Efficiency and Convergence Analysis of Numerical Algorithms for Solving Linear Systems
DOI:
https://doi.org/10.65405/9yyt5n66Keywords:
Linear systems and solves, numerical algorithms, convergence, Gaussian Elimination, conjugate gradient, preconditioningAbstract
The purpose of this work is to review numerical algorithms for the solution of linear systems and determine their convergence characteristics. Gaussian Elimination and LU Decomposition were found to be Direct methods while Jacobi Method, Gauss-Seidel Method and Conjugate Gradient Method were found to be Iterative methods.
The study called for constructing and solving linear systems by using matrices of different dimensions and characteristics. Performance measures such as, number of iterations, execution time and the error norm were computed in order to assess the efficiency of the algorithms. The results proved that direct methods are ideal for small to medium size systems since the percentage accuracy is much higher than the iterative methods taking reasonable amount of time for execution. However, when compared to the direct methods, the iterative methods were more efficient for large and sparse systems of equations and the conjugate gradient method was the most efficient and fastest.
This work also looked at certain effects like the condition numbers and distribution of eigenvalues and also the choice of the initial guess. This confirmed that preconditioning for instance boosts the convergence of iterative methods in a given structure.
This work can be used as a guide for choosing algorithms suited to system size and characteristics and as a practical guide on how to enhance the numerical performance. Moreover, it calls for the additional study and extension of numerical algorithms and liability of employing contemporary technologies including artificial intelligence.
Downloads
References
1. Ayesh, A., & Almahdawi, A. (2024). Evaluate the Efficiency of Techniques for Solving Systems of Linear Algebraic Equations in the Context of Distributed Information Processing Through the Utilization of Simulation. Journal of Distributed Systems, 5, 282–285.
2. Cullen, C. G. (2012). Matrices and linear transformations. Courier Corporation.
3. Golub, G. H., & Van Loan, C. F. (2013). Matrix Computations. Johns Hopkins University Press.
4. Hasanov, V. (2024). Iterative Methods for Solving a Linear Matrix Equation. Annual of Konstantin Preslavsky University of Shumen, Faculty of Mathematics and Informatics, XXV C, 3–14. https://doi.org/10.46687/VZHP2021
5. Ipsen, I. C. F. (2009). Numerical matrix analysis: Linear systems and least squares. Society for Industrial and Applied Mathematics.
6. Kishore Kumar, N., & Schneider, J. (2017). Literature survey on low rank approximation of matrices. Linear and Multilinear Algebra, 65(11), 2212–2244.
7. Kendell Atkinson, An introduction to Numerical Analysis,Wiley, 1989.
8. Tariq Alnnale. (2026).Predictive Governance in Digital Enterprises: An LSTM-Enhanced Deep Learning Framework for Economic Optimization of IT Incident Management Using Enriched Process Logs.DOI: https://doi.org/10.65405/dctw1z34.
9. Khosravi Dehdezi, E. (2024). Iterative Methods for Sparse Symmetric Multilinear Systems. Bulletin of the Iranian Mathematical Society, 50. https://doi.org/10.1007/s41980-024-00875-y
10. Lay, D. C. (2003). Linear algebra and its applications. Pearson Education India.
11. Richard L. Burden & J. Douglas Faires, Numerical Analysis, Brooks/Cole, 2011.
12. Saad, Y. (2003). Iterative methods for sparse linear systems. Society for Industrial and Applied Mathematics.
13. Sarlos, T. (2006, October). Improved approximation algorithms for large matrices via random projections. In 2006 47th annual IEEE symposium on foundations of computer science (FOCS'06) (pp. 143–152). IEEE.
14. Schneider, H., & Barker, G. P. (1989). Matrices and linear algebra. Courier Corporation.
15. Stoll, R. R. (2013). Linear algebra and matrix theory. Courier Corporation.
16. Strang, G. (2006). Linear algebra and its applications. Thomson Learning.
17. Zhang, Y., Li, M., & Liu, G. (2018). A Fourier Transform Analysis of Convergence Properties of Multigrid V-Cycle Algorithm. International Journal of Computational Methods, 16, 1844010. https://doi.org/10.1142/S0219876218440103.
