|
Geometric numerical integration (GNI)5 ECTS (englische Bezeichnung: Geometric numerical integration)
(Prüfungsordnungsmodul: Geometrische numerische Integration)
Modulverantwortliche/r: Sigrid Leyendecker Lehrende:
Sigrid Leyendecker, wissenschaftliche Mitarbeiter/innen
Start semester: |
SS 2019 | Duration: |
1 semester | Cycle: |
jährlich (SS) |
Präsenzzeit: |
60 Std. | Eigenstudium: |
90 Std. | Language: |
Englisch |
Lectures:
Empfohlene Voraussetzungen:
Mehrkörperdynamik
Inhalt:
In this lecture, numerical methods that preserve the geometric properties of the flow of a differential equation are presented. First, basic concepts of integration theory such as consistency and convergence are repeated. Several numerical integration methods (Runge-Kutta methods, collocation methods, partitioned methods, composition and splitting methods) are introduced. Conditions for the preservation of first integrals are derived and proven. After a brief introduction into symmetric methods, symplectic integrators for Lagrange and Hamilton systems are considered. Basic concepts such as Hamilton's principle, symplecticity, and Noether's theorem are introduced. A discrete formulation leads to the class of variational integrators which is equivalent to the class of symplectic methods. The symplectictiy leads to a more accurate long-time integration which is proven by concepts of backward error analysis and is demonstrated by means of numerical examples.
Lernziele und Kompetenzen:
- Wissen
- The students
are familiar with ‘Lagrange systems’ and ‘Hamiltonian systems’ and ‘Hamilton’s principle’
know the terms ‘ordinary differential equation’ and ‘analytic solution’
are familiar with ‘consistency’ and ‘convergence’ of a discrete evolution
know standard integrators to solve ordinary differential equations numerically (Runge-Kutta methods, collocation methods, composition and splitting methods…)
know symmetric integrators
are familiar with the terms ‘first integrals’ and ‘quadratic invariants’
are familiar with Noether’s theorem and symplecticity of the Hamilton flow
know symplectic integrators/variational integrators
know conservation properties of symplectic/variational integrators
are familiar with variational error analysis and backward error analysis - Anwenden
- The students
derive Lagrange- and Hamilton’s equations
determine invariants of dynamical systems
implement numerical integrators and solve the ordinary differential equations numerically
analyse the numerical solutions regarding accuracy, conservation of invariants, convergence, symmetry
Literatur:
E. Hairer, G. Wanner, and C. Lubich. Geometric Numerical Integration: Structure-Preserving Algorithms for Ordinary Differential Equations. Springer, 2004.
J. Marsden und T. Ratiu. Einführung in die Mechanik und Symmetrie. Eine grundlegende Darstellung klassischer mechanischer Systeme. Springer, 2001.
J. Marsden, and M. West. Discrete mechanics and variational integrators. Acta Numerica, pp. 357-514, 2001.
Verwendbarkeit des Moduls / Einpassung in den Musterstudienplan:
- Wirtschaftsingenieurwesen (Master of Science)
(Po-Vers. 2018w | TechFak | Wirtschaftsingenieurwesen (Master of Science) | Gesamtkonto | Studienrichtung Maschinenbau | 3. Wahlpflichtmodul + Vertiefungsmodul | 2.4 Höhere Mechanik | Vertiefungsmodul 2.4 Höhere Mechanik | Geometrische numerische Integration)
- Wirtschaftsingenieurwesen (Master of Science)
(Po-Vers. 2018w | TechFak | Wirtschaftsingenieurwesen (Master of Science) | Gesamtkonto | Studienrichtung Maschinenbau | 3. Wahlpflichtmodul + Vertiefungsmodul | 2.5 Höhere Mechanik | Vertiefungsmodul 2.5 Höhere Mechanik | Geometrische numerische Integration)
Dieses Modul ist daneben auch in den Studienfächern "Berufspädagogik Technik (Bachelor of Science)", "Berufspädagogik Technik (Master of Education)", "Computational Engineering (Rechnergestütztes Ingenieurwesen) (Master of Science)", "Maschinenbau (Master of Science)", "Mechatronik (Bachelor of Science)", "Mechatronik (Master of Science)", "Medizintechnik (Bachelor of Science)", "Medizintechnik (Master of Science)" verwendbar. Details
Studien-/Prüfungsleistungen:
Vorlesung + Übung Geometrische numerische Integration (Prüfungsnummer: 72771)
- Prüfungsleistung, mündliche Prüfung, Dauer (in Minuten): 30, benotet, 5 ECTS
- Anteil an der Berechnung der Modulnote: 100.0 %
- Erstablegung: SS 2019, 1. Wdh.: WS 2019/2020
1. Prüfer: | Sigrid Leyendecker |
|
|
|