VERIFIED NUMERICAL COMPUTATIONS FOR EIGENVALUES OF NON-COMMUTATIVE HARMONIC OSCILLATORS (Q4787845)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1850630
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | VERIFIED NUMERICAL COMPUTATIONS FOR EIGENVALUES OF NON-COMMUTATIVE HARMONIC OSCILLATORS |
scientific article; zbMATH DE number 1850630 |
Statements
VERIFIED NUMERICAL COMPUTATIONS FOR EIGENVALUES OF NON-COMMUTATIVE HARMONIC OSCILLATORS (English)
0 references
14 April 2003
0 references
numerical enclosure method
0 references
Hamiltonian
0 references
non-commutative harmonic oscillators
0 references
elliptic eigenvalue problems
0 references
error estimates
0 references
multiple eigenvalues
0 references
eigenfunctions
0 references
minimum-maximum principle
0 references
interval arithmetic
0 references
rounding arithmetic
0 references
0 references
0 references
0.6867161
0 references
0.6812159
0 references
0.67548025
0 references
0.67461854
0 references
0.6740448
0 references
0.6729364
0 references
0.6656903
0 references
The authors study the eigenvalue problem of the operator \[ Q_{(\alpha,\beta)}= I_{(\alpha,\beta)} \left(-\frac{1}{2}\frac{d^2}{dx^2} + \frac{x^2}{2}\right) + J \left(x\frac{d}{dx} + \frac{1}{2}\right),\quad x\in {\mathbb R}, \] where \[ I_{(\alpha,\beta)}=\begin{pmatrix} \alpha & 0\\ 0 & \beta \end{pmatrix},\quad J=\begin{pmatrix} 0 & -1\\ 1 & 0 \end{pmatrix} , \] and \(\alpha\) and \(\beta\) are real positive numbers such that \(\alpha\beta > 1\). This problem, especially when \(\alpha\neq\beta\), is not easy to solve unless numerical methods are employed. The authors together with some other author have already proposed a numerical enclosure method for elliptic eigenvalue problems [cf. \textit{K. Nagatou}, Computing 63, No. 2, 109-130 (1999; Zbl 0941.65108); \textit{K. Nagatou, N. Yamamoto} and \textit{M. T. Nakao}, Optimization 20, 543-565 (1999; Zbl 0938.65137)]. In this paper, using the complete orthonormal system in \(\text{ L}^2({\mathbb R})\) consisting of the Hermite functions, they apply that method to the coupling type eigenvalue problem described above in the unbounded domain. The paper is divided into five Sections. Section 1: Introduction; Section 2: Fixed point formulation and error estimates; Section 3: Verification procedure by computer; Section 4: Discussion of the enclosing for multiple eigenvalues making use of the parity of the eigenfunctions and the monotonicity of the eigenvalues. The monotonicity and the continuity of the eigenvalues are crucial, and they are proved by using the minimum-maximum principle. As a by-product one proves that the first eigenvalue is simple for sufficiently large \(\alpha\) and \(\beta\); Section 5: Numerical results, employing interval arithmetic based on the interval library PROFIL in order to avoid the effects of rounding errors.
0 references