Optimal error estimates of the unilateral contact problem in a curved and smooth boundary domain by the penalty method

2018 ◽  
Vol 40 (1) ◽  
pp. 729-763
Author(s):  
Ibrahima Dione

Abstract We consider linear finite elements to approximate the elasticity equations with unilateral contact boundary conditions, in a bounded two- or three-dimensional domain with curved and smooth boundary. We use the penalty method to weakly impose these boundary conditions. We establish an error estimate in the energy norm with respect to the mesh size $h$ and the penalty parameter $\varepsilon $. Assuming $\boldsymbol{H}^{\frac{3}{2}+\nu }\left (\varOmega \right )$ regularity of the solution, $0 < \nu \leq \frac{1}{2}$, we obtain an $\mathcal{O}\,(h^{\frac{1}{2}+\nu } + \varepsilon ^{\frac{1}{2}+\nu })$ convergence rate. Therefore, if the penalty parameter is chosen as $\varepsilon (h) := ch^{\theta }$ with $0 < \theta \leq 1$, we obtain an $\mathcal{O}\,(h^{\theta (\frac{1}{2}+\nu )})$ convergence rate. Thus, the optimal linear convergence rate is obtained when $\varepsilon $ behaves like $h$ (that is, $\theta = 1$) and $\nu = \frac{1}{2}$. We present a numerical example to illustrate the theoretical analysis.

2018 ◽  
Vol 52 (2) ◽  
pp. 543-566 ◽  
Author(s):  
Yongyong Cai ◽  
Yan Wang

A multiscale time integrator Fourier pseudospectral (MTI-FP) method is proposed and rigorously analyzed for the nonlinear Dirac equation (NLDE), which involves a dimensionless parameter ε ∈ (0, 1] inversely proportional to the speed of light. The solution to the NLDE propagates waves with wavelength O (ε2) and O (1) in time and space, respectively. In the nonrelativistic regime,i.e., 0 < ε ≪ 1, the rapid temporal oscillation causes significantly numerical burdens, making it quite challenging for designing and analyzing numerical methods with uniform error bounds inε ∈ (0, 1]. The key idea for designing the MTI-FP method is based on adopting a proper multiscale decomposition of the solution to the NLDE and applying the exponential wave integrator with appropriate numerical quadratures. Two independent error estimates are established for the proposed MTI-FP method as hm0+τ2/ε2andhm0 + τ2 + ε2, where his the mesh size, τis the time step and m0depends on the regularity of the solution. These two error bounds immediately suggest that the MTI-FP method converges uniformly and optimally in space with exponential convergence rate if the solution is smooth, and uniformly in time with linear convergence rate at O (τ) for all ε ∈ (0, 1] and optimally with quadratic convergence rate at O (τ2) in the regimes when either ε = O (1) or 0 < ε ≲ τ. Numerical results are reported to demonstrate that our error estimates are optimal and sharp.


2019 ◽  
Vol 2019 (1) ◽  
Author(s):  
Shijie Sun ◽  
Meiling Feng ◽  
Luoyi Shi

Abstract This paper considers an iterative algorithm of solving the multiple-sets split equality problem (MSSEP) whose step size is independent of the norm of the related operators, and investigates its sublinear and linear convergence rate. In particular, we present a notion of bounded Hölder regularity property for the MSSEP, which is a generalization of the well-known concept of bounded linear regularity property, and give several sufficient conditions to ensure it. Then we use this property to conclude the sublinear and linear convergence rate of the algorithm. In the end, some numerical experiments are provided to verify the validity of our consequences.


1997 ◽  
Vol 20 (2) ◽  
pp. 397-402 ◽  
Author(s):  
E. M. E. Zayed

The spectral functionΘ(t)=∑i=1∞exp(−tλj), where{λj}j=1∞are the eigenvalues of the negative Laplace-Beltrami operator−Δ, is studied for a compact Riemannian manifoldΩof dimension “k” with a smooth boundary∂Ω, where a finite number of piecewise impedance boundary conditions(∂∂ni+γi)u=0on the parts∂Ωi(i=1,…,m)of the boundary∂Ωcan be considered, such that∂Ω=∪i=1m∂Ωi, andγi(i=1,…,m)are assumed to be smooth functions which are not strictly positive.


Author(s):  
Ran Gu ◽  
Qiang Du

Abstract How to choose the step size of gradient descent method has been a popular subject of research. In this paper we propose a modified limited memory steepest descent method (MLMSD). In each iteration we propose a selection rule to pick a unique step size from a candidate set, which is calculated by Fletcher’s limited memory steepest descent method (LMSD), instead of going through all the step sizes in a sweep, as in Fletcher’s original LMSD algorithm. MLMSD is motivated by an inexact super-linear convergence rate analysis. The R-linear convergence of MLMSD is proved for a strictly convex quadratic minimization problem. Numerical tests are presented to show that our algorithm is efficient and robust.


Sign in / Sign up

Export Citation Format

Share Document