Coil optimization for electromagnetic levitation using a genetic like algorithmJournal of Applied Physics
Abstracthe technique of electromagnetic levitation (EML) provides a means for thermally processing an electrically conductive specimen in a containerless manner. For the investigation of metallicliquids and related melting or freezing transformations, the elimination of substrate-induced nucleation affords access to much higher undercooling than otherwise attainable. With heating and levitation both arising from the currents induced by the coil, the performance of any EML system depends on controlling the balance between lifting forces and heating effects, as influenced by the levitation coil geometry. In this work, a genetic algorithm is developed and utilized to optimize the design of electromagnetic levitation coils. The optimization is targeted specifically to reduce the steady-state temperature of the stably levitated metallic specimen. Reductions in temperature of nominally 70 K relative to that obtained with the initial design are achieved through coil optimization, and the results are compared with experiments foraluminum. Additionally, the optimization method is shown to be robust, generating a small range of converged results from a variety of initial starting conditions. While our optimizationcriterion was set to achieve the lowest possible sample temperature, the method is general and can be used to optimize for other criteria as well.
RightsCopyright 2013 American Institute of Physics. This article may be downloaded for personal use only. Any other use requires prior permission of the author and the American Institute of Physics.
Copyright OwnerAmerican Institute of Physics
Citation InformationZachary L. Royer, Carl Tackes, Richard A. LeSar and Ralph E. Napolitano. "Coil optimization for electromagnetic levitation using a genetic like algorithm" Journal of Applied Physics Vol. 113 Iss. 21 (2013) p. 214901
Available at: http://works.bepress.com/ralph_napolitano/8/