Studies of Big Data Processing at Linear Accelerator Sources Using Machine Learning


Studies of Big Data Processing at Linear Accelerator Sources Using Machine Learning

Bawatna, M.; Green, B. W.

In linear accelerator sources such as the electron beam of the super-conducting linear accelerator at the radiation source Electron Linear accelerator for beams with high Brilliance and low Emittance (ELBE), different kinds of secondary radiation can be produced for various research purposes from materials science up to medicine. A variety of different beam detectors generate a huge amount of data, which take a great deal of computing power to capture and analyse. In this contribution, we will discuss the possibilities of using Machine Learning method to solve the big data challenges. Moreover, we will present a technique that employ the machine learning strategy for the diagnostics of high-field terahertz pulses generated at the ELBE accelerator with extremely flexible parameters such as repetition rate, pulse form and polarization.

Keywords: Cloud computing; Machine learning; Big data; Deep neural networks

  • Contribution to proceedings
    Computer Science On-line Conference, CSOC 2020: Artificial Intelligence and Bioinspired Computational Methods, 20.-24.05.2020, Berlin, Germany
    Advances in Intelligent Systems and Computing, Volume 1225 AISC, 2020, Pages 450-460: Springer, 978-3-030-51970-4
    DOI: 10.1007/978-3-030-51971-1_37

Permalink: https://www.hzdr.de/publications/Publ-31417
Publ.-Id: 31417