A non-invasive online method for real-time determination of the Most Probable Energy (MPE) of industrial electron beams

Abstract

A non-invasive method for real-time determination
of the most probable energy (MPE) of industrial electron beams is
presented. The method is based on the use of beam energy
dispersion in a scanning magnetic field combined with local
detection of beam intensity using a thin collecting electrode. Due to
the lack of synchronization between the accelerator pulse
generator and the scanning magnet, the beam is sampled in an
asynchronous manner, allowing reconstruction of the spatial and
energy distribution through signal accumulation. Two methods for
MPE determination were investigated: local parabolic
approximation and a centroid-based approach. The latter was
found to be more robust under noisy industrial conditions and
provided superior repeatability. Application of the method to an
industrial linear accelerator demonstrated that the beam energy
exhibits significant temporal fluctuations. The proposed method
provides a direct, real-time measure of effective beam energy and
can be used as an operational parameter for accelerator tuning and
process control. It complements standard dosimetric techniques
by enabling continuous monitoring of beam energy under normal
operating conditions.