C++ Codes of Implicit Lu Algorithms for Absdll01

Xing Li,Ying Liu,Antonino Del Popolo

DOI: https://doi.org/10.48550/arXiv.math/0105167

2001-05-21

Abstract:This report is devoted to some C++ codes implementing the implicit LU class algorithms for solving linear determined, and undetermined systems with $n$ variables and $m$ equations. A main program used in part of the numerical test is given in the last section.

Numerical Analysis,Astrophysics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to develop and implement C++ code for implicit LU algorithms for solving linear deterministic and uncertain systems. Specifically, the paper focuses on:

1. **Solving linear systems**: The paper mainly discusses how to use the implicit LU algorithm to solve linear deterministic and uncertain systems with \(n\) variables and \(m\) equations. These systems can be represented as:
\[
Ax = b, \quad A\in\mathbb{R}^{m\times n}
\]
where \(A\) is the coefficient matrix, \(x\) is the unknown vector, and \(b\) is the known vector.

2. **Different versions of the implicit LU algorithm**: The paper describes in detail the implementation of three implicit LU algorithms:
- **Implicit LU algorithm without pivoting (iLUa)**: Applicable to regular matrices (i.e., all principal sub - matrices are non - singular), without pivot selection.
- **Implicit LU algorithm with column pivot selection (iLUaPivotC)**: Deals with non - regular matrices through column pivot selection and explicit column exchange, and adjusts the order of solution vector components.
- **Implicit LU algorithm with row pivot selection (iLUaPivotR)**: Deals with non - regular matrices through row pivot selection and explicit row exchange, without changing the order of solution vector components.

3. **Code implementation**: The paper provides C++ code implementations of these three implicit LU algorithms, using custom `CMatrix` and `CVector` classes to build the ABS software. The code implements the complete process from initialization to updating the solution vector and projection matrix.

4. **Numerical experiments**: To verify the effectiveness of the algorithms, the author conducted numerical experiments, testing different types of matrices (such as Micchelli - Fiedler matrices), and the results show that these algorithms are efficient.

In summary, the main purpose of this paper is to provide an efficient and reliable C++ implementation of the implicit LU algorithm for solving linear deterministic and uncertain systems, and to verify its effectiveness through numerical experiments.