@misc{9232703,
  abstract     = {{A Free-Electron Laser (FEL) is a light source capable of producing extremely brilliant light pulses over a wide spectral bandwidth, with pulse lengths ranging from a few hundred picoseconds down to the femtosecond time scale. These pulses are generated and amplified through the interaction of a relativistic electron beam with a co-propagating radiation field in an undulator, a process either starting from shot noise or from an external seed laser.
While the full FEL process requires complex simulations to be described, a theoretical model has been proposed that allows to predict the output of an externally seeded FEL based on the electron beam and seed laser parameters to a high accuracy. This model has been proven to be a useful diagnostic tool to retrieve electron beam and seed laser parameters that are otherwise hard to access. However, the model’s predictive accuracy fluctuates throughout the input parameter
space. It was the main goal of this Bachelor project to classify regions within a limited parameter subspace to yield "valid" or "invalid" predictions based on the accuracy of these predictions, and to further investigate probable reasons for the failure of the model for parameters that are significantly far from the nominal values.
The study was based on systematic comparison of the model’s predictions to simulations, accompanied by validation against experimental data. For the parameter configurations studied, the results indicated that in cases where the seed laser power strongly exceeds the nominal value, there is a large discrepancy (Δ > 20%) between the parameter values predicted by the model and those used in simulations. This is most likely due to saturation effects not yet included in the model and changes in the accumulated dispersion along the undulators that is assumed to be constant in the model.
Conversely, regions in which the model’s predictions are of high to intermediate accuracy (deviations Δ ≤ 10% and 10% < Δ ≤ 20%, respectively) were successfully determined. This was the case for constellations with low to intermediate values of both the energy spread of the electron beam (σE ≈ 30− 60 keV) and the seed laser power (P < 200 MW).
As expected, it was demonstrated that the model fails for parameter constellations yielding an initial bunching too low to overcome a startup from shot noise, a mechanism not included in the model. The studies show that in these cases, the pulse energy of the FEL is very low, allowing to identify unfavorable parameter sets a priori.}},
  author       = {{Linsner, Lena Emily}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Prediction of Seeded Free-Electron Laser Spectral Properties with a Time-dependent Theoretical Model}},
  year         = {{2026}},
}