Numerical computation of the coefficients in exponential fitting (Q2035512)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Numerical computation of the coefficients in exponential fitting |
scientific article; zbMATH DE number 7363167
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Numerical computation of the coefficients in exponential fitting |
scientific article; zbMATH DE number 7363167 |
Statements
Numerical computation of the coefficients in exponential fitting (English)
0 references
24 June 2021
0 references
Two classes of Numerov methods based on exponential fitting are considered to solve the Schrödinger equation. The first class of methods is defined by methods depending on three parameters that includes the well-known methods \(S_m\), \(m=0,1,2,3\), characterized by known analytical expressions for the coefficients. These relations are established in the literature under a convenient condition: the maximum number of parameters that can be different is only two. The second class is defined by methods depending on five parameters. For general Numerov methods based on exponential fitting dependent on three or five parameters, linear systems are established for the coefficients and no closed formulas are known. This problem is solved in this paper using subroutines accessible in the literature. The analysis of the accuracy of the new methods is presented.
0 references
Numerov method
0 references
Schrödinger equation
0 references
regularization
0 references
Coffey-Evans potential
0 references
boundary value problem
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references