Reconstruction of offsets of an electron gun using deep learning and an optimization algorithm

A photoelectron gun cavity is an electron source that every Energy Recovery Linac (ERL) requires. Its characteristics and precision determine the capabilities and performance of the ERL. Calibration of the electron gun is a crucial part of its initial setup, which requires a lot of time and experience. We present the first steps towards a tool that guides the electron gun operator, which knobs to turn in which position to achieve the desired properties. In this report, we determine the - typically difficult to identify - offsets between the simulation and the real-world device. We accomplish this by using machine learning and a global optimization algorithm.