Preconditioning of two-by-two block matrix systems with square matrix blocks, with applications

Owe Axelsson

Applications of Mathematics (2017)

  • Volume: 62, Issue: 6, page 537-559
  • ISSN: 0862-7940

Abstract

top
Two-by-two block matrices of special form with square matrix blocks arise in important applications, such as in optimal control of partial differential equations and in high order time integration methods. Two solution methods involving very efficient preconditioned matrices, one based on a Schur complement reduction of the given system and one based on a transformation matrix with a perturbation of one of the given matrix blocks are presented. The first method involves an additional inner solution with the pivot matrix block but gives a very tight condition number bound when applied for a time integration method. The second method does not involve this matrix block but only inner solutions with a linear combination of the pivot block and the off-diagonal matrix blocks. Both the methods give small condition number bounds that hold uniformly in all parameters involved in the problem, i.e. are fully robust. The paper presents shorter proofs, extended and new results compared to earlier publications.

How to cite

top

Axelsson, Owe. "Preconditioning of two-by-two block matrix systems with square matrix blocks, with applications." Applications of Mathematics 62.6 (2017): 537-559. <http://eudml.org/doc/294273>.

@article{Axelsson2017,
abstract = {Two-by-two block matrices of special form with square matrix blocks arise in important applications, such as in optimal control of partial differential equations and in high order time integration methods. Two solution methods involving very efficient preconditioned matrices, one based on a Schur complement reduction of the given system and one based on a transformation matrix with a perturbation of one of the given matrix blocks are presented. The first method involves an additional inner solution with the pivot matrix block but gives a very tight condition number bound when applied for a time integration method. The second method does not involve this matrix block but only inner solutions with a linear combination of the pivot block and the off-diagonal matrix blocks. Both the methods give small condition number bounds that hold uniformly in all parameters involved in the problem, i.e. are fully robust. The paper presents shorter proofs, extended and new results compared to earlier publications.},
author = {Axelsson, Owe},
journal = {Applications of Mathematics},
keywords = {preconditioning; Schur complement; transformation; optimal control; implicit time integration},
language = {eng},
number = {6},
pages = {537-559},
publisher = {Institute of Mathematics, Academy of Sciences of the Czech Republic},
title = {Preconditioning of two-by-two block matrix systems with square matrix blocks, with applications},
url = {http://eudml.org/doc/294273},
volume = {62},
year = {2017},
}

TY - JOUR
AU - Axelsson, Owe
TI - Preconditioning of two-by-two block matrix systems with square matrix blocks, with applications
JO - Applications of Mathematics
PY - 2017
PB - Institute of Mathematics, Academy of Sciences of the Czech Republic
VL - 62
IS - 6
SP - 537
EP - 559
AB - Two-by-two block matrices of special form with square matrix blocks arise in important applications, such as in optimal control of partial differential equations and in high order time integration methods. Two solution methods involving very efficient preconditioned matrices, one based on a Schur complement reduction of the given system and one based on a transformation matrix with a perturbation of one of the given matrix blocks are presented. The first method involves an additional inner solution with the pivot matrix block but gives a very tight condition number bound when applied for a time integration method. The second method does not involve this matrix block but only inner solutions with a linear combination of the pivot block and the off-diagonal matrix blocks. Both the methods give small condition number bounds that hold uniformly in all parameters involved in the problem, i.e. are fully robust. The paper presents shorter proofs, extended and new results compared to earlier publications.
LA - eng
KW - preconditioning; Schur complement; transformation; optimal control; implicit time integration
UR - http://eudml.org/doc/294273
ER -

References

top
  1. Axelsson, O., 10.1007/BF01933227, BIT, Nord. Tidskr. Inf.-behandl. 14 (1974), 279-287. (1974) Zbl0289.65028MR0391518DOI10.1007/BF01933227
  2. Axelsson, O., Blaheta, R., Kohut, R., 10.1002/nla.2015, Numer. Linear Algebra Appl. 22 (2015), 930-949. (2015) Zbl06604516MR3426322DOI10.1002/nla.2015
  3. Axelsson, O., Farouq, S., Neytcheva, M., 10.1007/s11075-016-0111-1, Numer. Algorithms 73 (2016), 631-663. (2016) Zbl1353.65059MR3564863DOI10.1007/s11075-016-0111-1
  4. Axelsson, O., Farouq, S., Neytcheva, M., 10.1007/s11075-016-0136-5, Numer. Algorithms 74 (2017), 19-37. (2017) Zbl1365.65167MR3590387DOI10.1007/s11075-016-0136-5
  5. Axelsson, O., Kucherov, A., 10.1002/1099-1506(200005)7:4<197::AID-NLA194>3.0.CO;2-S, Numer. Linear Algebra Appl. 7 (2000), 197-218. (2000) Zbl1051.65025MR1762967DOI10.1002/1099-1506(200005)7:4<197::AID-NLA194>3.0.CO;2-S
  6. Axelsson, O., Neytcheva, M., Ahmad, B., 10.1007/s11075-013-9764-1, Numer. Algorithms 66 (2014), 811-841. (2014) Zbl1307.65034MR3240302DOI10.1007/s11075-013-9764-1
  7. Axelsson, O., Vassilevski, P. S., 10.1137/0612048, SIAM J. Matrix Anal. Appl. 12 (1991), 625-644. (1991) Zbl0748.65028MR1121697DOI10.1137/0612048
  8. Bai, Z.-Z., 10.1007/s10665-013-9670-5, J. Eng. Math. 93 (2015), 41-60. (2015) Zbl1360.65089MR3386091DOI10.1007/s10665-013-9670-5
  9. Bai, Z.-Z., Benzi, M., Chen, F., Wang, Z.-Q., 10.1093/imanum/drs001, IMA J. Numer. Anal. 33 (2013), 343-369. (2013) Zbl1271.65100MR3020961DOI10.1093/imanum/drs001
  10. Bai, Z.-Z., Chen, F., Wang, Z.-Q., 10.1007/s11075-013-9696-9, Numer. Algorithms 62 (2013), 655-675. (2013) Zbl1267.65034MR3034831DOI10.1007/s11075-013-9696-9
  11. Brezzi, F., Fortin, M., 10.1007/978-1-4612-3172-1, Springer Series in Computational Mathematics 15, Springer, New York (1991). (1991) Zbl0788.73002MR1115205DOI10.1007/978-1-4612-3172-1
  12. Butcher, J. C., 10.1002/9780470753767, John Wiley & Sons, Chichester (2008). (2008) Zbl1167.65041MR2401398DOI10.1002/9780470753767
  13. Cahouet, J., Chabard, J.-P., 10.1002/fld.1650080802, Int. J. Numer. Methods Fluids 8 (1988), 869-895. (1988) Zbl0665.76038MR0953141DOI10.1002/fld.1650080802
  14. Greenbaum, A., Pták, V., Strakoš, Z., 10.1137/S0895479894275030, SIAM J. Matrix Anal. Appl. 17 (1996), 465-469. (1996) Zbl0857.65029MR1397238DOI10.1137/S0895479894275030
  15. Lions, J. L., 10.1137/1.9781611970616, CBMS-NSF Regional Conference Series in Applied Mathematics 6, Society for Industrial and Applied Mathematics, Philadelphia (1972). (1972) Zbl0275.49001MR0479526DOI10.1137/1.9781611970616
  16. Paige, C. C., Saunders, M. A., 10.1137/0712047, SIAM J. Numer. Anal. 12 (1975), 617-629. (1975) Zbl0319.65025MR0383715DOI10.1137/0712047
  17. Saad, Y., 10.1137/0914028, SIAM J. Sci. Comp. 14 (1993), 461-469. (1993) Zbl0780.65022MR1204241DOI10.1137/0914028

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.