Andreas Adelmann: Katalogdaten im Frühjahrssemester 2021 |
Name | Herr Dr. Andreas Adelmann |
Adresse | Universitätstrasse 6 CAB H 85.1 8092 Zürich SWITZERLAND |
Telefon | 044 632 75 22 |
andreaad@ethz.ch | |
URL | http://amas.web.psi.ch/people/aadelmann/index.html |
Departement | Physik |
Beziehung | Dozent |
Nummer | Titel | ECTS | Umfang | Dozierende | |
---|---|---|---|---|---|
401-5810-00L | Seminar in Physics for CSE | 4 KP | 2S | A. Adelmann | |
Kurzbeschreibung | In this seminar the students present a talk on an advanced topic in modern theoretical or computational physics. | ||||
Lernziel | |||||
402-0778-00L | Particle Accelerator Physics and Modeling II | 6 KP | 2V + 1U | A. Adelmann | |
Kurzbeschreibung | The effect of nonlinearities on the beam dynamics of charged particles will be discussed. For the nonlinear beam transport, Lie-Methods in combination with differential algebra (DA) and truncated power series (TPS) will be introduced. In the second part we will discuss surrogate model construction for such non-linear dynamical systems using neural networks and polynomial chaos expansion. | ||||
Lernziel | Models for nonlinear beam dynamics can be applied to new or existing particle accelerators. You create Python based surrogate models of dynamical systems, such as charged particle accelerators using Keras and Tensorflow. | ||||
Inhalt | - Symplectic Maps and Higher Order Beam Dynamics - Taylor Modells and Differential Algebra - Lie Methods - Normal Forms - Surrogate Models for dynamical systems - Surrogate model based neural networks - Surrogate model based polynomial chaos - Uncertanty quantification of dynamical systems | ||||
Skript | Lecture notes | ||||
Literatur | * Modern Map Methods in Particle Beam Physics M. Berz (http://bt.pa.msu.edu/pub/papers/AIEP108book/AIEP108book.pdf) | ||||
Voraussetzungen / Besonderes | Ideally Particle Accelerator Physics and Modelling 1 (PAM-1), however at the beginning of the semester, a crash course is offered introducing the minimum level of particle accelerator modeling needed to follow. This lecture is also suited for PhD. Students. |