Center for Extended Magnetohydrodynamic Modeling (CEMM) University of Utah SAPP 2007 Final Status Report

Sanderson, Allen R. ; Johnson, Christopher R.

University of Utah

关键词: Magnetic Islands;    Magnetic Fields;    43 Particle Accelerators;    Frequency Analysis;    Simulation;   

DOI  :  10.2172/1015186
RP-ID  :  Final Report
RP-ID  :  FG02-04ER25653
RP-ID  :  1015186

美国|英语

来源: UNT Digital Library

【 摘 要 】

During the third and final period of this grant, our goal was to refine the algorithmic approaches used to detect and visualize magnetic islands and their corresponding null points within both the NIMROD and M3D data sets. We refined our geometric approach, which gave a greater confidence in the accuracy of the Poincareplots created. The final results are best demonstrated through Figures 2-6 attached to the report. Technical details this work was reported in both the Physics and Visualization communities. The algorithms used to analyze the magnetic field lines and detect magnetic islands have been packaged into a library and were used within the SCIRun Problem Solving Environment which is being used by members of the CEMM for visualization. In addition, the library interface was developed so that it could be used by both the NIMROD and M3D codes directly. Thus allowing the fusion scientist to perform this analysis while their simulations were actively running. The use of the library for analysis and visualization was not limited to just within the CEMM SciDAC. Other groups such as the SciDAC for the Simulation of Wave Interactions with Magnetohydrodynamics using Silo code have used the tools for the analysis of their simulations, Figure 1. Though the funding of this project had concluded there is still much work to be performed on this analysis. The techniques developed are fast and robust when not in the presence of chaos. Magnetic field lines that are near the separatrices where chaos is most often present can be difficult to analyze yet these are the field lines that are greatest interest. We believe that investigating and developing techniques based on time frequency analysis may hold some promise. Two other issues that need to be address is the ability to automatically search for the magnetic islands and the ability to track the development of the magnetic islands over time. Our initial effort into automatically searching for the islands did not prove as robust as hoped and required more effort than could be allocated. These areas as well as other issues related to 'orbit analysis' are of interest to many members within of each of the Fusion SciDAC Centers and should be the subject of continuing SAPs such as this one.