Imposing Essential Boundary Conditions in Isogeometric Analysis with Nitsche’s Method

Abstract:

Isogeometric Analysis combines the NURBS technology and finite element approaches into a unified framework. Regrettably, the NURBS basis functions don’t interpolate at the control points, which lead to the difficulty for imposing the essential boundary conditions. A new approach inspired by Nitsche’s notion was proposed in order to remedy this issue. The essential boundary constraints are enforced by the consistent penalty terms. It has some notable advantages than the other methods, such as (i) the symmetric and positive definite bilinear formulation when the stiffness matrix is symmetric and the stabilization parameter is large enough; (ii) the well-conditioning coefficient matrix of the linear system. Finally, the numerical experiment was performed to verify the optimal rate of convergence of the present method.

• Periodical: Applied Mechanics and Materials (Volumes 121-126)
• Pages: 2779-2783
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.121-126.2779
• Online since: October 2011
• Authors: Tao Chen, Rong Mo, Zhong Wei Gong
• Keywords: Essential Boundary Condition, Isogeometric Analysis, Nitsche’s Method, Non-Uniform Rational B-Spline (NURBS), Stabilization Parameter

References

[1] T. Hughes, J.A. Cottrell and Y. Bazilevs: Comput Method Appl M Vol. 194 (2005), p.4135

[2] A. Embar, J. Dolbow and I. Harari: Int J Numer Meth Eng Vol. 83 (2010), p.877

[3] S. Fernandez-Mendez and A. Huerta: Comput Method Appl M Vol. 193 (2004), p.1257

[4] Y. Bazilevs and T. Hughes: Comput Fluids Vol. 36 (2007), p.12

[5] Y. Bazilevs, C. Michler, V.M. Calo and T. Hughes: Comput Method Appl M Vol. 196 (2007), p.4853

[6] M. Griebel and M.A. Schweitzer, in: Geometric Analysis and Nonlinear partial Differential Equations, edtied by Springer-Verlag, Berlin(2002)

[7] A. Embar, J. Dolbow and I. Harari: Int J Numer Meth Eng Vol. 83 (2010), p.877

[8] Y. Bazilevs, L.B. Da Veiga, J.A. Cottrell, T. Hughes and G. Sangalli: Mathematical Models and Methods in Applied Sciences Vol. 16 (2006), p.1031

DOI: 10.1142/s0218202506001455