U.S. Particle Accelerator School

Accelerator Simulation Using the Unified Accelerator Libraries (UAL) course

Sponsoring University:

Cornell University


Accelerator Simulation Using the Unified Accelerator Libraries (UAL)


Nikolay Malitsky, BNL and Richard Talman, Cornell University

Purpose and Audience
UAL is a simulation environment based on C++ that homogenizes pre-existing computer programs describing diverse physics such as nonlinear effects, field imperfections, space charge, longitudinal dynamics, electron cooling, etc. The purpose for the course is to make this simulation environment available to the students and to others.

Since the course is almost entirely computational, students should be comfortable with unix, text editing, and program execution. Students should also have at least a rudimentary understanding of accelerator concepts such as tunes, beta functions, chromaticity, and transfer matrices. Familiarity with a lattice design code like MAD or DIMAD will be helpful, even though lattice design will NOT be emphasized and design will largely be restricted to parameter variation in sample lattices.

Analysis and correction (as contrasted with initial design) of realistic "warts and all" accelerators is the main purpose of the UAL computational environment. The initial objective of the course is to make contact between analytic and numerical theory by confirming theoretical calculations of ideal lattice properties. Next is to introduce imperfections and/or space charge or other realistic complicating physical effects and to analyse their effects on lattice features such as tunes, tune spreads, dynamic apertures, particle loss due to resonance, and so on. And finally is to model and analyze some complex combinations of more than one physical effect, for example, field imperfections and space charge effects.

Instructional Method
There will be a one hour lecture (longer at first, shorter later) at the beginning of each day describing the plan for the day. There will also be a shorter get together in mid-afternoon to compare notes and iron out general difficulties. Otherwise the entire course will be spent in front of the computer, following the written (or verbal) course material with the instructors. Especially after the first day, to reflect an expectedly large range of student interests and previous experience, there will be a large degree of flexibility of actual tasks.

Course Content
Starting from one of a few very simple, largely FODO lattice templates, using MAD, the student will adjust its lengths, strengths, etc. to achieve given objectives. The theoretical lattice properties will then be confirmed numerically using UAL. Even at this level, for example because of chromaticity-correcting sextupoles, there are nonlinear effects that can be investigated using truncated maps. The extent to which these effects are magnified by more complicated lattice features, such as low beta sections can then be explored. Using the same, or a similar lattice, the effects of imperfections, such as magnetic field errors or misalignments will be investigated.

Depending on the interests of individual students, subsequent studies will be based on one of a few existing accelerator lattices that will be available---for example light source ring, damping ring, electron or proton ring or accelerator, or energy recovery linac. Students are invited to come equipped with their own particular ring, with the proviso that the lattice description be digest
ible in the UAL environment---this would need to be confirmed at least a week or so before the beginning of the course.

Subsequent studies will be variable and will reflect individual interests. Example topics are: dynamic aperture investigation and improvement, investigation of smoothing and decoupling algorithms, space charge effects, coherent synchrotron radiation, map tracking, beta squeeze performance, low emittance lattice properties, transition crossing, collimation, feedback performance, beam-beam tracking, etc. It may be possible for pairs sufficiently proficient with computers to employ the UAL parallel processing capability.

Reading Requirements
Familiarity with the document http://www0.bnl.gov/isd/documents/80139.pdf is the most useful prerequisite. Much reference material is available there. Course materials will be available at that location and at the USPAS web site well before the beginning of the course.

Credit Requirements
Students will be evaluated based on the descriptions of their computational investigations, including graphs and tables. Depending on how quickly the investigation specializes to an individualized topic, the preliminary, common, projects will count for as much as one half of the final grade. The rest of the grade will be based on a report describing a specialized investigation. Ideally this report could pass for an internal laboratory report concerning expected accelerator performance.