A neural network model of a quasiperiodic elliptically polarizing undulator in universal mode. (20th October 2022)

• Record Type: Journal Article
• Authors: Sheppard, Ryan; Baribeau, Cameron; Pedersen, Tor; Boland, Mark; Bertwistle, Drew
• Is Part Of: Journal of synchrotron radiation. Volume 29:Part 6(2022)
• Page Start: 1368
• Page End: 1375
• Publication Date: 2022-10-20
• Subjects: undulator -- neural network -- synchrotron radiation -- extreme ultraviolet; Synchrotron radiation -- Periodicals; Free electron lasers -- Periodicals; 539.73505
• DOI: 10.1107/S1600577522008554
• Languages: English
• ISSNs: 0909-0495

Abstract

Abstract : An approach based on machine learning to produce a fast‐executing model is introduced that predicts the polarization and energy of the radiated light produced at an insertion device. Abstract : Machine learning has recently been applied and deployed at several light source facilities in the domain of accelerator physics. Here, an approach based on machine learning to produce a fast‐executing model is introduced that predicts the polarization and energy of the radiated light produced at an insertion device. This paper demonstrates how a machine learning model can be trained on simulated data and later calibrated to a smaller, limited measured data set, a technique referred to as transfer learning. This result will enable users to efficiently determine the insertion device settings for achieving arbitrary beam characteristics.