DESIGN OPTIMIZATION FOR A SUPPORT FOR THE STORAGE RING QUADRUPOLE MAGNET IN A SYNCHROTRON RADIATION FACILITY

Abstract

High brilliance photon beam production requires high gradient magnets. High gradient magnets initiate large magnetic forces to be borne by supports to keep them in position. The objective of this study was to design a support for the CLS 2.0 quadrupole magnet that suppresses vibrations with the goals of the minimal amount of materials and low cost as compared to the existing system. The motivation of this study was associated with upgrading the CLS 2.0’s electron beam, specifically the beam size which will be more than a hundred times smaller than that of the current CLS. The optimization goals of the support design were : (1) Maximizing the natural frequency of the whole magnet system (magnet + supports) and (2) Minimizing the weight of the frame, while meeting the constraints: (1) Static deflection less than 10 microns. (2) Stress developed should be less than the yield stress of the frame material (3) Natural frequency of the system should be more than 50Hz. Such a problem, when translated to the optimization problem, is a large problem as too many design parameters are involved, which makes the “All-In-One (AIO)” strategy of optimization infeasible. This study adopted a divide-and-conquer strategy, i.e., to properly decompose the whole problem into a set of small problems and then optimize them separately. By applying this novel design process, the frame was successfully designed, and the verification showed satisfactory results. The contribution of this work lies in the field of computational design, and specifically, it provides a case demonstration of the divide-and-concur strategy usefulness while optimizing the design for large problems.