &minus;1 || 1 || &nbsp; || &nbsp; || &nbsp;|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 2 || &minus;3/2 || 2 || &minus;1/2 || &nbsp; || &nbsp;|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 3 || &minus;11/6 || 3 || &minus;3/2|| 1/3 || &nbsp; || &nbsp; || &nbsp; || &nbsp; || &nbsp;
|-
|| 4 || &minus;25/12 || 4 || &minus;3 || 4/3 || &minus;1/4|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 5 || &minus;137/60 || 5 || &minus;5 || 10/3 || &minus;5/4 || 1/5 || &nbsp; || &nbsp; || &nbsp;
|- style="border-bottom: 2px solid #aaa;"
|| 6 || &minus;49/20 || 6 || &minus;15/2 || 20/3 || &minus;15/4 || 6/5 || &minus;1/6 || &nbsp; || &nbsp;
|-
| rowspan="6" | 2
|| 1 || 1 || &minus;2 || 1 || &nbsp; || &nbsp;|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 2 || 2 || &minus;5 || 4 || &minus;1 || &nbsp;|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 3 || 35/12 || &minus;26/3 || 19/2 || &minus;14/3 || 11/12 || &nbsp; || &nbsp; || &nbsp; || &nbsp;
|-
|| 4 || 15/4 || &minus;77/6 || 107/6 || &minus;13 || 61/12 || &minus;5/6|| &nbsp; || &nbsp; || &nbsp;
|-
|| 5 || 203/45 || &minus;87/5 || 117/4 || &minus;254/9 || 33/2 || &minus;27/5 || 137/180 || &nbsp; || &nbsp;
|- style="border-bottom: 2px solid #aaa;"
|| 6 || 469/90 || &minus;223/10 || 879/20 || &minus;949/18 || 41 || &minus;201/10 || 1019/180 || &minus;7/10 || &nbsp;
|-
| rowspan="6" | 3
|| 1 || &minus;1 || 3 || &minus;3 || 1 || &nbsp;|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 2 || &minus;5/2 || 9 || &minus;12 || 7 || &minus;3/2|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 3 || &minus;17/4 || 71/4 || &minus;59/2 || 49/2 || &minus;41/4 || 7/4 || &nbsp; || &nbsp; || &nbsp;
|-
|| 4 || &minus;49/8 || 29 || &minus;461/8 || 62 || &minus;307/8 || 13 || &minus;15/8 || &nbsp; || &nbsp;
|-
|| 5 || &minus;967/120 || 638/15 || &minus;3929/40 || 389/3 || &minus;2545/24 || 268/5 || &minus;1849/120 || 29/15 || &nbsp;
|- style="border-bottom: 2px solid #aaa;"
|| 6 || &minus;801/80 || 349/6 || &minus;18353/120 || 2391/10 || &minus;1457/6 || 4891/30 || &minus;561/8 || 527/30 || &minus;469/240
|-
| rowspan="5" | 4
|| 1 || 1 || &minus;4 || 6 || &minus;4 || 1|| &nbsp;|| &nbsp; || &nbsp; || &nbsp;
|-
|| 2 || 3 || &minus;14 || 26 || &minus;24 || 11 || &minus;2 || &nbsp; || &nbsp; || &nbsp;
|-
|| 3 || 35/6 || &minus;31 || 137/2 || &minus;242/3 || 107/2 || &minus;19 || 17/6 || &nbsp; || &nbsp;
|-
|| 4 || 28/3 || &minus;111/2 || 142 || &minus;1219/6 || 176 || &minus;185/2 || 82/3 || &minus;7/2 || &nbsp;
|- style="border-bottom: 2px solid #aaa;"
|| 5 || 1069/80 || &minus;1316/15 || 15289/60 || &minus;2144/5 || 10993/24 || &minus;4772/15 || 2803/20 || &minus;536/15 || 967/240
|}
 
For example, the first derivative with a third-order accuracy and the second derivative with a second-order accuracy are
 
: <math>\displaystyle f'(x_{0}) \approx \displaystyle \frac{-\frac{11}{6}f(x_{0}) + 3f(x_{+1}) -\frac{3}{2}f(x_{+2}) +\frac{1}{3}f(x_{+3}) }{h_{x}} + O\left(h_{x}^3  \right), </math>
 
: <math>\displaystyle f''(x_{0}) \approx \displaystyle \frac{2f(x_{0}) - 5f(x_{+1}) + 4f(x_{+2}) - f(x_{+3}) }{h_{x}^2} + O\left(h_{x}^2  \right), </math>
 
while the corresponding backward approximations are given by
 
: <math>\displaystyle f'(x_{0}) \approx \displaystyle \frac{\frac{11}{6}f(x_{0}) - 3f(x_{-1}) +\frac{3}{2}f(x_{-2}) -\frac{1}{3}f(x_{-3}) }{h_{x}} + O\left(h_{x}^3  \right), </math>
 
: <math>\displaystyle f''(x_{0}) \approx \displaystyle \frac{2f(x_{0}) - 5f(x_{-1}) + 4f(x_{-2}) - f(x_{-3}) }{h_{x}^2} + O\left(h_{x}^2  \right). </math>
 
 
In general, to get the coefficients of the backward approximations, give all odd derivatives listed in the table the opposite sign, whereas for even derivatives the signs stay the same.
 
==See also==
* [[Finite difference method]]
* [[Finite difference]]
* [[Five-point stencil]]
 
== References ==
{{reflist}}
 
{{Numerical PDE}}
 
 
{{DEFAULTSORT:Finite Difference Coefficient}}
[[Category:Finite differences]]
[[Category:Numerical differential equations]]

Revision as of 01:15, 26 February 2014

56 year-old Property Manager Chang Righter from Lorette, likes to spend time beatboxing, penomet and towards the disabled. In the recent couple of months has paid a try to spots such as Banc d'Arguin National Park.

Also visit my web-site: penius Enlarment