Numerical solver of A(alpha)-stable for stiff ordinary differential equations

Rahimah, Jusoh and Aksah, Saufianim J. and Nooraini, Zainuddin and Zarina, Ibrahim (2023) Numerical solver of A(alpha)-stable for stiff ordinary differential equations. Engineering Letters, 31 (4). pp. 1574-1583. ISSN 1816-093X (Print); 1816-0948 (Online). (Published)

[img]
Preview
Pdf
Numerical Solver of alpha-Stable for Stiff Ordinary Differential Equations.pdf
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (1MB) | Preview

Abstract

In this paper, a numerical solver for stiff ordinary differential equations (ODEs) known as the Extended Singly Diagonally Implicit Block Backward Differentiation Formulas (ESDIBDDF) is developed. Objectives of this study are to analyse the A( )  − stability of the ESDIBBDF method and enhance its accuracy by employing a strategy that minimizes the error norm to optimize the values of free parameters. In addition to that, accuracy of the method is to be enhanced by approximating solutions by implementing extra functions to be evaluated. The formula is specifically designed in a lower triangular form with equal diagonal coefficients, enabling faster computation of numerical solutions. Numerical experiments are conducted to assess the efficiency of this method as a solver for stiff ODEs, comparing it with existing methods. The A( )  − stability analysis is verified and conditions for convergence are discussed. The conclusive works efficiently as an alternate solver for stiff ODEs. The research recommended extended application of the developed method to solve applied problems.

Item Type: Article
Additional Information: Indexed by Scopus
Uncontrolled Keywords: A(α)− stable; block multistep method; singly diagonally implicit; stiff ODEs
Subjects: Q Science > QA Mathematics
Faculty/Division: Center for Mathematical Science
Depositing User: Mrs Norsaini Abdul Samat
Date Deposited: 04 Jan 2024 07:05
Last Modified: 04 Jan 2024 07:05
URI: http://umpir.ump.edu.my/id/eprint/39861
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item