Title |
Performance Optimization of a Prototype Undulator U38 Using Multi-Objective Genetic Algorithm |
Authors |
- L.G. Yan, D.R. Deng, P. Li, D. Wupresenter
CAEP/IAE, Mianyang, Sichuan, People's Republic of China
|
Abstract |
Genetic Algorithm (GA) is one of the most excellent method to search the optimal solution of a problem, which has been applied to solve various problems. It is hard to estimate shim applied on raw undulator precisely. There are many methods have been developed to solve the problem. In this proceeding, we measured the magnetic field distribution of prototype undulator U38 and concluded the shim using multi-objective GA. The code was written with the language of Python and based on the package pyevolve. A multi-objective fitness function was setup to implement the multi-objective optimization. Experimentally,performances satisfied the requirements by shimming U38 three times. The trajectory center deviation, peak-to-peak error and phase error are reduced to 0.15 mm, 0.49% and 1°.
|
Funding |
The project of the national large-scale instrument development: 2011YQ130018; National Natural Science Foundation of China: 11505174, 11505173 and 11605190. |
Paper |
download TUPMF045.PDF [1.985 MB / 3 pages] |
Export |
download ※ BibTeX
※ LaTeX
※ Text/Word
※ RIS
※ EndNote |
Conference |
IPAC2018, Vancouver, BC, Canada |
Series |
International Particle Accelerator Conference (9th) |
Proceedings |
Link to full IPAC2018 Proccedings |
Session |
MC2 Poster Session |
Date |
01-May-18 16:00–17:30 |
Main Classification |
02 Photon Sources and Electron Accelerators |
Sub Classification |
T15 Undulators and Wigglers |
Keywords |
undulator, MMI, electron, laser, free-electron-laser |
Publisher |
JACoW Publishing, Geneva, Switzerland |
Editors |
Shane Koscielniak (TRIUMF, Vancouver, BC, Canada); Todd Satogata (JLab, Newport News, VA, USA); Volker RW Schaa (GSI, Darmstadt, Germany); Jana Thomson (TRIUMF, Vancouver, BC, Canada) |
ISBN |
978-3-95450-184-7 |
Published |
June 2018 |
Copyright |
|