U.S. Particle Accelerator School

Computational Methods in Electromagnetism course

Sponsoring University:

SUNY Stony Brook


Computational Methods in Electromagnetism


Kwok Ko, Zhenghai Li and Brian McCandless, SLAC

Numerical modeling has become an integral part of any accelerator design and in particular, electromagnetic (EM) modeling is a routine procedure in prototyping accelerator components. In this course, we will present the computational methods that form the basis for many of the numerical modeling tools in electromagnetism as applied to accelerator design. Electro- and magnetostatics, eddy currents, electromagnetics as well as space charge and wakefield effects will be covered, and the numerical algorithms (e.g. finite difference, finite element) appropriate for each of these areas will be examined. We will describe the main elements of a typical modeling tool: mesh generator, field solver and post-processor, and will compare their specific features in available code packages such as MAFIA, HFSS and others. Students will gain hands-on experience with EM software by designing accelerator components typical of those currently being considered for use in next-generation linear colliders and storage rings. The computer exercises will include discussions on the design principles and computational techniques that are applicable to the problem at hand. We will use codes developed at SLAC to introduce advanced topics such as parametric optimization and parallel processing as ways to tackle more complex designs with stringent requirements in resolution and accuracy. Prerequisites: Electromagnetism and Accelerator Physics, familiarity with numerical analysis and computers.