Namespaces
Variants
Actions

Difference between revisions of "Matrix factorization method"

From Encyclopedia of Mathematics
Jump to: navigation, search
(Importing text file)
 
m (tex encoded by computer)
 
Line 1: Line 1:
 +
<!--
 +
m0628101.png
 +
$#A+1 = 19 n = 0
 +
$#C+1 = 19 : ~/encyclopedia/old_files/data/M062/M.0602810 Matrix factorization method,
 +
Automatically converted into TeX, above some diagnostics.
 +
Please remove this comment and the {{TEX|auto}} line below,
 +
if TeX found to be correct.
 +
-->
 +
 +
{{TEX|auto}}
 +
{{TEX|done}}
 +
 
''matrix forward-backward substitution method''
 
''matrix forward-backward substitution method''
  
Line 5: Line 17:
 
The solution of the three-point difference scheme
 
The solution of the three-point difference scheme
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628101.png" /></td> </tr></table>
+
$$
 +
A _ {i} Y _ {i-} 1 - C _ {i} Y _ {i} + B _ {i} Y _ {i+} 1  = \
 +
- F _ {i} ,\ \
 +
i = 1 \dots N- 1 ,
 +
$$
  
where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628102.png" /> is an unknown grid vector, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628103.png" /> is the right-hand side vector and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628104.png" /> are given square matrices, under the boundary conditions
+
where $  Y _ {i} = \{ y _ {1,i} \dots y _ {n,i} \} $
 +
is an unknown grid vector, $  F _ {i} $
 +
is the right-hand side vector and $  A _ {i} , B _ {i} , C _ {i} $
 +
are given square matrices, under the boundary conditions
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628105.png" /></td> </tr></table>
+
$$
 +
- C _ {0} Y _ {0} + B _ {0} Y _ {1}  = \
 +
- F _ {0} ,\ \
 +
A _ {N} Y _ {N-} 1 - C _ {N} Y _ {N}  = - F _ {N} ,
 +
$$
  
 
is sought for, as in the scalar case, in the form
 
is sought for, as in the scalar case, in the form
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628106.png" /></td> <td valign="top" style="width:5%;text-align:right;">(*)</td></tr></table>
+
$$ \tag{* }
 +
Y _ {i}  = R _ {i+} 1 Y _ {i+} 1 + Q _ {i+} 1 ,\ \
 +
i = 0 \dots N- 1.
 +
$$
  
The coefficients (the matrix <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628107.png" /> and the vector <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628108.png" />) are determined from the recurrence relations ( "forward substitution" )
+
The coefficients (the matrix $  R _ {i+} 1 $
 +
and the vector $  Q _ {i+} 1 $)  
 +
are determined from the recurrence relations ( "forward substitution" )
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m0628109.png" /></td> </tr></table>
+
$$
 +
R _ {i+} 1  = ( C _ {i} - A _ {i} R _ {i} )  ^ {-} 1 B _ {i} ,
 +
$$
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281010.png" /></td> </tr></table>
+
$$
 +
Q _ {i+} 1  = ( C _ {i} - A _ {i} R _ {i} )  ^ {-} 1
 +
( A _ {i} Q _ {i} + F _ {i} ),\  i = 1 \dots N- 1,
 +
$$
  
while <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281011.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281012.png" /> are given by the left boundary condition:
+
while $  R _ {1} $
 +
and $  Q _ {1} $
 +
are given by the left boundary condition:
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281013.png" /></td> </tr></table>
+
$$
 +
R _ {1}  = C _ {0}  ^ {-} 1 B _ {0} ,\ \
 +
Q _ {1}  = C _ {0}  ^ {-} 1 F _ {0} .
 +
$$
  
The <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281014.png" /> are calculated by formula (*) ( "backward substitution" ), and
+
The $  Y _ {i} $
 +
are calculated by formula (*) ( "backward substitution" ), and
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281015.png" /></td> </tr></table>
+
$$
 +
Y _ {N}  = ( C _ {N} - A _ {N} R _ {N} )  ^ {-} 1 ( A _ {N} Q _ {N} + F _ {N} ).
 +
$$
  
 
There is stability of this method to rounding errors under the conditions
 
There is stability of this method to rounding errors under the conditions
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281016.png" /></td> </tr></table>
+
$$
 +
\| C _ {0}  ^ {-} 1 B _ {0} \|  < 1,\ \
 +
\| C _ {N}  ^ {-} 1 A _ {N} \|  < 1,
 +
$$
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281017.png" /></td> </tr></table>
+
$$
 +
\| C _ {i}  ^ {-} 1 B _ {i} \| + \| C _ {i}  ^ {-} 1 A _ {i} \|  < 1 ,\  i = 1 \dots N- 1,
 +
$$
  
which implies that <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281018.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m062/m062810/m06281019.png" /> (see [[#References|[1]]]). A different form of the stability conditions is also available (see [[#References|[2]]], [[#References|[3]]]). The matrix factorization method is applied also to two-point difference schemes (see [[#References|[3]]]). A variant in which inversion of matrices is replaced by orthogonalization is also used (see [[#References|[4]]]).
+
which implies that $  \| R _ {i} \| < 1 $,  
 +
$  i = 1 \dots N $(
 +
see [[#References|[1]]]). A different form of the stability conditions is also available (see [[#References|[2]]], [[#References|[3]]]). The matrix factorization method is applied also to two-point difference schemes (see [[#References|[3]]]). A variant in which inversion of matrices is replaced by orthogonalization is also used (see [[#References|[4]]]).
  
 
====References====
 
====References====
 
<table><TR><TD valign="top">[1]</TD> <TD valign="top">  A.A. Samarskii,  "Theorie der Differenzverfahren" , Akad. Verlagsgesell. Geest u. Portig K.-D.  (1984)  (Translated from Russian)</TD></TR><TR><TD valign="top">[2]</TD> <TD valign="top">  V.V. Ogneva,  "The  "sweep"  method for the solution of difference equations"  ''USSR Comp. Math. Math. Phys.'' , '''7''' :  4  (1967)  pp. 113–126  ''Zh. Vychisl. Mat. i Mat. Fiz.'' , '''7''' :  4  (1967)  pp. 803–812</TD></TR><TR><TD valign="top">[3]</TD> <TD valign="top">  A.A. Samarskii,  E.S. Nikolaev,  "Numerical methods for grid equations" , '''1–2''' , Birkhäuser  (1989)  (Translated from Russian)</TD></TR><TR><TD valign="top">[4]</TD> <TD valign="top">  S.K. Godunov,  "A method of orthogonal successive substitution for the solution of systems of difference equations"  ''USSR Comp. Math. Math. Phys.'' , '''2''' :  6  (1962)  pp. 1151–1165  ''Zh. Vychisl. Mat. i Mat. Fiz.'' , '''2''' :  6  (1962)  pp. 972–982</TD></TR><TR><TD valign="top">[5]</TD> <TD valign="top">  E.L. Wachspress,  "Iterative solution of elliptic systems and applications to the neutron diffusion equations of reactor physics" , Prentice-Hall  (1966)</TD></TR></table>
 
<table><TR><TD valign="top">[1]</TD> <TD valign="top">  A.A. Samarskii,  "Theorie der Differenzverfahren" , Akad. Verlagsgesell. Geest u. Portig K.-D.  (1984)  (Translated from Russian)</TD></TR><TR><TD valign="top">[2]</TD> <TD valign="top">  V.V. Ogneva,  "The  "sweep"  method for the solution of difference equations"  ''USSR Comp. Math. Math. Phys.'' , '''7''' :  4  (1967)  pp. 113–126  ''Zh. Vychisl. Mat. i Mat. Fiz.'' , '''7''' :  4  (1967)  pp. 803–812</TD></TR><TR><TD valign="top">[3]</TD> <TD valign="top">  A.A. Samarskii,  E.S. Nikolaev,  "Numerical methods for grid equations" , '''1–2''' , Birkhäuser  (1989)  (Translated from Russian)</TD></TR><TR><TD valign="top">[4]</TD> <TD valign="top">  S.K. Godunov,  "A method of orthogonal successive substitution for the solution of systems of difference equations"  ''USSR Comp. Math. Math. Phys.'' , '''2''' :  6  (1962)  pp. 1151–1165  ''Zh. Vychisl. Mat. i Mat. Fiz.'' , '''2''' :  6  (1962)  pp. 972–982</TD></TR><TR><TD valign="top">[5]</TD> <TD valign="top">  E.L. Wachspress,  "Iterative solution of elliptic systems and applications to the neutron diffusion equations of reactor physics" , Prentice-Hall  (1966)</TD></TR></table>
 
 
  
 
====Comments====
 
====Comments====
 
  
 
====References====
 
====References====
 
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  I. [I. Babushka] Babuška,  M. Práger,  E. Vitásek,  "Numerical processes in differential equations" , Interscience  (1966)</TD></TR></table>
 
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  I. [I. Babushka] Babuška,  M. Práger,  E. Vitásek,  "Numerical processes in differential equations" , Interscience  (1966)</TD></TR></table>

Latest revision as of 08:00, 6 June 2020


matrix forward-backward substitution method

A method for solving finite-difference systems that approximate boundary value problems for systems of ordinary differential equations in one-dimensional problems, and for elliptic equations in two-dimensional problems.

The solution of the three-point difference scheme

$$ A _ {i} Y _ {i-} 1 - C _ {i} Y _ {i} + B _ {i} Y _ {i+} 1 = \ - F _ {i} ,\ \ i = 1 \dots N- 1 , $$

where $ Y _ {i} = \{ y _ {1,i} \dots y _ {n,i} \} $ is an unknown grid vector, $ F _ {i} $ is the right-hand side vector and $ A _ {i} , B _ {i} , C _ {i} $ are given square matrices, under the boundary conditions

$$ - C _ {0} Y _ {0} + B _ {0} Y _ {1} = \ - F _ {0} ,\ \ A _ {N} Y _ {N-} 1 - C _ {N} Y _ {N} = - F _ {N} , $$

is sought for, as in the scalar case, in the form

$$ \tag{* } Y _ {i} = R _ {i+} 1 Y _ {i+} 1 + Q _ {i+} 1 ,\ \ i = 0 \dots N- 1. $$

The coefficients (the matrix $ R _ {i+} 1 $ and the vector $ Q _ {i+} 1 $) are determined from the recurrence relations ( "forward substitution" )

$$ R _ {i+} 1 = ( C _ {i} - A _ {i} R _ {i} ) ^ {-} 1 B _ {i} , $$

$$ Q _ {i+} 1 = ( C _ {i} - A _ {i} R _ {i} ) ^ {-} 1 ( A _ {i} Q _ {i} + F _ {i} ),\ i = 1 \dots N- 1, $$

while $ R _ {1} $ and $ Q _ {1} $ are given by the left boundary condition:

$$ R _ {1} = C _ {0} ^ {-} 1 B _ {0} ,\ \ Q _ {1} = C _ {0} ^ {-} 1 F _ {0} . $$

The $ Y _ {i} $ are calculated by formula (*) ( "backward substitution" ), and

$$ Y _ {N} = ( C _ {N} - A _ {N} R _ {N} ) ^ {-} 1 ( A _ {N} Q _ {N} + F _ {N} ). $$

There is stability of this method to rounding errors under the conditions

$$ \| C _ {0} ^ {-} 1 B _ {0} \| < 1,\ \ \| C _ {N} ^ {-} 1 A _ {N} \| < 1, $$

$$ \| C _ {i} ^ {-} 1 B _ {i} \| + \| C _ {i} ^ {-} 1 A _ {i} \| < 1 ,\ i = 1 \dots N- 1, $$

which implies that $ \| R _ {i} \| < 1 $, $ i = 1 \dots N $( see [1]). A different form of the stability conditions is also available (see [2], [3]). The matrix factorization method is applied also to two-point difference schemes (see [3]). A variant in which inversion of matrices is replaced by orthogonalization is also used (see [4]).

References

[1] A.A. Samarskii, "Theorie der Differenzverfahren" , Akad. Verlagsgesell. Geest u. Portig K.-D. (1984) (Translated from Russian)
[2] V.V. Ogneva, "The "sweep" method for the solution of difference equations" USSR Comp. Math. Math. Phys. , 7 : 4 (1967) pp. 113–126 Zh. Vychisl. Mat. i Mat. Fiz. , 7 : 4 (1967) pp. 803–812
[3] A.A. Samarskii, E.S. Nikolaev, "Numerical methods for grid equations" , 1–2 , Birkhäuser (1989) (Translated from Russian)
[4] S.K. Godunov, "A method of orthogonal successive substitution for the solution of systems of difference equations" USSR Comp. Math. Math. Phys. , 2 : 6 (1962) pp. 1151–1165 Zh. Vychisl. Mat. i Mat. Fiz. , 2 : 6 (1962) pp. 972–982
[5] E.L. Wachspress, "Iterative solution of elliptic systems and applications to the neutron diffusion equations of reactor physics" , Prentice-Hall (1966)

Comments

References

[a1] I. [I. Babushka] Babuška, M. Práger, E. Vitásek, "Numerical processes in differential equations" , Interscience (1966)
How to Cite This Entry:
Matrix factorization method. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Matrix_factorization_method&oldid=47794
This article was adapted from an original article by T.A. Germogenova (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article