三维磁场计算的有限元神经网络模型

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (654 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract In order to solve the problem of finite element method (FEM) with computational cost, this paper proposes a novel finite-element neural network (FENN) embedding a finite-element model in neural network parallel architecture. The novel network carried out parallel distributed calculation of FEM for solving 3D magnetic field while conjugate gradient (CG) method is used as the learning algorithm. The vector plot of magnetic field intensity and magnetic potential contour lines of cross section are obtained. The results indicate that, compared with FEM, FENN divides complex calculation into parallel distributed nodes, simplifies the network of FEM, and is well-behaved magnetic field calculation model with rapidness and accuracy.

Key words： FEM FENN parallel distributed calculation magnetic field analysis conjugate gradient

Received: 28 November 2011 Published: 20 March 2014

PACS:	TP183
	TM153

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Xu Chao
	Wang Changlong
	Sheng Hui
	Yuan Xichao

Cite this article:

Xu Chao,Wang Changlong,Sheng Hui等. FENN Model for 3-D Magnetic Field Calculation[J]. Transactions of China Electrotechnical Society, 2012, 27(11): 125-132.

URL:

http://dgjsxb.ces-transaction.com/EN/ OR http://dgjsxb.ces-transaction.com/EN/Y2012/V27/I11/125

[1] 陈磊, 柴凤, 裴宇龙, 等. 新型旋转电磁热机三维电磁场方程的求解及相关电磁参数的研究[J]. 电工技术学报, 2011, 26(6): 147-153.
[2] 梁慧敏, 由佳欣, 叶雪荣, 等. 基于三维磁场仿真分析的含永磁继电器等效磁路模型的建立[J]. 电工技术学报, 2011, 26(1): 46-50.
[3] 任仪, 赵延文, 聂在平. 基于高阶叠层矢量基函数的时域电磁场积分方法[J]. 电子学报, 2008, 36(3): 516-519.
[4] Ji F, Wang C, Sun S, et al. Application of 3-D FEM in the simulation analysis for MFL signals [J]. Insight, 2009, 51(1):32-35.
[5] 纪凤珠, 王长龙, 陈正阁, 等. 基于三维有限元法的漏磁场分析[J]. 兵工学报, 2007, 28(7): 876-879.
[6] 厉天威, 阮江军, 黄道春, 等. 大规模电磁场数值计算中并行迭代方法的比较[J]. 电工技术学报, 2007, 22(8): 166-173.
[7] Takeuchi J, Kosugi Y. Neural network representation of the finite element method [J]. Neural Networks, 1994, 7(2): 389-395.
[8] Capizzi G, Coco S, Laudani A. A multilayer perceptron neural model for the differentiation of laplacian 3-D finite-element solutions [J]. IEEE Transactions on Magnetics, 2003, 39(3): 1277-1280.
[9] Capizzi G, Coco S, Giuffrida C, et al. A neural network approach for the differentiation of numerical solutions of 3-D electromagnetic problems [J]. IEEE Transactions on Magnetics, 2004, 40(2): 953-956.
[10] Yamashita H, Kowata N, Cingoski V, et al. Direct solution method for finite element analysis using Hopfield neural network [J]. IEEE Transactions on Magnetics, 1995, 31(3): 1964-1967.
[11] Ramuhalli P, Udpa L, Udpa S S. Neural network-based inversion algorithms in magnetic flux leakage nondestructive evaluation [J]. Journal of Applied Physics, 2003, 93(10): 8274-8276.
[12] Ramuhalli P, Udpa L, Udpa S S. Finite-element neural networks for solving differential equations [J]. IEEE Transactions on Neural Networks, 2005, 16(6): 1381-1392.
[13] 金建铭. 电磁场有限元方法[M]. 王建国, 译. 西安:西安电子科技大学出版社, 1998.
[14] 周国华, 肖昌汉, 刘胜道, 等. 基于六面体单元表面磁场积分法求解三维静磁场[J]. 电工技术学报, 2009, 24(3): 1-7.
[15] 刘瑜, 梁正, 杨梓强. 电磁场FDTD算法的容错分析与实现[J]. 电子学报, 2008, 36(9): 1766-1770.
[16] 曹印妮, 张东来, 徐殿国. 三维漏磁场的钢丝绳局部损伤分析算法研究[J]. 电子学报, 2007, 35(6): 1170-1173.
[17] 赵镜红, 张俊洪, 方芳, 等. 径向充磁圆筒永磁直线电机磁场和推力解析计算[J]. 电工技术学报, 2011, 26(7): 154-160.
[18] 徐游. 电磁学[M]. 北京: 科学出版社, 2004.