LU-decomposition and numerical structure for solving large sparse nonsymmetric linear systems

Computers and Mathematics with Applications

Volumn 43, Issue 1-2, 2002, Pages 131-155

  Mittal, R C ^a   Al Kurdi, A ^a  

^a INDIAN INSTITUTE OF TECHNOLOGY ROORKEE   (India)

Author keywords

Boolean matrix; Cramer's rule; Determinant; Doolittle's method; Fill ins; Incomplete LU decomposition; Nonsymmetric sparse linear system; Numerical structure; Permanent; Storage scheme

Indexed keywords

BOOLEAN ALGEBRA; LINEAR SYSTEMS; LINEARIZATION; MATHEMATICAL MODELS; MATRIX ALGEBRA; PROBLEM SOLVING;

BOOLEAN MATRIX; CRAMER RULE; DOOLITTLE METHOD; INCOMPLETE LU-DECOMPOSITION; NONSYMMETRIC LINEAR SYSTEMS; NUMERICAL STRUCTURE;

NUMERICAL METHODS;

EID: 0036143108     PISSN: 08981221     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0898-1221(01)00279-6     Document Type: Article

References (19)

1. A node-addition model for symbolic factorization
2. A compact row structure storage scheme for Cholesky factors using elimination trees
3. A class of first order factorization
4. Conjugate gradient methods and LU-preconditioning of non-symmetric systems with arbitrary patterns
5. A robust incomplete cholesky-conjugate gradient algorithm
6. Solving sparse symmetric sets of linear equations by preconditioned conjugate gradient method
7. Preconditioning techniques for non-symmetric matrices with applications to temperature calculations of cooled concrete
8. On the sensitivity of the LU-factorization