%0 Report %1 citeulike:11810082 %A Carnes, Brian R. %A Copps, Kevin D. %C Albuquerque %D 2008 %K 68w01-algorithms-general 80-04-heat-transfer-explicit-machine-computation-and-programs 80m10-heat-transfer-finite-element-methods %N SAND2008-2607 %T Thermal Contact Algorithms in SIERRA Mechanics %U http://prod.sandia.gov/techlib/access-control.cgi/2008/082607.pdf %X We examine algorithms for the finite element approximation of thermal contact models. We focus on the implementation of thermal contact algorithms in SIERRA Mechanics. Following the mathematical formulation of models for tied contact and resistance contact, we present three numerical algorithms: ) the multi-point constraint (MPC) algorithm, ) a resistance algorithm, and ) a new generalized algorithm. We compare and contrast both the correctness and performance of the algorithms in three test problems. We tabulate the convergence rates of global norms of the temperature solution on sequentially refined meshes. We present the results of a parameter study of the effect of contact search tolerances. We outline best practices in using the software for predictive simulations, and suggest future improvements to the implementation.

@techreport{citeulike:11810082, abstract = {We examine algorithms for the finite element approximation of thermal contact models. We focus on the implementation of thermal contact algorithms in SIERRA Mechanics. Following the mathematical formulation of models for tied contact and resistance contact, we present three numerical algorithms: ) the multi-point constraint (MPC) algorithm, ) a resistance algorithm, and ) a new generalized algorithm. We compare and contrast both the correctness and performance of the algorithms in three test problems. We tabulate the convergence rates of global norms of the temperature solution on sequentially refined meshes. We present the results of a parameter study of the effect of contact search tolerances. We outline best practices in using the software for predictive simulations, and suggest future improvements to the implementation.}, added-at = {2017-06-29T07:13:07.000+0200}, address = {Albuquerque}, author = {Carnes, Brian R. and Copps, Kevin D.}, biburl = {https://www.bibsonomy.org/bibtex/25329cb6d88298aa5144afc77ffee0af4/gdmcbain}, citeulike-article-id = {11810082}, citeulike-attachment-1 = {carnes_08_thermal.pdf; /pdf/user/gdmcbain/article/11810082/856518/carnes_08_thermal.pdf; cb5299c19380af8d462178b4c7c96d82ee1ad351}, citeulike-linkout-0 = {http://prod.sandia.gov/techlib/access-control.cgi/2008/082607.pdf}, comment = {(private-note)mentioned by marc.youcef@insa-lyon.fr 2012-08-17 ---=note-separator=--- Makes some interesting and important general points about verification in Section 6.1: \% First, without an exact ''analytical'' solution, we cannot be entirely certain of the error rates in these average temperatures, other global quantities, or even global norms. Second, the exact solution to this problem likely has multiple singularities, because of the reentrant corners and edges, the discontinuous material properties, and the jump in the boundary condition on \$\Gamma\_{D2}\$ at initial time. Because the solution is not smooth, we do not know the expected order of convergence in either global quantities or norms, which makes it impossible to make conclusions about whether the error rates in Figure 9 are close to expected asymptotic rates. Finally, the transient natural of the problem leads to difficulties in decoupling interactions between spatial and temporal error sources. \% We claim that this problem, while important, is not adequate for a code verification study, where we are simply interested in verifying convergence rates under optimal conditions. ... \% Instead of trying to disentangle the sources of numerical error other than the contact algorithm in a 3D transient problem, we chose instead to study a simpler problem. next we examine a 1D contact problem with an exact analytical solution and observe convergence rates for each of the contact algorithms, verifying if the rates are optimal.}, file = {carnes_08_thermal.pdf}, institution = {Sandia National Laboratories}, interhash = {050c83baead6ee7e87b48348d75a66f7}, intrahash = {5329cb6d88298aa5144afc77ffee0af4}, keywords = {68w01-algorithms-general 80-04-heat-transfer-explicit-machine-computation-and-programs 80m10-heat-transfer-finite-element-methods}, number = {SAND2008-2607}, posted-at = {2012-11-29 16:33:51}, priority = {2}, timestamp = {2019-04-09T06:21:05.000+0200}, title = {Thermal Contact Algorithms in {SIERRA} Mechanics}, type = {Sandia Report}, url = {http://prod.sandia.gov/techlib/access-control.cgi/2008/082607.pdf}, year = 2008 }