Research on EBE-FEM Parallel Algorithm Combined with Fast Color Marking Method Based on CUDA Platform

作者

  • Xiuke Yan School of Electrical Engineering Shenyang University of Technology, Shenyang, 110870, China
  • Xvdong Ren School of Electrical Engineering Shenyang University of Technology, Shenyang, 110870, China
  • Jinpeng Lan School of Electrical Engineering Shenyang University of Technology, Shenyang, 110870, China
  • Ziyan Ren School of Electrical Engineering Shenyang University of Technology, Shenyang, 110870, China
  • Yanli Zhang School of Electrical Engineering Shenyang University of Technology, Shenyang, 110870, China

##plugins.pubIds.doi.readerDisplayName##:

https://doi.org/10.13052/2021.ACES.J.361003

关键词:

─ EBE-FEM, parallel computation, fast color marking, element grouping

摘要

The element-by-element finite element method (EBE-FEM) parallel algorithm has been realized on Compute Unified Device Architecture (CUDA) platform in this paper. An improved fast color marking method (FCM) combined with tabu search algorithm is proposed to solve the problem that the elements sharing a node wait for accessing the same memory space in parallel computation. The elements in the same color can be processed at the same time without waiting. This method can get more even color grouping faster than the classical coloring method (CCM).  Combining it with the EBE parallel algorithm can achieve faster element-level operations. The validity and accuracy of the method has been verified by comparing the computed results with the analytical solution of the magnetic field produced by the solenoid. The parallel program is applied to analyze the main magnetic field of a single-phase transformer, which shows higher speedup performance.

##plugins.generic.usageStats.downloads##

##plugins.generic.usageStats.noStats##

##submission.authorBiographies##

##submission.authorWithAffiliation##

Xiuke Yan (correspondence author) Ms. Yan received her B.S degree, M.S degree, and Ph.D. degree in electrical engineering from Shenyang University of Technology, China, in 1996, 1999, and 2005, respectively. She is currently a professor in Shenyang University of Technology. Her research interests include numerical analysis of coupled field and optimization design of electrical equipment, parallel algorithm research of finite element method.

##submission.authorWithAffiliation##

Xvdong Ren Mr. Ren received a B.S degree in electrical engineering from Shandong Jianzhu University, China, in 2018. Currently studying at Shenyang University of Technology, studying for a Ph.D. in electrical engineering, and his research direction is numerical analysis and optimization of engineering electromagnetic fields.

参考

M. Macedonia, “The GPU enters computing's mainstream,” IEEE Computer, vol. 36, no. 10, pp. 106-108, Oct. 2003.

N. Godel, N. Nunn, T. Warburton and M. Clemens, “Accelerating multi GPU based discontinuous Galerkin FEM computations for electromagnetic radio frequency problems,” ACES Journal, vol. 25, no. 4, pp. 331-338, Apr. 2010.

T. Hughus, I. Levit and J. Winget, “An element-by-element solution algorithm for problems of structural and solid mechanics,” Computer Methods in Applied Mechanics and Engineering, vol. 36, no. 2, pp. 241-254, Feb. 1983.

W. Dongyang, Y. Xiuke, T. Renyuan, X. Dexin and R. Ziyan, “Parallel realization of element by element analysis of eddy current field based on graphic processing unit,” ACES Journal, vol. 33, no. 2, pp. 168-171, Feb. 2018.

C. Cecka, A. J. Lew and E. Darve, “Assembly of finite element methods on graphics processors,” International Journal for Numerical Methods in Engineering, vol. 85, no. 5, pp. 40-669, Aug. 2011.

Z. Yan, Y. Xiuke, R, Xvdong, W. Sheng, W. Dongyang, “Parallel implementation and branch optimization of EBE-FEM based on CUDA platform,” ACES Journal, vol. 35, no. 6, pp. 595-600, Jun. 2020.

Y. Jararweh, M. Jarrah, A. Bousselham and S. Hariri, “GPU-based personal supercomputing,” Applied Electrical Engineering and Computing Technologies (AEECT) IEEE, Jan. 2014.

I. Kiss, S. Gyimothy, Z. Badics and J. Pavo, “Parallel Realization of the Element-by-Element FEM Technique by CUDA,” IEEE Transactions on Magnetics, vol. 48, no. 2, pp. 507-510, Feb. 2012.

M. kronbichler, K. Ljungkvist, “Multigrid for matrix-free high-order finite element computations on graphics processors,” ACM Transactions on Parallel Computing, vol. 6, no. 1, pp. 3-34, May. 2019.

G. F. Carey, E. Barragy, R. Mclay and M. Sharma, “Element-by-element vector and parallel computa-tions,” Communications in Numerical Methods in Engineering, vol. 4, no. 3, pp. 299-307, 1988.

K. Kormann, M. Kronbichler, “Parallel Finite Element Operator Application: Graph Partitioning and Coloring,” IEEE Seventh International Conference on E-Science, Dec. 2011.

A. Czumaj, K. Jansen, M. Friedhelm and I. Schiermeyer, “Algorithmic Graph Theory,” Cambridge University Press, 1985.

D. Komatitsch, D. Michea and G. Erlebacher, “Porting a high-order finite-element earthquake modeling application to NVIDIA graphics cards using CUDA,” Journal of Parallel and Distributed Computing, vol. 69, no. 5, pp. 451-460, Jan. 2009.

U. Kiran, D. Sharma and S. Singh Gautam, “GPU-warp based finite element matrices generation and assembly using coloring method,” Journal of Computational Design and Engineering, vol. 4, no. 4, pp. 705-718, Nov. 2018.

Y. Xiuke, H. Xiaoyu, W. Dongyaong, X. Dexin, B. Baodong and R. Ziyan, “Research on preconditioned conjugate gradient method based on EBE-FEM and the application in electromagnetic field analysis,” IEEE trasactions on Magnetics, Jan. 2017.

G. Chunlei, T. Hubing, “Development and application of a fast multipole method in a hybrid FEM/MoM field solver,” ACES Journal, vol. 19, no. 3, pp. 126-134, Nov. 2004.

##submission.downloads##

已出版

2021-11-21

##submission.howToCite##

[1]
X. Yan, X. . Ren, J. . Lan, Z. . Ren, 和 Y. . Zhang, 《Research on EBE-FEM Parallel Algorithm Combined with Fast Color Marking Method Based on CUDA Platform》, ACES Journal, 卷 36, 期 10, 页 1281–1287, 11月 2021.

期

2021: Vol 36 Iss 10

栏目

General Submission