This work examines the multi-pass steering of six electron beams in an FFA arc ranging from approximately 10.5 GeV to 22 GeV. Shown here is an algorithm based on singular value decomposition (SVD) to successfully steer all six beams through the arc given precise knowledge of all beam positions at each of one hundred and one diagnostic locations with one hundred individual corrector magnets: that is successive application of SVD to different 100 × 101 response matrices—one for each beam energy. Further, a machine learning scheme is developed which only requires knowledge of the energy-averaged beam position at each location to provide equivalent steering. Extension of this scheme to other beam optics quantities as well as transverse and longitudinal coupling is explored.

It is currently planned to increase the energy of the CEBAF recirculating linear accelerator to 20 GeV or more by adding two new recirculating arcs that contain multiple new energy passes. The beam is continuous (CW), so no field ramping is desired, making this a fixed-field accelerator (FFA). The wide energy range requires a low dispersion lattice that can be created with high-gradient permanent magnets. One constraint is the existing tunnel radius in relation to the fields achievable by practically-sized permanent magnets. Thus, searching for the most efficient implementation in terms of magnet material volume is important. In this paper, a lattice cell search and optimization is conducted that evaluates cells by the magnet volume per unit length, with the permanent magnet designs also produced via an automated code. The new lattice cells are compared to the previous manually designed cell.

This work examines the multi-pass steering of six electron beams in an FFA arc ranging from approximately 10.5 GeV to 22 GeV. Shown here is an algorithm based on singular value decomposition (SVD) to successfully steer all six beams through the arc given precise knowledge of all beam positions at each of one hundred and one diagnostic locations with one hundred individual corrector magnets: that is successive application of SVD to different 100 × 101 response matrices—one for each beam energy. Further, a machine learning scheme is developed which only requires knowledge of the energy-averaged beam position at each location to provide equivalent steering. Extension of this scheme to other beam optics quantities as well as transverse and longitudinal coupling is explored.

As Thomas Jefferson National Accelerator Facility (Jefferson Lab) looks toward the future, we are considering expanding our energy reach by using Fixed-Field Alternating Gradient (FFA) technology. Significant efforts have been made to design a hybrid accelerator which combines conventional recirculating electron LINAC design with permanent magnet-based FFA technology to increase the number of beam recirculations, and thus the energy. In an effort to further this progress, Jefferson Lab awarded a Laboratory Directed Research and Development (LDRD) grant to focus not on the design, but on detailed simulations of the designs created by the larger collaboration. This document will summarize the work performed during this LDRD, and direct the reader to other proceedings which describe elements of the work in greater detail.