FEL2017 - List of Authors (Scheinker, A.)

Paper	Title	Page
TUB04	Recent On-Line Taper Optimization on LCLS	229
	J. Wu, X. Huang, T.O. Raubenheimer SLAC, Menlo Park, California, USA A. Scheinker LANL, Los Alamos, New Mexico, USA
	Funding: The work was supported by the US Department of Energy (DOE) under contract DE-AC02-76SF00515 and the US DOE Office of Science Early Career Research Program grant FWP-2013-SLAC-100164. High-brightness XFELs are demanding for many users, in particular for certain types of imaging applications. Self-seeding XFELs can respond to a heavily tapered undulator more effectively, therefore seeded tapered FELs are considered as a path to high-power FELs in the terawatts level. Due to many effects, including the synchrotron motion, the optimization of the taper profile is intrinsically multi-dimensional and computationally expensive. With an operating XFEL, such as LCLS, the on-line optimization becomes more economical than numerical simulation. Here we report recent on-line taper optimization on LCLS taking full advantages of nonlinear optimizers as well as up-to-date development of artificial intelligence: deep machine learning and neural networks.
	Slides TUB04 [8.227 MB]
DOI •	reference for this paper ※ https://doi.org/10.18429/JACoW-FEL2017-TUB04
Export •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEP036	Adaptive Feedback for Automatic Phase-Space Tuning of Electron Beams in Advanced XFELs	496
	A. Scheinker LANL, Los Alamos, New Mexico, USA D.K. Bohler SLAC, Menlo Park, California, USA
	Particle accelerators are extremely complex having thousands of coupled, nonlinear components which include magnets, laser sources, and radio frequency (RF) accelerating cavities. Many of these components are time-varying. One example is the RF systems which experience unpredictable temperature-based perturbations resulting in frequency and phase shifts. In order to provide users with their desired beam and thereby light properties, LCLS sometimes requires up to 6 hours of manual, experience-based hand tuning of parameters by operators and beam physicists, during a total of 12 hours of beam time provided for the user. Even standard operational changes can require hours to switch between user setups. The main goal of this work is to study model-independent feedback control approaches which can work together with physics-based controls to make overall machine performance more robust, enable faster tuning (seconds to minutes instead of hours), and optimize performance in real time in response to un-modeled time variation and disturbances.
DOI •	reference for this paper ※ https://doi.org/10.18429/JACoW-FEL2017-WEP036
Export •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

Paper

Title

Page

Recent On-Line Taper Optimization on LCLS

229

J. Wu, X. Huang, T.O. Raubenheimer
SLAC, Menlo Park, California, USA
A. Scheinker
LANL, Los Alamos, New Mexico, USA

Funding: The work was supported by the US Department of Energy (DOE) under contract DE-AC02-76SF00515 and the US DOE Office of Science Early Career Research Program grant FWP-2013-SLAC-100164.
High-brightness XFELs are demanding for many users, in particular for certain types of imaging applications. Self-seeding XFELs can respond to a heavily tapered undulator more effectively, therefore seeded tapered FELs are considered as a path to high-power FELs in the terawatts level. Due to many effects, including the synchrotron motion, the optimization of the taper profile is intrinsically multi-dimensional and computationally expensive. With an operating XFEL, such as LCLS, the on-line optimization becomes more economical than numerical simulation. Here we report recent on-line taper optimization on LCLS taking full advantages of nonlinear optimizers as well as up-to-date development of artificial intelligence: deep machine learning and neural networks.

Slides TUB04 [8.227 MB]

DOI •

reference for this paper ※ https://doi.org/10.18429/JACoW-FEL2017-TUB04

Export •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEP036

Adaptive Feedback for Automatic Phase-Space Tuning of Electron Beams in Advanced XFELs

496

A. Scheinker
LANL, Los Alamos, New Mexico, USA
D.K. Bohler
SLAC, Menlo Park, California, USA

Particle accelerators are extremely complex having thousands of coupled, nonlinear components which include magnets, laser sources, and radio frequency (RF) accelerating cavities. Many of these components are time-varying. One example is the RF systems which experience unpredictable temperature-based perturbations resulting in frequency and phase shifts. In order to provide users with their desired beam and thereby light properties, LCLS sometimes requires up to 6 hours of manual, experience-based hand tuning of parameters by operators and beam physicists, during a total of 12 hours of beam time provided for the user. Even standard operational changes can require hours to switch between user setups. The main goal of this work is to study model-independent feedback control approaches which can work together with physics-based controls to make overall machine performance more robust, enable faster tuning (seconds to minutes instead of hours), and optimize performance in real time in response to un-modeled time variation and disturbances.

DOI •

reference for this paper ※ https://doi.org/10.18429/JACoW-FEL2017-WEP036

Export •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)