Design Optimization of Nested Bitter Magnets

Abstract

An optimization technique that utilizes a genetic algorithm is used to minimize the electrical power produced by the conducting disks in resistive Bitter magnets. The resulting values from the optimization give the conditional design parameters necessary for a preliminary magnet design. This precursor design method can be employed for either split or single solenoid nested designs. This method is used to design a 10 T, 10 s magnet at the University of Maryland, Baltimore County for plasma physics research. The design parameters from the optimization are used to generate computer-aided design models that are examined with finite element analysis software to verify safety and performance of the design. The magnet design then progresses to a series of thermal and mechanical tests to ensure safe operating conditions.