Solution of Large Stochastic Finite Element Problems-Application to ECT-NDT

• Published in: IEEE transactions on magnetics Vol. 49; no. 5; pp. 1605 - 1608
• Main Authors: Beddek, K., Clenet, S., Moreau, O., Le Menach, Y.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: New York, NY IEEE 01.05.2013; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Cross-disciplinary physics: materials science; rheology; Electric power; Electromagnetism; Engineering Sciences; Exact sciences and technology; Finite element analysis; Finite Element Method; Iterative methods; iterative solver; Linear systems; Magnetism; Materials science; Mathematical analysis; Mathematical models; Methodology; Nondestructive testing; Other topics in materials science; Physics; Polynomials; Probes; quasistatic fields; Random variables; Solvers; Spectra; Statistics; stochastic model; Stochastic processes; Stochasticity; uncertainty quantification; Vectors; Finite element method; Stochastic model; uncertainty quantification; iterative solver; quasistatic fields; Modelling; Application; Magnetic properties; Uncertainty Quantification; Stochastic Model; Finite Element Method; Iterative Solver; Quasi Static Fields
• ISSN: 0018-9464; 1941-0069
• DOI: 10.1109/TMAG.2013.2240665

This paper describes an efficient bloc iterative solver for the spectral stochastic finite element method (SSFEM). The SSFEM was widely used to quantify the effect of input data uncertainties on the outputs of finite element models. The bloc iterative solver allows reducing computational cost of the SSFEM. The method is applied on an industrial nondestructive testing (NDT) problem. The numerical performances of the method are compared with those of the nonintrusive spectral projection (NISP).